| 1 4 4 1 2 2 2 1 2 2 2 1 2 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 | // SPDX-License-Identifier: GPL-2.0-only /* * * YeAH TCP * * For further details look at: * https://web.archive.org/web/20080316215752/http://wil.cs.caltech.edu/pfldnet2007/paper/YeAH_TCP.pdf * */ #include <linux/mm.h> #include <linux/module.h> #include <linux/skbuff.h> #include <linux/inet_diag.h> #include <net/tcp.h> #include "tcp_vegas.h" #define TCP_YEAH_ALPHA 80 /* number of packets queued at the bottleneck */ #define TCP_YEAH_GAMMA 1 /* fraction of queue to be removed per rtt */ #define TCP_YEAH_DELTA 3 /* log minimum fraction of cwnd to be removed on loss */ #define TCP_YEAH_EPSILON 1 /* log maximum fraction to be removed on early decongestion */ #define TCP_YEAH_PHY 8 /* maximum delta from base */ #define TCP_YEAH_RHO 16 /* minimum number of consecutive rtt to consider competition on loss */ #define TCP_YEAH_ZETA 50 /* minimum number of state switches to reset reno_count */ #define TCP_SCALABLE_AI_CNT 100U /* YeAH variables */ struct yeah { struct vegas vegas; /* must be first */ /* YeAH */ u32 lastQ; u32 doing_reno_now; u32 reno_count; u32 fast_count; }; static void tcp_yeah_init(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct yeah *yeah = inet_csk_ca(sk); tcp_vegas_init(sk); yeah->doing_reno_now = 0; yeah->lastQ = 0; yeah->reno_count = 2; /* Ensure the MD arithmetic works. This is somewhat pedantic, * since I don't think we will see a cwnd this large. :) */ tp->snd_cwnd_clamp = min_t(u32, tp->snd_cwnd_clamp, 0xffffffff/128); } static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct yeah *yeah = inet_csk_ca(sk); if (!tcp_is_cwnd_limited(sk)) return; if (tcp_in_slow_start(tp)) { acked = tcp_slow_start(tp, acked); if (!acked) goto do_vegas; } if (!yeah->doing_reno_now) { /* Scalable */ tcp_cong_avoid_ai(tp, min(tcp_snd_cwnd(tp), TCP_SCALABLE_AI_CNT), acked); } else { /* Reno */ tcp_cong_avoid_ai(tp, tcp_snd_cwnd(tp), acked); } /* The key players are v_vegas.beg_snd_una and v_beg_snd_nxt. * * These are so named because they represent the approximate values * of snd_una and snd_nxt at the beginning of the current RTT. More * precisely, they represent the amount of data sent during the RTT. * At the end of the RTT, when we receive an ACK for v_beg_snd_nxt, * we will calculate that (v_beg_snd_nxt - v_vegas.beg_snd_una) outstanding * bytes of data have been ACKed during the course of the RTT, giving * an "actual" rate of: * * (v_beg_snd_nxt - v_vegas.beg_snd_una) / (rtt duration) * * Unfortunately, v_vegas.beg_snd_una is not exactly equal to snd_una, * because delayed ACKs can cover more than one segment, so they * don't line up yeahly with the boundaries of RTTs. * * Another unfortunate fact of life is that delayed ACKs delay the * advance of the left edge of our send window, so that the number * of bytes we send in an RTT is often less than our cwnd will allow. * So we keep track of our cwnd separately, in v_beg_snd_cwnd. */ do_vegas: if (after(ack, yeah->vegas.beg_snd_nxt)) { /* We do the Vegas calculations only if we got enough RTT * samples that we can be reasonably sure that we got * at least one RTT sample that wasn't from a delayed ACK. * If we only had 2 samples total, * then that means we're getting only 1 ACK per RTT, which * means they're almost certainly delayed ACKs. * If we have 3 samples, we should be OK. */ if (yeah->vegas.cntRTT > 2) { u32 rtt, queue; u64 bw; /* We have enough RTT samples, so, using the Vegas * algorithm, we determine if we should increase or * decrease cwnd, and by how much. */ /* Pluck out the RTT we are using for the Vegas * calculations. This is the min RTT seen during the * last RTT. Taking the min filters out the effects * of delayed ACKs, at the cost of noticing congestion * a bit later. */ rtt = yeah->vegas.minRTT; /* Compute excess number of packets above bandwidth * Avoid doing full 64 bit divide. */ bw = tcp_snd_cwnd(tp); bw *= rtt - yeah->vegas.baseRTT; do_div(bw, rtt); queue = bw; if (queue > TCP_YEAH_ALPHA || rtt - yeah->vegas.baseRTT > (yeah->vegas.baseRTT / TCP_YEAH_PHY)) { if (queue > TCP_YEAH_ALPHA && tcp_snd_cwnd(tp) > yeah->reno_count) { u32 reduction = min(queue / TCP_YEAH_GAMMA , tcp_snd_cwnd(tp) >> TCP_YEAH_EPSILON); tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) - reduction); tcp_snd_cwnd_set(tp, max(tcp_snd_cwnd(tp), yeah->reno_count)); tp->snd_ssthresh = tcp_snd_cwnd(tp); } if (yeah->reno_count <= 2) yeah->reno_count = max(tcp_snd_cwnd(tp)>>1, 2U); else yeah->reno_count++; yeah->doing_reno_now = min(yeah->doing_reno_now + 1, 0xffffffU); } else { yeah->fast_count++; if (yeah->fast_count > TCP_YEAH_ZETA) { yeah->reno_count = 2; yeah->fast_count = 0; } yeah->doing_reno_now = 0; } yeah->lastQ = queue; } /* Save the extent of the current window so we can use this * at the end of the next RTT. */ yeah->vegas.beg_snd_una = yeah->vegas.beg_snd_nxt; yeah->vegas.beg_snd_nxt = tp->snd_nxt; yeah->vegas.beg_snd_cwnd = tcp_snd_cwnd(tp); /* Wipe the slate clean for the next RTT. */ yeah->vegas.cntRTT = 0; yeah->vegas.minRTT = 0x7fffffff; } } static u32 tcp_yeah_ssthresh(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); struct yeah *yeah = inet_csk_ca(sk); u32 reduction; if (yeah->doing_reno_now < TCP_YEAH_RHO) { reduction = yeah->lastQ; reduction = min(reduction, max(tcp_snd_cwnd(tp)>>1, 2U)); reduction = max(reduction, tcp_snd_cwnd(tp) >> TCP_YEAH_DELTA); } else reduction = max(tcp_snd_cwnd(tp)>>1, 2U); yeah->fast_count = 0; yeah->reno_count = max(yeah->reno_count>>1, 2U); return max_t(int, tcp_snd_cwnd(tp) - reduction, 2); } static struct tcp_congestion_ops tcp_yeah __read_mostly = { .init = tcp_yeah_init, .ssthresh = tcp_yeah_ssthresh, .undo_cwnd = tcp_reno_undo_cwnd, .cong_avoid = tcp_yeah_cong_avoid, .set_state = tcp_vegas_state, .cwnd_event = tcp_vegas_cwnd_event, .get_info = tcp_vegas_get_info, .pkts_acked = tcp_vegas_pkts_acked, .owner = THIS_MODULE, .name = "yeah", }; static int __init tcp_yeah_register(void) { BUILD_BUG_ON(sizeof(struct yeah) > ICSK_CA_PRIV_SIZE); tcp_register_congestion_control(&tcp_yeah); return 0; } static void __exit tcp_yeah_unregister(void) { tcp_unregister_congestion_control(&tcp_yeah); } module_init(tcp_yeah_register); module_exit(tcp_yeah_unregister); MODULE_AUTHOR("Angelo P. Castellani"); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("YeAH TCP"); |
| 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 | // SPDX-License-Identifier: GPL-2.0-or-later /* * Etoms Et61x151 GPL Linux driver by Michel Xhaard (09/09/2004) * * V4L2 by Jean-Francois Moine <http://moinejf.free.fr> */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define MODULE_NAME "etoms" #include "gspca.h" MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>"); MODULE_DESCRIPTION("Etoms USB Camera Driver"); MODULE_LICENSE("GPL"); /* specific webcam descriptor */ struct sd { struct gspca_dev gspca_dev; /* !! must be the first item */ unsigned char autogain; char sensor; #define SENSOR_PAS106 0 #define SENSOR_TAS5130CXX 1 signed char ag_cnt; #define AG_CNT_START 13 }; static const struct v4l2_pix_format vga_mode[] = { {320, 240, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE, .bytesperline = 320, .sizeimage = 320 * 240, .colorspace = V4L2_COLORSPACE_SRGB, .priv = 1}, /* {640, 480, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE, .bytesperline = 640, .sizeimage = 640 * 480, .colorspace = V4L2_COLORSPACE_SRGB, .priv = 0}, */ }; static const struct v4l2_pix_format sif_mode[] = { {176, 144, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE, .bytesperline = 176, .sizeimage = 176 * 144, .colorspace = V4L2_COLORSPACE_SRGB, .priv = 1}, {352, 288, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE, .bytesperline = 352, .sizeimage = 352 * 288, .colorspace = V4L2_COLORSPACE_SRGB, .priv = 0}, }; #define ETOMS_ALT_SIZE_1000 12 #define ET_GPIO_DIR_CTRL 0x04 /* Control IO bit[0..5] (0 in 1 out) */ #define ET_GPIO_OUT 0x05 /* Only IO data */ #define ET_GPIO_IN 0x06 /* Read Only IO data */ #define ET_RESET_ALL 0x03 #define ET_ClCK 0x01 #define ET_CTRL 0x02 /* enable i2c OutClck Powerdown mode */ #define ET_COMP 0x12 /* Compression register */ #define ET_MAXQt 0x13 #define ET_MINQt 0x14 #define ET_COMP_VAL0 0x02 #define ET_COMP_VAL1 0x03 #define ET_REG1d 0x1d #define ET_REG1e 0x1e #define ET_REG1f 0x1f #define ET_REG20 0x20 #define ET_REG21 0x21 #define ET_REG22 0x22 #define ET_REG23 0x23 #define ET_REG24 0x24 #define ET_REG25 0x25 /* base registers for luma calculation */ #define ET_LUMA_CENTER 0x39 #define ET_G_RED 0x4d #define ET_G_GREEN1 0x4e #define ET_G_BLUE 0x4f #define ET_G_GREEN2 0x50 #define ET_G_GR_H 0x51 #define ET_G_GB_H 0x52 #define ET_O_RED 0x34 #define ET_O_GREEN1 0x35 #define ET_O_BLUE 0x36 #define ET_O_GREEN2 0x37 #define ET_SYNCHRO 0x68 #define ET_STARTX 0x69 #define ET_STARTY 0x6a #define ET_WIDTH_LOW 0x6b #define ET_HEIGTH_LOW 0x6c #define ET_W_H_HEIGTH 0x6d #define ET_REG6e 0x6e /* OBW */ #define ET_REG6f 0x6f /* OBW */ #define ET_REG70 0x70 /* OBW_AWB */ #define ET_REG71 0x71 /* OBW_AWB */ #define ET_REG72 0x72 /* OBW_AWB */ #define ET_REG73 0x73 /* Clkdelay ns */ #define ET_REG74 0x74 /* test pattern */ #define ET_REG75 0x75 /* test pattern */ #define ET_I2C_CLK 0x8c #define ET_PXL_CLK 0x60 #define ET_I2C_BASE 0x89 #define ET_I2C_COUNT 0x8a #define ET_I2C_PREFETCH 0x8b #define ET_I2C_REG 0x88 #define ET_I2C_DATA7 0x87 #define ET_I2C_DATA6 0x86 #define ET_I2C_DATA5 0x85 #define ET_I2C_DATA4 0x84 #define ET_I2C_DATA3 0x83 #define ET_I2C_DATA2 0x82 #define ET_I2C_DATA1 0x81 #define ET_I2C_DATA0 0x80 #define PAS106_REG2 0x02 /* pxlClk = systemClk/(reg2) */ #define PAS106_REG3 0x03 /* line/frame H [11..4] */ #define PAS106_REG4 0x04 /* line/frame L [3..0] */ #define PAS106_REG5 0x05 /* exposure time line offset(default 5) */ #define PAS106_REG6 0x06 /* exposure time pixel offset(default 6) */ #define PAS106_REG7 0x07 /* signbit Dac (default 0) */ #define PAS106_REG9 0x09 #define PAS106_REG0e 0x0e /* global gain [4..0](default 0x0e) */ #define PAS106_REG13 0x13 /* end i2c write */ static const __u8 GainRGBG[] = { 0x80, 0x80, 0x80, 0x80, 0x00, 0x00 }; static const __u8 I2c2[] = { 0x08, 0x08, 0x08, 0x08, 0x0d }; static const __u8 I2c3[] = { 0x12, 0x05 }; static const __u8 I2c4[] = { 0x41, 0x08 }; /* read 'len' bytes to gspca_dev->usb_buf */ static void reg_r(struct gspca_dev *gspca_dev, __u16 index, __u16 len) { struct usb_device *dev = gspca_dev->dev; if (len > USB_BUF_SZ) { gspca_err(gspca_dev, "reg_r: buffer overflow\n"); return; } usb_control_msg(dev, usb_rcvctrlpipe(dev, 0), 0, USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_INTERFACE, 0, index, gspca_dev->usb_buf, len, 500); gspca_dbg(gspca_dev, D_USBI, "reg read [%02x] -> %02x ..\n", index, gspca_dev->usb_buf[0]); } static void reg_w_val(struct gspca_dev *gspca_dev, __u16 index, __u8 val) { struct usb_device *dev = gspca_dev->dev; gspca_dev->usb_buf[0] = val; usb_control_msg(dev, usb_sndctrlpipe(dev, 0), 0, USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_INTERFACE, 0, index, gspca_dev->usb_buf, 1, 500); } static void reg_w(struct gspca_dev *gspca_dev, __u16 index, const __u8 *buffer, __u16 len) { struct usb_device *dev = gspca_dev->dev; if (len > USB_BUF_SZ) { pr_err("reg_w: buffer overflow\n"); return; } gspca_dbg(gspca_dev, D_USBO, "reg write [%02x] = %02x..\n", index, *buffer); memcpy(gspca_dev->usb_buf, buffer, len); usb_control_msg(dev, usb_sndctrlpipe(dev, 0), 0, USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_INTERFACE, 0, index, gspca_dev->usb_buf, len, 500); } static int i2c_w(struct gspca_dev *gspca_dev, __u8 reg, const __u8 *buffer, int len, __u8 mode) { /* buffer should be [D0..D7] */ __u8 ptchcount; /* set the base address */ reg_w_val(gspca_dev, ET_I2C_BASE, 0x40); /* sensor base for the pas106 */ /* set count and prefetch */ ptchcount = ((len & 0x07) << 4) | (mode & 0x03); reg_w_val(gspca_dev, ET_I2C_COUNT, ptchcount); /* set the register base */ reg_w_val(gspca_dev, ET_I2C_REG, reg); while (--len >= 0) reg_w_val(gspca_dev, ET_I2C_DATA0 + len, buffer[len]); return 0; } static int i2c_r(struct gspca_dev *gspca_dev, __u8 reg) { /* set the base address */ reg_w_val(gspca_dev, ET_I2C_BASE, 0x40); /* sensor base for the pas106 */ /* set count and prefetch (cnd: 4 bits - mode: 4 bits) */ reg_w_val(gspca_dev, ET_I2C_COUNT, 0x11); reg_w_val(gspca_dev, ET_I2C_REG, reg); /* set the register base */ reg_w_val(gspca_dev, ET_I2C_PREFETCH, 0x02); /* prefetch */ reg_w_val(gspca_dev, ET_I2C_PREFETCH, 0x00); reg_r(gspca_dev, ET_I2C_DATA0, 1); /* read one byte */ return 0; } static int Et_WaitStatus(struct gspca_dev *gspca_dev) { int retry = 10; while (retry--) { reg_r(gspca_dev, ET_ClCK, 1); if (gspca_dev->usb_buf[0] != 0) return 1; } return 0; } static int et_video(struct gspca_dev *gspca_dev, int on) { int ret; reg_w_val(gspca_dev, ET_GPIO_OUT, on ? 0x10 /* startvideo - set Bit5 */ : 0); /* stopvideo */ ret = Et_WaitStatus(gspca_dev); if (ret != 0) gspca_err(gspca_dev, "timeout video on/off\n"); return ret; } static void Et_init2(struct gspca_dev *gspca_dev) { __u8 value; static const __u8 FormLine[] = { 0x84, 0x03, 0x14, 0xf4, 0x01, 0x05 }; gspca_dbg(gspca_dev, D_STREAM, "Open Init2 ET\n"); reg_w_val(gspca_dev, ET_GPIO_DIR_CTRL, 0x2f); reg_w_val(gspca_dev, ET_GPIO_OUT, 0x10); reg_r(gspca_dev, ET_GPIO_IN, 1); reg_w_val(gspca_dev, ET_ClCK, 0x14); /* 0x14 // 0x16 enabled pattern */ reg_w_val(gspca_dev, ET_CTRL, 0x1b); /* compression et subsampling */ if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv) value = ET_COMP_VAL1; /* 320 */ else value = ET_COMP_VAL0; /* 640 */ reg_w_val(gspca_dev, ET_COMP, value); reg_w_val(gspca_dev, ET_MAXQt, 0x1f); reg_w_val(gspca_dev, ET_MINQt, 0x04); /* undocumented registers */ reg_w_val(gspca_dev, ET_REG1d, 0xff); reg_w_val(gspca_dev, ET_REG1e, 0xff); reg_w_val(gspca_dev, ET_REG1f, 0xff); reg_w_val(gspca_dev, ET_REG20, 0x35); reg_w_val(gspca_dev, ET_REG21, 0x01); reg_w_val(gspca_dev, ET_REG22, 0x00); reg_w_val(gspca_dev, ET_REG23, 0xff); reg_w_val(gspca_dev, ET_REG24, 0xff); reg_w_val(gspca_dev, ET_REG25, 0x0f); /* colors setting */ reg_w_val(gspca_dev, 0x30, 0x11); /* 0x30 */ reg_w_val(gspca_dev, 0x31, 0x40); reg_w_val(gspca_dev, 0x32, 0x00); reg_w_val(gspca_dev, ET_O_RED, 0x00); /* 0x34 */ reg_w_val(gspca_dev, ET_O_GREEN1, 0x00); reg_w_val(gspca_dev, ET_O_BLUE, 0x00); reg_w_val(gspca_dev, ET_O_GREEN2, 0x00); /*************/ reg_w_val(gspca_dev, ET_G_RED, 0x80); /* 0x4d */ reg_w_val(gspca_dev, ET_G_GREEN1, 0x80); reg_w_val(gspca_dev, ET_G_BLUE, 0x80); reg_w_val(gspca_dev, ET_G_GREEN2, 0x80); reg_w_val(gspca_dev, ET_G_GR_H, 0x00); reg_w_val(gspca_dev, ET_G_GB_H, 0x00); /* 0x52 */ /* Window control registers */ reg_w_val(gspca_dev, 0x61, 0x80); /* use cmc_out */ reg_w_val(gspca_dev, 0x62, 0x02); reg_w_val(gspca_dev, 0x63, 0x03); reg_w_val(gspca_dev, 0x64, 0x14); reg_w_val(gspca_dev, 0x65, 0x0e); reg_w_val(gspca_dev, 0x66, 0x02); reg_w_val(gspca_dev, 0x67, 0x02); /**************************************/ reg_w_val(gspca_dev, ET_SYNCHRO, 0x8f); /* 0x68 */ reg_w_val(gspca_dev, ET_STARTX, 0x69); /* 0x6a //0x69 */ reg_w_val(gspca_dev, ET_STARTY, 0x0d); /* 0x0d //0x0c */ reg_w_val(gspca_dev, ET_WIDTH_LOW, 0x80); reg_w_val(gspca_dev, ET_HEIGTH_LOW, 0xe0); reg_w_val(gspca_dev, ET_W_H_HEIGTH, 0x60); /* 6d */ reg_w_val(gspca_dev, ET_REG6e, 0x86); reg_w_val(gspca_dev, ET_REG6f, 0x01); reg_w_val(gspca_dev, ET_REG70, 0x26); reg_w_val(gspca_dev, ET_REG71, 0x7a); reg_w_val(gspca_dev, ET_REG72, 0x01); /* Clock Pattern registers ***************** */ reg_w_val(gspca_dev, ET_REG73, 0x00); reg_w_val(gspca_dev, ET_REG74, 0x18); /* 0x28 */ reg_w_val(gspca_dev, ET_REG75, 0x0f); /* 0x01 */ /**********************************************/ reg_w_val(gspca_dev, 0x8a, 0x20); reg_w_val(gspca_dev, 0x8d, 0x0f); reg_w_val(gspca_dev, 0x8e, 0x08); /**************************************/ reg_w_val(gspca_dev, 0x03, 0x08); reg_w_val(gspca_dev, ET_PXL_CLK, 0x03); reg_w_val(gspca_dev, 0x81, 0xff); reg_w_val(gspca_dev, 0x80, 0x00); reg_w_val(gspca_dev, 0x81, 0xff); reg_w_val(gspca_dev, 0x80, 0x20); reg_w_val(gspca_dev, 0x03, 0x01); reg_w_val(gspca_dev, 0x03, 0x00); reg_w_val(gspca_dev, 0x03, 0x08); /********************************************/ /* reg_r(gspca_dev, ET_I2C_BASE, 1); always 0x40 as the pas106 ??? */ /* set the sensor */ if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv) value = 0x04; /* 320 */ else /* 640 */ value = 0x1e; /* 0x17 * setting PixelClock * 0x03 mean 24/(3+1) = 6 Mhz * 0x05 -> 24/(5+1) = 4 Mhz * 0x0b -> 24/(11+1) = 2 Mhz * 0x17 -> 24/(23+1) = 1 Mhz */ reg_w_val(gspca_dev, ET_PXL_CLK, value); /* now set by fifo the FormatLine setting */ reg_w(gspca_dev, 0x62, FormLine, 6); /* set exposure times [ 0..0x78] 0->longvalue 0x78->shortvalue */ reg_w_val(gspca_dev, 0x81, 0x47); /* 0x47; */ reg_w_val(gspca_dev, 0x80, 0x40); /* 0x40; */ /* Pedro change */ /* Brightness change Brith+ decrease value */ /* Brigth- increase value */ /* original value = 0x70; */ reg_w_val(gspca_dev, 0x81, 0x30); /* 0x20; - set brightness */ reg_w_val(gspca_dev, 0x80, 0x20); /* 0x20; */ } static void setbrightness(struct gspca_dev *gspca_dev, s32 val) { int i; for (i = 0; i < 4; i++) reg_w_val(gspca_dev, ET_O_RED + i, val); } static void setcontrast(struct gspca_dev *gspca_dev, s32 val) { __u8 RGBG[] = { 0x80, 0x80, 0x80, 0x80, 0x00, 0x00 }; memset(RGBG, val, sizeof(RGBG) - 2); reg_w(gspca_dev, ET_G_RED, RGBG, 6); } static void setcolors(struct gspca_dev *gspca_dev, s32 val) { struct sd *sd = (struct sd *) gspca_dev; __u8 I2cc[] = { 0x05, 0x02, 0x02, 0x05, 0x0d }; __u8 i2cflags = 0x01; /* __u8 green = 0; */ I2cc[3] = val; /* red */ I2cc[0] = 15 - val; /* blue */ /* green = 15 - ((((7*I2cc[0]) >> 2 ) + I2cc[3]) >> 1); */ /* I2cc[1] = I2cc[2] = green; */ if (sd->sensor == SENSOR_PAS106) { i2c_w(gspca_dev, PAS106_REG13, &i2cflags, 1, 3); i2c_w(gspca_dev, PAS106_REG9, I2cc, sizeof I2cc, 1); } } static s32 getcolors(struct gspca_dev *gspca_dev) { struct sd *sd = (struct sd *) gspca_dev; if (sd->sensor == SENSOR_PAS106) { /* i2c_r(gspca_dev, PAS106_REG9); * blue */ i2c_r(gspca_dev, PAS106_REG9 + 3); /* red */ return gspca_dev->usb_buf[0] & 0x0f; } return 0; } static void setautogain(struct gspca_dev *gspca_dev) { struct sd *sd = (struct sd *) gspca_dev; if (sd->autogain) sd->ag_cnt = AG_CNT_START; else sd->ag_cnt = -1; } static void Et_init1(struct gspca_dev *gspca_dev) { __u8 value; /* __u8 I2c0 [] = {0x0a, 0x12, 0x05, 0x22, 0xac, 0x00, 0x01, 0x00}; */ __u8 I2c0[] = { 0x0a, 0x12, 0x05, 0x6d, 0xcd, 0x00, 0x01, 0x00 }; /* try 1/120 0x6d 0xcd 0x40 */ /* __u8 I2c0 [] = {0x0a, 0x12, 0x05, 0xfe, 0xfe, 0xc0, 0x01, 0x00}; * 1/60000 hmm ?? */ gspca_dbg(gspca_dev, D_STREAM, "Open Init1 ET\n\n"); reg_w_val(gspca_dev, ET_GPIO_DIR_CTRL, 7); reg_r(gspca_dev, ET_GPIO_IN, 1); reg_w_val(gspca_dev, ET_RESET_ALL, 1); reg_w_val(gspca_dev, ET_RESET_ALL, 0); reg_w_val(gspca_dev, ET_ClCK, 0x10); reg_w_val(gspca_dev, ET_CTRL, 0x19); /* compression et subsampling */ if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv) value = ET_COMP_VAL1; else value = ET_COMP_VAL0; gspca_dbg(gspca_dev, D_STREAM, "Open mode %d Compression %d\n", gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv, value); reg_w_val(gspca_dev, ET_COMP, value); reg_w_val(gspca_dev, ET_MAXQt, 0x1d); reg_w_val(gspca_dev, ET_MINQt, 0x02); /* undocumented registers */ reg_w_val(gspca_dev, ET_REG1d, 0xff); reg_w_val(gspca_dev, ET_REG1e, 0xff); reg_w_val(gspca_dev, ET_REG1f, 0xff); reg_w_val(gspca_dev, ET_REG20, 0x35); reg_w_val(gspca_dev, ET_REG21, 0x01); reg_w_val(gspca_dev, ET_REG22, 0x00); reg_w_val(gspca_dev, ET_REG23, 0xf7); reg_w_val(gspca_dev, ET_REG24, 0xff); reg_w_val(gspca_dev, ET_REG25, 0x07); /* colors setting */ reg_w_val(gspca_dev, ET_G_RED, 0x80); reg_w_val(gspca_dev, ET_G_GREEN1, 0x80); reg_w_val(gspca_dev, ET_G_BLUE, 0x80); reg_w_val(gspca_dev, ET_G_GREEN2, 0x80); reg_w_val(gspca_dev, ET_G_GR_H, 0x00); reg_w_val(gspca_dev, ET_G_GB_H, 0x00); /* Window control registers */ reg_w_val(gspca_dev, ET_SYNCHRO, 0xf0); reg_w_val(gspca_dev, ET_STARTX, 0x56); /* 0x56 */ reg_w_val(gspca_dev, ET_STARTY, 0x05); /* 0x04 */ reg_w_val(gspca_dev, ET_WIDTH_LOW, 0x60); reg_w_val(gspca_dev, ET_HEIGTH_LOW, 0x20); reg_w_val(gspca_dev, ET_W_H_HEIGTH, 0x50); reg_w_val(gspca_dev, ET_REG6e, 0x86); reg_w_val(gspca_dev, ET_REG6f, 0x01); reg_w_val(gspca_dev, ET_REG70, 0x86); reg_w_val(gspca_dev, ET_REG71, 0x14); reg_w_val(gspca_dev, ET_REG72, 0x00); /* Clock Pattern registers */ reg_w_val(gspca_dev, ET_REG73, 0x00); reg_w_val(gspca_dev, ET_REG74, 0x00); reg_w_val(gspca_dev, ET_REG75, 0x0a); reg_w_val(gspca_dev, ET_I2C_CLK, 0x04); reg_w_val(gspca_dev, ET_PXL_CLK, 0x01); /* set the sensor */ if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv) { I2c0[0] = 0x06; i2c_w(gspca_dev, PAS106_REG2, I2c0, sizeof I2c0, 1); i2c_w(gspca_dev, PAS106_REG9, I2c2, sizeof I2c2, 1); value = 0x06; i2c_w(gspca_dev, PAS106_REG2, &value, 1, 1); i2c_w(gspca_dev, PAS106_REG3, I2c3, sizeof I2c3, 1); /* value = 0x1f; */ value = 0x04; i2c_w(gspca_dev, PAS106_REG0e, &value, 1, 1); } else { I2c0[0] = 0x0a; i2c_w(gspca_dev, PAS106_REG2, I2c0, sizeof I2c0, 1); i2c_w(gspca_dev, PAS106_REG9, I2c2, sizeof I2c2, 1); value = 0x0a; i2c_w(gspca_dev, PAS106_REG2, &value, 1, 1); i2c_w(gspca_dev, PAS106_REG3, I2c3, sizeof I2c3, 1); value = 0x04; /* value = 0x10; */ i2c_w(gspca_dev, PAS106_REG0e, &value, 1, 1); /* bit 2 enable bit 1:2 select 0 1 2 3 value = 0x07; * curve 0 * i2c_w(gspca_dev, PAS106_REG0f, &value, 1, 1); */ } /* value = 0x01; */ /* value = 0x22; */ /* i2c_w(gspca_dev, PAS106_REG5, &value, 1, 1); */ /* magnetude and sign bit for DAC */ i2c_w(gspca_dev, PAS106_REG7, I2c4, sizeof I2c4, 1); /* now set by fifo the whole colors setting */ reg_w(gspca_dev, ET_G_RED, GainRGBG, 6); setcolors(gspca_dev, getcolors(gspca_dev)); } /* this function is called at probe time */ static int sd_config(struct gspca_dev *gspca_dev, const struct usb_device_id *id) { struct sd *sd = (struct sd *) gspca_dev; struct cam *cam; cam = &gspca_dev->cam; sd->sensor = id->driver_info; if (sd->sensor == SENSOR_PAS106) { cam->cam_mode = sif_mode; cam->nmodes = ARRAY_SIZE(sif_mode); } else { cam->cam_mode = vga_mode; cam->nmodes = ARRAY_SIZE(vga_mode); } sd->ag_cnt = -1; return 0; } /* this function is called at probe and resume time */ static int sd_init(struct gspca_dev *gspca_dev) { struct sd *sd = (struct sd *) gspca_dev; if (sd->sensor == SENSOR_PAS106) Et_init1(gspca_dev); else Et_init2(gspca_dev); reg_w_val(gspca_dev, ET_RESET_ALL, 0x08); et_video(gspca_dev, 0); /* video off */ return 0; } /* -- start the camera -- */ static int sd_start(struct gspca_dev *gspca_dev) { struct sd *sd = (struct sd *) gspca_dev; if (sd->sensor == SENSOR_PAS106) Et_init1(gspca_dev); else Et_init2(gspca_dev); setautogain(gspca_dev); reg_w_val(gspca_dev, ET_RESET_ALL, 0x08); et_video(gspca_dev, 1); /* video on */ return 0; } static void sd_stopN(struct gspca_dev *gspca_dev) { et_video(gspca_dev, 0); /* video off */ } static __u8 Et_getgainG(struct gspca_dev *gspca_dev) { struct sd *sd = (struct sd *) gspca_dev; if (sd->sensor == SENSOR_PAS106) { i2c_r(gspca_dev, PAS106_REG0e); gspca_dbg(gspca_dev, D_CONF, "Etoms gain G %d\n", gspca_dev->usb_buf[0]); return gspca_dev->usb_buf[0]; } return 0x1f; } static void Et_setgainG(struct gspca_dev *gspca_dev, __u8 gain) { struct sd *sd = (struct sd *) gspca_dev; if (sd->sensor == SENSOR_PAS106) { __u8 i2cflags = 0x01; i2c_w(gspca_dev, PAS106_REG13, &i2cflags, 1, 3); i2c_w(gspca_dev, PAS106_REG0e, &gain, 1, 1); } } #define BLIMIT(bright) \ (u8)((bright > 0x1f) ? 0x1f : ((bright < 4) ? 3 : bright)) #define LIMIT(color) \ (u8)((color > 0xff) ? 0xff : ((color < 0) ? 0 : color)) static void do_autogain(struct gspca_dev *gspca_dev) { struct sd *sd = (struct sd *) gspca_dev; __u8 luma; __u8 luma_mean = 128; __u8 luma_delta = 20; __u8 spring = 4; int Gbright; __u8 r, g, b; if (sd->ag_cnt < 0) return; if (--sd->ag_cnt >= 0) return; sd->ag_cnt = AG_CNT_START; Gbright = Et_getgainG(gspca_dev); reg_r(gspca_dev, ET_LUMA_CENTER, 4); g = (gspca_dev->usb_buf[0] + gspca_dev->usb_buf[3]) >> 1; r = gspca_dev->usb_buf[1]; b = gspca_dev->usb_buf[2]; r = ((r << 8) - (r << 4) - (r << 3)) >> 10; b = ((b << 7) >> 10); g = ((g << 9) + (g << 7) + (g << 5)) >> 10; luma = LIMIT(r + g + b); gspca_dbg(gspca_dev, D_FRAM, "Etoms luma G %d\n", luma); if (luma < luma_mean - luma_delta || luma > luma_mean + luma_delta) { Gbright += (luma_mean - luma) >> spring; Gbright = BLIMIT(Gbright); gspca_dbg(gspca_dev, D_FRAM, "Etoms Gbright %d\n", Gbright); Et_setgainG(gspca_dev, (__u8) Gbright); } } #undef BLIMIT #undef LIMIT static void sd_pkt_scan(struct gspca_dev *gspca_dev, u8 *data, /* isoc packet */ int len) /* iso packet length */ { int seqframe; seqframe = data[0] & 0x3f; len = (int) (((data[0] & 0xc0) << 2) | data[1]); if (seqframe == 0x3f) { gspca_dbg(gspca_dev, D_FRAM, "header packet found datalength %d !!\n", len); gspca_dbg(gspca_dev, D_FRAM, "G %d R %d G %d B %d", data[2], data[3], data[4], data[5]); data += 30; /* don't change datalength as the chips provided it */ gspca_frame_add(gspca_dev, LAST_PACKET, NULL, 0); gspca_frame_add(gspca_dev, FIRST_PACKET, data, len); return; } if (len) { data += 8; gspca_frame_add(gspca_dev, INTER_PACKET, data, len); } else { /* Drop Packet */ gspca_dev->last_packet_type = DISCARD_PACKET; } } static int sd_s_ctrl(struct v4l2_ctrl *ctrl) { struct gspca_dev *gspca_dev = container_of(ctrl->handler, struct gspca_dev, ctrl_handler); struct sd *sd = (struct sd *)gspca_dev; gspca_dev->usb_err = 0; if (!gspca_dev->streaming) return 0; switch (ctrl->id) { case V4L2_CID_BRIGHTNESS: setbrightness(gspca_dev, ctrl->val); break; case V4L2_CID_CONTRAST: setcontrast(gspca_dev, ctrl->val); break; case V4L2_CID_SATURATION: setcolors(gspca_dev, ctrl->val); break; case V4L2_CID_AUTOGAIN: sd->autogain = ctrl->val; setautogain(gspca_dev); break; } return gspca_dev->usb_err; } static const struct v4l2_ctrl_ops sd_ctrl_ops = { .s_ctrl = sd_s_ctrl, }; static int sd_init_controls(struct gspca_dev *gspca_dev) { struct sd *sd = (struct sd *)gspca_dev; struct v4l2_ctrl_handler *hdl = &gspca_dev->ctrl_handler; gspca_dev->vdev.ctrl_handler = hdl; v4l2_ctrl_handler_init(hdl, 4); v4l2_ctrl_new_std(hdl, &sd_ctrl_ops, V4L2_CID_BRIGHTNESS, 1, 127, 1, 63); v4l2_ctrl_new_std(hdl, &sd_ctrl_ops, V4L2_CID_CONTRAST, 0, 255, 1, 127); if (sd->sensor == SENSOR_PAS106) v4l2_ctrl_new_std(hdl, &sd_ctrl_ops, V4L2_CID_SATURATION, 0, 15, 1, 7); v4l2_ctrl_new_std(hdl, &sd_ctrl_ops, V4L2_CID_AUTOGAIN, 0, 1, 1, 1); if (hdl->error) { pr_err("Could not initialize controls\n"); return hdl->error; } return 0; } /* sub-driver description */ static const struct sd_desc sd_desc = { .name = MODULE_NAME, .config = sd_config, .init = sd_init, .init_controls = sd_init_controls, .start = sd_start, .stopN = sd_stopN, .pkt_scan = sd_pkt_scan, .dq_callback = do_autogain, }; /* -- module initialisation -- */ static const struct usb_device_id device_table[] = { {USB_DEVICE(0x102c, 0x6151), .driver_info = SENSOR_PAS106}, {USB_DEVICE(0x102c, 0x6251), .driver_info = SENSOR_TAS5130CXX}, {} }; MODULE_DEVICE_TABLE(usb, device_table); /* -- device connect -- */ static int sd_probe(struct usb_interface *intf, const struct usb_device_id *id) { return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd), THIS_MODULE); } static struct usb_driver sd_driver = { .name = MODULE_NAME, .id_table = device_table, .probe = sd_probe, .disconnect = gspca_disconnect, #ifdef CONFIG_PM .suspend = gspca_suspend, .resume = gspca_resume, .reset_resume = gspca_resume, #endif }; module_usb_driver(sd_driver); |
| 459 2 2 2 2 2 4 4 4 4 40 26 4 1 1 15 13 7 1 4 1 1 1 1 13 2 11 11 11 5 6 6 5 5 5 1 4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 | // SPDX-License-Identifier: GPL-2.0 #include <linux/kernel.h> #include <linux/errno.h> #include <linux/fs.h> #include <linux/file.h> #include <linux/io_uring.h> #include <uapi/linux/io_uring.h> #include "../kernel/futex/futex.h" #include "io_uring.h" #include "alloc_cache.h" #include "futex.h" struct io_futex { struct file *file; void __user *uaddr; unsigned long futex_val; unsigned long futex_mask; u32 futex_flags; unsigned int futex_nr; bool futexv_unqueued; }; struct io_futex_data { struct futex_q q; struct io_kiocb *req; }; struct io_futexv_data { unsigned long owned; struct futex_vector futexv[]; }; #define IO_FUTEX_ALLOC_CACHE_MAX 32 bool io_futex_cache_init(struct io_ring_ctx *ctx) { return io_alloc_cache_init(&ctx->futex_cache, IO_FUTEX_ALLOC_CACHE_MAX, sizeof(struct io_futex_data), 0); } void io_futex_cache_free(struct io_ring_ctx *ctx) { io_alloc_cache_free(&ctx->futex_cache, kfree); } static void __io_futex_complete(struct io_tw_req tw_req, io_tw_token_t tw) { hlist_del_init(&tw_req.req->hash_node); io_req_task_complete(tw_req, tw); } static void io_futex_complete(struct io_tw_req tw_req, io_tw_token_t tw) { struct io_kiocb *req = tw_req.req; struct io_ring_ctx *ctx = req->ctx; io_tw_lock(ctx, tw); io_cache_free(&ctx->futex_cache, req->async_data); io_req_async_data_clear(req, 0); __io_futex_complete(tw_req, tw); } static void io_futexv_complete(struct io_tw_req tw_req, io_tw_token_t tw) { struct io_kiocb *req = tw_req.req; struct io_futex *iof = io_kiocb_to_cmd(req, struct io_futex); struct io_futexv_data *ifd = req->async_data; io_tw_lock(req->ctx, tw); if (!iof->futexv_unqueued) { int res; res = futex_unqueue_multiple(ifd->futexv, iof->futex_nr); if (res != -1) io_req_set_res(req, res, 0); } io_req_async_data_free(req); __io_futex_complete(tw_req, tw); } static bool io_futexv_claim(struct io_futexv_data *ifd) { if (test_bit(0, &ifd->owned) || test_and_set_bit_lock(0, &ifd->owned)) return false; return true; } static bool __io_futex_cancel(struct io_kiocb *req) { /* futex wake already done or in progress */ if (req->opcode == IORING_OP_FUTEX_WAIT) { struct io_futex_data *ifd = req->async_data; if (!futex_unqueue(&ifd->q)) return false; req->io_task_work.func = io_futex_complete; } else { struct io_futexv_data *ifd = req->async_data; if (!io_futexv_claim(ifd)) return false; req->io_task_work.func = io_futexv_complete; } hlist_del_init(&req->hash_node); io_req_set_res(req, -ECANCELED, 0); io_req_task_work_add(req); return true; } int io_futex_cancel(struct io_ring_ctx *ctx, struct io_cancel_data *cd, unsigned int issue_flags) { return io_cancel_remove(ctx, cd, issue_flags, &ctx->futex_list, __io_futex_cancel); } bool io_futex_remove_all(struct io_ring_ctx *ctx, struct io_uring_task *tctx, bool cancel_all) { return io_cancel_remove_all(ctx, tctx, &ctx->futex_list, cancel_all, __io_futex_cancel); } int io_futex_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_futex *iof = io_kiocb_to_cmd(req, struct io_futex); u32 flags; if (unlikely(sqe->len || sqe->futex_flags || sqe->buf_index || sqe->file_index)) return -EINVAL; iof->uaddr = u64_to_user_ptr(READ_ONCE(sqe->addr)); iof->futex_val = READ_ONCE(sqe->addr2); iof->futex_mask = READ_ONCE(sqe->addr3); flags = READ_ONCE(sqe->fd); if (flags & ~FUTEX2_VALID_MASK) return -EINVAL; iof->futex_flags = futex2_to_flags(flags); if (!futex_flags_valid(iof->futex_flags)) return -EINVAL; if (!futex_validate_input(iof->futex_flags, iof->futex_val) || !futex_validate_input(iof->futex_flags, iof->futex_mask)) return -EINVAL; /* Mark as inflight, so file exit cancelation will find it */ io_req_track_inflight(req); return 0; } static void io_futex_wakev_fn(struct wake_q_head *wake_q, struct futex_q *q) { struct io_kiocb *req = q->wake_data; struct io_futexv_data *ifd = req->async_data; if (!io_futexv_claim(ifd)) return; if (unlikely(!__futex_wake_mark(q))) return; io_req_set_res(req, 0, 0); req->io_task_work.func = io_futexv_complete; io_req_task_work_add(req); } int io_futexv_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_futex *iof = io_kiocb_to_cmd(req, struct io_futex); struct io_futexv_data *ifd; int ret; /* No flags or mask supported for waitv */ if (unlikely(sqe->fd || sqe->buf_index || sqe->file_index || sqe->addr2 || sqe->futex_flags || sqe->addr3)) return -EINVAL; iof->uaddr = u64_to_user_ptr(READ_ONCE(sqe->addr)); iof->futex_nr = READ_ONCE(sqe->len); if (!iof->futex_nr || iof->futex_nr > FUTEX_WAITV_MAX) return -EINVAL; ifd = kzalloc(struct_size_t(struct io_futexv_data, futexv, iof->futex_nr), GFP_KERNEL_ACCOUNT); if (!ifd) return -ENOMEM; ret = futex_parse_waitv(ifd->futexv, iof->uaddr, iof->futex_nr, io_futex_wakev_fn, req); if (ret) { kfree(ifd); return ret; } /* Mark as inflight, so file exit cancelation will find it */ io_req_track_inflight(req); iof->futexv_unqueued = 0; req->flags |= REQ_F_ASYNC_DATA; req->async_data = ifd; return 0; } static void io_futex_wake_fn(struct wake_q_head *wake_q, struct futex_q *q) { struct io_futex_data *ifd = container_of(q, struct io_futex_data, q); struct io_kiocb *req = ifd->req; if (unlikely(!__futex_wake_mark(q))) return; io_req_set_res(req, 0, 0); req->io_task_work.func = io_futex_complete; io_req_task_work_add(req); } int io_futexv_wait(struct io_kiocb *req, unsigned int issue_flags) { struct io_futex *iof = io_kiocb_to_cmd(req, struct io_futex); struct io_futexv_data *ifd = req->async_data; struct io_ring_ctx *ctx = req->ctx; int ret, woken = -1; io_ring_submit_lock(ctx, issue_flags); ret = futex_wait_multiple_setup(ifd->futexv, iof->futex_nr, &woken); /* * Error case, ret is < 0. Mark the request as failed. */ if (unlikely(ret < 0)) { io_ring_submit_unlock(ctx, issue_flags); req_set_fail(req); io_req_set_res(req, ret, 0); io_req_async_data_free(req); return IOU_COMPLETE; } /* * 0 return means that we successfully setup the waiters, and that * nobody triggered a wakeup while we were doing so. If the wakeup * happened post setup, the task_work will be run post this issue and * under the submission lock. 1 means We got woken while setting up, * let that side do the completion. Note that * futex_wait_multiple_setup() will have unqueued all the futexes in * this case. Mark us as having done that already, since this is * different from normal wakeup. */ if (!ret) { /* * If futex_wait_multiple_setup() returns 0 for a * successful setup, then the task state will not be * runnable. This is fine for the sync syscall, as * it'll be blocking unless we already got one of the * futexes woken, but it obviously won't work for an * async invocation. Mark us runnable again. */ __set_current_state(TASK_RUNNING); hlist_add_head(&req->hash_node, &ctx->futex_list); } else { iof->futexv_unqueued = 1; if (woken != -1) io_req_set_res(req, woken, 0); } io_ring_submit_unlock(ctx, issue_flags); return IOU_ISSUE_SKIP_COMPLETE; } int io_futex_wait(struct io_kiocb *req, unsigned int issue_flags) { struct io_futex *iof = io_kiocb_to_cmd(req, struct io_futex); struct io_ring_ctx *ctx = req->ctx; struct io_futex_data *ifd = NULL; int ret; if (!iof->futex_mask) { ret = -EINVAL; goto done; } io_ring_submit_lock(ctx, issue_flags); ifd = io_cache_alloc(&ctx->futex_cache, GFP_NOWAIT); if (!ifd) { ret = -ENOMEM; goto done_unlock; } req->flags |= REQ_F_ASYNC_DATA; req->async_data = ifd; ifd->q = futex_q_init; ifd->q.bitset = iof->futex_mask; ifd->q.wake = io_futex_wake_fn; ifd->req = req; ret = futex_wait_setup(iof->uaddr, iof->futex_val, iof->futex_flags, &ifd->q, NULL, NULL); if (!ret) { hlist_add_head(&req->hash_node, &ctx->futex_list); io_ring_submit_unlock(ctx, issue_flags); return IOU_ISSUE_SKIP_COMPLETE; } done_unlock: io_ring_submit_unlock(ctx, issue_flags); done: if (ret < 0) req_set_fail(req); io_req_set_res(req, ret, 0); io_req_async_data_free(req); return IOU_COMPLETE; } int io_futex_wake(struct io_kiocb *req, unsigned int issue_flags) { struct io_futex *iof = io_kiocb_to_cmd(req, struct io_futex); int ret; /* * Strict flags - ensure that waking 0 futexes yields a 0 result. * See commit 43adf8449510 ("futex: FLAGS_STRICT") for details. */ ret = futex_wake(iof->uaddr, FLAGS_STRICT | iof->futex_flags, iof->futex_val, iof->futex_mask); if (ret < 0) req_set_fail(req); io_req_set_res(req, ret, 0); return IOU_COMPLETE; } |
| 12 12 11 7 7 6 7 5 5 4 408 407 11 403 407 172 292 286 286 316 317 314 188 142 1 149 32 139 54 141 141 1 299 300 219 146 144 402 403 402 403 310 178 188 55 291 6 6 293 295 269 244 106 269 287 285 9 100 108 274 29 130 139 287 286 286 144 269 287 286 193 149 149 149 242 243 260 1 1 1 1 1 1 1 1 12 12 295 295 295 213 88 295 19 294 294 16 1 11 5 18 14 18 18 11 7 11 7 18 8 1 11 11 2 2 11 11 11 11 11 11 1 1 11 2 9 11 11 11 1 11 11 1 302 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 246 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387 4388 | // SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Red Hat. All rights reserved. */ #include <linux/pagemap.h> #include <linux/sched.h> #include <linux/sched/signal.h> #include <linux/slab.h> #include <linux/math64.h> #include <linux/ratelimit.h> #include <linux/error-injection.h> #include <linux/sched/mm.h> #include <linux/string_choices.h> #include "extent-tree.h" #include "fs.h" #include "messages.h" #include "misc.h" #include "free-space-cache.h" #include "transaction.h" #include "disk-io.h" #include "extent_io.h" #include "space-info.h" #include "block-group.h" #include "discard.h" #include "subpage.h" #include "inode-item.h" #include "accessors.h" #include "file-item.h" #include "file.h" #include "super.h" #include "relocation.h" #define BITS_PER_BITMAP (PAGE_SIZE * 8UL) #define MAX_CACHE_BYTES_PER_GIG SZ_64K #define FORCE_EXTENT_THRESHOLD SZ_1M static struct kmem_cache *btrfs_free_space_cachep; static struct kmem_cache *btrfs_free_space_bitmap_cachep; struct btrfs_trim_range { u64 start; u64 bytes; struct list_head list; }; static int link_free_space(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info); static void unlink_free_space(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, bool update_stat); static int search_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *bitmap_info, u64 *offset, u64 *bytes, bool for_alloc); static void free_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *bitmap_info); static void bitmap_clear_bits(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, u64 offset, u64 bytes, bool update_stats); static void btrfs_crc32c_final(u32 crc, u8 *result) { put_unaligned_le32(~crc, result); } static void __btrfs_remove_free_space_cache(struct btrfs_free_space_ctl *ctl) { struct btrfs_free_space *info; struct rb_node *node; while ((node = rb_last(&ctl->free_space_offset)) != NULL) { info = rb_entry(node, struct btrfs_free_space, offset_index); if (!info->bitmap) { unlink_free_space(ctl, info, true); kmem_cache_free(btrfs_free_space_cachep, info); } else { free_bitmap(ctl, info); } cond_resched_lock(&ctl->tree_lock); } } static struct inode *__lookup_free_space_inode(struct btrfs_root *root, struct btrfs_path *path, u64 offset) { struct btrfs_key key; struct btrfs_key location; struct btrfs_disk_key disk_key; struct btrfs_free_space_header *header; struct extent_buffer *leaf; struct btrfs_inode *inode; unsigned nofs_flag; int ret; key.objectid = BTRFS_FREE_SPACE_OBJECTID; key.type = 0; key.offset = offset; ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) return ERR_PTR(ret); if (ret > 0) { btrfs_release_path(path); return ERR_PTR(-ENOENT); } leaf = path->nodes[0]; header = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_free_space_header); btrfs_free_space_key(leaf, header, &disk_key); btrfs_disk_key_to_cpu(&location, &disk_key); btrfs_release_path(path); /* * We are often under a trans handle at this point, so we need to make * sure NOFS is set to keep us from deadlocking. */ nofs_flag = memalloc_nofs_save(); inode = btrfs_iget_path(location.objectid, root, path); btrfs_release_path(path); memalloc_nofs_restore(nofs_flag); if (IS_ERR(inode)) return ERR_CAST(inode); mapping_set_gfp_mask(inode->vfs_inode.i_mapping, mapping_gfp_constraint(inode->vfs_inode.i_mapping, ~(__GFP_FS | __GFP_HIGHMEM))); return &inode->vfs_inode; } struct inode *lookup_free_space_inode(struct btrfs_block_group *block_group, struct btrfs_path *path) { struct btrfs_fs_info *fs_info = block_group->fs_info; struct inode *inode = NULL; u32 flags = BTRFS_INODE_NODATASUM | BTRFS_INODE_NODATACOW; spin_lock(&block_group->lock); if (block_group->inode) inode = igrab(&block_group->inode->vfs_inode); spin_unlock(&block_group->lock); if (inode) return inode; inode = __lookup_free_space_inode(fs_info->tree_root, path, block_group->start); if (IS_ERR(inode)) return inode; spin_lock(&block_group->lock); if (!((BTRFS_I(inode)->flags & flags) == flags)) { btrfs_info(fs_info, "Old style space inode found, converting."); BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM | BTRFS_INODE_NODATACOW; block_group->disk_cache_state = BTRFS_DC_CLEAR; } if (!test_and_set_bit(BLOCK_GROUP_FLAG_IREF, &block_group->runtime_flags)) block_group->inode = BTRFS_I(igrab(inode)); spin_unlock(&block_group->lock); return inode; } static int __create_free_space_inode(struct btrfs_root *root, struct btrfs_trans_handle *trans, struct btrfs_path *path, u64 ino, u64 offset) { struct btrfs_key key; struct btrfs_disk_key disk_key; struct btrfs_free_space_header *header; struct btrfs_inode_item *inode_item; struct extent_buffer *leaf; /* We inline CRCs for the free disk space cache */ const u64 flags = BTRFS_INODE_NOCOMPRESS | BTRFS_INODE_PREALLOC | BTRFS_INODE_NODATASUM | BTRFS_INODE_NODATACOW; int ret; ret = btrfs_insert_empty_inode(trans, root, path, ino); if (ret) return ret; leaf = path->nodes[0]; inode_item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_inode_item); btrfs_item_key(leaf, &disk_key, path->slots[0]); memzero_extent_buffer(leaf, (unsigned long)inode_item, sizeof(*inode_item)); btrfs_set_inode_generation(leaf, inode_item, trans->transid); btrfs_set_inode_size(leaf, inode_item, 0); btrfs_set_inode_nbytes(leaf, inode_item, 0); btrfs_set_inode_uid(leaf, inode_item, 0); btrfs_set_inode_gid(leaf, inode_item, 0); btrfs_set_inode_mode(leaf, inode_item, S_IFREG | 0600); btrfs_set_inode_flags(leaf, inode_item, flags); btrfs_set_inode_nlink(leaf, inode_item, 1); btrfs_set_inode_transid(leaf, inode_item, trans->transid); btrfs_set_inode_block_group(leaf, inode_item, offset); btrfs_release_path(path); key.objectid = BTRFS_FREE_SPACE_OBJECTID; key.type = 0; key.offset = offset; ret = btrfs_insert_empty_item(trans, root, path, &key, sizeof(struct btrfs_free_space_header)); if (ret < 0) { btrfs_release_path(path); return ret; } leaf = path->nodes[0]; header = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_free_space_header); memzero_extent_buffer(leaf, (unsigned long)header, sizeof(*header)); btrfs_set_free_space_key(leaf, header, &disk_key); btrfs_release_path(path); return 0; } int create_free_space_inode(struct btrfs_trans_handle *trans, struct btrfs_block_group *block_group, struct btrfs_path *path) { int ret; u64 ino; ret = btrfs_get_free_objectid(trans->fs_info->tree_root, &ino); if (ret < 0) return ret; return __create_free_space_inode(trans->fs_info->tree_root, trans, path, ino, block_group->start); } /* * inode is an optional sink: if it is NULL, btrfs_remove_free_space_inode * handles lookup, otherwise it takes ownership and iputs the inode. * Don't reuse an inode pointer after passing it into this function. */ int btrfs_remove_free_space_inode(struct btrfs_trans_handle *trans, struct inode *inode, struct btrfs_block_group *block_group) { BTRFS_PATH_AUTO_FREE(path); struct btrfs_key key; int ret = 0; path = btrfs_alloc_path(); if (!path) return -ENOMEM; if (!inode) inode = lookup_free_space_inode(block_group, path); if (IS_ERR(inode)) { if (PTR_ERR(inode) != -ENOENT) ret = PTR_ERR(inode); return ret; } ret = btrfs_orphan_add(trans, BTRFS_I(inode)); if (ret) { btrfs_add_delayed_iput(BTRFS_I(inode)); return ret; } clear_nlink(inode); /* One for the block groups ref */ spin_lock(&block_group->lock); if (test_and_clear_bit(BLOCK_GROUP_FLAG_IREF, &block_group->runtime_flags)) { block_group->inode = NULL; spin_unlock(&block_group->lock); iput(inode); } else { spin_unlock(&block_group->lock); } /* One for the lookup ref */ btrfs_add_delayed_iput(BTRFS_I(inode)); key.objectid = BTRFS_FREE_SPACE_OBJECTID; key.type = 0; key.offset = block_group->start; ret = btrfs_search_slot(trans, trans->fs_info->tree_root, &key, path, -1, 1); if (ret) { if (ret > 0) ret = 0; return ret; } return btrfs_del_item(trans, trans->fs_info->tree_root, path); } int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans, struct btrfs_block_group *block_group, struct inode *vfs_inode) { struct btrfs_truncate_control control = { .inode = BTRFS_I(vfs_inode), .new_size = 0, .ino = btrfs_ino(BTRFS_I(vfs_inode)), .min_type = BTRFS_EXTENT_DATA_KEY, .clear_extent_range = true, }; struct btrfs_inode *inode = BTRFS_I(vfs_inode); struct btrfs_root *root = inode->root; struct extent_state *cached_state = NULL; int ret = 0; bool locked = false; if (block_group) { BTRFS_PATH_AUTO_FREE(path); path = btrfs_alloc_path(); if (!path) { ret = -ENOMEM; goto fail; } locked = true; mutex_lock(&trans->transaction->cache_write_mutex); if (!list_empty(&block_group->io_list)) { list_del_init(&block_group->io_list); btrfs_wait_cache_io(trans, block_group, path); btrfs_put_block_group(block_group); } /* * now that we've truncated the cache away, its no longer * setup or written */ spin_lock(&block_group->lock); block_group->disk_cache_state = BTRFS_DC_CLEAR; spin_unlock(&block_group->lock); } btrfs_i_size_write(inode, 0); truncate_pagecache(vfs_inode, 0); btrfs_lock_extent(&inode->io_tree, 0, (u64)-1, &cached_state); btrfs_drop_extent_map_range(inode, 0, (u64)-1, false); /* * We skip the throttling logic for free space cache inodes, so we don't * need to check for -EAGAIN. */ ret = btrfs_truncate_inode_items(trans, root, &control); inode_sub_bytes(&inode->vfs_inode, control.sub_bytes); btrfs_inode_safe_disk_i_size_write(inode, control.last_size); btrfs_unlock_extent(&inode->io_tree, 0, (u64)-1, &cached_state); if (ret) goto fail; ret = btrfs_update_inode(trans, inode); fail: if (locked) mutex_unlock(&trans->transaction->cache_write_mutex); if (ret) btrfs_abort_transaction(trans, ret); return ret; } static void readahead_cache(struct inode *inode) { struct file_ra_state ra; pgoff_t last_index; file_ra_state_init(&ra, inode->i_mapping); last_index = (i_size_read(inode) - 1) >> PAGE_SHIFT; page_cache_sync_readahead(inode->i_mapping, &ra, NULL, 0, last_index); } static int io_ctl_init(struct btrfs_io_ctl *io_ctl, struct inode *inode, int write) { int num_pages; num_pages = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); /* Make sure we can fit our crcs and generation into the first page */ if (write && (num_pages * sizeof(u32) + sizeof(u64)) > PAGE_SIZE) return -ENOSPC; memset(io_ctl, 0, sizeof(struct btrfs_io_ctl)); io_ctl->pages = kcalloc(num_pages, sizeof(struct page *), GFP_NOFS); if (!io_ctl->pages) return -ENOMEM; io_ctl->num_pages = num_pages; io_ctl->fs_info = inode_to_fs_info(inode); io_ctl->inode = inode; return 0; } ALLOW_ERROR_INJECTION(io_ctl_init, ERRNO); static void io_ctl_free(struct btrfs_io_ctl *io_ctl) { kfree(io_ctl->pages); io_ctl->pages = NULL; } static void io_ctl_unmap_page(struct btrfs_io_ctl *io_ctl) { if (io_ctl->cur) { io_ctl->cur = NULL; io_ctl->orig = NULL; } } static void io_ctl_map_page(struct btrfs_io_ctl *io_ctl, int clear) { ASSERT(io_ctl->index < io_ctl->num_pages); io_ctl->page = io_ctl->pages[io_ctl->index++]; io_ctl->cur = page_address(io_ctl->page); io_ctl->orig = io_ctl->cur; io_ctl->size = PAGE_SIZE; if (clear) clear_page(io_ctl->cur); } static void io_ctl_drop_pages(struct btrfs_io_ctl *io_ctl) { int i; io_ctl_unmap_page(io_ctl); for (i = 0; i < io_ctl->num_pages; i++) { if (io_ctl->pages[i]) { btrfs_folio_clear_checked(io_ctl->fs_info, page_folio(io_ctl->pages[i]), page_offset(io_ctl->pages[i]), PAGE_SIZE); unlock_page(io_ctl->pages[i]); put_page(io_ctl->pages[i]); } } } static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, bool uptodate) { struct folio *folio; struct inode *inode = io_ctl->inode; gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping); int i; for (i = 0; i < io_ctl->num_pages; i++) { int ret; folio = __filemap_get_folio(inode->i_mapping, i, FGP_LOCK | FGP_ACCESSED | FGP_CREAT, mask); if (IS_ERR(folio)) { io_ctl_drop_pages(io_ctl); return PTR_ERR(folio); } ret = set_folio_extent_mapped(folio); if (ret < 0) { folio_unlock(folio); folio_put(folio); io_ctl_drop_pages(io_ctl); return ret; } io_ctl->pages[i] = &folio->page; if (uptodate && !folio_test_uptodate(folio)) { btrfs_read_folio(NULL, folio); folio_lock(folio); if (folio->mapping != inode->i_mapping) { btrfs_err(BTRFS_I(inode)->root->fs_info, "free space cache page truncated"); io_ctl_drop_pages(io_ctl); return -EIO; } if (!folio_test_uptodate(folio)) { btrfs_err(BTRFS_I(inode)->root->fs_info, "error reading free space cache"); io_ctl_drop_pages(io_ctl); return -EIO; } } } for (i = 0; i < io_ctl->num_pages; i++) clear_page_dirty_for_io(io_ctl->pages[i]); return 0; } static void io_ctl_set_generation(struct btrfs_io_ctl *io_ctl, u64 generation) { io_ctl_map_page(io_ctl, 1); /* * Skip the csum areas. If we don't check crcs then we just have a * 64bit chunk at the front of the first page. */ io_ctl->cur += (sizeof(u32) * io_ctl->num_pages); io_ctl->size -= sizeof(u64) + (sizeof(u32) * io_ctl->num_pages); put_unaligned_le64(generation, io_ctl->cur); io_ctl->cur += sizeof(u64); } static int io_ctl_check_generation(struct btrfs_io_ctl *io_ctl, u64 generation) { u64 cache_gen; /* * Skip the crc area. If we don't check crcs then we just have a 64bit * chunk at the front of the first page. */ io_ctl->cur += sizeof(u32) * io_ctl->num_pages; io_ctl->size -= sizeof(u64) + (sizeof(u32) * io_ctl->num_pages); cache_gen = get_unaligned_le64(io_ctl->cur); if (cache_gen != generation) { btrfs_err_rl(io_ctl->fs_info, "space cache generation (%llu) does not match inode (%llu)", cache_gen, generation); io_ctl_unmap_page(io_ctl); return -EIO; } io_ctl->cur += sizeof(u64); return 0; } static void io_ctl_set_crc(struct btrfs_io_ctl *io_ctl, int index) { u32 *tmp; u32 crc = ~(u32)0; unsigned offset = 0; if (index == 0) offset = sizeof(u32) * io_ctl->num_pages; crc = crc32c(crc, io_ctl->orig + offset, PAGE_SIZE - offset); btrfs_crc32c_final(crc, (u8 *)&crc); io_ctl_unmap_page(io_ctl); tmp = page_address(io_ctl->pages[0]); tmp += index; *tmp = crc; } static int io_ctl_check_crc(struct btrfs_io_ctl *io_ctl, int index) { u32 *tmp, val; u32 crc = ~(u32)0; unsigned offset = 0; if (index == 0) offset = sizeof(u32) * io_ctl->num_pages; tmp = page_address(io_ctl->pages[0]); tmp += index; val = *tmp; io_ctl_map_page(io_ctl, 0); crc = crc32c(crc, io_ctl->orig + offset, PAGE_SIZE - offset); btrfs_crc32c_final(crc, (u8 *)&crc); if (val != crc) { btrfs_err_rl(io_ctl->fs_info, "csum mismatch on free space cache"); io_ctl_unmap_page(io_ctl); return -EIO; } return 0; } static int io_ctl_add_entry(struct btrfs_io_ctl *io_ctl, u64 offset, u64 bytes, void *bitmap) { struct btrfs_free_space_entry *entry; if (!io_ctl->cur) return -ENOSPC; entry = io_ctl->cur; put_unaligned_le64(offset, &entry->offset); put_unaligned_le64(bytes, &entry->bytes); entry->type = (bitmap) ? BTRFS_FREE_SPACE_BITMAP : BTRFS_FREE_SPACE_EXTENT; io_ctl->cur += sizeof(struct btrfs_free_space_entry); io_ctl->size -= sizeof(struct btrfs_free_space_entry); if (io_ctl->size >= sizeof(struct btrfs_free_space_entry)) return 0; io_ctl_set_crc(io_ctl, io_ctl->index - 1); /* No more pages to map */ if (io_ctl->index >= io_ctl->num_pages) return 0; /* map the next page */ io_ctl_map_page(io_ctl, 1); return 0; } static int io_ctl_add_bitmap(struct btrfs_io_ctl *io_ctl, void *bitmap) { if (!io_ctl->cur) return -ENOSPC; /* * If we aren't at the start of the current page, unmap this one and * map the next one if there is any left. */ if (io_ctl->cur != io_ctl->orig) { io_ctl_set_crc(io_ctl, io_ctl->index - 1); if (io_ctl->index >= io_ctl->num_pages) return -ENOSPC; io_ctl_map_page(io_ctl, 0); } copy_page(io_ctl->cur, bitmap); io_ctl_set_crc(io_ctl, io_ctl->index - 1); if (io_ctl->index < io_ctl->num_pages) io_ctl_map_page(io_ctl, 0); return 0; } static void io_ctl_zero_remaining_pages(struct btrfs_io_ctl *io_ctl) { /* * If we're not on the boundary we know we've modified the page and we * need to crc the page. */ if (io_ctl->cur != io_ctl->orig) io_ctl_set_crc(io_ctl, io_ctl->index - 1); else io_ctl_unmap_page(io_ctl); while (io_ctl->index < io_ctl->num_pages) { io_ctl_map_page(io_ctl, 1); io_ctl_set_crc(io_ctl, io_ctl->index - 1); } } static int io_ctl_read_entry(struct btrfs_io_ctl *io_ctl, struct btrfs_free_space *entry, u8 *type) { struct btrfs_free_space_entry *e; int ret; if (!io_ctl->cur) { ret = io_ctl_check_crc(io_ctl, io_ctl->index); if (ret) return ret; } e = io_ctl->cur; entry->offset = get_unaligned_le64(&e->offset); entry->bytes = get_unaligned_le64(&e->bytes); *type = e->type; io_ctl->cur += sizeof(struct btrfs_free_space_entry); io_ctl->size -= sizeof(struct btrfs_free_space_entry); if (io_ctl->size >= sizeof(struct btrfs_free_space_entry)) return 0; io_ctl_unmap_page(io_ctl); return 0; } static int io_ctl_read_bitmap(struct btrfs_io_ctl *io_ctl, struct btrfs_free_space *entry) { int ret; ret = io_ctl_check_crc(io_ctl, io_ctl->index); if (ret) return ret; copy_page(entry->bitmap, io_ctl->cur); io_ctl_unmap_page(io_ctl); return 0; } static void recalculate_thresholds(struct btrfs_free_space_ctl *ctl) { struct btrfs_block_group *block_group = ctl->block_group; u64 max_bytes; u64 bitmap_bytes; u64 extent_bytes; u64 size = block_group->length; u64 bytes_per_bg = BITS_PER_BITMAP * ctl->unit; u64 max_bitmaps = div64_u64(size + bytes_per_bg - 1, bytes_per_bg); max_bitmaps = max_t(u64, max_bitmaps, 1); if (ctl->total_bitmaps > max_bitmaps) btrfs_err(block_group->fs_info, "invalid free space control: bg start=%llu len=%llu total_bitmaps=%u unit=%u max_bitmaps=%llu bytes_per_bg=%llu", block_group->start, block_group->length, ctl->total_bitmaps, ctl->unit, max_bitmaps, bytes_per_bg); ASSERT(ctl->total_bitmaps <= max_bitmaps); /* * We are trying to keep the total amount of memory used per 1GiB of * space to be MAX_CACHE_BYTES_PER_GIG. However, with a reclamation * mechanism of pulling extents >= FORCE_EXTENT_THRESHOLD out of * bitmaps, we may end up using more memory than this. */ if (size < SZ_1G) max_bytes = MAX_CACHE_BYTES_PER_GIG; else max_bytes = MAX_CACHE_BYTES_PER_GIG * div_u64(size, SZ_1G); bitmap_bytes = ctl->total_bitmaps * ctl->unit; /* * we want the extent entry threshold to always be at most 1/2 the max * bytes we can have, or whatever is less than that. */ extent_bytes = max_bytes - bitmap_bytes; extent_bytes = min_t(u64, extent_bytes, max_bytes >> 1); ctl->extents_thresh = div_u64(extent_bytes, sizeof(struct btrfs_free_space)); } static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, struct btrfs_free_space_ctl *ctl, struct btrfs_path *path, u64 offset) { struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_free_space_header *header; struct extent_buffer *leaf; struct btrfs_io_ctl io_ctl; struct btrfs_key key; struct btrfs_free_space *e, *n; LIST_HEAD(bitmaps); u64 num_entries; u64 num_bitmaps; u64 generation; u8 type; int ret = 0; /* Nothing in the space cache, goodbye */ if (!i_size_read(inode)) return 0; key.objectid = BTRFS_FREE_SPACE_OBJECTID; key.type = 0; key.offset = offset; ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) return 0; else if (ret > 0) { btrfs_release_path(path); return 0; } ret = -1; leaf = path->nodes[0]; header = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_free_space_header); num_entries = btrfs_free_space_entries(leaf, header); num_bitmaps = btrfs_free_space_bitmaps(leaf, header); generation = btrfs_free_space_generation(leaf, header); btrfs_release_path(path); if (!BTRFS_I(inode)->generation) { btrfs_info(fs_info, "the free space cache file (%llu) is invalid, skip it", offset); return 0; } if (BTRFS_I(inode)->generation != generation) { btrfs_err(fs_info, "free space inode generation (%llu) did not match free space cache generation (%llu)", BTRFS_I(inode)->generation, generation); return 0; } if (!num_entries) return 0; ret = io_ctl_init(&io_ctl, inode, 0); if (ret) return ret; readahead_cache(inode); ret = io_ctl_prepare_pages(&io_ctl, true); if (ret) goto out; ret = io_ctl_check_crc(&io_ctl, 0); if (ret) goto free_cache; ret = io_ctl_check_generation(&io_ctl, generation); if (ret) goto free_cache; while (num_entries) { e = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS); if (!e) { ret = -ENOMEM; goto free_cache; } ret = io_ctl_read_entry(&io_ctl, e, &type); if (ret) { kmem_cache_free(btrfs_free_space_cachep, e); goto free_cache; } if (!e->bytes) { ret = -1; kmem_cache_free(btrfs_free_space_cachep, e); goto free_cache; } if (type == BTRFS_FREE_SPACE_EXTENT) { spin_lock(&ctl->tree_lock); ret = link_free_space(ctl, e); spin_unlock(&ctl->tree_lock); if (ret) { btrfs_err(fs_info, "Duplicate entries in free space cache, dumping"); kmem_cache_free(btrfs_free_space_cachep, e); goto free_cache; } } else { ASSERT(num_bitmaps); num_bitmaps--; e->bitmap = kmem_cache_zalloc( btrfs_free_space_bitmap_cachep, GFP_NOFS); if (!e->bitmap) { ret = -ENOMEM; kmem_cache_free( btrfs_free_space_cachep, e); goto free_cache; } spin_lock(&ctl->tree_lock); ret = link_free_space(ctl, e); if (ret) { spin_unlock(&ctl->tree_lock); btrfs_err(fs_info, "Duplicate entries in free space cache, dumping"); kmem_cache_free(btrfs_free_space_bitmap_cachep, e->bitmap); kmem_cache_free(btrfs_free_space_cachep, e); goto free_cache; } ctl->total_bitmaps++; recalculate_thresholds(ctl); spin_unlock(&ctl->tree_lock); list_add_tail(&e->list, &bitmaps); } num_entries--; } io_ctl_unmap_page(&io_ctl); /* * We add the bitmaps at the end of the entries in order that * the bitmap entries are added to the cache. */ list_for_each_entry_safe(e, n, &bitmaps, list) { list_del_init(&e->list); ret = io_ctl_read_bitmap(&io_ctl, e); if (ret) goto free_cache; } io_ctl_drop_pages(&io_ctl); ret = 1; out: io_ctl_free(&io_ctl); return ret; free_cache: io_ctl_drop_pages(&io_ctl); spin_lock(&ctl->tree_lock); __btrfs_remove_free_space_cache(ctl); spin_unlock(&ctl->tree_lock); goto out; } static int copy_free_space_cache(struct btrfs_block_group *block_group, struct btrfs_free_space_ctl *ctl) { struct btrfs_free_space *info; struct rb_node *n; int ret = 0; while (!ret && (n = rb_first(&ctl->free_space_offset)) != NULL) { info = rb_entry(n, struct btrfs_free_space, offset_index); if (!info->bitmap) { const u64 offset = info->offset; const u64 bytes = info->bytes; unlink_free_space(ctl, info, true); spin_unlock(&ctl->tree_lock); kmem_cache_free(btrfs_free_space_cachep, info); ret = btrfs_add_free_space(block_group, offset, bytes); spin_lock(&ctl->tree_lock); } else { u64 offset = info->offset; u64 bytes = ctl->unit; ret = search_bitmap(ctl, info, &offset, &bytes, false); if (ret == 0) { bitmap_clear_bits(ctl, info, offset, bytes, true); spin_unlock(&ctl->tree_lock); ret = btrfs_add_free_space(block_group, offset, bytes); spin_lock(&ctl->tree_lock); } else { free_bitmap(ctl, info); ret = 0; } } cond_resched_lock(&ctl->tree_lock); } return ret; } static struct lock_class_key btrfs_free_space_inode_key; int load_free_space_cache(struct btrfs_block_group *block_group) { struct btrfs_fs_info *fs_info = block_group->fs_info; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space_ctl tmp_ctl = {}; struct inode *inode; struct btrfs_path *path; int ret = 0; bool matched; u64 used = block_group->used; /* * Because we could potentially discard our loaded free space, we want * to load everything into a temporary structure first, and then if it's * valid copy it all into the actual free space ctl. */ btrfs_init_free_space_ctl(block_group, &tmp_ctl); /* * If this block group has been marked to be cleared for one reason or * another then we can't trust the on disk cache, so just return. */ spin_lock(&block_group->lock); if (block_group->disk_cache_state != BTRFS_DC_WRITTEN) { spin_unlock(&block_group->lock); return 0; } spin_unlock(&block_group->lock); path = btrfs_alloc_path(); if (!path) return 0; path->search_commit_root = true; path->skip_locking = true; /* * We must pass a path with search_commit_root set to btrfs_iget in * order to avoid a deadlock when allocating extents for the tree root. * * When we are COWing an extent buffer from the tree root, when looking * for a free extent, at extent-tree.c:find_free_extent(), we can find * block group without its free space cache loaded. When we find one * we must load its space cache which requires reading its free space * cache's inode item from the root tree. If this inode item is located * in the same leaf that we started COWing before, then we end up in * deadlock on the extent buffer (trying to read lock it when we * previously write locked it). * * It's safe to read the inode item using the commit root because * block groups, once loaded, stay in memory forever (until they are * removed) as well as their space caches once loaded. New block groups * once created get their ->cached field set to BTRFS_CACHE_FINISHED so * we will never try to read their inode item while the fs is mounted. */ inode = lookup_free_space_inode(block_group, path); if (IS_ERR(inode)) { btrfs_free_path(path); return 0; } /* We may have converted the inode and made the cache invalid. */ spin_lock(&block_group->lock); if (block_group->disk_cache_state != BTRFS_DC_WRITTEN) { spin_unlock(&block_group->lock); btrfs_free_path(path); goto out; } spin_unlock(&block_group->lock); /* * Reinitialize the class of struct inode's mapping->invalidate_lock for * free space inodes to prevent false positives related to locks for normal * inodes. */ lockdep_set_class(&(&inode->i_data)->invalidate_lock, &btrfs_free_space_inode_key); ret = __load_free_space_cache(fs_info->tree_root, inode, &tmp_ctl, path, block_group->start); btrfs_free_path(path); if (ret <= 0) goto out; matched = (tmp_ctl.free_space == (block_group->length - used - block_group->bytes_super)); if (matched) { spin_lock(&tmp_ctl.tree_lock); ret = copy_free_space_cache(block_group, &tmp_ctl); spin_unlock(&tmp_ctl.tree_lock); /* * ret == 1 means we successfully loaded the free space cache, * so we need to re-set it here. */ if (ret == 0) ret = 1; } else { /* * We need to call the _locked variant so we don't try to update * the discard counters. */ spin_lock(&tmp_ctl.tree_lock); __btrfs_remove_free_space_cache(&tmp_ctl); spin_unlock(&tmp_ctl.tree_lock); btrfs_warn(fs_info, "block group %llu has wrong amount of free space", block_group->start); ret = -1; } out: if (ret < 0) { /* This cache is bogus, make sure it gets cleared */ spin_lock(&block_group->lock); block_group->disk_cache_state = BTRFS_DC_CLEAR; spin_unlock(&block_group->lock); ret = 0; btrfs_warn(fs_info, "failed to load free space cache for block group %llu, rebuilding it now", block_group->start); } spin_lock(&ctl->tree_lock); btrfs_discard_update_discardable(block_group); spin_unlock(&ctl->tree_lock); iput(inode); return ret; } static noinline_for_stack int write_cache_extent_entries(struct btrfs_io_ctl *io_ctl, struct btrfs_free_space_ctl *ctl, struct btrfs_block_group *block_group, int *entries, int *bitmaps, struct list_head *bitmap_list) { int ret; struct btrfs_free_cluster *cluster = NULL; struct btrfs_free_cluster *cluster_locked = NULL; struct rb_node *node = rb_first(&ctl->free_space_offset); struct btrfs_trim_range *trim_entry; /* Get the cluster for this block_group if it exists */ if (!list_empty(&block_group->cluster_list)) { cluster = list_first_entry(&block_group->cluster_list, struct btrfs_free_cluster, block_group_list); } if (!node && cluster) { cluster_locked = cluster; spin_lock(&cluster_locked->lock); node = rb_first(&cluster->root); cluster = NULL; } /* Write out the extent entries */ while (node) { struct btrfs_free_space *e; e = rb_entry(node, struct btrfs_free_space, offset_index); *entries += 1; ret = io_ctl_add_entry(io_ctl, e->offset, e->bytes, e->bitmap); if (ret) goto fail; if (e->bitmap) { list_add_tail(&e->list, bitmap_list); *bitmaps += 1; } node = rb_next(node); if (!node && cluster) { node = rb_first(&cluster->root); cluster_locked = cluster; spin_lock(&cluster_locked->lock); cluster = NULL; } } if (cluster_locked) { spin_unlock(&cluster_locked->lock); cluster_locked = NULL; } /* * Make sure we don't miss any range that was removed from our rbtree * because trimming is running. Otherwise after a umount+mount (or crash * after committing the transaction) we would leak free space and get * an inconsistent free space cache report from fsck. */ list_for_each_entry(trim_entry, &ctl->trimming_ranges, list) { ret = io_ctl_add_entry(io_ctl, trim_entry->start, trim_entry->bytes, NULL); if (ret) goto fail; *entries += 1; } return 0; fail: if (cluster_locked) spin_unlock(&cluster_locked->lock); return -ENOSPC; } static noinline_for_stack int update_cache_item(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode, struct btrfs_path *path, u64 offset, int entries, int bitmaps) { struct btrfs_key key; struct btrfs_free_space_header *header; struct extent_buffer *leaf; int ret; key.objectid = BTRFS_FREE_SPACE_OBJECTID; key.type = 0; key.offset = offset; ret = btrfs_search_slot(trans, root, &key, path, 0, 1); if (ret < 0) { btrfs_clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, inode->i_size - 1, EXTENT_DELALLOC, NULL); return ret; } leaf = path->nodes[0]; if (ret > 0) { struct btrfs_key found_key; ASSERT(path->slots[0]); path->slots[0]--; btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); if (found_key.objectid != BTRFS_FREE_SPACE_OBJECTID || found_key.offset != offset) { btrfs_clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, inode->i_size - 1, EXTENT_DELALLOC, NULL); btrfs_release_path(path); return -ENOENT; } } BTRFS_I(inode)->generation = trans->transid; header = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_free_space_header); btrfs_set_free_space_entries(leaf, header, entries); btrfs_set_free_space_bitmaps(leaf, header, bitmaps); btrfs_set_free_space_generation(leaf, header, trans->transid); btrfs_release_path(path); return 0; } static noinline_for_stack int write_pinned_extent_entries( struct btrfs_trans_handle *trans, struct btrfs_block_group *block_group, struct btrfs_io_ctl *io_ctl, int *entries) { u64 start, extent_start, extent_end, len; const u64 block_group_end = btrfs_block_group_end(block_group); struct extent_io_tree *unpin = NULL; int ret; /* * We want to add any pinned extents to our free space cache * so we don't leak the space * * We shouldn't have switched the pinned extents yet so this is the * right one */ unpin = &trans->transaction->pinned_extents; start = block_group->start; while (start < block_group_end) { if (!btrfs_find_first_extent_bit(unpin, start, &extent_start, &extent_end, EXTENT_DIRTY, NULL)) return 0; /* This pinned extent is out of our range */ if (extent_start >= block_group_end) return 0; extent_start = max(extent_start, start); extent_end = min(block_group_end, extent_end + 1); len = extent_end - extent_start; *entries += 1; ret = io_ctl_add_entry(io_ctl, extent_start, len, NULL); if (ret) return -ENOSPC; start = extent_end; } return 0; } static noinline_for_stack int write_bitmap_entries(struct btrfs_io_ctl *io_ctl, struct list_head *bitmap_list) { struct btrfs_free_space *entry, *next; int ret; /* Write out the bitmaps */ list_for_each_entry_safe(entry, next, bitmap_list, list) { ret = io_ctl_add_bitmap(io_ctl, entry->bitmap); if (ret) return -ENOSPC; list_del_init(&entry->list); } return 0; } static int flush_dirty_cache(struct inode *inode) { int ret; ret = btrfs_wait_ordered_range(BTRFS_I(inode), 0, (u64)-1); if (ret) btrfs_clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, inode->i_size - 1, EXTENT_DELALLOC, NULL); return ret; } static void noinline_for_stack cleanup_bitmap_list(struct list_head *bitmap_list) { struct btrfs_free_space *entry, *next; list_for_each_entry_safe(entry, next, bitmap_list, list) list_del_init(&entry->list); } static void noinline_for_stack cleanup_write_cache_enospc(struct inode *inode, struct btrfs_io_ctl *io_ctl, struct extent_state **cached_state) { io_ctl_drop_pages(io_ctl); btrfs_unlock_extent(&BTRFS_I(inode)->io_tree, 0, i_size_read(inode) - 1, cached_state); } static int __btrfs_wait_cache_io(struct btrfs_root *root, struct btrfs_trans_handle *trans, struct btrfs_block_group *block_group, struct btrfs_io_ctl *io_ctl, struct btrfs_path *path, u64 offset) { int ret; struct inode *inode = io_ctl->inode; if (!inode) return 0; /* Flush the dirty pages in the cache file. */ ret = flush_dirty_cache(inode); if (ret) goto out; /* Update the cache item to tell everyone this cache file is valid. */ ret = update_cache_item(trans, root, inode, path, offset, io_ctl->entries, io_ctl->bitmaps); out: if (ret) { invalidate_inode_pages2(inode->i_mapping); BTRFS_I(inode)->generation = 0; if (block_group) btrfs_debug(root->fs_info, "failed to write free space cache for block group %llu error %d", block_group->start, ret); } btrfs_update_inode(trans, BTRFS_I(inode)); if (block_group) { /* the dirty list is protected by the dirty_bgs_lock */ spin_lock(&trans->transaction->dirty_bgs_lock); /* the disk_cache_state is protected by the block group lock */ spin_lock(&block_group->lock); /* * only mark this as written if we didn't get put back on * the dirty list while waiting for IO. Otherwise our * cache state won't be right, and we won't get written again */ if (!ret && list_empty(&block_group->dirty_list)) block_group->disk_cache_state = BTRFS_DC_WRITTEN; else if (ret) block_group->disk_cache_state = BTRFS_DC_ERROR; spin_unlock(&block_group->lock); spin_unlock(&trans->transaction->dirty_bgs_lock); io_ctl->inode = NULL; iput(inode); } return ret; } int btrfs_wait_cache_io(struct btrfs_trans_handle *trans, struct btrfs_block_group *block_group, struct btrfs_path *path) { return __btrfs_wait_cache_io(block_group->fs_info->tree_root, trans, block_group, &block_group->io_ctl, path, block_group->start); } /* * Write out cached info to an inode. * * @inode: freespace inode we are writing out * @ctl: free space cache we are going to write out * @block_group: block_group for this cache if it belongs to a block_group * @io_ctl: holds context for the io * @trans: the trans handle * * This function writes out a free space cache struct to disk for quick recovery * on mount. This will return 0 if it was successful in writing the cache out, * or an errno if it was not. */ static int __btrfs_write_out_cache(struct inode *inode, struct btrfs_free_space_ctl *ctl, struct btrfs_block_group *block_group, struct btrfs_trans_handle *trans) { struct btrfs_io_ctl *io_ctl = &block_group->io_ctl; struct extent_state *cached_state = NULL; LIST_HEAD(bitmap_list); int entries = 0; int bitmaps = 0; int ret; int must_iput = 0; int i_size; if (!i_size_read(inode)) return -EIO; WARN_ON(io_ctl->pages); ret = io_ctl_init(io_ctl, inode, 1); if (ret) return ret; if (block_group->flags & BTRFS_BLOCK_GROUP_DATA) { down_write(&block_group->data_rwsem); spin_lock(&block_group->lock); if (block_group->delalloc_bytes) { block_group->disk_cache_state = BTRFS_DC_WRITTEN; spin_unlock(&block_group->lock); up_write(&block_group->data_rwsem); BTRFS_I(inode)->generation = 0; ret = 0; must_iput = 1; goto out; } spin_unlock(&block_group->lock); } /* Lock all pages first so we can lock the extent safely. */ ret = io_ctl_prepare_pages(io_ctl, false); if (ret) goto out_unlock; btrfs_lock_extent(&BTRFS_I(inode)->io_tree, 0, i_size_read(inode) - 1, &cached_state); io_ctl_set_generation(io_ctl, trans->transid); mutex_lock(&ctl->cache_writeout_mutex); /* Write out the extent entries in the free space cache */ spin_lock(&ctl->tree_lock); ret = write_cache_extent_entries(io_ctl, ctl, block_group, &entries, &bitmaps, &bitmap_list); if (ret) goto out_nospc_locked; /* * Some spaces that are freed in the current transaction are pinned, * they will be added into free space cache after the transaction is * committed, we shouldn't lose them. * * If this changes while we are working we'll get added back to * the dirty list and redo it. No locking needed */ ret = write_pinned_extent_entries(trans, block_group, io_ctl, &entries); if (ret) goto out_nospc_locked; /* * At last, we write out all the bitmaps and keep cache_writeout_mutex * locked while doing it because a concurrent trim can be manipulating * or freeing the bitmap. */ ret = write_bitmap_entries(io_ctl, &bitmap_list); spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); if (ret) goto out_nospc; /* Zero out the rest of the pages just to make sure */ io_ctl_zero_remaining_pages(io_ctl); /* Everything is written out, now we dirty the pages in the file. */ i_size = i_size_read(inode); for (int i = 0; i < round_up(i_size, PAGE_SIZE) / PAGE_SIZE; i++) { u64 dirty_start = i * PAGE_SIZE; u64 dirty_len = min_t(u64, dirty_start + PAGE_SIZE, i_size) - dirty_start; ret = btrfs_dirty_folio(BTRFS_I(inode), page_folio(io_ctl->pages[i]), dirty_start, dirty_len, &cached_state, false); if (ret < 0) goto out_nospc; } if (block_group->flags & BTRFS_BLOCK_GROUP_DATA) up_write(&block_group->data_rwsem); /* * Release the pages and unlock the extent, we will flush * them out later */ io_ctl_drop_pages(io_ctl); io_ctl_free(io_ctl); btrfs_unlock_extent(&BTRFS_I(inode)->io_tree, 0, i_size_read(inode) - 1, &cached_state); /* * at this point the pages are under IO and we're happy, * The caller is responsible for waiting on them and updating * the cache and the inode */ io_ctl->entries = entries; io_ctl->bitmaps = bitmaps; ret = btrfs_fdatawrite_range(BTRFS_I(inode), 0, (u64)-1); if (ret) goto out; return 0; out_nospc_locked: cleanup_bitmap_list(&bitmap_list); spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); out_nospc: cleanup_write_cache_enospc(inode, io_ctl, &cached_state); out_unlock: if (block_group->flags & BTRFS_BLOCK_GROUP_DATA) up_write(&block_group->data_rwsem); out: io_ctl->inode = NULL; io_ctl_free(io_ctl); if (ret) { invalidate_inode_pages2(inode->i_mapping); BTRFS_I(inode)->generation = 0; } btrfs_update_inode(trans, BTRFS_I(inode)); if (must_iput) iput(inode); return ret; } int btrfs_write_out_cache(struct btrfs_trans_handle *trans, struct btrfs_block_group *block_group, struct btrfs_path *path) { struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct inode *inode; int ret = 0; spin_lock(&block_group->lock); if (block_group->disk_cache_state < BTRFS_DC_SETUP) { spin_unlock(&block_group->lock); return 0; } spin_unlock(&block_group->lock); inode = lookup_free_space_inode(block_group, path); if (IS_ERR(inode)) return 0; ret = __btrfs_write_out_cache(inode, ctl, block_group, trans); if (ret) { btrfs_debug(fs_info, "failed to write free space cache for block group %llu error %d", block_group->start, ret); spin_lock(&block_group->lock); block_group->disk_cache_state = BTRFS_DC_ERROR; spin_unlock(&block_group->lock); block_group->io_ctl.inode = NULL; iput(inode); } /* * if ret == 0 the caller is expected to call btrfs_wait_cache_io * to wait for IO and put the inode */ return ret; } static inline unsigned long offset_to_bit(u64 bitmap_start, u32 unit, u64 offset) { ASSERT(offset >= bitmap_start); offset -= bitmap_start; return (unsigned long)(div_u64(offset, unit)); } static inline unsigned long bytes_to_bits(u64 bytes, u32 unit) { return (unsigned long)(div_u64(bytes, unit)); } static inline u64 offset_to_bitmap(struct btrfs_free_space_ctl *ctl, u64 offset) { u64 bitmap_start; u64 bytes_per_bitmap; bytes_per_bitmap = BITS_PER_BITMAP * ctl->unit; bitmap_start = offset - ctl->start; bitmap_start = div64_u64(bitmap_start, bytes_per_bitmap); bitmap_start *= bytes_per_bitmap; bitmap_start += ctl->start; return bitmap_start; } static int tree_insert_offset(struct btrfs_free_space_ctl *ctl, struct btrfs_free_cluster *cluster, struct btrfs_free_space *new_entry) { struct rb_root *root; struct rb_node **p; struct rb_node *parent = NULL; lockdep_assert_held(&ctl->tree_lock); if (cluster) { lockdep_assert_held(&cluster->lock); root = &cluster->root; } else { root = &ctl->free_space_offset; } p = &root->rb_node; while (*p) { struct btrfs_free_space *info; parent = *p; info = rb_entry(parent, struct btrfs_free_space, offset_index); if (new_entry->offset < info->offset) { p = &(*p)->rb_left; } else if (new_entry->offset > info->offset) { p = &(*p)->rb_right; } else { /* * we could have a bitmap entry and an extent entry * share the same offset. If this is the case, we want * the extent entry to always be found first if we do a * linear search through the tree, since we want to have * the quickest allocation time, and allocating from an * extent is faster than allocating from a bitmap. So * if we're inserting a bitmap and we find an entry at * this offset, we want to go right, or after this entry * logically. If we are inserting an extent and we've * found a bitmap, we want to go left, or before * logically. */ if (new_entry->bitmap) { if (info->bitmap) { WARN_ON_ONCE(1); return -EEXIST; } p = &(*p)->rb_right; } else { if (!info->bitmap) { WARN_ON_ONCE(1); return -EEXIST; } p = &(*p)->rb_left; } } } rb_link_node(&new_entry->offset_index, parent, p); rb_insert_color(&new_entry->offset_index, root); return 0; } /* * This is a little subtle. We *only* have ->max_extent_size set if we actually * searched through the bitmap and figured out the largest ->max_extent_size, * otherwise it's 0. In the case that it's 0 we don't want to tell the * allocator the wrong thing, we want to use the actual real max_extent_size * we've found already if it's larger, or we want to use ->bytes. * * This matters because find_free_space() will skip entries who's ->bytes is * less than the required bytes. So if we didn't search down this bitmap, we * may pick some previous entry that has a smaller ->max_extent_size than we * have. For example, assume we have two entries, one that has * ->max_extent_size set to 4K and ->bytes set to 1M. A second entry hasn't set * ->max_extent_size yet, has ->bytes set to 8K and it's contiguous. We will * call into find_free_space(), and return with max_extent_size == 4K, because * that first bitmap entry had ->max_extent_size set, but the second one did * not. If instead we returned 8K we'd come in searching for 8K, and find the * 8K contiguous range. * * Consider the other case, we have 2 8K chunks in that second entry and still * don't have ->max_extent_size set. We'll return 16K, and the next time the * allocator comes in it'll fully search our second bitmap, and this time it'll * get an uptodate value of 8K as the maximum chunk size. Then we'll get the * right allocation the next loop through. */ static inline u64 get_max_extent_size(const struct btrfs_free_space *entry) { if (entry->bitmap && entry->max_extent_size) return entry->max_extent_size; return entry->bytes; } /* * We want the largest entry to be leftmost, so this is inverted from what you'd * normally expect. */ static bool entry_less(struct rb_node *node, const struct rb_node *parent) { const struct btrfs_free_space *entry, *exist; entry = rb_entry(node, struct btrfs_free_space, bytes_index); exist = rb_entry(parent, struct btrfs_free_space, bytes_index); return get_max_extent_size(exist) < get_max_extent_size(entry); } /* * searches the tree for the given offset. * * fuzzy - If this is set, then we are trying to make an allocation, and we just * want a section that has at least bytes size and comes at or after the given * offset. */ static struct btrfs_free_space * tree_search_offset(struct btrfs_free_space_ctl *ctl, u64 offset, int bitmap_only, int fuzzy) { struct rb_node *n = ctl->free_space_offset.rb_node; struct btrfs_free_space *entry = NULL, *prev = NULL; lockdep_assert_held(&ctl->tree_lock); /* find entry that is closest to the 'offset' */ while (n) { entry = rb_entry(n, struct btrfs_free_space, offset_index); prev = entry; if (offset < entry->offset) n = n->rb_left; else if (offset > entry->offset) n = n->rb_right; else break; entry = NULL; } if (bitmap_only) { if (!entry) return NULL; if (entry->bitmap) return entry; /* * bitmap entry and extent entry may share same offset, * in that case, bitmap entry comes after extent entry. */ n = rb_next(n); if (!n) return NULL; entry = rb_entry(n, struct btrfs_free_space, offset_index); if (entry->offset != offset) return NULL; WARN_ON(!entry->bitmap); return entry; } else if (entry) { if (entry->bitmap) { /* * if previous extent entry covers the offset, * we should return it instead of the bitmap entry */ n = rb_prev(&entry->offset_index); if (n) { prev = rb_entry(n, struct btrfs_free_space, offset_index); if (!prev->bitmap && prev->offset + prev->bytes > offset) entry = prev; } } return entry; } if (!prev) return NULL; /* find last entry before the 'offset' */ entry = prev; if (entry->offset > offset) { n = rb_prev(&entry->offset_index); if (n) { entry = rb_entry(n, struct btrfs_free_space, offset_index); ASSERT(entry->offset <= offset); } else { if (fuzzy) return entry; else return NULL; } } if (entry->bitmap) { n = rb_prev(&entry->offset_index); if (n) { prev = rb_entry(n, struct btrfs_free_space, offset_index); if (!prev->bitmap && prev->offset + prev->bytes > offset) return prev; } if (entry->offset + BITS_PER_BITMAP * ctl->unit > offset) return entry; } else if (entry->offset + entry->bytes > offset) return entry; if (!fuzzy) return NULL; while (1) { n = rb_next(&entry->offset_index); if (!n) return NULL; entry = rb_entry(n, struct btrfs_free_space, offset_index); if (entry->bitmap) { if (entry->offset + BITS_PER_BITMAP * ctl->unit > offset) break; } else { if (entry->offset + entry->bytes > offset) break; } } return entry; } static inline void unlink_free_space(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, bool update_stat) { lockdep_assert_held(&ctl->tree_lock); rb_erase(&info->offset_index, &ctl->free_space_offset); rb_erase_cached(&info->bytes_index, &ctl->free_space_bytes); ctl->free_extents--; if (!info->bitmap && !btrfs_free_space_trimmed(info)) { ctl->discardable_extents[BTRFS_STAT_CURR]--; ctl->discardable_bytes[BTRFS_STAT_CURR] -= info->bytes; } if (update_stat) ctl->free_space -= info->bytes; } static int link_free_space(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info) { int ret = 0; lockdep_assert_held(&ctl->tree_lock); ASSERT(info->bytes || info->bitmap); ret = tree_insert_offset(ctl, NULL, info); if (ret) return ret; rb_add_cached(&info->bytes_index, &ctl->free_space_bytes, entry_less); if (!info->bitmap && !btrfs_free_space_trimmed(info)) { ctl->discardable_extents[BTRFS_STAT_CURR]++; ctl->discardable_bytes[BTRFS_STAT_CURR] += info->bytes; } ctl->free_space += info->bytes; ctl->free_extents++; return ret; } static void relink_bitmap_entry(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info) { ASSERT(info->bitmap); /* * If our entry is empty it's because we're on a cluster and we don't * want to re-link it into our ctl bytes index. */ if (RB_EMPTY_NODE(&info->bytes_index)) return; lockdep_assert_held(&ctl->tree_lock); rb_erase_cached(&info->bytes_index, &ctl->free_space_bytes); rb_add_cached(&info->bytes_index, &ctl->free_space_bytes, entry_less); } static inline void bitmap_clear_bits(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, u64 offset, u64 bytes, bool update_stat) { unsigned long start, count, end; int extent_delta = -1; start = offset_to_bit(info->offset, ctl->unit, offset); count = bytes_to_bits(bytes, ctl->unit); end = start + count; ASSERT(end <= BITS_PER_BITMAP); bitmap_clear(info->bitmap, start, count); info->bytes -= bytes; if (info->max_extent_size > ctl->unit) info->max_extent_size = 0; relink_bitmap_entry(ctl, info); if (start && test_bit(start - 1, info->bitmap)) extent_delta++; if (end < BITS_PER_BITMAP && test_bit(end, info->bitmap)) extent_delta++; info->bitmap_extents += extent_delta; if (!btrfs_free_space_trimmed(info)) { ctl->discardable_extents[BTRFS_STAT_CURR] += extent_delta; ctl->discardable_bytes[BTRFS_STAT_CURR] -= bytes; } if (update_stat) ctl->free_space -= bytes; } static void btrfs_bitmap_set_bits(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, u64 offset, u64 bytes) { unsigned long start, count, end; int extent_delta = 1; start = offset_to_bit(info->offset, ctl->unit, offset); count = bytes_to_bits(bytes, ctl->unit); end = start + count; ASSERT(end <= BITS_PER_BITMAP); bitmap_set(info->bitmap, start, count); /* * We set some bytes, we have no idea what the max extent size is * anymore. */ info->max_extent_size = 0; info->bytes += bytes; ctl->free_space += bytes; relink_bitmap_entry(ctl, info); if (start && test_bit(start - 1, info->bitmap)) extent_delta--; if (end < BITS_PER_BITMAP && test_bit(end, info->bitmap)) extent_delta--; info->bitmap_extents += extent_delta; if (!btrfs_free_space_trimmed(info)) { ctl->discardable_extents[BTRFS_STAT_CURR] += extent_delta; ctl->discardable_bytes[BTRFS_STAT_CURR] += bytes; } } /* * If we can not find suitable extent, we will use bytes to record * the size of the max extent. */ static int search_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *bitmap_info, u64 *offset, u64 *bytes, bool for_alloc) { unsigned long found_bits = 0; unsigned long max_bits = 0; unsigned long bits, i; unsigned long next_zero; unsigned long extent_bits; /* * Skip searching the bitmap if we don't have a contiguous section that * is large enough for this allocation. */ if (for_alloc && bitmap_info->max_extent_size && bitmap_info->max_extent_size < *bytes) { *bytes = bitmap_info->max_extent_size; return -1; } i = offset_to_bit(bitmap_info->offset, ctl->unit, max_t(u64, *offset, bitmap_info->offset)); bits = bytes_to_bits(*bytes, ctl->unit); for_each_set_bit_from(i, bitmap_info->bitmap, BITS_PER_BITMAP) { if (for_alloc && bits == 1) { found_bits = 1; break; } next_zero = find_next_zero_bit(bitmap_info->bitmap, BITS_PER_BITMAP, i); extent_bits = next_zero - i; if (extent_bits >= bits) { found_bits = extent_bits; break; } else if (extent_bits > max_bits) { max_bits = extent_bits; } i = next_zero; } if (found_bits) { *offset = (u64)(i * ctl->unit) + bitmap_info->offset; *bytes = (u64)(found_bits) * ctl->unit; return 0; } *bytes = (u64)(max_bits) * ctl->unit; bitmap_info->max_extent_size = *bytes; relink_bitmap_entry(ctl, bitmap_info); return -1; } /* Cache the size of the max extent in bytes */ static struct btrfs_free_space * find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes, unsigned long align, u64 *max_extent_size, bool use_bytes_index) { struct btrfs_free_space *entry; struct rb_node *node; u64 tmp; u64 align_off; int ret; if (!ctl->free_space_offset.rb_node) return NULL; again: if (use_bytes_index) { node = rb_first_cached(&ctl->free_space_bytes); } else { entry = tree_search_offset(ctl, offset_to_bitmap(ctl, *offset), 0, 1); if (!entry) return NULL; node = &entry->offset_index; } for (; node; node = rb_next(node)) { if (use_bytes_index) entry = rb_entry(node, struct btrfs_free_space, bytes_index); else entry = rb_entry(node, struct btrfs_free_space, offset_index); /* * If we are using the bytes index then all subsequent entries * in this tree are going to be < bytes, so simply set the max * extent size and exit the loop. * * If we're using the offset index then we need to keep going * through the rest of the tree. */ if (entry->bytes < *bytes) { *max_extent_size = max(get_max_extent_size(entry), *max_extent_size); if (use_bytes_index) break; continue; } /* make sure the space returned is big enough * to match our requested alignment */ if (*bytes >= align) { tmp = entry->offset - ctl->start + align - 1; tmp = div64_u64(tmp, align); tmp = tmp * align + ctl->start; align_off = tmp - entry->offset; } else { align_off = 0; tmp = entry->offset; } /* * We don't break here if we're using the bytes index because we * may have another entry that has the correct alignment that is * the right size, so we don't want to miss that possibility. * At worst this adds another loop through the logic, but if we * broke here we could prematurely ENOSPC. */ if (entry->bytes < *bytes + align_off) { *max_extent_size = max(get_max_extent_size(entry), *max_extent_size); continue; } if (entry->bitmap) { struct rb_node *old_next = rb_next(node); u64 size = *bytes; ret = search_bitmap(ctl, entry, &tmp, &size, true); if (!ret) { *offset = tmp; *bytes = size; return entry; } else { *max_extent_size = max(get_max_extent_size(entry), *max_extent_size); } /* * The bitmap may have gotten re-arranged in the space * index here because the max_extent_size may have been * updated. Start from the beginning again if this * happened. */ if (use_bytes_index && old_next != rb_next(node)) goto again; continue; } *offset = tmp; *bytes = entry->bytes - align_off; return entry; } return NULL; } static void add_new_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, u64 offset) { info->offset = offset_to_bitmap(ctl, offset); info->bytes = 0; info->bitmap_extents = 0; INIT_LIST_HEAD(&info->list); link_free_space(ctl, info); ctl->total_bitmaps++; recalculate_thresholds(ctl); } static void free_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *bitmap_info) { /* * Normally when this is called, the bitmap is completely empty. However, * if we are blowing up the free space cache for one reason or another * via __btrfs_remove_free_space_cache(), then it may not be freed and * we may leave stats on the table. */ if (bitmap_info->bytes && !btrfs_free_space_trimmed(bitmap_info)) { ctl->discardable_extents[BTRFS_STAT_CURR] -= bitmap_info->bitmap_extents; ctl->discardable_bytes[BTRFS_STAT_CURR] -= bitmap_info->bytes; } unlink_free_space(ctl, bitmap_info, true); kmem_cache_free(btrfs_free_space_bitmap_cachep, bitmap_info->bitmap); kmem_cache_free(btrfs_free_space_cachep, bitmap_info); ctl->total_bitmaps--; recalculate_thresholds(ctl); } static noinline int remove_from_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *bitmap_info, u64 *offset, u64 *bytes) { u64 end; u64 search_start, search_bytes; int ret; again: end = bitmap_info->offset + (u64)(BITS_PER_BITMAP * ctl->unit) - 1; /* * We need to search for bits in this bitmap. We could only cover some * of the extent in this bitmap thanks to how we add space, so we need * to search for as much as it as we can and clear that amount, and then * go searching for the next bit. */ search_start = *offset; search_bytes = ctl->unit; search_bytes = min(search_bytes, end - search_start + 1); ret = search_bitmap(ctl, bitmap_info, &search_start, &search_bytes, false); if (ret < 0 || search_start != *offset) return -EINVAL; /* We may have found more bits than what we need */ search_bytes = min(search_bytes, *bytes); /* Cannot clear past the end of the bitmap */ search_bytes = min(search_bytes, end - search_start + 1); bitmap_clear_bits(ctl, bitmap_info, search_start, search_bytes, true); *offset += search_bytes; *bytes -= search_bytes; if (*bytes) { struct rb_node *next = rb_next(&bitmap_info->offset_index); if (!bitmap_info->bytes) free_bitmap(ctl, bitmap_info); /* * no entry after this bitmap, but we still have bytes to * remove, so something has gone wrong. */ if (!next) return -EINVAL; bitmap_info = rb_entry(next, struct btrfs_free_space, offset_index); /* * if the next entry isn't a bitmap we need to return to let the * extent stuff do its work. */ if (!bitmap_info->bitmap) return -EAGAIN; /* * Ok the next item is a bitmap, but it may not actually hold * the information for the rest of this free space stuff, so * look for it, and if we don't find it return so we can try * everything over again. */ search_start = *offset; search_bytes = ctl->unit; ret = search_bitmap(ctl, bitmap_info, &search_start, &search_bytes, false); if (ret < 0 || search_start != *offset) return -EAGAIN; goto again; } else if (!bitmap_info->bytes) free_bitmap(ctl, bitmap_info); return 0; } static u64 add_bytes_to_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, u64 offset, u64 bytes, enum btrfs_trim_state trim_state) { u64 bytes_to_set = 0; u64 end; /* * This is a tradeoff to make bitmap trim state minimal. We mark the * whole bitmap untrimmed if at any point we add untrimmed regions. */ if (trim_state == BTRFS_TRIM_STATE_UNTRIMMED) { if (btrfs_free_space_trimmed(info)) { ctl->discardable_extents[BTRFS_STAT_CURR] += info->bitmap_extents; ctl->discardable_bytes[BTRFS_STAT_CURR] += info->bytes; } info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; } end = info->offset + (u64)(BITS_PER_BITMAP * ctl->unit); bytes_to_set = min(end - offset, bytes); btrfs_bitmap_set_bits(ctl, info, offset, bytes_to_set); return bytes_to_set; } static bool use_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info) { struct btrfs_block_group *block_group = ctl->block_group; struct btrfs_fs_info *fs_info = block_group->fs_info; bool forced = false; #ifdef CONFIG_BTRFS_DEBUG if (btrfs_should_fragment_free_space(block_group)) forced = true; #endif /* This is a way to reclaim large regions from the bitmaps. */ if (!forced && info->bytes >= FORCE_EXTENT_THRESHOLD) return false; /* * If we are below the extents threshold then we can add this as an * extent, and don't have to deal with the bitmap */ if (!forced && ctl->free_extents < ctl->extents_thresh) { /* * If this block group has some small extents we don't want to * use up all of our free slots in the cache with them, we want * to reserve them to larger extents, however if we have plenty * of cache left then go ahead and add them, no sense in adding * the overhead of a bitmap if we don't have to. */ if (info->bytes <= fs_info->sectorsize * 8) { if (ctl->free_extents * 3 <= ctl->extents_thresh) return false; } else { return false; } } /* * The original block groups from mkfs can be really small, like 8 * megabytes, so don't bother with a bitmap for those entries. However * some block groups can be smaller than what a bitmap would cover but * are still large enough that they could overflow the 32k memory limit, * so allow those block groups to still be allowed to have a bitmap * entry. */ if (((BITS_PER_BITMAP * ctl->unit) >> 1) > block_group->length) return false; return true; } static const struct btrfs_free_space_op free_space_op = { .use_bitmap = use_bitmap, }; static int insert_into_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info) { struct btrfs_free_space *bitmap_info; struct btrfs_block_group *block_group = NULL; int added = 0; u64 bytes, offset, bytes_added; enum btrfs_trim_state trim_state; int ret; bytes = info->bytes; offset = info->offset; trim_state = info->trim_state; if (!ctl->op->use_bitmap(ctl, info)) return 0; if (ctl->op == &free_space_op) block_group = ctl->block_group; again: /* * Since we link bitmaps right into the cluster we need to see if we * have a cluster here, and if so and it has our bitmap we need to add * the free space to that bitmap. */ if (block_group && !list_empty(&block_group->cluster_list)) { struct btrfs_free_cluster *cluster; struct rb_node *node; struct btrfs_free_space *entry; cluster = list_first_entry(&block_group->cluster_list, struct btrfs_free_cluster, block_group_list); spin_lock(&cluster->lock); node = rb_first(&cluster->root); if (!node) { spin_unlock(&cluster->lock); goto no_cluster_bitmap; } entry = rb_entry(node, struct btrfs_free_space, offset_index); if (!entry->bitmap) { spin_unlock(&cluster->lock); goto no_cluster_bitmap; } if (entry->offset == offset_to_bitmap(ctl, offset)) { bytes_added = add_bytes_to_bitmap(ctl, entry, offset, bytes, trim_state); bytes -= bytes_added; offset += bytes_added; } spin_unlock(&cluster->lock); if (!bytes) { ret = 1; goto out; } } no_cluster_bitmap: bitmap_info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), 1, 0); if (!bitmap_info) { ASSERT(added == 0); goto new_bitmap; } bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes, trim_state); bytes -= bytes_added; offset += bytes_added; added = 0; if (!bytes) { ret = 1; goto out; } else goto again; new_bitmap: if (info && info->bitmap) { add_new_bitmap(ctl, info, offset); added = 1; info = NULL; goto again; } else { spin_unlock(&ctl->tree_lock); /* no pre-allocated info, allocate a new one */ if (!info) { info = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS); if (!info) { spin_lock(&ctl->tree_lock); ret = -ENOMEM; goto out; } } /* allocate the bitmap */ info->bitmap = kmem_cache_zalloc(btrfs_free_space_bitmap_cachep, GFP_NOFS); info->trim_state = BTRFS_TRIM_STATE_TRIMMED; spin_lock(&ctl->tree_lock); if (!info->bitmap) { ret = -ENOMEM; goto out; } goto again; } out: if (info) { if (info->bitmap) kmem_cache_free(btrfs_free_space_bitmap_cachep, info->bitmap); kmem_cache_free(btrfs_free_space_cachep, info); } return ret; } /* * Free space merging rules: * 1) Merge trimmed areas together * 2) Let untrimmed areas coalesce with trimmed areas * 3) Always pull neighboring regions from bitmaps * * The above rules are for when we merge free space based on btrfs_trim_state. * Rules 2 and 3 are subtle because they are suboptimal, but are done for the * same reason: to promote larger extent regions which makes life easier for * find_free_extent(). Rule 2 enables coalescing based on the common path * being returning free space from btrfs_finish_extent_commit(). So when free * space is trimmed, it will prevent aggregating trimmed new region and * untrimmed regions in the rb_tree. Rule 3 is purely to obtain larger extents * and provide find_free_extent() with the largest extents possible hoping for * the reuse path. */ static bool try_merge_free_space(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, bool update_stat) { struct btrfs_free_space *left_info = NULL; struct btrfs_free_space *right_info; bool merged = false; u64 offset = info->offset; u64 bytes = info->bytes; const bool is_trimmed = btrfs_free_space_trimmed(info); struct rb_node *right_prev = NULL; /* * first we want to see if there is free space adjacent to the range we * are adding, if there is remove that struct and add a new one to * cover the entire range */ right_info = tree_search_offset(ctl, offset + bytes, 0, 0); if (right_info) right_prev = rb_prev(&right_info->offset_index); if (right_prev) left_info = rb_entry(right_prev, struct btrfs_free_space, offset_index); else if (!right_info) left_info = tree_search_offset(ctl, offset - 1, 0, 0); /* See try_merge_free_space() comment. */ if (right_info && !right_info->bitmap && (!is_trimmed || btrfs_free_space_trimmed(right_info))) { unlink_free_space(ctl, right_info, update_stat); info->bytes += right_info->bytes; kmem_cache_free(btrfs_free_space_cachep, right_info); merged = true; } /* See try_merge_free_space() comment. */ if (left_info && !left_info->bitmap && left_info->offset + left_info->bytes == offset && (!is_trimmed || btrfs_free_space_trimmed(left_info))) { unlink_free_space(ctl, left_info, update_stat); info->offset = left_info->offset; info->bytes += left_info->bytes; kmem_cache_free(btrfs_free_space_cachep, left_info); merged = true; } return merged; } static bool steal_from_bitmap_to_end(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, bool update_stat) { struct btrfs_free_space *bitmap; unsigned long i; unsigned long j; const u64 end = info->offset + info->bytes; const u64 bitmap_offset = offset_to_bitmap(ctl, end); u64 bytes; bitmap = tree_search_offset(ctl, bitmap_offset, 1, 0); if (!bitmap) return false; i = offset_to_bit(bitmap->offset, ctl->unit, end); j = find_next_zero_bit(bitmap->bitmap, BITS_PER_BITMAP, i); if (j == i) return false; bytes = (j - i) * ctl->unit; info->bytes += bytes; /* See try_merge_free_space() comment. */ if (!btrfs_free_space_trimmed(bitmap)) info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; bitmap_clear_bits(ctl, bitmap, end, bytes, update_stat); if (!bitmap->bytes) free_bitmap(ctl, bitmap); return true; } static bool steal_from_bitmap_to_front(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, bool update_stat) { struct btrfs_free_space *bitmap; u64 bitmap_offset; unsigned long i; unsigned long j; unsigned long prev_j; u64 bytes; bitmap_offset = offset_to_bitmap(ctl, info->offset); /* If we're on a boundary, try the previous logical bitmap. */ if (bitmap_offset == info->offset) { if (info->offset == 0) return false; bitmap_offset = offset_to_bitmap(ctl, info->offset - 1); } bitmap = tree_search_offset(ctl, bitmap_offset, 1, 0); if (!bitmap) return false; i = offset_to_bit(bitmap->offset, ctl->unit, info->offset) - 1; j = 0; prev_j = (unsigned long)-1; for_each_clear_bit_from(j, bitmap->bitmap, BITS_PER_BITMAP) { if (j > i) break; prev_j = j; } if (prev_j == i) return false; if (prev_j == (unsigned long)-1) bytes = (i + 1) * ctl->unit; else bytes = (i - prev_j) * ctl->unit; info->offset -= bytes; info->bytes += bytes; /* See try_merge_free_space() comment. */ if (!btrfs_free_space_trimmed(bitmap)) info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; bitmap_clear_bits(ctl, bitmap, info->offset, bytes, update_stat); if (!bitmap->bytes) free_bitmap(ctl, bitmap); return true; } /* * We prefer always to allocate from extent entries, both for clustered and * non-clustered allocation requests. So when attempting to add a new extent * entry, try to see if there's adjacent free space in bitmap entries, and if * there is, migrate that space from the bitmaps to the extent. * Like this we get better chances of satisfying space allocation requests * because we attempt to satisfy them based on a single cache entry, and never * on 2 or more entries - even if the entries represent a contiguous free space * region (e.g. 1 extent entry + 1 bitmap entry starting where the extent entry * ends). */ static void steal_from_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *info, bool update_stat) { /* * Only work with disconnected entries, as we can change their offset, * and must be extent entries. */ ASSERT(!info->bitmap); ASSERT(RB_EMPTY_NODE(&info->offset_index)); if (ctl->total_bitmaps > 0) { bool stole_end; bool stole_front = false; stole_end = steal_from_bitmap_to_end(ctl, info, update_stat); if (ctl->total_bitmaps > 0) stole_front = steal_from_bitmap_to_front(ctl, info, update_stat); if (stole_end || stole_front) try_merge_free_space(ctl, info, update_stat); } } static int __btrfs_add_free_space(struct btrfs_block_group *block_group, u64 offset, u64 bytes, enum btrfs_trim_state trim_state) { struct btrfs_fs_info *fs_info = block_group->fs_info; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *info; int ret = 0; u64 filter_bytes = bytes; ASSERT(!btrfs_is_zoned(fs_info)); info = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS); if (!info) return -ENOMEM; info->offset = offset; info->bytes = bytes; info->trim_state = trim_state; RB_CLEAR_NODE(&info->offset_index); RB_CLEAR_NODE(&info->bytes_index); spin_lock(&ctl->tree_lock); if (try_merge_free_space(ctl, info, true)) goto link; /* * There was no extent directly to the left or right of this new * extent then we know we're going to have to allocate a new extent, so * before we do that see if we need to drop this into a bitmap */ ret = insert_into_bitmap(ctl, info); if (ret < 0) { goto out; } else if (ret) { ret = 0; goto out; } link: /* * Only steal free space from adjacent bitmaps if we're sure we're not * going to add the new free space to existing bitmap entries - because * that would mean unnecessary work that would be reverted. Therefore * attempt to steal space from bitmaps if we're adding an extent entry. */ steal_from_bitmap(ctl, info, true); filter_bytes = max(filter_bytes, info->bytes); ret = link_free_space(ctl, info); if (ret) kmem_cache_free(btrfs_free_space_cachep, info); out: btrfs_discard_update_discardable(block_group); spin_unlock(&ctl->tree_lock); if (ret) { btrfs_crit(fs_info, "unable to add free space :%d", ret); ASSERT(ret != -EEXIST); } if (trim_state != BTRFS_TRIM_STATE_TRIMMED) { btrfs_discard_check_filter(block_group, filter_bytes); btrfs_discard_queue_work(&fs_info->discard_ctl, block_group); } return ret; } static int __btrfs_add_free_space_zoned(struct btrfs_block_group *block_group, u64 bytenr, u64 size, bool used) { struct btrfs_space_info *sinfo = block_group->space_info; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; u64 offset = bytenr - block_group->start; u64 to_free, to_unusable; int bg_reclaim_threshold = 0; bool initial; u64 reclaimable_unusable; spin_lock(&block_group->lock); initial = ((size == block_group->length) && (block_group->alloc_offset == 0)); WARN_ON(!initial && offset + size > block_group->zone_capacity); if (!initial) bg_reclaim_threshold = READ_ONCE(sinfo->bg_reclaim_threshold); if (!used) to_free = size; else if (initial) to_free = block_group->zone_capacity; else if (offset >= block_group->alloc_offset) to_free = size; else if (offset + size <= block_group->alloc_offset) to_free = 0; else to_free = offset + size - block_group->alloc_offset; to_unusable = size - to_free; spin_lock(&ctl->tree_lock); ctl->free_space += to_free; spin_unlock(&ctl->tree_lock); /* * If the block group is read-only, we should account freed space into * bytes_readonly. */ if (!block_group->ro) { block_group->zone_unusable += to_unusable; WARN_ON(block_group->zone_unusable > block_group->length); } if (!used) { block_group->alloc_offset -= size; } reclaimable_unusable = block_group->zone_unusable - (block_group->length - block_group->zone_capacity); /* All the region is now unusable. Mark it as unused and reclaim */ if (block_group->zone_unusable == block_group->length) { btrfs_mark_bg_unused(block_group); } else if (bg_reclaim_threshold && reclaimable_unusable >= mult_perc(block_group->zone_capacity, bg_reclaim_threshold)) { btrfs_mark_bg_to_reclaim(block_group); } spin_unlock(&block_group->lock); return 0; } int btrfs_add_free_space(struct btrfs_block_group *block_group, u64 bytenr, u64 size) { enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED; if (block_group->flags & BTRFS_BLOCK_GROUP_REMAPPED) return 0; if (btrfs_is_zoned(block_group->fs_info)) return __btrfs_add_free_space_zoned(block_group, bytenr, size, true); if (btrfs_test_opt(block_group->fs_info, DISCARD_SYNC)) trim_state = BTRFS_TRIM_STATE_TRIMMED; return __btrfs_add_free_space(block_group, bytenr, size, trim_state); } int btrfs_add_free_space_unused(struct btrfs_block_group *block_group, u64 bytenr, u64 size) { if (btrfs_is_zoned(block_group->fs_info)) return __btrfs_add_free_space_zoned(block_group, bytenr, size, false); return btrfs_add_free_space(block_group, bytenr, size); } /* * This is a subtle distinction because when adding free space back in general, * we want it to be added as untrimmed for async. But in the case where we add * it on loading of a block group, we want to consider it trimmed. */ int btrfs_add_free_space_async_trimmed(struct btrfs_block_group *block_group, u64 bytenr, u64 size) { enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED; if (btrfs_is_zoned(block_group->fs_info)) return __btrfs_add_free_space_zoned(block_group, bytenr, size, true); if (btrfs_test_opt(block_group->fs_info, DISCARD_SYNC) || btrfs_test_opt(block_group->fs_info, DISCARD_ASYNC)) trim_state = BTRFS_TRIM_STATE_TRIMMED; return __btrfs_add_free_space(block_group, bytenr, size, trim_state); } int btrfs_remove_free_space(struct btrfs_block_group *block_group, u64 offset, u64 bytes) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *info; int ret; bool re_search = false; if (btrfs_is_zoned(block_group->fs_info)) { /* * This can happen with conventional zones when replaying log. * Since the allocation info of tree-log nodes are not recorded * to the extent-tree, calculate_alloc_pointer() failed to * advance the allocation pointer after last allocated tree log * node blocks. * * This function is called from * btrfs_pin_extent_for_log_replay() when replaying the log. * Advance the pointer not to overwrite the tree-log nodes. */ if (block_group->start + block_group->alloc_offset < offset + bytes) { block_group->alloc_offset = offset + bytes - block_group->start; } return 0; } spin_lock(&ctl->tree_lock); again: ret = 0; if (!bytes) goto out_lock; info = tree_search_offset(ctl, offset, 0, 0); if (!info) { /* * oops didn't find an extent that matched the space we wanted * to remove, look for a bitmap instead */ info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), 1, 0); if (!info) { /* * If we found a partial bit of our free space in a * bitmap but then couldn't find the other part this may * be a problem, so WARN about it. */ WARN_ON(re_search); goto out_lock; } } re_search = false; if (!info->bitmap) { unlink_free_space(ctl, info, true); if (offset == info->offset) { u64 to_free = min(bytes, info->bytes); info->bytes -= to_free; info->offset += to_free; if (info->bytes) { ret = link_free_space(ctl, info); WARN_ON(ret); } else { kmem_cache_free(btrfs_free_space_cachep, info); } offset += to_free; bytes -= to_free; goto again; } else { u64 old_end = info->bytes + info->offset; info->bytes = offset - info->offset; ret = link_free_space(ctl, info); WARN_ON(ret); if (ret) goto out_lock; /* Not enough bytes in this entry to satisfy us */ if (old_end < offset + bytes) { bytes -= old_end - offset; offset = old_end; goto again; } else if (old_end == offset + bytes) { /* all done */ goto out_lock; } spin_unlock(&ctl->tree_lock); ret = __btrfs_add_free_space(block_group, offset + bytes, old_end - (offset + bytes), info->trim_state); WARN_ON(ret); return ret; } } ret = remove_from_bitmap(ctl, info, &offset, &bytes); if (ret == -EAGAIN) { re_search = true; goto again; } out_lock: btrfs_discard_update_discardable(block_group); spin_unlock(&ctl->tree_lock); return ret; } void btrfs_dump_free_space(struct btrfs_block_group *block_group, u64 bytes) { struct btrfs_fs_info *fs_info = block_group->fs_info; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *info; struct rb_node *n; int count = 0; /* * Zoned btrfs does not use free space tree and cluster. Just print * out the free space after the allocation offset. */ if (btrfs_is_zoned(fs_info)) { btrfs_info(fs_info, "free space %llu active %d", block_group->zone_capacity - block_group->alloc_offset, test_bit(BLOCK_GROUP_FLAG_ZONE_IS_ACTIVE, &block_group->runtime_flags)); return; } spin_lock(&ctl->tree_lock); for (n = rb_first(&ctl->free_space_offset); n; n = rb_next(n)) { info = rb_entry(n, struct btrfs_free_space, offset_index); if (info->bytes >= bytes && !block_group->ro) count++; btrfs_crit(fs_info, "entry offset %llu, bytes %llu, bitmap %s", info->offset, info->bytes, str_yes_no(info->bitmap)); } spin_unlock(&ctl->tree_lock); btrfs_info(fs_info, "block group has cluster?: %s", str_no_yes(list_empty(&block_group->cluster_list))); btrfs_info(fs_info, "%d free space entries at or bigger than %llu bytes", count, bytes); } void btrfs_init_free_space_ctl(struct btrfs_block_group *block_group, struct btrfs_free_space_ctl *ctl) { struct btrfs_fs_info *fs_info = block_group->fs_info; spin_lock_init(&ctl->tree_lock); ctl->unit = fs_info->sectorsize; ctl->start = block_group->start; ctl->block_group = block_group; ctl->op = &free_space_op; ctl->free_space_bytes = RB_ROOT_CACHED; INIT_LIST_HEAD(&ctl->trimming_ranges); mutex_init(&ctl->cache_writeout_mutex); /* * we only want to have 32k of ram per block group for keeping * track of free space, and if we pass 1/2 of that we want to * start converting things over to using bitmaps */ ctl->extents_thresh = (SZ_32K / 2) / sizeof(struct btrfs_free_space); } /* * for a given cluster, put all of its extents back into the free * space cache. If the block group passed doesn't match the block group * pointed to by the cluster, someone else raced in and freed the * cluster already. In that case, we just return without changing anything */ static void __btrfs_return_cluster_to_free_space( struct btrfs_block_group *block_group, struct btrfs_free_cluster *cluster) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct rb_node *node; lockdep_assert_held(&ctl->tree_lock); spin_lock(&cluster->lock); if (cluster->block_group != block_group) { spin_unlock(&cluster->lock); return; } cluster->block_group = NULL; cluster->window_start = 0; list_del_init(&cluster->block_group_list); node = rb_first(&cluster->root); while (node) { struct btrfs_free_space *entry; entry = rb_entry(node, struct btrfs_free_space, offset_index); node = rb_next(&entry->offset_index); rb_erase(&entry->offset_index, &cluster->root); RB_CLEAR_NODE(&entry->offset_index); if (!entry->bitmap) { /* Merging treats extents as if they were new */ if (!btrfs_free_space_trimmed(entry)) { ctl->discardable_extents[BTRFS_STAT_CURR]--; ctl->discardable_bytes[BTRFS_STAT_CURR] -= entry->bytes; } try_merge_free_space(ctl, entry, false); steal_from_bitmap(ctl, entry, false); /* As we insert directly, update these statistics */ if (!btrfs_free_space_trimmed(entry)) { ctl->discardable_extents[BTRFS_STAT_CURR]++; ctl->discardable_bytes[BTRFS_STAT_CURR] += entry->bytes; } } tree_insert_offset(ctl, NULL, entry); rb_add_cached(&entry->bytes_index, &ctl->free_space_bytes, entry_less); } cluster->root = RB_ROOT; spin_unlock(&cluster->lock); btrfs_put_block_group(block_group); } void btrfs_remove_free_space_cache(struct btrfs_block_group *block_group) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_cluster *cluster; struct list_head *head; spin_lock(&ctl->tree_lock); while ((head = block_group->cluster_list.next) != &block_group->cluster_list) { cluster = list_entry(head, struct btrfs_free_cluster, block_group_list); WARN_ON(cluster->block_group != block_group); __btrfs_return_cluster_to_free_space(block_group, cluster); cond_resched_lock(&ctl->tree_lock); } __btrfs_remove_free_space_cache(ctl); btrfs_discard_update_discardable(block_group); spin_unlock(&ctl->tree_lock); } /* * Walk @block_group's free space rb_tree to determine if everything is trimmed. */ bool btrfs_is_free_space_trimmed(struct btrfs_block_group *block_group) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *info; struct rb_node *node; bool ret = true; if (block_group->flags & BTRFS_BLOCK_GROUP_REMAPPED && !test_bit(BLOCK_GROUP_FLAG_STRIPE_REMOVAL_PENDING, &block_group->runtime_flags) && block_group->identity_remap_count == 0) { return true; } spin_lock(&ctl->tree_lock); node = rb_first(&ctl->free_space_offset); while (node) { info = rb_entry(node, struct btrfs_free_space, offset_index); if (!btrfs_free_space_trimmed(info)) { ret = false; break; } node = rb_next(node); } spin_unlock(&ctl->tree_lock); return ret; } u64 btrfs_find_space_for_alloc(struct btrfs_block_group *block_group, u64 offset, u64 bytes, u64 empty_size, u64 *max_extent_size) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_discard_ctl *discard_ctl = &block_group->fs_info->discard_ctl; struct btrfs_free_space *entry = NULL; u64 bytes_search = bytes + empty_size; u64 ret = 0; u64 align_gap = 0; u64 align_gap_len = 0; enum btrfs_trim_state align_gap_trim_state = BTRFS_TRIM_STATE_UNTRIMMED; bool use_bytes_index = (offset == block_group->start); ASSERT(!btrfs_is_zoned(block_group->fs_info)); spin_lock(&ctl->tree_lock); entry = find_free_space(ctl, &offset, &bytes_search, block_group->full_stripe_len, max_extent_size, use_bytes_index); if (!entry) goto out; ret = offset; if (entry->bitmap) { bitmap_clear_bits(ctl, entry, offset, bytes, true); if (!btrfs_free_space_trimmed(entry)) atomic64_add(bytes, &discard_ctl->discard_bytes_saved); if (!entry->bytes) free_bitmap(ctl, entry); } else { unlink_free_space(ctl, entry, true); align_gap_len = offset - entry->offset; align_gap = entry->offset; align_gap_trim_state = entry->trim_state; if (!btrfs_free_space_trimmed(entry)) atomic64_add(bytes, &discard_ctl->discard_bytes_saved); entry->offset = offset + bytes; WARN_ON(entry->bytes < bytes + align_gap_len); entry->bytes -= bytes + align_gap_len; if (!entry->bytes) kmem_cache_free(btrfs_free_space_cachep, entry); else link_free_space(ctl, entry); } out: btrfs_discard_update_discardable(block_group); spin_unlock(&ctl->tree_lock); if (align_gap_len) __btrfs_add_free_space(block_group, align_gap, align_gap_len, align_gap_trim_state); return ret; } /* * given a cluster, put all of its extents back into the free space * cache. If a block group is passed, this function will only free * a cluster that belongs to the passed block group. * * Otherwise, it'll get a reference on the block group pointed to by the * cluster and remove the cluster from it. */ void btrfs_return_cluster_to_free_space( struct btrfs_block_group *block_group, struct btrfs_free_cluster *cluster) { struct btrfs_free_space_ctl *ctl; /* first, get a safe pointer to the block group */ spin_lock(&cluster->lock); if (!block_group) { block_group = cluster->block_group; if (!block_group) { spin_unlock(&cluster->lock); return; } } else if (cluster->block_group != block_group) { /* someone else has already freed it don't redo their work */ spin_unlock(&cluster->lock); return; } btrfs_get_block_group(block_group); spin_unlock(&cluster->lock); ctl = block_group->free_space_ctl; /* now return any extents the cluster had on it */ spin_lock(&ctl->tree_lock); __btrfs_return_cluster_to_free_space(block_group, cluster); spin_unlock(&ctl->tree_lock); btrfs_discard_queue_work(&block_group->fs_info->discard_ctl, block_group); /* finally drop our ref */ btrfs_put_block_group(block_group); } static u64 btrfs_alloc_from_bitmap(struct btrfs_block_group *block_group, struct btrfs_free_cluster *cluster, struct btrfs_free_space *entry, u64 bytes, u64 min_start, u64 *max_extent_size) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; int ret2; u64 search_start = cluster->window_start; u64 search_bytes = bytes; u64 ret = 0; search_start = min_start; search_bytes = bytes; ret2 = search_bitmap(ctl, entry, &search_start, &search_bytes, true); if (ret2) { *max_extent_size = max(get_max_extent_size(entry), *max_extent_size); return 0; } ret = search_start; bitmap_clear_bits(ctl, entry, ret, bytes, false); return ret; } /* * given a cluster, try to allocate 'bytes' from it, returns 0 * if it couldn't find anything suitably large, or a logical disk offset * if things worked out */ u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group, struct btrfs_free_cluster *cluster, u64 bytes, u64 min_start, u64 *max_extent_size) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_discard_ctl *discard_ctl = &block_group->fs_info->discard_ctl; struct btrfs_free_space *entry = NULL; struct rb_node *node; u64 ret = 0; ASSERT(!btrfs_is_zoned(block_group->fs_info)); spin_lock(&cluster->lock); if (bytes > cluster->max_size) goto out; if (cluster->block_group != block_group) goto out; node = rb_first(&cluster->root); if (!node) goto out; entry = rb_entry(node, struct btrfs_free_space, offset_index); while (1) { if (entry->bytes < bytes) *max_extent_size = max(get_max_extent_size(entry), *max_extent_size); if (entry->bytes < bytes || (!entry->bitmap && entry->offset < min_start)) { node = rb_next(&entry->offset_index); if (!node) break; entry = rb_entry(node, struct btrfs_free_space, offset_index); continue; } if (entry->bitmap) { ret = btrfs_alloc_from_bitmap(block_group, cluster, entry, bytes, cluster->window_start, max_extent_size); if (ret == 0) { node = rb_next(&entry->offset_index); if (!node) break; entry = rb_entry(node, struct btrfs_free_space, offset_index); continue; } cluster->window_start += bytes; } else { ret = entry->offset; entry->offset += bytes; entry->bytes -= bytes; } break; } out: spin_unlock(&cluster->lock); if (!ret) return 0; spin_lock(&ctl->tree_lock); if (!btrfs_free_space_trimmed(entry)) atomic64_add(bytes, &discard_ctl->discard_bytes_saved); ctl->free_space -= bytes; if (!entry->bitmap && !btrfs_free_space_trimmed(entry)) ctl->discardable_bytes[BTRFS_STAT_CURR] -= bytes; spin_lock(&cluster->lock); if (entry->bytes == 0) { rb_erase(&entry->offset_index, &cluster->root); ctl->free_extents--; if (entry->bitmap) { kmem_cache_free(btrfs_free_space_bitmap_cachep, entry->bitmap); ctl->total_bitmaps--; recalculate_thresholds(ctl); } else if (!btrfs_free_space_trimmed(entry)) { ctl->discardable_extents[BTRFS_STAT_CURR]--; } kmem_cache_free(btrfs_free_space_cachep, entry); } spin_unlock(&cluster->lock); spin_unlock(&ctl->tree_lock); return ret; } static int btrfs_bitmap_cluster(struct btrfs_block_group *block_group, struct btrfs_free_space *entry, struct btrfs_free_cluster *cluster, u64 offset, u64 bytes, u64 cont1_bytes, u64 min_bytes) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; unsigned long next_zero; unsigned long i; unsigned long want_bits; unsigned long min_bits; unsigned long found_bits; unsigned long max_bits = 0; unsigned long start = 0; unsigned long total_found = 0; int ret; lockdep_assert_held(&ctl->tree_lock); i = offset_to_bit(entry->offset, ctl->unit, max_t(u64, offset, entry->offset)); want_bits = bytes_to_bits(bytes, ctl->unit); min_bits = bytes_to_bits(min_bytes, ctl->unit); /* * Don't bother looking for a cluster in this bitmap if it's heavily * fragmented. */ if (entry->max_extent_size && entry->max_extent_size < cont1_bytes) return -ENOSPC; again: found_bits = 0; for_each_set_bit_from(i, entry->bitmap, BITS_PER_BITMAP) { next_zero = find_next_zero_bit(entry->bitmap, BITS_PER_BITMAP, i); if (next_zero - i >= min_bits) { found_bits = next_zero - i; if (found_bits > max_bits) max_bits = found_bits; break; } if (next_zero - i > max_bits) max_bits = next_zero - i; i = next_zero; } if (!found_bits) { entry->max_extent_size = (u64)max_bits * ctl->unit; return -ENOSPC; } if (!total_found) { start = i; cluster->max_size = 0; } total_found += found_bits; if (cluster->max_size < found_bits * ctl->unit) cluster->max_size = found_bits * ctl->unit; if (total_found < want_bits || cluster->max_size < cont1_bytes) { i = next_zero + 1; goto again; } cluster->window_start = start * ctl->unit + entry->offset; rb_erase(&entry->offset_index, &ctl->free_space_offset); rb_erase_cached(&entry->bytes_index, &ctl->free_space_bytes); /* * We need to know if we're currently on the normal space index when we * manipulate the bitmap so that we know we need to remove and re-insert * it into the space_index tree. Clear the bytes_index node here so the * bitmap manipulation helpers know not to mess with the space_index * until this bitmap entry is added back into the normal cache. */ RB_CLEAR_NODE(&entry->bytes_index); ret = tree_insert_offset(ctl, cluster, entry); ASSERT(!ret); /* -EEXIST; Logic error */ trace_btrfs_setup_cluster(block_group, cluster, total_found * ctl->unit, 1); return 0; } /* * This searches the block group for just extents to fill the cluster with. * Try to find a cluster with at least bytes total bytes, at least one * extent of cont1_bytes, and other clusters of at least min_bytes. */ static noinline int setup_cluster_no_bitmap(struct btrfs_block_group *block_group, struct btrfs_free_cluster *cluster, struct list_head *bitmaps, u64 offset, u64 bytes, u64 cont1_bytes, u64 min_bytes) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *first = NULL; struct btrfs_free_space *entry = NULL; struct btrfs_free_space *last; struct rb_node *node; u64 window_free; u64 max_extent; u64 total_size = 0; lockdep_assert_held(&ctl->tree_lock); entry = tree_search_offset(ctl, offset, 0, 1); if (!entry) return -ENOSPC; /* * We don't want bitmaps, so just move along until we find a normal * extent entry. */ while (entry->bitmap || entry->bytes < min_bytes) { if (entry->bitmap && list_empty(&entry->list)) list_add_tail(&entry->list, bitmaps); node = rb_next(&entry->offset_index); if (!node) return -ENOSPC; entry = rb_entry(node, struct btrfs_free_space, offset_index); } window_free = entry->bytes; max_extent = entry->bytes; first = entry; last = entry; for (node = rb_next(&entry->offset_index); node; node = rb_next(&entry->offset_index)) { entry = rb_entry(node, struct btrfs_free_space, offset_index); if (entry->bitmap) { if (list_empty(&entry->list)) list_add_tail(&entry->list, bitmaps); continue; } if (entry->bytes < min_bytes) continue; last = entry; window_free += entry->bytes; if (entry->bytes > max_extent) max_extent = entry->bytes; } if (window_free < bytes || max_extent < cont1_bytes) return -ENOSPC; cluster->window_start = first->offset; node = &first->offset_index; /* * now we've found our entries, pull them out of the free space * cache and put them into the cluster rbtree */ do { int ret; entry = rb_entry(node, struct btrfs_free_space, offset_index); node = rb_next(&entry->offset_index); if (entry->bitmap || entry->bytes < min_bytes) continue; rb_erase(&entry->offset_index, &ctl->free_space_offset); rb_erase_cached(&entry->bytes_index, &ctl->free_space_bytes); ret = tree_insert_offset(ctl, cluster, entry); total_size += entry->bytes; ASSERT(!ret); /* -EEXIST; Logic error */ } while (node && entry != last); cluster->max_size = max_extent; trace_btrfs_setup_cluster(block_group, cluster, total_size, 0); return 0; } /* * This specifically looks for bitmaps that may work in the cluster, we assume * that we have already failed to find extents that will work. */ static noinline int setup_cluster_bitmap(struct btrfs_block_group *block_group, struct btrfs_free_cluster *cluster, struct list_head *bitmaps, u64 offset, u64 bytes, u64 cont1_bytes, u64 min_bytes) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *entry = NULL; int ret = -ENOSPC; u64 bitmap_offset = offset_to_bitmap(ctl, offset); if (ctl->total_bitmaps == 0) return -ENOSPC; /* * The bitmap that covers offset won't be in the list unless offset * is just its start offset. */ if (!list_empty(bitmaps)) entry = list_first_entry(bitmaps, struct btrfs_free_space, list); if (!entry || entry->offset != bitmap_offset) { entry = tree_search_offset(ctl, bitmap_offset, 1, 0); if (entry && list_empty(&entry->list)) list_add(&entry->list, bitmaps); } list_for_each_entry(entry, bitmaps, list) { if (entry->bytes < bytes) continue; ret = btrfs_bitmap_cluster(block_group, entry, cluster, offset, bytes, cont1_bytes, min_bytes); if (!ret) return 0; } /* * The bitmaps list has all the bitmaps that record free space * starting after offset, so no more search is required. */ return -ENOSPC; } /* * here we try to find a cluster of blocks in a block group. The goal * is to find at least bytes+empty_size. * We might not find them all in one contiguous area. * * returns zero and sets up cluster if things worked out, otherwise * it returns -enospc */ int btrfs_find_space_cluster(struct btrfs_block_group *block_group, struct btrfs_free_cluster *cluster, u64 offset, u64 bytes, u64 empty_size) { struct btrfs_fs_info *fs_info = block_group->fs_info; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *entry, *tmp; LIST_HEAD(bitmaps); u64 min_bytes; u64 cont1_bytes; int ret; /* * Choose the minimum extent size we'll require for this * cluster. For SSD_SPREAD, don't allow any fragmentation. * For metadata, allow allocates with smaller extents. For * data, keep it dense. */ if (btrfs_test_opt(fs_info, SSD_SPREAD)) { cont1_bytes = bytes + empty_size; min_bytes = cont1_bytes; } else if (block_group->flags & BTRFS_BLOCK_GROUP_METADATA) { cont1_bytes = bytes; min_bytes = fs_info->sectorsize; } else { cont1_bytes = max(bytes, (bytes + empty_size) >> 2); min_bytes = fs_info->sectorsize; } spin_lock(&ctl->tree_lock); /* * If we know we don't have enough space to make a cluster don't even * bother doing all the work to try and find one. */ if (ctl->free_space < bytes) { spin_unlock(&ctl->tree_lock); return -ENOSPC; } spin_lock(&cluster->lock); /* someone already found a cluster, hooray */ if (cluster->block_group) { ret = 0; goto out; } trace_btrfs_find_cluster(block_group, offset, bytes, empty_size, min_bytes); ret = setup_cluster_no_bitmap(block_group, cluster, &bitmaps, offset, bytes + empty_size, cont1_bytes, min_bytes); if (ret) ret = setup_cluster_bitmap(block_group, cluster, &bitmaps, offset, bytes + empty_size, cont1_bytes, min_bytes); /* Clear our temporary list */ list_for_each_entry_safe(entry, tmp, &bitmaps, list) list_del_init(&entry->list); if (!ret) { btrfs_get_block_group(block_group); list_add_tail(&cluster->block_group_list, &block_group->cluster_list); cluster->block_group = block_group; } else { trace_btrfs_failed_cluster_setup(block_group); } out: spin_unlock(&cluster->lock); spin_unlock(&ctl->tree_lock); return ret; } /* * simple code to zero out a cluster */ void btrfs_init_free_cluster(struct btrfs_free_cluster *cluster) { spin_lock_init(&cluster->lock); spin_lock_init(&cluster->refill_lock); cluster->root = RB_ROOT; cluster->max_size = 0; cluster->fragmented = false; INIT_LIST_HEAD(&cluster->block_group_list); cluster->block_group = NULL; } static int do_trimming(struct btrfs_block_group *block_group, u64 *total_trimmed, u64 start, u64 bytes, u64 reserved_start, u64 reserved_bytes, enum btrfs_trim_state reserved_trim_state, struct btrfs_trim_range *trim_entry) { struct btrfs_space_info *space_info = block_group->space_info; struct btrfs_fs_info *fs_info = block_group->fs_info; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; int ret; bool bg_ro; const u64 end = start + bytes; const u64 reserved_end = reserved_start + reserved_bytes; enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED; u64 trimmed = 0; spin_lock(&space_info->lock); spin_lock(&block_group->lock); bg_ro = block_group->ro; if (!bg_ro) { block_group->reserved += reserved_bytes; spin_unlock(&block_group->lock); space_info->bytes_reserved += reserved_bytes; } else { spin_unlock(&block_group->lock); } spin_unlock(&space_info->lock); ret = btrfs_discard_extent(fs_info, start, bytes, &trimmed, false); if (!ret) { *total_trimmed += trimmed; trim_state = BTRFS_TRIM_STATE_TRIMMED; } mutex_lock(&ctl->cache_writeout_mutex); if (reserved_start < start) __btrfs_add_free_space(block_group, reserved_start, start - reserved_start, reserved_trim_state); if (end < reserved_end) __btrfs_add_free_space(block_group, end, reserved_end - end, reserved_trim_state); __btrfs_add_free_space(block_group, start, bytes, trim_state); list_del(&trim_entry->list); mutex_unlock(&ctl->cache_writeout_mutex); if (!bg_ro) { spin_lock(&space_info->lock); spin_lock(&block_group->lock); bg_ro = block_group->ro; block_group->reserved -= reserved_bytes; spin_unlock(&block_group->lock); space_info->bytes_reserved -= reserved_bytes; if (bg_ro) space_info->bytes_readonly += reserved_bytes; spin_unlock(&space_info->lock); } return ret; } /* * If @async is set, then we will trim 1 region and return. */ static int trim_no_bitmap(struct btrfs_block_group *block_group, u64 *total_trimmed, u64 start, u64 end, u64 minlen, bool async) { struct btrfs_discard_ctl *discard_ctl = &block_group->fs_info->discard_ctl; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *entry; struct rb_node *node; int ret = 0; u64 extent_start; u64 extent_bytes; enum btrfs_trim_state extent_trim_state; u64 bytes; const u64 max_discard_size = READ_ONCE(discard_ctl->max_discard_size); while (start < end) { struct btrfs_trim_range trim_entry; mutex_lock(&ctl->cache_writeout_mutex); spin_lock(&ctl->tree_lock); if (ctl->free_space < minlen) goto out_unlock; entry = tree_search_offset(ctl, start, 0, 1); if (!entry) goto out_unlock; /* Skip bitmaps and if async, already trimmed entries */ while (entry->bitmap || (async && btrfs_free_space_trimmed(entry))) { node = rb_next(&entry->offset_index); if (!node) goto out_unlock; entry = rb_entry(node, struct btrfs_free_space, offset_index); } if (entry->offset >= end) goto out_unlock; extent_start = entry->offset; extent_bytes = entry->bytes; extent_trim_state = entry->trim_state; if (async) { start = entry->offset; bytes = entry->bytes; if (bytes < minlen) { spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); goto next; } unlink_free_space(ctl, entry, true); /* * Let bytes = BTRFS_MAX_DISCARD_SIZE + X. * If X < BTRFS_ASYNC_DISCARD_MIN_FILTER, we won't trim * X when we come back around. So trim it now. */ if (max_discard_size && bytes >= (max_discard_size + BTRFS_ASYNC_DISCARD_MIN_FILTER)) { bytes = max_discard_size; extent_bytes = max_discard_size; entry->offset += max_discard_size; entry->bytes -= max_discard_size; link_free_space(ctl, entry); } else { kmem_cache_free(btrfs_free_space_cachep, entry); } } else { start = max(start, extent_start); bytes = min(extent_start + extent_bytes, end) - start; if (bytes < minlen) { spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); goto next; } unlink_free_space(ctl, entry, true); kmem_cache_free(btrfs_free_space_cachep, entry); } spin_unlock(&ctl->tree_lock); trim_entry.start = extent_start; trim_entry.bytes = extent_bytes; list_add_tail(&trim_entry.list, &ctl->trimming_ranges); mutex_unlock(&ctl->cache_writeout_mutex); ret = do_trimming(block_group, total_trimmed, start, bytes, extent_start, extent_bytes, extent_trim_state, &trim_entry); if (ret) { block_group->discard_cursor = start + bytes; break; } next: start += bytes; block_group->discard_cursor = start; if (async && *total_trimmed) break; if (btrfs_trim_interrupted()) { ret = -ERESTARTSYS; break; } cond_resched(); } return ret; out_unlock: block_group->discard_cursor = btrfs_block_group_end(block_group); spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); return ret; } void btrfs_trim_fully_remapped_block_group(struct btrfs_block_group *bg) { struct btrfs_fs_info *fs_info = bg->fs_info; struct btrfs_discard_ctl *discard_ctl = &fs_info->discard_ctl; int ret = 0; u64 bytes, trimmed; const u64 max_discard_size = READ_ONCE(discard_ctl->max_discard_size); u64 end = btrfs_block_group_end(bg); if (!test_bit(BLOCK_GROUP_FLAG_STRIPE_REMOVAL_PENDING, &bg->runtime_flags)) { bg->discard_cursor = end; if (bg->used == 0) { spin_lock(&fs_info->unused_bgs_lock); if (!list_empty(&bg->bg_list)) { list_del_init(&bg->bg_list); btrfs_put_block_group(bg); } spin_unlock(&fs_info->unused_bgs_lock); btrfs_mark_bg_unused(bg); } return; } bytes = end - bg->discard_cursor; if (max_discard_size && bytes >= (max_discard_size + BTRFS_ASYNC_DISCARD_MIN_FILTER)) bytes = max_discard_size; ret = btrfs_discard_extent(fs_info, bg->discard_cursor, bytes, &trimmed, false); if (ret) return; bg->discard_cursor += trimmed; if (bg->discard_cursor < end) return; btrfs_complete_bg_remapping(bg); } /* * If we break out of trimming a bitmap prematurely, we should reset the * trimming bit. In a rather contrived case, it's possible to race here so * reset the state to BTRFS_TRIM_STATE_UNTRIMMED. * * start = start of bitmap * end = near end of bitmap * * Thread 1: Thread 2: * trim_bitmaps(start) * trim_bitmaps(end) * end_trimming_bitmap() * reset_trimming_bitmap() */ static void reset_trimming_bitmap(struct btrfs_free_space_ctl *ctl, u64 offset) { struct btrfs_free_space *entry; spin_lock(&ctl->tree_lock); entry = tree_search_offset(ctl, offset, 1, 0); if (entry) { if (btrfs_free_space_trimmed(entry)) { ctl->discardable_extents[BTRFS_STAT_CURR] += entry->bitmap_extents; ctl->discardable_bytes[BTRFS_STAT_CURR] += entry->bytes; } entry->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; } spin_unlock(&ctl->tree_lock); } static void end_trimming_bitmap(struct btrfs_free_space_ctl *ctl, struct btrfs_free_space *entry) { if (btrfs_free_space_trimming_bitmap(entry)) { entry->trim_state = BTRFS_TRIM_STATE_TRIMMED; ctl->discardable_extents[BTRFS_STAT_CURR] -= entry->bitmap_extents; ctl->discardable_bytes[BTRFS_STAT_CURR] -= entry->bytes; } } /* * If @async is set, then we will trim 1 region and return. */ static int trim_bitmaps(struct btrfs_block_group *block_group, u64 *total_trimmed, u64 start, u64 end, u64 minlen, u64 maxlen, bool async) { struct btrfs_discard_ctl *discard_ctl = &block_group->fs_info->discard_ctl; struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; struct btrfs_free_space *entry; int ret = 0; int ret2; u64 bytes; u64 offset = offset_to_bitmap(ctl, start); const u64 max_discard_size = READ_ONCE(discard_ctl->max_discard_size); while (offset < end) { bool next_bitmap = false; struct btrfs_trim_range trim_entry; mutex_lock(&ctl->cache_writeout_mutex); spin_lock(&ctl->tree_lock); if (ctl->free_space < minlen) { block_group->discard_cursor = btrfs_block_group_end(block_group); spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); break; } entry = tree_search_offset(ctl, offset, 1, 0); /* * Bitmaps are marked trimmed lossily now to prevent constant * discarding of the same bitmap (the reason why we are bound * by the filters). So, retrim the block group bitmaps when we * are preparing to punt to the unused_bgs list. This uses * @minlen to determine if we are in BTRFS_DISCARD_INDEX_UNUSED * which is the only discard index which sets minlen to 0. */ if (!entry || (async && minlen && start == offset && btrfs_free_space_trimmed(entry))) { spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); next_bitmap = true; goto next; } /* * Async discard bitmap trimming begins at by setting the start * to be key.objectid and the offset_to_bitmap() aligns to the * start of the bitmap. This lets us know we are fully * scanning the bitmap rather than only some portion of it. */ if (start == offset) entry->trim_state = BTRFS_TRIM_STATE_TRIMMING; bytes = minlen; ret2 = search_bitmap(ctl, entry, &start, &bytes, false); if (ret2 || start >= end) { /* * We lossily consider a bitmap trimmed if we only skip * over regions <= BTRFS_ASYNC_DISCARD_MIN_FILTER. */ if (ret2 && minlen <= BTRFS_ASYNC_DISCARD_MIN_FILTER) end_trimming_bitmap(ctl, entry); else entry->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); next_bitmap = true; goto next; } /* * We already trimmed a region, but are using the locking above * to reset the trim_state. */ if (async && *total_trimmed) { spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); return ret; } bytes = min(bytes, end - start); if (bytes < minlen || (async && maxlen && bytes > maxlen)) { spin_unlock(&ctl->tree_lock); mutex_unlock(&ctl->cache_writeout_mutex); goto next; } /* * Let bytes = BTRFS_MAX_DISCARD_SIZE + X. * If X < @minlen, we won't trim X when we come back around. * So trim it now. We differ here from trimming extents as we * don't keep individual state per bit. */ if (async && max_discard_size && bytes > (max_discard_size + minlen)) bytes = max_discard_size; bitmap_clear_bits(ctl, entry, start, bytes, true); if (entry->bytes == 0) free_bitmap(ctl, entry); spin_unlock(&ctl->tree_lock); trim_entry.start = start; trim_entry.bytes = bytes; list_add_tail(&trim_entry.list, &ctl->trimming_ranges); mutex_unlock(&ctl->cache_writeout_mutex); ret = do_trimming(block_group, total_trimmed, start, bytes, start, bytes, 0, &trim_entry); if (ret) { reset_trimming_bitmap(ctl, offset); block_group->discard_cursor = btrfs_block_group_end(block_group); break; } next: if (next_bitmap) { offset += BITS_PER_BITMAP * ctl->unit; start = offset; } else { start += bytes; } block_group->discard_cursor = start; if (btrfs_trim_interrupted()) { if (start != offset) reset_trimming_bitmap(ctl, offset); ret = -ERESTARTSYS; break; } cond_resched(); } if (offset >= end) block_group->discard_cursor = end; return ret; } int btrfs_trim_block_group(struct btrfs_block_group *block_group, u64 *trimmed, u64 start, u64 end, u64 minlen) { struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; int ret; u64 rem = 0; ASSERT(!btrfs_is_zoned(block_group->fs_info)); *trimmed = 0; spin_lock(&block_group->lock); if (test_bit(BLOCK_GROUP_FLAG_REMOVED, &block_group->runtime_flags)) { spin_unlock(&block_group->lock); return 0; } btrfs_freeze_block_group(block_group); spin_unlock(&block_group->lock); ret = trim_no_bitmap(block_group, trimmed, start, end, minlen, false); if (ret) goto out; ret = trim_bitmaps(block_group, trimmed, start, end, minlen, 0, false); div64_u64_rem(end, BITS_PER_BITMAP * ctl->unit, &rem); /* If we ended in the middle of a bitmap, reset the trimming flag */ if (rem) reset_trimming_bitmap(ctl, offset_to_bitmap(ctl, end)); out: btrfs_unfreeze_block_group(block_group); return ret; } int btrfs_trim_block_group_extents(struct btrfs_block_group *block_group, u64 *trimmed, u64 start, u64 end, u64 minlen, bool async) { int ret; *trimmed = 0; spin_lock(&block_group->lock); if (test_bit(BLOCK_GROUP_FLAG_REMOVED, &block_group->runtime_flags)) { spin_unlock(&block_group->lock); return 0; } btrfs_freeze_block_group(block_group); spin_unlock(&block_group->lock); ret = trim_no_bitmap(block_group, trimmed, start, end, minlen, async); btrfs_unfreeze_block_group(block_group); return ret; } int btrfs_trim_block_group_bitmaps(struct btrfs_block_group *block_group, u64 *trimmed, u64 start, u64 end, u64 minlen, u64 maxlen, bool async) { int ret; *trimmed = 0; spin_lock(&block_group->lock); if (test_bit(BLOCK_GROUP_FLAG_REMOVED, &block_group->runtime_flags)) { spin_unlock(&block_group->lock); return 0; } btrfs_freeze_block_group(block_group); spin_unlock(&block_group->lock); ret = trim_bitmaps(block_group, trimmed, start, end, minlen, maxlen, async); btrfs_unfreeze_block_group(block_group); return ret; } bool btrfs_free_space_cache_v1_active(struct btrfs_fs_info *fs_info) { return btrfs_super_cache_generation(fs_info->super_copy); } static int cleanup_free_space_cache_v1(struct btrfs_fs_info *fs_info, struct btrfs_trans_handle *trans) { struct btrfs_block_group *block_group; struct rb_node *node; btrfs_info(fs_info, "cleaning free space cache v1"); node = rb_first_cached(&fs_info->block_group_cache_tree); while (node) { int ret; block_group = rb_entry(node, struct btrfs_block_group, cache_node); ret = btrfs_remove_free_space_inode(trans, NULL, block_group); if (ret) return ret; node = rb_next(node); } return 0; } int btrfs_set_free_space_cache_v1_active(struct btrfs_fs_info *fs_info, bool active) { struct btrfs_trans_handle *trans; int ret; /* * update_super_roots will appropriately set or unset * super_copy->cache_generation based on SPACE_CACHE and * BTRFS_FS_CLEANUP_SPACE_CACHE_V1. For this reason, we need a * transaction commit whether we are enabling space cache v1 and don't * have any other work to do, or are disabling it and removing free * space inodes. */ trans = btrfs_start_transaction(fs_info->tree_root, 0); if (IS_ERR(trans)) return PTR_ERR(trans); if (!active) { set_bit(BTRFS_FS_CLEANUP_SPACE_CACHE_V1, &fs_info->flags); ret = cleanup_free_space_cache_v1(fs_info, trans); if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); btrfs_end_transaction(trans); goto out; } } ret = btrfs_commit_transaction(trans); out: clear_bit(BTRFS_FS_CLEANUP_SPACE_CACHE_V1, &fs_info->flags); return ret; } int __init btrfs_free_space_init(void) { btrfs_free_space_cachep = KMEM_CACHE(btrfs_free_space, 0); if (!btrfs_free_space_cachep) return -ENOMEM; btrfs_free_space_bitmap_cachep = kmem_cache_create("btrfs_free_space_bitmap", PAGE_SIZE, PAGE_SIZE, 0, NULL); if (!btrfs_free_space_bitmap_cachep) { kmem_cache_destroy(btrfs_free_space_cachep); return -ENOMEM; } return 0; } void __cold btrfs_free_space_exit(void) { kmem_cache_destroy(btrfs_free_space_cachep); kmem_cache_destroy(btrfs_free_space_bitmap_cachep); } #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS /* * Use this if you need to make a bitmap or extent entry specifically, it * doesn't do any of the merging that add_free_space does, this acts a lot like * how the free space cache loading stuff works, so you can get really weird * configurations. */ int test_add_free_space_entry(struct btrfs_block_group *cache, u64 offset, u64 bytes, bool bitmap) { struct btrfs_free_space_ctl *ctl = cache->free_space_ctl; struct btrfs_free_space *info = NULL, *bitmap_info; void *map = NULL; enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_TRIMMED; u64 bytes_added; int ret; again: if (!info) { info = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS); if (!info) return -ENOMEM; } if (!bitmap) { spin_lock(&ctl->tree_lock); info->offset = offset; info->bytes = bytes; info->max_extent_size = 0; ret = link_free_space(ctl, info); spin_unlock(&ctl->tree_lock); if (ret) kmem_cache_free(btrfs_free_space_cachep, info); return ret; } if (!map) { map = kmem_cache_zalloc(btrfs_free_space_bitmap_cachep, GFP_NOFS); if (!map) { kmem_cache_free(btrfs_free_space_cachep, info); return -ENOMEM; } } spin_lock(&ctl->tree_lock); bitmap_info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), 1, 0); if (!bitmap_info) { info->bitmap = map; map = NULL; add_new_bitmap(ctl, info, offset); bitmap_info = info; info = NULL; } bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes, trim_state); bytes -= bytes_added; offset += bytes_added; spin_unlock(&ctl->tree_lock); if (bytes) goto again; if (info) kmem_cache_free(btrfs_free_space_cachep, info); if (map) kmem_cache_free(btrfs_free_space_bitmap_cachep, map); return 0; } /* * Checks to see if the given range is in the free space cache. This is really * just used to check the absence of space, so if there is free space in the * range at all we will return 1. */ int test_check_exists(struct btrfs_block_group *cache, u64 offset, u64 bytes) { struct btrfs_free_space_ctl *ctl = cache->free_space_ctl; struct btrfs_free_space *info; int ret = 0; spin_lock(&ctl->tree_lock); info = tree_search_offset(ctl, offset, 0, 0); if (!info) { info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), 1, 0); if (!info) goto out; } have_info: if (info->bitmap) { u64 bit_off, bit_bytes; struct rb_node *n; struct btrfs_free_space *tmp; bit_off = offset; bit_bytes = ctl->unit; ret = search_bitmap(ctl, info, &bit_off, &bit_bytes, false); if (!ret) { if (bit_off == offset) { ret = 1; goto out; } else if (bit_off > offset && offset + bytes > bit_off) { ret = 1; goto out; } } n = rb_prev(&info->offset_index); while (n) { tmp = rb_entry(n, struct btrfs_free_space, offset_index); if (tmp->offset + tmp->bytes < offset) break; if (offset + bytes < tmp->offset) { n = rb_prev(&tmp->offset_index); continue; } info = tmp; goto have_info; } n = rb_next(&info->offset_index); while (n) { tmp = rb_entry(n, struct btrfs_free_space, offset_index); if (offset + bytes < tmp->offset) break; if (tmp->offset + tmp->bytes < offset) { n = rb_next(&tmp->offset_index); continue; } info = tmp; goto have_info; } ret = 0; goto out; } if (info->offset == offset) { ret = 1; goto out; } if (offset > info->offset && offset < info->offset + info->bytes) ret = 1; out: spin_unlock(&ctl->tree_lock); return ret; } #endif /* CONFIG_BTRFS_FS_RUN_SANITY_TESTS */ |
| 8 8 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 | /* * Copyright © 2008 Intel Corporation * Copyright © 2016 Collabora Ltd * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), * to deal in the Software without restriction, including without limitation * the rights to use, copy, modify, merge, publish, distribute, sublicense, * and/or sell copies of the Software, and to permit persons to whom the * Software is furnished to do so, subject to the following conditions: * * The above copyright notice and this permission notice (including the next * paragraph) shall be included in all copies or substantial portions of the * Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS * IN THE SOFTWARE. * * Based on code from the i915 driver. * Original author: Damien Lespiau <damien.lespiau@intel.com> * */ #include <linux/circ_buf.h> #include <linux/ctype.h> #include <linux/debugfs.h> #include <linux/export.h> #include <linux/poll.h> #include <linux/uaccess.h> #include <drm/drm_crtc.h> #include <drm/drm_debugfs_crc.h> #include <drm/drm_drv.h> #include <drm/drm_print.h> #include "drm_internal.h" /** * DOC: CRC ABI * * DRM device drivers can provide to userspace CRC information of each frame as * it reached a given hardware component (a CRC sampling "source"). * * Userspace can control generation of CRCs in a given CRTC by writing to the * file dri/0/crtc-N/crc/control in debugfs, with N being the :ref:`index of * the CRTC<crtc_index>`. Accepted values are source names (which are * driver-specific) and the "auto" keyword, which will let the driver select a * default source of frame CRCs for this CRTC. * * Once frame CRC generation is enabled, userspace can capture them by reading * the dri/0/crtc-N/crc/data file. Each line in that file contains the frame * number in the first field and then a number of unsigned integer fields * containing the CRC data. Fields are separated by a single space and the number * of CRC fields is source-specific. * * Note that though in some cases the CRC is computed in a specified way and on * the frame contents as supplied by userspace (eDP 1.3), in general the CRC * computation is performed in an unspecified way and on frame contents that have * been already processed in also an unspecified way and thus userspace cannot * rely on being able to generate matching CRC values for the frame contents that * it submits. In this general case, the maximum userspace can do is to compare * the reported CRCs of frames that should have the same contents. * * On the driver side the implementation effort is minimal, drivers only need to * implement &drm_crtc_funcs.set_crc_source and &drm_crtc_funcs.verify_crc_source. * The debugfs files are automatically set up if those vfuncs are set. CRC samples * need to be captured in the driver by calling drm_crtc_add_crc_entry(). * Depending on the driver and HW requirements, &drm_crtc_funcs.set_crc_source * may result in a commit (even a full modeset). * * CRC results must be reliable across non-full-modeset atomic commits, so if a * commit via DRM_IOCTL_MODE_ATOMIC would disable or otherwise interfere with * CRC generation, then the driver must mark that commit as a full modeset * (drm_atomic_crtc_needs_modeset() should return true). As a result, to ensure * consistent results, generic userspace must re-setup CRC generation after a * legacy SETCRTC or an atomic commit with DRM_MODE_ATOMIC_ALLOW_MODESET. */ static int crc_control_show(struct seq_file *m, void *data) { struct drm_crtc *crtc = m->private; if (crtc->funcs->get_crc_sources) { size_t count; const char *const *sources = crtc->funcs->get_crc_sources(crtc, &count); size_t values_cnt; int i; if (count == 0 || !sources) goto out; for (i = 0; i < count; i++) if (!crtc->funcs->verify_crc_source(crtc, sources[i], &values_cnt)) { if (strcmp(sources[i], crtc->crc.source)) seq_printf(m, "%s\n", sources[i]); else seq_printf(m, "%s*\n", sources[i]); } } return 0; out: seq_printf(m, "%s*\n", crtc->crc.source); return 0; } static int crc_control_open(struct inode *inode, struct file *file) { struct drm_crtc *crtc = inode->i_private; return single_open(file, crc_control_show, crtc); } static ssize_t crc_control_write(struct file *file, const char __user *ubuf, size_t len, loff_t *offp) { struct seq_file *m = file->private_data; struct drm_crtc *crtc = m->private; struct drm_crtc_crc *crc = &crtc->crc; char *source; size_t values_cnt; int ret; if (len == 0) return 0; if (len > PAGE_SIZE - 1) { DRM_DEBUG_KMS("Expected < %lu bytes into crtc crc control\n", PAGE_SIZE); return -E2BIG; } source = memdup_user_nul(ubuf, len); if (IS_ERR(source)) return PTR_ERR(source); if (source[len - 1] == '\n') source[len - 1] = '\0'; ret = crtc->funcs->verify_crc_source(crtc, source, &values_cnt); if (ret) { kfree(source); return ret; } spin_lock_irq(&crc->lock); if (crc->opened) { spin_unlock_irq(&crc->lock); kfree(source); return -EBUSY; } kfree(crc->source); crc->source = source; spin_unlock_irq(&crc->lock); *offp += len; return len; } static const struct file_operations drm_crtc_crc_control_fops = { .owner = THIS_MODULE, .open = crc_control_open, .read = seq_read, .llseek = seq_lseek, .release = single_release, .write = crc_control_write }; static int crtc_crc_data_count(struct drm_crtc_crc *crc) { assert_spin_locked(&crc->lock); return CIRC_CNT(crc->head, crc->tail, DRM_CRC_ENTRIES_NR); } static void crtc_crc_cleanup(struct drm_crtc_crc *crc) { kfree(crc->entries); crc->overflow = false; crc->entries = NULL; crc->head = 0; crc->tail = 0; crc->values_cnt = 0; crc->opened = false; } static int crtc_crc_open(struct inode *inode, struct file *filep) { struct drm_crtc *crtc = inode->i_private; struct drm_crtc_crc *crc = &crtc->crc; struct drm_crtc_crc_entry *entries = NULL; size_t values_cnt; int ret = 0; if (drm_drv_uses_atomic_modeset(crtc->dev)) { ret = drm_modeset_lock_single_interruptible(&crtc->mutex); if (ret) return ret; if (!crtc->state->active) ret = -EIO; drm_modeset_unlock(&crtc->mutex); if (ret) return ret; } ret = crtc->funcs->verify_crc_source(crtc, crc->source, &values_cnt); if (ret) return ret; if (WARN_ON(values_cnt > DRM_MAX_CRC_NR)) return -EINVAL; if (WARN_ON(values_cnt == 0)) return -EINVAL; entries = kcalloc(DRM_CRC_ENTRIES_NR, sizeof(*entries), GFP_KERNEL); if (!entries) return -ENOMEM; spin_lock_irq(&crc->lock); if (!crc->opened) { crc->opened = true; crc->entries = entries; crc->values_cnt = values_cnt; } else { ret = -EBUSY; } spin_unlock_irq(&crc->lock); if (ret) { kfree(entries); return ret; } ret = crtc->funcs->set_crc_source(crtc, crc->source); if (ret) goto err; return 0; err: spin_lock_irq(&crc->lock); crtc_crc_cleanup(crc); spin_unlock_irq(&crc->lock); return ret; } static int crtc_crc_release(struct inode *inode, struct file *filep) { struct drm_crtc *crtc = filep->f_inode->i_private; struct drm_crtc_crc *crc = &crtc->crc; /* terminate the infinite while loop if 'drm_dp_aux_crc_work' running */ spin_lock_irq(&crc->lock); crc->opened = false; spin_unlock_irq(&crc->lock); crtc->funcs->set_crc_source(crtc, NULL); spin_lock_irq(&crc->lock); crtc_crc_cleanup(crc); spin_unlock_irq(&crc->lock); return 0; } /* * 1 frame field of 10 chars plus a number of CRC fields of 10 chars each, space * separated, with a newline at the end and null-terminated. */ #define LINE_LEN(values_cnt) (10 + 11 * values_cnt + 1 + 1) #define MAX_LINE_LEN (LINE_LEN(DRM_MAX_CRC_NR)) static ssize_t crtc_crc_read(struct file *filep, char __user *user_buf, size_t count, loff_t *pos) { struct drm_crtc *crtc = filep->f_inode->i_private; struct drm_crtc_crc *crc = &crtc->crc; struct drm_crtc_crc_entry *entry; char buf[MAX_LINE_LEN]; int ret, i; spin_lock_irq(&crc->lock); if (!crc->source) { spin_unlock_irq(&crc->lock); return 0; } /* Nothing to read? */ while (crtc_crc_data_count(crc) == 0) { if (filep->f_flags & O_NONBLOCK) { spin_unlock_irq(&crc->lock); return -EAGAIN; } ret = wait_event_interruptible_lock_irq(crc->wq, crtc_crc_data_count(crc), crc->lock); if (ret) { spin_unlock_irq(&crc->lock); return ret; } } /* We know we have an entry to be read */ entry = &crc->entries[crc->tail]; if (count < LINE_LEN(crc->values_cnt)) { spin_unlock_irq(&crc->lock); return -EINVAL; } BUILD_BUG_ON_NOT_POWER_OF_2(DRM_CRC_ENTRIES_NR); crc->tail = (crc->tail + 1) & (DRM_CRC_ENTRIES_NR - 1); spin_unlock_irq(&crc->lock); if (entry->has_frame_counter) sprintf(buf, "0x%08x", entry->frame); else sprintf(buf, "XXXXXXXXXX"); for (i = 0; i < crc->values_cnt; i++) sprintf(buf + 10 + i * 11, " 0x%08x", entry->crcs[i]); sprintf(buf + 10 + crc->values_cnt * 11, "\n"); if (copy_to_user(user_buf, buf, LINE_LEN(crc->values_cnt))) return -EFAULT; return LINE_LEN(crc->values_cnt); } static __poll_t crtc_crc_poll(struct file *file, poll_table *wait) { struct drm_crtc *crtc = file->f_inode->i_private; struct drm_crtc_crc *crc = &crtc->crc; __poll_t ret = 0; poll_wait(file, &crc->wq, wait); spin_lock_irq(&crc->lock); if (crc->source && crtc_crc_data_count(crc)) ret |= EPOLLIN | EPOLLRDNORM; spin_unlock_irq(&crc->lock); return ret; } static const struct file_operations drm_crtc_crc_data_fops = { .owner = THIS_MODULE, .open = crtc_crc_open, .read = crtc_crc_read, .poll = crtc_crc_poll, .release = crtc_crc_release, }; void drm_debugfs_crtc_crc_add(struct drm_crtc *crtc) { struct dentry *crc_ent; if (!crtc->funcs->set_crc_source || !crtc->funcs->verify_crc_source) return; crc_ent = debugfs_create_dir("crc", crtc->debugfs_entry); debugfs_create_file("control", S_IRUGO | S_IWUSR, crc_ent, crtc, &drm_crtc_crc_control_fops); debugfs_create_file("data", S_IRUGO, crc_ent, crtc, &drm_crtc_crc_data_fops); } /** * drm_crtc_add_crc_entry - Add entry with CRC information for a frame * @crtc: CRTC to which the frame belongs * @has_frame: whether this entry has a frame number to go with * @frame: number of the frame these CRCs are about * @crcs: array of CRC values, with length matching #drm_crtc_crc.values_cnt * * For each frame, the driver polls the source of CRCs for new data and calls * this function to add them to the buffer from where userspace reads. */ int drm_crtc_add_crc_entry(struct drm_crtc *crtc, bool has_frame, uint32_t frame, uint32_t *crcs) { struct drm_crtc_crc *crc = &crtc->crc; struct drm_crtc_crc_entry *entry; int head, tail; unsigned long flags; spin_lock_irqsave(&crc->lock, flags); /* Caller may not have noticed yet that userspace has stopped reading */ if (!crc->entries) { spin_unlock_irqrestore(&crc->lock, flags); return -EINVAL; } head = crc->head; tail = crc->tail; if (CIRC_SPACE(head, tail, DRM_CRC_ENTRIES_NR) < 1) { bool was_overflow = crc->overflow; crc->overflow = true; spin_unlock_irqrestore(&crc->lock, flags); if (!was_overflow) DRM_ERROR("Overflow of CRC buffer, userspace reads too slow.\n"); return -ENOBUFS; } entry = &crc->entries[head]; entry->frame = frame; entry->has_frame_counter = has_frame; memcpy(&entry->crcs, crcs, sizeof(*crcs) * crc->values_cnt); head = (head + 1) & (DRM_CRC_ENTRIES_NR - 1); crc->head = head; spin_unlock_irqrestore(&crc->lock, flags); wake_up_interruptible(&crc->wq); return 0; } EXPORT_SYMBOL_GPL(drm_crtc_add_crc_entry); |
| 71078 | 1 2 3 4 5 6 7 8 9 10 11 | /* SPDX-License-Identifier: GPL-2.0 */ #include <asm/processor.h> static inline int phys_addr_valid(resource_size_t addr) { #ifdef CONFIG_PHYS_ADDR_T_64BIT return !(addr >> boot_cpu_data.x86_phys_bits); #else return 1; #endif } |
| 32 587 706 32 32 706 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 | #ifndef _LINUX_GENERIC_RADIX_TREE_H #define _LINUX_GENERIC_RADIX_TREE_H /** * DOC: Generic radix trees/sparse arrays * * Very simple and minimalistic, supporting arbitrary size entries up to * GENRADIX_NODE_SIZE. * * A genradix is defined with the type it will store, like so: * * static GENRADIX(struct foo) foo_genradix; * * The main operations are: * * - genradix_init(radix) - initialize an empty genradix * * - genradix_free(radix) - free all memory owned by the genradix and * reinitialize it * * - genradix_ptr(radix, idx) - gets a pointer to the entry at idx, returning * NULL if that entry does not exist * * - genradix_ptr_alloc(radix, idx, gfp) - gets a pointer to an entry, * allocating it if necessary * * - genradix_for_each(radix, iter, p) - iterate over each entry in a genradix * * The radix tree allocates one page of entries at a time, so entries may exist * that were never explicitly allocated - they will be initialized to all * zeroes. * * Internally, a genradix is just a radix tree of pages, and indexing works in * terms of byte offsets. The wrappers in this header file use sizeof on the * type the radix contains to calculate a byte offset from the index - see * __idx_to_offset. */ #include <asm/page.h> #include <linux/bug.h> #include <linux/limits.h> #include <linux/log2.h> #include <linux/math.h> #include <linux/slab.h> #include <linux/types.h> struct genradix_root; #define GENRADIX_NODE_SHIFT 9 #define GENRADIX_NODE_SIZE (1U << GENRADIX_NODE_SHIFT) #define GENRADIX_ARY (GENRADIX_NODE_SIZE / sizeof(struct genradix_node *)) #define GENRADIX_ARY_SHIFT ilog2(GENRADIX_ARY) /* depth that's needed for a genradix that can address up to ULONG_MAX: */ #define GENRADIX_MAX_DEPTH \ DIV_ROUND_UP(BITS_PER_LONG - GENRADIX_NODE_SHIFT, GENRADIX_ARY_SHIFT) #define GENRADIX_DEPTH_MASK \ ((unsigned long) (roundup_pow_of_two(GENRADIX_MAX_DEPTH + 1) - 1)) static inline int genradix_depth_shift(unsigned depth) { return GENRADIX_NODE_SHIFT + GENRADIX_ARY_SHIFT * depth; } /* * Returns size (of data, in bytes) that a tree of a given depth holds: */ static inline size_t genradix_depth_size(unsigned depth) { return 1UL << genradix_depth_shift(depth); } static inline unsigned genradix_root_to_depth(struct genradix_root *r) { return (unsigned long) r & GENRADIX_DEPTH_MASK; } static inline struct genradix_node *genradix_root_to_node(struct genradix_root *r) { return (void *) ((unsigned long) r & ~GENRADIX_DEPTH_MASK); } struct __genradix { struct genradix_root *root; }; struct genradix_node { union { /* Interior node: */ struct genradix_node *children[GENRADIX_ARY]; /* Leaf: */ u8 data[GENRADIX_NODE_SIZE]; }; }; static inline struct genradix_node *genradix_alloc_node(gfp_t gfp_mask) { return kzalloc(GENRADIX_NODE_SIZE, gfp_mask); } static inline void genradix_free_node(struct genradix_node *node) { kfree(node); } /* * NOTE: currently, sizeof(_type) must not be larger than GENRADIX_NODE_SIZE: */ #define __GENRADIX_INITIALIZER \ { \ .tree = { \ .root = NULL, \ } \ } /* * We use a 0 size array to stash the type we're storing without taking any * space at runtime - then the various accessor macros can use typeof() to get * to it for casts/sizeof - we also force the alignment so that storing a type * with a ridiculous alignment doesn't blow up the alignment or size of the * genradix. */ #define GENRADIX(_type) \ struct { \ struct __genradix tree; \ _type type[0] __aligned(1); \ } #define DEFINE_GENRADIX(_name, _type) \ GENRADIX(_type) _name = __GENRADIX_INITIALIZER /** * genradix_init - initialize a genradix * @_radix: genradix to initialize * * Does not fail */ #define genradix_init(_radix) \ do { \ *(_radix) = (typeof(*_radix)) __GENRADIX_INITIALIZER; \ } while (0) void __genradix_free(struct __genradix *); /** * genradix_free: free all memory owned by a genradix * @_radix: the genradix to free * * After freeing, @_radix will be reinitialized and empty */ #define genradix_free(_radix) __genradix_free(&(_radix)->tree) static inline size_t __idx_to_offset(size_t idx, size_t obj_size) { if (__builtin_constant_p(obj_size)) BUILD_BUG_ON(obj_size > GENRADIX_NODE_SIZE); else BUG_ON(obj_size > GENRADIX_NODE_SIZE); if (!is_power_of_2(obj_size)) { size_t objs_per_page = GENRADIX_NODE_SIZE / obj_size; return (idx / objs_per_page) * GENRADIX_NODE_SIZE + (idx % objs_per_page) * obj_size; } else { return idx * obj_size; } } #define __genradix_cast(_radix) (typeof((_radix)->type[0]) *) #define __genradix_obj_size(_radix) sizeof((_radix)->type[0]) #define __genradix_objs_per_page(_radix) \ (GENRADIX_NODE_SIZE / sizeof((_radix)->type[0])) #define __genradix_page_remainder(_radix) \ (GENRADIX_NODE_SIZE % sizeof((_radix)->type[0])) #define __genradix_idx_to_offset(_radix, _idx) \ __idx_to_offset(_idx, __genradix_obj_size(_radix)) static inline void *__genradix_ptr_inlined(struct __genradix *radix, size_t offset) { struct genradix_root *r = READ_ONCE(radix->root); struct genradix_node *n = genradix_root_to_node(r); unsigned level = genradix_root_to_depth(r); unsigned shift = genradix_depth_shift(level); if (unlikely(ilog2(offset) >= genradix_depth_shift(level))) return NULL; while (n && shift > GENRADIX_NODE_SHIFT) { shift -= GENRADIX_ARY_SHIFT; n = n->children[offset >> shift]; offset &= (1UL << shift) - 1; } return n ? &n->data[offset] : NULL; } #define genradix_ptr_inlined(_radix, _idx) \ (__genradix_cast(_radix) \ __genradix_ptr_inlined(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx))) void *__genradix_ptr(struct __genradix *, size_t); /** * genradix_ptr - get a pointer to a genradix entry * @_radix: genradix to access * @_idx: index to fetch * * Returns a pointer to entry at @_idx, or NULL if that entry does not exist. */ #define genradix_ptr(_radix, _idx) \ (__genradix_cast(_radix) \ __genradix_ptr(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx))) void *__genradix_ptr_alloc(struct __genradix *, size_t, struct genradix_node **, gfp_t); #define genradix_ptr_alloc_inlined(_radix, _idx, _gfp) \ (__genradix_cast(_radix) \ (__genradix_ptr_inlined(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx)) ?: \ __genradix_ptr_alloc(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx), \ NULL, _gfp))) #define genradix_ptr_alloc_preallocated_inlined(_radix, _idx, _new_node, _gfp)\ (__genradix_cast(_radix) \ (__genradix_ptr_inlined(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx)) ?: \ __genradix_ptr_alloc(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx), \ _new_node, _gfp))) /** * genradix_ptr_alloc - get a pointer to a genradix entry, allocating it * if necessary * @_radix: genradix to access * @_idx: index to fetch * @_gfp: gfp mask * * Returns a pointer to entry at @_idx, or NULL on allocation failure */ #define genradix_ptr_alloc(_radix, _idx, _gfp) \ (__genradix_cast(_radix) \ __genradix_ptr_alloc(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx), \ NULL, _gfp)) #define genradix_ptr_alloc_preallocated(_radix, _idx, _new_node, _gfp)\ (__genradix_cast(_radix) \ __genradix_ptr_alloc(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx), \ _new_node, _gfp)) struct genradix_iter { size_t offset; size_t pos; }; /** * genradix_iter_init - initialize a genradix_iter * @_radix: genradix that will be iterated over * @_idx: index to start iterating from */ #define genradix_iter_init(_radix, _idx) \ ((struct genradix_iter) { \ .pos = (_idx), \ .offset = __genradix_idx_to_offset((_radix), (_idx)),\ }) void *__genradix_iter_peek(struct genradix_iter *, struct __genradix *, size_t); /** * genradix_iter_peek - get first entry at or above iterator's current * position * @_iter: a genradix_iter * @_radix: genradix being iterated over * * If no more entries exist at or above @_iter's current position, returns NULL */ #define genradix_iter_peek(_iter, _radix) \ (__genradix_cast(_radix) \ __genradix_iter_peek(_iter, &(_radix)->tree, \ __genradix_objs_per_page(_radix))) void *__genradix_iter_peek_prev(struct genradix_iter *, struct __genradix *, size_t, size_t); /** * genradix_iter_peek_prev - get first entry at or below iterator's current * position * @_iter: a genradix_iter * @_radix: genradix being iterated over * * If no more entries exist at or below @_iter's current position, returns NULL */ #define genradix_iter_peek_prev(_iter, _radix) \ (__genradix_cast(_radix) \ __genradix_iter_peek_prev(_iter, &(_radix)->tree, \ __genradix_objs_per_page(_radix), \ __genradix_obj_size(_radix) + \ __genradix_page_remainder(_radix))) static inline void __genradix_iter_advance(struct genradix_iter *iter, size_t obj_size) { if (iter->offset + obj_size < iter->offset) { iter->offset = SIZE_MAX; iter->pos = SIZE_MAX; return; } iter->offset += obj_size; if (!is_power_of_2(obj_size) && (iter->offset & (GENRADIX_NODE_SIZE - 1)) + obj_size > GENRADIX_NODE_SIZE) iter->offset = round_up(iter->offset, GENRADIX_NODE_SIZE); iter->pos++; } #define genradix_iter_advance(_iter, _radix) \ __genradix_iter_advance(_iter, __genradix_obj_size(_radix)) static inline void __genradix_iter_rewind(struct genradix_iter *iter, size_t obj_size) { if (iter->offset == 0 || iter->offset == SIZE_MAX) { iter->offset = SIZE_MAX; return; } if ((iter->offset & (GENRADIX_NODE_SIZE - 1)) == 0) iter->offset -= GENRADIX_NODE_SIZE % obj_size; iter->offset -= obj_size; iter->pos--; } #define genradix_iter_rewind(_iter, _radix) \ __genradix_iter_rewind(_iter, __genradix_obj_size(_radix)) #define genradix_for_each_from(_radix, _iter, _p, _start) \ for (_iter = genradix_iter_init(_radix, _start); \ (_p = genradix_iter_peek(&_iter, _radix)) != NULL; \ genradix_iter_advance(&_iter, _radix)) /** * genradix_for_each - iterate over entry in a genradix * @_radix: genradix to iterate over * @_iter: a genradix_iter to track current position * @_p: pointer to genradix entry type * * On every iteration, @_p will point to the current entry, and @_iter.pos * will be the current entry's index. */ #define genradix_for_each(_radix, _iter, _p) \ genradix_for_each_from(_radix, _iter, _p, 0) #define genradix_last_pos(_radix) \ (SIZE_MAX / GENRADIX_NODE_SIZE * __genradix_objs_per_page(_radix) - 1) /** * genradix_for_each_reverse - iterate over entry in a genradix, reverse order * @_radix: genradix to iterate over * @_iter: a genradix_iter to track current position * @_p: pointer to genradix entry type * * On every iteration, @_p will point to the current entry, and @_iter.pos * will be the current entry's index. */ #define genradix_for_each_reverse(_radix, _iter, _p) \ for (_iter = genradix_iter_init(_radix, genradix_last_pos(_radix));\ (_p = genradix_iter_peek_prev(&_iter, _radix)) != NULL;\ genradix_iter_rewind(&_iter, _radix)) int __genradix_prealloc(struct __genradix *, size_t, gfp_t); /** * genradix_prealloc - preallocate entries in a generic radix tree * @_radix: genradix to preallocate * @_nr: number of entries to preallocate * @_gfp: gfp mask * * Returns 0 on success, -ENOMEM on failure */ #define genradix_prealloc(_radix, _nr, _gfp) \ __genradix_prealloc(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _nr + 1),\ _gfp) #endif /* _LINUX_GENERIC_RADIX_TREE_H */ |
| 6 6 6 6 6 6 6 5 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 7 5 1 1 1 1 1 1 1 6 6 2 6 6 1 1 6 2 1 2 7 7 7 6 6 7 7 2 2 2 7 8 1 1 6 6 6 6 6 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 | // SPDX-License-Identifier: GPL-2.0-only /* * Driver for the VoIP USB phones with CM109 chipsets. * * Copyright (C) 2007 - 2008 Alfred E. Heggestad <aeh@db.org> */ /* * Tested devices: * - Komunikate KIP1000 * - Genius G-talk * - Allied-Telesis Corega USBPH01 * - ... * * This driver is based on the yealink.c driver * * Thanks to: * - Authors of yealink.c * - Thomas Reitmayr * - Oliver Neukum for good review comments and code * - Shaun Jackman <sjackman@gmail.com> for Genius G-talk keymap * - Dmitry Torokhov for valuable input and review * * Todo: * - Read/write EEPROM */ #include <linux/kernel.h> #include <linux/init.h> #include <linux/slab.h> #include <linux/module.h> #include <linux/moduleparam.h> #include <linux/rwsem.h> #include <linux/usb/input.h> #define DRIVER_VERSION "20080805" #define DRIVER_AUTHOR "Alfred E. Heggestad" #define DRIVER_DESC "CM109 phone driver" static char *phone = "kip1000"; module_param(phone, charp, S_IRUSR); MODULE_PARM_DESC(phone, "Phone name {kip1000, gtalk, usbph01, atcom}"); enum { /* HID Registers */ HID_IR0 = 0x00, /* Record/Playback-mute button, Volume up/down */ HID_IR1 = 0x01, /* GPI, generic registers or EEPROM_DATA0 */ HID_IR2 = 0x02, /* Generic registers or EEPROM_DATA1 */ HID_IR3 = 0x03, /* Generic registers or EEPROM_CTRL */ HID_OR0 = 0x00, /* Mapping control, buzzer, SPDIF (offset 0x04) */ HID_OR1 = 0x01, /* GPO - General Purpose Output */ HID_OR2 = 0x02, /* Set GPIO to input/output mode */ HID_OR3 = 0x03, /* SPDIF status channel or EEPROM_CTRL */ /* HID_IR0 */ RECORD_MUTE = 1 << 3, PLAYBACK_MUTE = 1 << 2, VOLUME_DOWN = 1 << 1, VOLUME_UP = 1 << 0, /* HID_OR0 */ /* bits 7-6 0: HID_OR1-2 are used for GPO; HID_OR0, 3 are used for buzzer and SPDIF 1: HID_OR0-3 are used as generic HID registers 2: Values written to HID_OR0-3 are also mapped to MCU_CTRL, EEPROM_DATA0-1, EEPROM_CTRL (see Note) 3: Reserved */ HID_OR_GPO_BUZ_SPDIF = 0 << 6, HID_OR_GENERIC_HID_REG = 1 << 6, HID_OR_MAP_MCU_EEPROM = 2 << 6, BUZZER_ON = 1 << 5, /* up to 256 normal keys, up to 15 special key combinations */ KEYMAP_SIZE = 256 + 15, }; /* CM109 protocol packet */ struct cm109_ctl_packet { u8 byte[4]; } __attribute__ ((packed)); enum { USB_PKT_LEN = sizeof(struct cm109_ctl_packet) }; /* CM109 device structure */ struct cm109_dev { struct input_dev *idev; /* input device */ struct usb_device *udev; /* usb device */ struct usb_interface *intf; /* irq input channel */ struct cm109_ctl_packet *irq_data; dma_addr_t irq_dma; struct urb *urb_irq; /* control output channel */ struct cm109_ctl_packet *ctl_data; dma_addr_t ctl_dma; struct usb_ctrlrequest *ctl_req; struct urb *urb_ctl; /* * The 3 bitfields below are protected by ctl_submit_lock. * They have to be separate since they are accessed from IRQ * context. */ unsigned irq_urb_pending:1; /* irq_urb is in flight */ unsigned ctl_urb_pending:1; /* ctl_urb is in flight */ unsigned buzzer_pending:1; /* need to issue buzz command */ spinlock_t ctl_submit_lock; unsigned char buzzer_state; /* on/off */ /* flags */ unsigned open:1; unsigned resetting:1; unsigned shutdown:1; /* This mutex protects writes to the above flags */ struct mutex pm_mutex; unsigned short keymap[KEYMAP_SIZE]; char phys[64]; /* physical device path */ int key_code; /* last reported key */ int keybit; /* 0=new scan 1,2,4,8=scan columns */ u8 gpi; /* Cached value of GPI (high nibble) */ }; /****************************************************************************** * CM109 key interface *****************************************************************************/ static unsigned short special_keymap(int code) { if (code > 0xff) { switch (code - 0xff) { case RECORD_MUTE: return KEY_MICMUTE; case PLAYBACK_MUTE: return KEY_MUTE; case VOLUME_DOWN: return KEY_VOLUMEDOWN; case VOLUME_UP: return KEY_VOLUMEUP; } } return KEY_RESERVED; } /* Map device buttons to internal key events. * * The "up" and "down" keys, are symbolised by arrows on the button. * The "pickup" and "hangup" keys are symbolised by a green and red phone * on the button. Komunikate KIP1000 Keyboard Matrix -> -- 1 -- 2 -- 3 --> GPI pin 4 (0x10) | | | | <- -- 4 -- 5 -- 6 --> GPI pin 5 (0x20) | | | | END - 7 -- 8 -- 9 --> GPI pin 6 (0x40) | | | | OK -- * -- 0 -- # --> GPI pin 7 (0x80) | | | | /|\ /|\ /|\ /|\ | | | | GPO pin: 3 2 1 0 0x8 0x4 0x2 0x1 */ static unsigned short keymap_kip1000(int scancode) { switch (scancode) { /* phone key: */ case 0x82: return KEY_NUMERIC_0; /* 0 */ case 0x14: return KEY_NUMERIC_1; /* 1 */ case 0x12: return KEY_NUMERIC_2; /* 2 */ case 0x11: return KEY_NUMERIC_3; /* 3 */ case 0x24: return KEY_NUMERIC_4; /* 4 */ case 0x22: return KEY_NUMERIC_5; /* 5 */ case 0x21: return KEY_NUMERIC_6; /* 6 */ case 0x44: return KEY_NUMERIC_7; /* 7 */ case 0x42: return KEY_NUMERIC_8; /* 8 */ case 0x41: return KEY_NUMERIC_9; /* 9 */ case 0x81: return KEY_NUMERIC_POUND; /* # */ case 0x84: return KEY_NUMERIC_STAR; /* * */ case 0x88: return KEY_ENTER; /* pickup */ case 0x48: return KEY_ESC; /* hangup */ case 0x28: return KEY_LEFT; /* IN */ case 0x18: return KEY_RIGHT; /* OUT */ default: return special_keymap(scancode); } } /* Contributed by Shaun Jackman <sjackman@gmail.com> Genius G-Talk keyboard matrix 0 1 2 3 4: 0 4 8 Talk 5: 1 5 9 End 6: 2 6 # Up 7: 3 7 * Down */ static unsigned short keymap_gtalk(int scancode) { switch (scancode) { case 0x11: return KEY_NUMERIC_0; case 0x21: return KEY_NUMERIC_1; case 0x41: return KEY_NUMERIC_2; case 0x81: return KEY_NUMERIC_3; case 0x12: return KEY_NUMERIC_4; case 0x22: return KEY_NUMERIC_5; case 0x42: return KEY_NUMERIC_6; case 0x82: return KEY_NUMERIC_7; case 0x14: return KEY_NUMERIC_8; case 0x24: return KEY_NUMERIC_9; case 0x44: return KEY_NUMERIC_POUND; /* # */ case 0x84: return KEY_NUMERIC_STAR; /* * */ case 0x18: return KEY_ENTER; /* Talk (green handset) */ case 0x28: return KEY_ESC; /* End (red handset) */ case 0x48: return KEY_UP; /* Menu up (rocker switch) */ case 0x88: return KEY_DOWN; /* Menu down (rocker switch) */ default: return special_keymap(scancode); } } /* * Keymap for Allied-Telesis Corega USBPH01 * http://www.alliedtelesis-corega.com/2/1344/1437/1360/chprd.html * * Contributed by july@nat.bg */ static unsigned short keymap_usbph01(int scancode) { switch (scancode) { case 0x11: return KEY_NUMERIC_0; /* 0 */ case 0x21: return KEY_NUMERIC_1; /* 1 */ case 0x41: return KEY_NUMERIC_2; /* 2 */ case 0x81: return KEY_NUMERIC_3; /* 3 */ case 0x12: return KEY_NUMERIC_4; /* 4 */ case 0x22: return KEY_NUMERIC_5; /* 5 */ case 0x42: return KEY_NUMERIC_6; /* 6 */ case 0x82: return KEY_NUMERIC_7; /* 7 */ case 0x14: return KEY_NUMERIC_8; /* 8 */ case 0x24: return KEY_NUMERIC_9; /* 9 */ case 0x44: return KEY_NUMERIC_POUND; /* # */ case 0x84: return KEY_NUMERIC_STAR; /* * */ case 0x18: return KEY_ENTER; /* pickup */ case 0x28: return KEY_ESC; /* hangup */ case 0x48: return KEY_LEFT; /* IN */ case 0x88: return KEY_RIGHT; /* OUT */ default: return special_keymap(scancode); } } /* * Keymap for ATCom AU-100 * http://www.atcom.cn/products.html * http://www.packetizer.com/products/au100/ * http://www.voip-info.org/wiki/view/AU-100 * * Contributed by daniel@gimpelevich.san-francisco.ca.us */ static unsigned short keymap_atcom(int scancode) { switch (scancode) { /* phone key: */ case 0x82: return KEY_NUMERIC_0; /* 0 */ case 0x11: return KEY_NUMERIC_1; /* 1 */ case 0x12: return KEY_NUMERIC_2; /* 2 */ case 0x14: return KEY_NUMERIC_3; /* 3 */ case 0x21: return KEY_NUMERIC_4; /* 4 */ case 0x22: return KEY_NUMERIC_5; /* 5 */ case 0x24: return KEY_NUMERIC_6; /* 6 */ case 0x41: return KEY_NUMERIC_7; /* 7 */ case 0x42: return KEY_NUMERIC_8; /* 8 */ case 0x44: return KEY_NUMERIC_9; /* 9 */ case 0x84: return KEY_NUMERIC_POUND; /* # */ case 0x81: return KEY_NUMERIC_STAR; /* * */ case 0x18: return KEY_ENTER; /* pickup */ case 0x28: return KEY_ESC; /* hangup */ case 0x48: return KEY_LEFT; /* left arrow */ case 0x88: return KEY_RIGHT; /* right arrow */ default: return special_keymap(scancode); } } static unsigned short (*keymap)(int) = keymap_kip1000; /* * Completes a request by converting the data into events for the * input subsystem. */ static void report_key(struct cm109_dev *dev, int key) { struct input_dev *idev = dev->idev; if (dev->key_code >= 0) { /* old key up */ input_report_key(idev, dev->key_code, 0); } dev->key_code = key; if (key >= 0) { /* new valid key */ input_report_key(idev, key, 1); } input_sync(idev); } /* * Converts data of special key presses (volume, mute) into events * for the input subsystem, sends press-n-release for mute keys. */ static void cm109_report_special(struct cm109_dev *dev) { static const u8 autorelease = RECORD_MUTE | PLAYBACK_MUTE; struct input_dev *idev = dev->idev; u8 data = dev->irq_data->byte[HID_IR0]; unsigned short keycode; int i; for (i = 0; i < 4; i++) { keycode = dev->keymap[0xff + BIT(i)]; if (keycode == KEY_RESERVED) continue; input_report_key(idev, keycode, data & BIT(i)); if (data & autorelease & BIT(i)) { input_sync(idev); input_report_key(idev, keycode, 0); } } input_sync(idev); } /****************************************************************************** * CM109 usb communication interface *****************************************************************************/ static void cm109_submit_buzz_toggle(struct cm109_dev *dev) { int error; if (dev->buzzer_state) dev->ctl_data->byte[HID_OR0] |= BUZZER_ON; else dev->ctl_data->byte[HID_OR0] &= ~BUZZER_ON; error = usb_submit_urb(dev->urb_ctl, GFP_ATOMIC); if (error) dev_err(&dev->intf->dev, "%s: usb_submit_urb (urb_ctl) failed %d\n", __func__, error); } static void cm109_submit_ctl(struct cm109_dev *dev) { int error; guard(spinlock_irqsave)(&dev->ctl_submit_lock); dev->irq_urb_pending = 0; if (unlikely(dev->shutdown)) return; if (dev->buzzer_state) dev->ctl_data->byte[HID_OR0] |= BUZZER_ON; else dev->ctl_data->byte[HID_OR0] &= ~BUZZER_ON; dev->ctl_data->byte[HID_OR1] = dev->keybit; dev->ctl_data->byte[HID_OR2] = dev->keybit; dev->buzzer_pending = 0; dev->ctl_urb_pending = 1; error = usb_submit_urb(dev->urb_ctl, GFP_ATOMIC); if (error) dev_err(&dev->intf->dev, "%s: usb_submit_urb (urb_ctl) failed %d\n", __func__, error); } /* * IRQ handler */ static void cm109_urb_irq_callback(struct urb *urb) { struct cm109_dev *dev = urb->context; const int status = urb->status; dev_dbg(&dev->intf->dev, "### URB IRQ: [0x%02x 0x%02x 0x%02x 0x%02x] keybit=0x%02x\n", dev->irq_data->byte[0], dev->irq_data->byte[1], dev->irq_data->byte[2], dev->irq_data->byte[3], dev->keybit); if (status) { if (status == -ESHUTDOWN) return; dev_err_ratelimited(&dev->intf->dev, "%s: urb status %d\n", __func__, status); goto out; } /* Special keys */ cm109_report_special(dev); /* Scan key column */ if (dev->keybit == 0xf) { /* Any changes ? */ if ((dev->gpi & 0xf0) == (dev->irq_data->byte[HID_IR1] & 0xf0)) goto out; dev->gpi = dev->irq_data->byte[HID_IR1] & 0xf0; dev->keybit = 0x1; } else { report_key(dev, dev->keymap[dev->irq_data->byte[HID_IR1]]); dev->keybit <<= 1; if (dev->keybit > 0x8) dev->keybit = 0xf; } out: cm109_submit_ctl(dev); } static void cm109_urb_ctl_callback(struct urb *urb) { struct cm109_dev *dev = urb->context; const int status = urb->status; int error; dev_dbg(&dev->intf->dev, "### URB CTL: [0x%02x 0x%02x 0x%02x 0x%02x]\n", dev->ctl_data->byte[0], dev->ctl_data->byte[1], dev->ctl_data->byte[2], dev->ctl_data->byte[3]); if (status) { if (status == -ESHUTDOWN) return; dev_err_ratelimited(&dev->intf->dev, "%s: urb status %d\n", __func__, status); } guard(spinlock_irqsave)(&dev->ctl_submit_lock); dev->ctl_urb_pending = 0; if (unlikely(dev->shutdown)) return; if (dev->buzzer_pending || status) { dev->buzzer_pending = 0; dev->ctl_urb_pending = 1; cm109_submit_buzz_toggle(dev); } else if (likely(!dev->irq_urb_pending)) { /* ask for key data */ dev->irq_urb_pending = 1; error = usb_submit_urb(dev->urb_irq, GFP_ATOMIC); if (error) dev_err(&dev->intf->dev, "%s: usb_submit_urb (urb_irq) failed %d\n", __func__, error); } } static void cm109_toggle_buzzer_async(struct cm109_dev *dev) { guard(spinlock_irqsave)(&dev->ctl_submit_lock); if (dev->ctl_urb_pending) { /* URB completion will resubmit */ dev->buzzer_pending = 1; } else { dev->ctl_urb_pending = 1; cm109_submit_buzz_toggle(dev); } } static void cm109_toggle_buzzer_sync(struct cm109_dev *dev, int on) { int error; if (on) dev->ctl_data->byte[HID_OR0] |= BUZZER_ON; else dev->ctl_data->byte[HID_OR0] &= ~BUZZER_ON; error = usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0), dev->ctl_req->bRequest, dev->ctl_req->bRequestType, le16_to_cpu(dev->ctl_req->wValue), le16_to_cpu(dev->ctl_req->wIndex), dev->ctl_data, USB_PKT_LEN, USB_CTRL_SET_TIMEOUT); if (error < 0 && error != -EINTR) dev_err(&dev->intf->dev, "%s: usb_control_msg() failed %d\n", __func__, error); } static void cm109_stop_traffic(struct cm109_dev *dev) { dev->shutdown = 1; /* * Make sure other CPUs see this */ smp_wmb(); usb_kill_urb(dev->urb_ctl); usb_kill_urb(dev->urb_irq); cm109_toggle_buzzer_sync(dev, 0); dev->shutdown = 0; smp_wmb(); } static void cm109_restore_state(struct cm109_dev *dev) { if (dev->open) { /* * Restore buzzer state. * This will also kick regular URB submission */ cm109_toggle_buzzer_async(dev); } } /****************************************************************************** * input event interface *****************************************************************************/ static int cm109_input_open(struct input_dev *idev) { struct cm109_dev *dev = input_get_drvdata(idev); int error; error = usb_autopm_get_interface(dev->intf); if (error < 0) { dev_err(&idev->dev, "%s - cannot autoresume, result %d\n", __func__, error); return error; } scoped_guard(mutex, &dev->pm_mutex) { dev->buzzer_state = 0; dev->key_code = -1; /* no keys pressed */ dev->keybit = 0xf; /* issue INIT */ dev->ctl_data->byte[HID_OR0] = HID_OR_GPO_BUZ_SPDIF; dev->ctl_data->byte[HID_OR1] = dev->keybit; dev->ctl_data->byte[HID_OR2] = dev->keybit; dev->ctl_data->byte[HID_OR3] = 0x00; dev->ctl_urb_pending = 1; error = usb_submit_urb(dev->urb_ctl, GFP_KERNEL); if (!error) { dev->open = 1; return 0; } } dev->ctl_urb_pending = 0; usb_autopm_put_interface(dev->intf); dev_err(&dev->intf->dev, "%s: usb_submit_urb (urb_ctl) failed %d\n", __func__, error); return error; } static void cm109_input_close(struct input_dev *idev) { struct cm109_dev *dev = input_get_drvdata(idev); scoped_guard(mutex, &dev->pm_mutex) { /* * Once we are here event delivery is stopped so we * don't need to worry about someone starting buzzer * again */ cm109_stop_traffic(dev); dev->open = 0; } usb_autopm_put_interface(dev->intf); } static int cm109_input_ev(struct input_dev *idev, unsigned int type, unsigned int code, int value) { struct cm109_dev *dev = input_get_drvdata(idev); dev_dbg(&dev->intf->dev, "input_ev: type=%u code=%u value=%d\n", type, code, value); if (type != EV_SND) return -EINVAL; switch (code) { case SND_TONE: case SND_BELL: dev->buzzer_state = !!value; if (!dev->resetting) cm109_toggle_buzzer_async(dev); return 0; default: return -EINVAL; } } /****************************************************************************** * Linux interface and usb initialisation *****************************************************************************/ struct driver_info { char *name; }; static const struct driver_info info_cm109 = { .name = "CM109 USB driver", }; enum { VENDOR_ID = 0x0d8c, /* C-Media Electronics */ PRODUCT_ID_CM109 = 0x000e, /* CM109 defines range 0x0008 - 0x000f */ }; /* table of devices that work with this driver */ static const struct usb_device_id cm109_usb_table[] = { { .match_flags = USB_DEVICE_ID_MATCH_DEVICE | USB_DEVICE_ID_MATCH_INT_INFO, .idVendor = VENDOR_ID, .idProduct = PRODUCT_ID_CM109, .bInterfaceClass = USB_CLASS_HID, .bInterfaceSubClass = 0, .bInterfaceProtocol = 0, .driver_info = (kernel_ulong_t) &info_cm109 }, /* you can add more devices here with product ID 0x0008 - 0x000f */ { } }; static void cm109_usb_cleanup(struct cm109_dev *dev) { kfree(dev->ctl_req); usb_free_coherent(dev->udev, USB_PKT_LEN, dev->ctl_data, dev->ctl_dma); usb_free_coherent(dev->udev, USB_PKT_LEN, dev->irq_data, dev->irq_dma); usb_free_urb(dev->urb_irq); /* parameter validation in core/urb */ usb_free_urb(dev->urb_ctl); /* parameter validation in core/urb */ kfree(dev); } static void cm109_usb_disconnect(struct usb_interface *interface) { struct cm109_dev *dev = usb_get_intfdata(interface); usb_set_intfdata(interface, NULL); input_unregister_device(dev->idev); cm109_usb_cleanup(dev); } static int cm109_usb_probe(struct usb_interface *intf, const struct usb_device_id *id) { struct usb_device *udev = interface_to_usbdev(intf); struct driver_info *nfo = (struct driver_info *)id->driver_info; struct usb_host_interface *interface; struct usb_endpoint_descriptor *endpoint; struct cm109_dev *dev; struct input_dev *input_dev = NULL; int ret, pipe, i; int error = -ENOMEM; interface = intf->cur_altsetting; if (interface->desc.bNumEndpoints < 1) return -ENODEV; endpoint = &interface->endpoint[0].desc; if (!usb_endpoint_is_int_in(endpoint)) return -ENODEV; dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return -ENOMEM; spin_lock_init(&dev->ctl_submit_lock); mutex_init(&dev->pm_mutex); dev->udev = udev; dev->intf = intf; dev->idev = input_dev = input_allocate_device(); if (!input_dev) goto err_out; /* allocate usb buffers */ dev->irq_data = usb_alloc_coherent(udev, USB_PKT_LEN, GFP_KERNEL, &dev->irq_dma); if (!dev->irq_data) goto err_out; dev->ctl_data = usb_alloc_coherent(udev, USB_PKT_LEN, GFP_KERNEL, &dev->ctl_dma); if (!dev->ctl_data) goto err_out; dev->ctl_req = kmalloc(sizeof(*(dev->ctl_req)), GFP_KERNEL); if (!dev->ctl_req) goto err_out; /* allocate urb structures */ dev->urb_irq = usb_alloc_urb(0, GFP_KERNEL); if (!dev->urb_irq) goto err_out; dev->urb_ctl = usb_alloc_urb(0, GFP_KERNEL); if (!dev->urb_ctl) goto err_out; /* get a handle to the interrupt data pipe */ pipe = usb_rcvintpipe(udev, endpoint->bEndpointAddress); ret = usb_maxpacket(udev, pipe); if (ret != USB_PKT_LEN) dev_err(&intf->dev, "invalid payload size %d, expected %d\n", ret, USB_PKT_LEN); /* initialise irq urb */ usb_fill_int_urb(dev->urb_irq, udev, pipe, dev->irq_data, USB_PKT_LEN, cm109_urb_irq_callback, dev, endpoint->bInterval); dev->urb_irq->transfer_dma = dev->irq_dma; dev->urb_irq->transfer_flags |= URB_NO_TRANSFER_DMA_MAP; dev->urb_irq->dev = udev; /* initialise ctl urb */ dev->ctl_req->bRequestType = USB_TYPE_CLASS | USB_RECIP_INTERFACE | USB_DIR_OUT; dev->ctl_req->bRequest = USB_REQ_SET_CONFIGURATION; dev->ctl_req->wValue = cpu_to_le16(0x200); dev->ctl_req->wIndex = cpu_to_le16(interface->desc.bInterfaceNumber); dev->ctl_req->wLength = cpu_to_le16(USB_PKT_LEN); usb_fill_control_urb(dev->urb_ctl, udev, usb_sndctrlpipe(udev, 0), (void *)dev->ctl_req, dev->ctl_data, USB_PKT_LEN, cm109_urb_ctl_callback, dev); dev->urb_ctl->transfer_dma = dev->ctl_dma; dev->urb_ctl->transfer_flags |= URB_NO_TRANSFER_DMA_MAP; dev->urb_ctl->dev = udev; /* find out the physical bus location */ usb_make_path(udev, dev->phys, sizeof(dev->phys)); strlcat(dev->phys, "/input0", sizeof(dev->phys)); /* register settings for the input device */ input_dev->name = nfo->name; input_dev->phys = dev->phys; usb_to_input_id(udev, &input_dev->id); input_dev->dev.parent = &intf->dev; input_set_drvdata(input_dev, dev); input_dev->open = cm109_input_open; input_dev->close = cm109_input_close; input_dev->event = cm109_input_ev; input_dev->keycode = dev->keymap; input_dev->keycodesize = sizeof(unsigned char); input_dev->keycodemax = ARRAY_SIZE(dev->keymap); input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_SND); input_dev->sndbit[0] = BIT_MASK(SND_BELL) | BIT_MASK(SND_TONE); /* register available key events */ for (i = 0; i < KEYMAP_SIZE; i++) { unsigned short k = keymap(i); dev->keymap[i] = k; __set_bit(k, input_dev->keybit); } __clear_bit(KEY_RESERVED, input_dev->keybit); error = input_register_device(dev->idev); if (error) goto err_out; usb_set_intfdata(intf, dev); return 0; err_out: input_free_device(input_dev); cm109_usb_cleanup(dev); return error; } static int cm109_usb_suspend(struct usb_interface *intf, pm_message_t message) { struct cm109_dev *dev = usb_get_intfdata(intf); dev_info(&intf->dev, "cm109: usb_suspend (event=%d)\n", message.event); guard(mutex)(&dev->pm_mutex); cm109_stop_traffic(dev); return 0; } static int cm109_usb_resume(struct usb_interface *intf) { struct cm109_dev *dev = usb_get_intfdata(intf); dev_info(&intf->dev, "cm109: usb_resume\n"); guard(mutex)(&dev->pm_mutex); cm109_restore_state(dev); return 0; } static int cm109_usb_pre_reset(struct usb_interface *intf) { struct cm109_dev *dev = usb_get_intfdata(intf); mutex_lock(&dev->pm_mutex); /* * Make sure input events don't try to toggle buzzer * while we are resetting */ dev->resetting = 1; smp_wmb(); cm109_stop_traffic(dev); return 0; } static int cm109_usb_post_reset(struct usb_interface *intf) { struct cm109_dev *dev = usb_get_intfdata(intf); dev->resetting = 0; smp_wmb(); cm109_restore_state(dev); mutex_unlock(&dev->pm_mutex); return 0; } static struct usb_driver cm109_driver = { .name = "cm109", .probe = cm109_usb_probe, .disconnect = cm109_usb_disconnect, .suspend = cm109_usb_suspend, .resume = cm109_usb_resume, .reset_resume = cm109_usb_resume, .pre_reset = cm109_usb_pre_reset, .post_reset = cm109_usb_post_reset, .id_table = cm109_usb_table, .supports_autosuspend = 1, }; static int __init cm109_select_keymap(void) { /* Load the phone keymap */ if (!strcasecmp(phone, "kip1000")) { keymap = keymap_kip1000; printk(KERN_INFO KBUILD_MODNAME ": " "Keymap for Komunikate KIP1000 phone loaded\n"); } else if (!strcasecmp(phone, "gtalk")) { keymap = keymap_gtalk; printk(KERN_INFO KBUILD_MODNAME ": " "Keymap for Genius G-talk phone loaded\n"); } else if (!strcasecmp(phone, "usbph01")) { keymap = keymap_usbph01; printk(KERN_INFO KBUILD_MODNAME ": " "Keymap for Allied-Telesis Corega USBPH01 phone loaded\n"); } else if (!strcasecmp(phone, "atcom")) { keymap = keymap_atcom; printk(KERN_INFO KBUILD_MODNAME ": " "Keymap for ATCom AU-100 phone loaded\n"); } else { printk(KERN_ERR KBUILD_MODNAME ": " "Unsupported phone: %s\n", phone); return -EINVAL; } return 0; } static int __init cm109_init(void) { int err; err = cm109_select_keymap(); if (err) return err; err = usb_register(&cm109_driver); if (err) return err; printk(KERN_INFO KBUILD_MODNAME ": " DRIVER_DESC ": " DRIVER_VERSION " (C) " DRIVER_AUTHOR "\n"); return 0; } static void __exit cm109_exit(void) { usb_deregister(&cm109_driver); } module_init(cm109_init); module_exit(cm109_exit); MODULE_DEVICE_TABLE(usb, cm109_usb_table); MODULE_AUTHOR(DRIVER_AUTHOR); MODULE_DESCRIPTION(DRIVER_DESC); MODULE_LICENSE("GPL"); |
| 42 43 40 13 14 13 5 4 9 6 3 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 | // SPDX-License-Identifier: GPL-2.0-only /* * AppArmor security module * * This file contains basic common functions used in AppArmor * * Copyright (C) 1998-2008 Novell/SUSE * Copyright 2009-2010 Canonical Ltd. */ #include <linux/ctype.h> #include <linux/mm.h> #include <linux/slab.h> #include <linux/string.h> #include <linux/vmalloc.h> #include "include/audit.h" #include "include/apparmor.h" #include "include/lib.h" #include "include/perms.h" #include "include/policy.h" struct aa_perms nullperms; struct aa_perms allperms = { .allow = ALL_PERMS_MASK, .quiet = ALL_PERMS_MASK, .hide = ALL_PERMS_MASK }; struct val_table_ent { const char *str; int value; }; static struct val_table_ent debug_values_table[] = { { "N", DEBUG_NONE }, { "none", DEBUG_NONE }, { "n", DEBUG_NONE }, { "0", DEBUG_NONE }, { "all", DEBUG_ALL }, { "Y", DEBUG_ALL }, { "y", DEBUG_ALL }, { "1", DEBUG_ALL }, { "abs_root", DEBUG_LABEL_ABS_ROOT }, { "label", DEBUG_LABEL }, { "domain", DEBUG_DOMAIN }, { "policy", DEBUG_POLICY }, { "interface", DEBUG_INTERFACE }, { "unpack", DEBUG_UNPACK }, { "tags", DEBUG_TAGS }, { NULL, 0 } }; static struct val_table_ent *val_table_find_ent(struct val_table_ent *table, const char *name, size_t len) { struct val_table_ent *entry; for (entry = table; entry->str != NULL; entry++) { if (strncmp(entry->str, name, len) == 0 && strlen(entry->str) == len) return entry; } return NULL; } int aa_parse_debug_params(const char *str) { struct val_table_ent *ent; const char *next; int val = 0; do { size_t n = strcspn(str, "\r\n,"); next = str + n; ent = val_table_find_ent(debug_values_table, str, next - str); if (ent) val |= ent->value; else AA_DEBUG(DEBUG_INTERFACE, "unknown debug type '%.*s'", (int)(next - str), str); str = next + 1; } while (*next != 0); return val; } /** * val_mask_to_str - convert a perm mask to its short string * @str: character buffer to store string in (at least 10 characters) * @size: size of the @str buffer * @table: NUL-terminated character buffer of permission characters (NOT NULL) * @mask: permission mask to convert */ static int val_mask_to_str(char *str, size_t size, const struct val_table_ent *table, u32 mask) { const struct val_table_ent *ent; int total = 0; for (ent = table; ent->str; ent++) { if (ent->value && (ent->value & mask) == ent->value) { int len = scnprintf(str, size, "%s%s", total ? "," : "", ent->str); size -= len; str += len; total += len; mask &= ~ent->value; } } return total; } int aa_print_debug_params(char *buffer) { if (!aa_g_debug) return sprintf(buffer, "N"); return val_mask_to_str(buffer, PAGE_SIZE, debug_values_table, aa_g_debug); } bool aa_resize_str_table(struct aa_str_table *t, int newsize, gfp_t gfp) { struct aa_str_table_ent *n; int i; if (t->size == newsize) return true; n = kcalloc(newsize, sizeof(*n), gfp); if (!n) return false; for (i = 0; i < min(t->size, newsize); i++) n[i] = t->table[i]; for (; i < t->size; i++) kfree_sensitive(t->table[i].strs); if (newsize > t->size) memset(&n[t->size], 0, (newsize-t->size)*sizeof(*n)); kfree_sensitive(t->table); t->table = n; t->size = newsize; return true; } /** * aa_destroy_str_table - free entries str table * @t: the string table to free (MAYBE NULL) */ void aa_destroy_str_table(struct aa_str_table *t) { int i; if (t) { if (!t->table) return; for (i = 0; i < t->size; i++) kfree_sensitive(t->table[i].strs); kfree_sensitive(t->table); t->table = NULL; t->size = 0; } } /** * skipn_spaces - Removes leading whitespace from @str. * @str: The string to be stripped. * @n: length of str to parse, will stop at \0 if encountered before n * * Returns a pointer to the first non-whitespace character in @str. * if all whitespace will return NULL */ const char *skipn_spaces(const char *str, size_t n) { for (; n && isspace(*str); --n) ++str; if (n) return (char *)str; return NULL; } const char *aa_splitn_fqname(const char *fqname, size_t n, const char **ns_name, size_t *ns_len) { const char *end = fqname + n; const char *name = skipn_spaces(fqname, n); *ns_name = NULL; *ns_len = 0; if (!name) return NULL; if (name[0] == ':') { char *split = strnchr(&name[1], end - &name[1], ':'); *ns_name = skipn_spaces(&name[1], end - &name[1]); if (!*ns_name) return NULL; if (split) { *ns_len = split - *ns_name; if (*ns_len == 0) *ns_name = NULL; split++; if (end - split > 1 && strncmp(split, "//", 2) == 0) split += 2; name = skipn_spaces(split, end - split); } else { /* a ns name without a following profile is allowed */ name = NULL; *ns_len = end - *ns_name; } } if (name && *name == 0) name = NULL; return name; } /** * aa_info_message - log a none profile related status message * @str: message to log */ void aa_info_message(const char *str) { if (audit_enabled) { DEFINE_AUDIT_DATA(ad, LSM_AUDIT_DATA_NONE, AA_CLASS_NONE, NULL); ad.info = str; aa_audit_msg(AUDIT_APPARMOR_STATUS, &ad, NULL); } printk(KERN_INFO "AppArmor: %s\n", str); } __counted char *aa_str_alloc(int size, gfp_t gfp) { struct counted_str *str; str = kmalloc(struct_size(str, name, size), gfp); if (!str) return NULL; kref_init(&str->count); return str->name; } void aa_str_kref(struct kref *kref) { kfree(container_of(kref, struct counted_str, count)); } const char aa_file_perm_chrs[] = "xwracd km l "; const char *aa_file_perm_names[] = { "exec", "write", "read", "append", "create", "delete", "open", "rename", "setattr", "getattr", "setcred", "getcred", "chmod", "chown", "chgrp", "lock", "mmap", "mprot", "link", "snapshot", "unknown", "unknown", "unknown", "unknown", "unknown", "unknown", "unknown", "unknown", "stack", "change_onexec", "change_profile", "change_hat", }; /** * aa_perm_mask_to_str - convert a perm mask to its short string * @str: character buffer to store string in (at least 10 characters) * @str_size: size of the @str buffer * @chrs: NUL-terminated character buffer of permission characters * @mask: permission mask to convert */ void aa_perm_mask_to_str(char *str, size_t str_size, const char *chrs, u32 mask) { unsigned int i, perm = 1; size_t num_chrs = strlen(chrs); for (i = 0; i < num_chrs; perm <<= 1, i++) { if (mask & perm) { /* Ensure that one byte is left for NUL-termination */ if (WARN_ON_ONCE(str_size <= 1)) break; *str++ = chrs[i]; str_size--; } } *str = '\0'; } void aa_audit_perm_names(struct audit_buffer *ab, const char * const *names, u32 mask) { const char *fmt = "%s"; unsigned int i, perm = 1; bool prev = false; for (i = 0; i < 32; perm <<= 1, i++) { if (mask & perm) { audit_log_format(ab, fmt, names[i]); if (!prev) { prev = true; fmt = " %s"; } } } } void aa_audit_perm_mask(struct audit_buffer *ab, u32 mask, const char *chrs, u32 chrsmask, const char * const *names, u32 namesmask) { char str[33]; audit_log_format(ab, "\""); if ((mask & chrsmask) && chrs) { aa_perm_mask_to_str(str, sizeof(str), chrs, mask & chrsmask); mask &= ~chrsmask; audit_log_format(ab, "%s", str); if (mask & namesmask) audit_log_format(ab, " "); } if ((mask & namesmask) && names) aa_audit_perm_names(ab, names, mask & namesmask); audit_log_format(ab, "\""); } /** * aa_apply_modes_to_perms - apply namespace and profile flags to perms * @profile: that perms where computed from * @perms: perms to apply mode modifiers to * * TODO: split into profile and ns based flags for when accumulating perms */ void aa_apply_modes_to_perms(struct aa_profile *profile, struct aa_perms *perms) { switch (AUDIT_MODE(profile)) { case AUDIT_ALL: perms->audit = ALL_PERMS_MASK; fallthrough; case AUDIT_NOQUIET: perms->quiet = 0; break; case AUDIT_QUIET: perms->audit = 0; fallthrough; case AUDIT_QUIET_DENIED: perms->quiet = ALL_PERMS_MASK; break; } if (KILL_MODE(profile)) perms->kill = ALL_PERMS_MASK; else if (COMPLAIN_MODE(profile)) perms->complain = ALL_PERMS_MASK; else if (USER_MODE(profile)) perms->prompt = ALL_PERMS_MASK; } void aa_profile_match_label(struct aa_profile *profile, struct aa_ruleset *rules, struct aa_label *label, int type, u32 request, struct aa_perms *perms) { /* TODO: doesn't yet handle extended types */ aa_state_t state; state = aa_dfa_next(rules->policy->dfa, rules->policy->start[AA_CLASS_LABEL], type); aa_label_match(profile, rules, label, state, false, request, perms); } /** * aa_check_perms - do audit mode selection based on perms set * @profile: profile being checked * @perms: perms computed for the request * @request: requested perms * @ad: initialized audit structure (MAY BE NULL if not auditing) * @cb: callback fn for type specific fields (MAY BE NULL) * * Returns: 0 if permission else error code * * Note: profile audit modes need to be set before calling by setting the * perm masks appropriately. * * If not auditing then complain mode is not enabled and the * error code will indicate whether there was an explicit deny * with a positive value. */ int aa_check_perms(struct aa_profile *profile, struct aa_perms *perms, u32 request, struct apparmor_audit_data *ad, void (*cb)(struct audit_buffer *, void *)) { int type, error; u32 denied = request & (~perms->allow | perms->deny); if (likely(!denied)) { /* mask off perms that are not being force audited */ request &= perms->audit; if (!request || !ad) return 0; type = AUDIT_APPARMOR_AUDIT; error = 0; } else { error = -EACCES; if (denied & perms->kill) type = AUDIT_APPARMOR_KILL; else if (denied == (denied & perms->complain)) type = AUDIT_APPARMOR_ALLOWED; else type = AUDIT_APPARMOR_DENIED; if (denied == (denied & perms->hide)) error = -ENOENT; denied &= ~perms->quiet; if (!ad || !denied) return error; } if (ad) { ad->subj_label = &profile->label; ad->request = request; ad->denied = denied; ad->error = error; aa_audit_msg(type, ad, cb); } if (type == AUDIT_APPARMOR_ALLOWED) error = 0; return error; } /** * aa_policy_init - initialize a policy structure * @policy: policy to initialize (NOT NULL) * @prefix: prefix name if any is required. (MAYBE NULL) * @name: name of the policy, init will make a copy of it (NOT NULL) * @gfp: allocation mode * * Note: this fn creates a copy of strings passed in * * Returns: true if policy init successful */ bool aa_policy_init(struct aa_policy *policy, const char *prefix, const char *name, gfp_t gfp) { char *hname; size_t hname_sz; hname_sz = (prefix ? strlen(prefix) + 2 : 0) + strlen(name) + 1; /* freed by policy_free */ hname = aa_str_alloc(hname_sz, gfp); if (!hname) return false; if (prefix) scnprintf(hname, hname_sz, "%s//%s", prefix, name); else strscpy(hname, name, hname_sz); policy->hname = hname; /* base.name is a substring of fqname */ policy->name = basename(policy->hname); INIT_LIST_HEAD(&policy->list); INIT_LIST_HEAD(&policy->profiles); return true; } /** * aa_policy_destroy - free the elements referenced by @policy * @policy: policy that is to have its elements freed (NOT NULL) */ void aa_policy_destroy(struct aa_policy *policy) { AA_BUG(on_list_rcu(&policy->profiles)); AA_BUG(on_list_rcu(&policy->list)); /* don't free name as its a subset of hname */ aa_put_str(policy->hname); } |
| 2 2 2 2 2 2 2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 | /* SPDX-License-Identifier: GPL-2.0-or-later */ /* * Glue Code for the AVX/AES-NI/GFNI assembler implementation of the ARIA Cipher * * Copyright (c) 2022 Taehee Yoo <ap420073@gmail.com> */ #include <crypto/algapi.h> #include <crypto/aria.h> #include <linux/crypto.h> #include <linux/err.h> #include <linux/export.h> #include <linux/module.h> #include <linux/types.h> #include "ecb_cbc_helpers.h" #include "aria-avx.h" asmlinkage void aria_aesni_avx_encrypt_16way(const void *ctx, u8 *dst, const u8 *src); EXPORT_SYMBOL_GPL(aria_aesni_avx_encrypt_16way); asmlinkage void aria_aesni_avx_decrypt_16way(const void *ctx, u8 *dst, const u8 *src); EXPORT_SYMBOL_GPL(aria_aesni_avx_decrypt_16way); asmlinkage void aria_aesni_avx_ctr_crypt_16way(const void *ctx, u8 *dst, const u8 *src, u8 *keystream, u8 *iv); EXPORT_SYMBOL_GPL(aria_aesni_avx_ctr_crypt_16way); asmlinkage void aria_aesni_avx_gfni_encrypt_16way(const void *ctx, u8 *dst, const u8 *src); EXPORT_SYMBOL_GPL(aria_aesni_avx_gfni_encrypt_16way); asmlinkage void aria_aesni_avx_gfni_decrypt_16way(const void *ctx, u8 *dst, const u8 *src); EXPORT_SYMBOL_GPL(aria_aesni_avx_gfni_decrypt_16way); asmlinkage void aria_aesni_avx_gfni_ctr_crypt_16way(const void *ctx, u8 *dst, const u8 *src, u8 *keystream, u8 *iv); EXPORT_SYMBOL_GPL(aria_aesni_avx_gfni_ctr_crypt_16way); static struct aria_avx_ops aria_ops; struct aria_avx_request_ctx { u8 keystream[ARIA_AESNI_PARALLEL_BLOCK_SIZE]; }; static int ecb_do_encrypt(struct skcipher_request *req, const u32 *rkey) { ECB_WALK_START(req, ARIA_BLOCK_SIZE, ARIA_AESNI_PARALLEL_BLOCKS); ECB_BLOCK(ARIA_AESNI_PARALLEL_BLOCKS, aria_ops.aria_encrypt_16way); ECB_BLOCK(1, aria_encrypt); ECB_WALK_END(); } static int ecb_do_decrypt(struct skcipher_request *req, const u32 *rkey) { ECB_WALK_START(req, ARIA_BLOCK_SIZE, ARIA_AESNI_PARALLEL_BLOCKS); ECB_BLOCK(ARIA_AESNI_PARALLEL_BLOCKS, aria_ops.aria_decrypt_16way); ECB_BLOCK(1, aria_decrypt); ECB_WALK_END(); } static int aria_avx_ecb_encrypt(struct skcipher_request *req) { struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); struct aria_ctx *ctx = crypto_skcipher_ctx(tfm); return ecb_do_encrypt(req, ctx->enc_key[0]); } static int aria_avx_ecb_decrypt(struct skcipher_request *req) { struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); struct aria_ctx *ctx = crypto_skcipher_ctx(tfm); return ecb_do_decrypt(req, ctx->dec_key[0]); } static int aria_avx_set_key(struct crypto_skcipher *tfm, const u8 *key, unsigned int keylen) { return aria_set_key(&tfm->base, key, keylen); } static int aria_avx_ctr_encrypt(struct skcipher_request *req) { struct aria_avx_request_ctx *req_ctx = skcipher_request_ctx(req); struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); struct aria_ctx *ctx = crypto_skcipher_ctx(tfm); struct skcipher_walk walk; unsigned int nbytes; int err; err = skcipher_walk_virt(&walk, req, false); while ((nbytes = walk.nbytes) > 0) { const u8 *src = walk.src.virt.addr; u8 *dst = walk.dst.virt.addr; while (nbytes >= ARIA_AESNI_PARALLEL_BLOCK_SIZE) { kernel_fpu_begin(); aria_ops.aria_ctr_crypt_16way(ctx, dst, src, &req_ctx->keystream[0], walk.iv); kernel_fpu_end(); dst += ARIA_AESNI_PARALLEL_BLOCK_SIZE; src += ARIA_AESNI_PARALLEL_BLOCK_SIZE; nbytes -= ARIA_AESNI_PARALLEL_BLOCK_SIZE; } while (nbytes >= ARIA_BLOCK_SIZE) { memcpy(&req_ctx->keystream[0], walk.iv, ARIA_BLOCK_SIZE); crypto_inc(walk.iv, ARIA_BLOCK_SIZE); aria_encrypt(ctx, &req_ctx->keystream[0], &req_ctx->keystream[0]); crypto_xor_cpy(dst, src, &req_ctx->keystream[0], ARIA_BLOCK_SIZE); dst += ARIA_BLOCK_SIZE; src += ARIA_BLOCK_SIZE; nbytes -= ARIA_BLOCK_SIZE; } if (walk.nbytes == walk.total && nbytes > 0) { memcpy(&req_ctx->keystream[0], walk.iv, ARIA_BLOCK_SIZE); crypto_inc(walk.iv, ARIA_BLOCK_SIZE); aria_encrypt(ctx, &req_ctx->keystream[0], &req_ctx->keystream[0]); crypto_xor_cpy(dst, src, &req_ctx->keystream[0], nbytes); dst += nbytes; src += nbytes; nbytes = 0; } err = skcipher_walk_done(&walk, nbytes); } return err; } static int aria_avx_init_tfm(struct crypto_skcipher *tfm) { crypto_skcipher_set_reqsize(tfm, sizeof(struct aria_avx_request_ctx)); return 0; } static struct skcipher_alg aria_algs[] = { { .base.cra_name = "ecb(aria)", .base.cra_driver_name = "ecb-aria-avx", .base.cra_priority = 400, .base.cra_blocksize = ARIA_BLOCK_SIZE, .base.cra_ctxsize = sizeof(struct aria_ctx), .base.cra_module = THIS_MODULE, .min_keysize = ARIA_MIN_KEY_SIZE, .max_keysize = ARIA_MAX_KEY_SIZE, .setkey = aria_avx_set_key, .encrypt = aria_avx_ecb_encrypt, .decrypt = aria_avx_ecb_decrypt, }, { .base.cra_name = "ctr(aria)", .base.cra_driver_name = "ctr-aria-avx", .base.cra_priority = 400, .base.cra_blocksize = 1, .base.cra_ctxsize = sizeof(struct aria_ctx), .base.cra_module = THIS_MODULE, .min_keysize = ARIA_MIN_KEY_SIZE, .max_keysize = ARIA_MAX_KEY_SIZE, .ivsize = ARIA_BLOCK_SIZE, .chunksize = ARIA_BLOCK_SIZE, .walksize = 16 * ARIA_BLOCK_SIZE, .setkey = aria_avx_set_key, .encrypt = aria_avx_ctr_encrypt, .decrypt = aria_avx_ctr_encrypt, .init = aria_avx_init_tfm, } }; static int __init aria_avx_init(void) { const char *feature_name; if (!boot_cpu_has(X86_FEATURE_AVX) || !boot_cpu_has(X86_FEATURE_AES) || !boot_cpu_has(X86_FEATURE_OSXSAVE)) { pr_info("AVX or AES-NI instructions are not detected.\n"); return -ENODEV; } if (!cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, &feature_name)) { pr_info("CPU feature '%s' is not supported.\n", feature_name); return -ENODEV; } if (boot_cpu_has(X86_FEATURE_GFNI)) { aria_ops.aria_encrypt_16way = aria_aesni_avx_gfni_encrypt_16way; aria_ops.aria_decrypt_16way = aria_aesni_avx_gfni_decrypt_16way; aria_ops.aria_ctr_crypt_16way = aria_aesni_avx_gfni_ctr_crypt_16way; } else { aria_ops.aria_encrypt_16way = aria_aesni_avx_encrypt_16way; aria_ops.aria_decrypt_16way = aria_aesni_avx_decrypt_16way; aria_ops.aria_ctr_crypt_16way = aria_aesni_avx_ctr_crypt_16way; } return crypto_register_skciphers(aria_algs, ARRAY_SIZE(aria_algs)); } static void __exit aria_avx_exit(void) { crypto_unregister_skciphers(aria_algs, ARRAY_SIZE(aria_algs)); } module_init(aria_avx_init); module_exit(aria_avx_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Taehee Yoo <ap420073@gmail.com>"); MODULE_DESCRIPTION("ARIA Cipher Algorithm, AVX/AES-NI/GFNI optimized"); MODULE_ALIAS_CRYPTO("aria"); MODULE_ALIAS_CRYPTO("aria-aesni-avx"); |
| 11393 9865 5311 209 229 4 4 9 2983 1424 1 263 25 1153 1155 487 867 2986 2983 2976 20 20 19 51 51 51 51 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 | // SPDX-License-Identifier: GPL-2.0 /* * Kernel internal schedule timeout and sleeping functions */ #include <linux/delay.h> #include <linux/jiffies.h> #include <linux/timer.h> #include <linux/sched/signal.h> #include <linux/sched/debug.h> #include "tick-internal.h" /* * Since schedule_timeout()'s timer is defined on the stack, it must store * the target task on the stack as well. */ struct process_timer { struct timer_list timer; struct task_struct *task; }; static void process_timeout(struct timer_list *t) { struct process_timer *timeout = timer_container_of(timeout, t, timer); wake_up_process(timeout->task); } /** * schedule_timeout - sleep until timeout * @timeout: timeout value in jiffies * * Make the current task sleep until @timeout jiffies have elapsed. * The function behavior depends on the current task state * (see also set_current_state() description): * * %TASK_RUNNING - the scheduler is called, but the task does not sleep * at all. That happens because sched_submit_work() does nothing for * tasks in %TASK_RUNNING state. * * %TASK_UNINTERRUPTIBLE - at least @timeout jiffies are guaranteed to * pass before the routine returns unless the current task is explicitly * woken up, (e.g. by wake_up_process()). * * %TASK_INTERRUPTIBLE - the routine may return early if a signal is * delivered to the current task or the current task is explicitly woken * up. * * The current task state is guaranteed to be %TASK_RUNNING when this * routine returns. * * Specifying a @timeout value of %MAX_SCHEDULE_TIMEOUT will schedule * the CPU away without a bound on the timeout. In this case the return * value will be %MAX_SCHEDULE_TIMEOUT. * * Returns: 0 when the timer has expired otherwise the remaining time in * jiffies will be returned. In all cases the return value is guaranteed * to be non-negative. */ signed long __sched schedule_timeout(signed long timeout) { struct process_timer timer; unsigned long expire; switch (timeout) { case MAX_SCHEDULE_TIMEOUT: /* * These two special cases are useful to be comfortable * in the caller. Nothing more. We could take * MAX_SCHEDULE_TIMEOUT from one of the negative value * but I' d like to return a valid offset (>=0) to allow * the caller to do everything it want with the retval. */ schedule(); goto out; default: /* * Another bit of PARANOID. Note that the retval will be * 0 since no piece of kernel is supposed to do a check * for a negative retval of schedule_timeout() (since it * should never happens anyway). You just have the printk() * that will tell you if something is gone wrong and where. */ if (timeout < 0) { pr_err("%s: wrong timeout value %lx\n", __func__, timeout); dump_stack(); __set_current_state(TASK_RUNNING); goto out; } } expire = timeout + jiffies; timer.task = current; timer_setup_on_stack(&timer.timer, process_timeout, 0); timer.timer.expires = expire; add_timer(&timer.timer); schedule(); timer_delete_sync(&timer.timer); /* Remove the timer from the object tracker */ timer_destroy_on_stack(&timer.timer); timeout = expire - jiffies; out: return timeout < 0 ? 0 : timeout; } EXPORT_SYMBOL(schedule_timeout); /* * __set_current_state() can be used in schedule_timeout_*() functions, because * schedule_timeout() calls schedule() unconditionally. */ /** * schedule_timeout_interruptible - sleep until timeout (interruptible) * @timeout: timeout value in jiffies * * See schedule_timeout() for details. * * Task state is set to TASK_INTERRUPTIBLE before starting the timeout. */ signed long __sched schedule_timeout_interruptible(signed long timeout) { __set_current_state(TASK_INTERRUPTIBLE); return schedule_timeout(timeout); } EXPORT_SYMBOL(schedule_timeout_interruptible); /** * schedule_timeout_killable - sleep until timeout (killable) * @timeout: timeout value in jiffies * * See schedule_timeout() for details. * * Task state is set to TASK_KILLABLE before starting the timeout. */ signed long __sched schedule_timeout_killable(signed long timeout) { __set_current_state(TASK_KILLABLE); return schedule_timeout(timeout); } EXPORT_SYMBOL(schedule_timeout_killable); /** * schedule_timeout_uninterruptible - sleep until timeout (uninterruptible) * @timeout: timeout value in jiffies * * See schedule_timeout() for details. * * Task state is set to TASK_UNINTERRUPTIBLE before starting the timeout. */ signed long __sched schedule_timeout_uninterruptible(signed long timeout) { __set_current_state(TASK_UNINTERRUPTIBLE); return schedule_timeout(timeout); } EXPORT_SYMBOL(schedule_timeout_uninterruptible); /** * schedule_timeout_idle - sleep until timeout (idle) * @timeout: timeout value in jiffies * * See schedule_timeout() for details. * * Task state is set to TASK_IDLE before starting the timeout. It is similar to * schedule_timeout_uninterruptible(), except this task will not contribute to * load average. */ signed long __sched schedule_timeout_idle(signed long timeout) { __set_current_state(TASK_IDLE); return schedule_timeout(timeout); } EXPORT_SYMBOL(schedule_timeout_idle); /** * schedule_hrtimeout_range_clock - sleep until timeout * @expires: timeout value (ktime_t) * @delta: slack in expires timeout (ktime_t) * @mode: timer mode * @clock_id: timer clock to be used * * Details are explained in schedule_hrtimeout_range() function description as * this function is commonly used. */ int __sched schedule_hrtimeout_range_clock(ktime_t *expires, u64 delta, const enum hrtimer_mode mode, clockid_t clock_id) { struct hrtimer_sleeper t; /* * Optimize when a zero timeout value is given. It does not * matter whether this is an absolute or a relative time. */ if (expires && *expires == 0) { __set_current_state(TASK_RUNNING); return 0; } /* * A NULL parameter means "infinite" */ if (!expires) { schedule(); return -EINTR; } hrtimer_setup_sleeper_on_stack(&t, clock_id, mode); hrtimer_set_expires_range_ns(&t.timer, *expires, delta); hrtimer_sleeper_start_expires(&t, mode); if (likely(t.task)) schedule(); hrtimer_cancel(&t.timer); destroy_hrtimer_on_stack(&t.timer); __set_current_state(TASK_RUNNING); return !t.task ? 0 : -EINTR; } EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock); /** * schedule_hrtimeout_range - sleep until timeout * @expires: timeout value (ktime_t) * @delta: slack in expires timeout (ktime_t) * @mode: timer mode * * Make the current task sleep until the given expiry time has * elapsed. The routine will return immediately unless * the current task state has been set (see set_current_state()). * * The @delta argument gives the kernel the freedom to schedule the * actual wakeup to a time that is both power and performance friendly * for regular (non RT/DL) tasks. * The kernel give the normal best effort behavior for "@expires+@delta", * but may decide to fire the timer earlier, but no earlier than @expires. * * You can set the task state as follows - * * %TASK_UNINTERRUPTIBLE - at least @timeout time is guaranteed to * pass before the routine returns unless the current task is explicitly * woken up, (e.g. by wake_up_process()). * * %TASK_INTERRUPTIBLE - the routine may return early if a signal is * delivered to the current task or the current task is explicitly woken * up. * * The current task state is guaranteed to be TASK_RUNNING when this * routine returns. * * Returns: 0 when the timer has expired. If the task was woken before the * timer expired by a signal (only possible in state TASK_INTERRUPTIBLE) or * by an explicit wakeup, it returns -EINTR. */ int __sched schedule_hrtimeout_range(ktime_t *expires, u64 delta, const enum hrtimer_mode mode) { return schedule_hrtimeout_range_clock(expires, delta, mode, CLOCK_MONOTONIC); } EXPORT_SYMBOL_GPL(schedule_hrtimeout_range); /** * schedule_hrtimeout - sleep until timeout * @expires: timeout value (ktime_t) * @mode: timer mode * * See schedule_hrtimeout_range() for details. @delta argument of * schedule_hrtimeout_range() is set to 0 and has therefore no impact. */ int __sched schedule_hrtimeout(ktime_t *expires, const enum hrtimer_mode mode) { return schedule_hrtimeout_range(expires, 0, mode); } EXPORT_SYMBOL_GPL(schedule_hrtimeout); /** * msleep - sleep safely even with waitqueue interruptions * @msecs: Requested sleep duration in milliseconds * * msleep() uses jiffy based timeouts for the sleep duration. Because of the * design of the timer wheel, the maximum additional percentage delay (slack) is * 12.5%. This is only valid for timers which will end up in level 1 or a higher * level of the timer wheel. For explanation of those 12.5% please check the * detailed description about the basics of the timer wheel. * * The slack of timers which will end up in level 0 depends on sleep duration * (msecs) and HZ configuration and can be calculated in the following way (with * the timer wheel design restriction that the slack is not less than 12.5%): * * ``slack = MSECS_PER_TICK / msecs`` * * When the allowed slack of the callsite is known, the calculation could be * turned around to find the minimal allowed sleep duration to meet the * constraints. For example: * * * ``HZ=1000`` with ``slack=25%``: ``MSECS_PER_TICK / slack = 1 / (1/4) = 4``: * all sleep durations greater or equal 4ms will meet the constraints. * * ``HZ=1000`` with ``slack=12.5%``: ``MSECS_PER_TICK / slack = 1 / (1/8) = 8``: * all sleep durations greater or equal 8ms will meet the constraints. * * ``HZ=250`` with ``slack=25%``: ``MSECS_PER_TICK / slack = 4 / (1/4) = 16``: * all sleep durations greater or equal 16ms will meet the constraints. * * ``HZ=250`` with ``slack=12.5%``: ``MSECS_PER_TICK / slack = 4 / (1/8) = 32``: * all sleep durations greater or equal 32ms will meet the constraints. * * See also the signal aware variant msleep_interruptible(). */ void msleep(unsigned int msecs) { unsigned long timeout = msecs_to_jiffies(msecs); while (timeout) timeout = schedule_timeout_uninterruptible(timeout); } EXPORT_SYMBOL(msleep); /** * msleep_interruptible - sleep waiting for signals * @msecs: Requested sleep duration in milliseconds * * See msleep() for some basic information. * * The difference between msleep() and msleep_interruptible() is that the sleep * could be interrupted by a signal delivery and then returns early. * * Returns: The remaining time of the sleep duration transformed to msecs (see * schedule_timeout() for details). */ unsigned long msleep_interruptible(unsigned int msecs) { unsigned long timeout = msecs_to_jiffies(msecs); while (timeout && !signal_pending(current)) timeout = schedule_timeout_interruptible(timeout); return jiffies_to_msecs(timeout); } EXPORT_SYMBOL(msleep_interruptible); /** * usleep_range_state - Sleep for an approximate time in a given state * @min: Minimum time in usecs to sleep * @max: Maximum time in usecs to sleep * @state: State of the current task that will be while sleeping * * usleep_range_state() sleeps at least for the minimum specified time but not * longer than the maximum specified amount of time. The range might reduce * power usage by allowing hrtimers to coalesce an already scheduled interrupt * with this hrtimer. In the worst case, an interrupt is scheduled for the upper * bound. * * The sleeping task is set to the specified state before starting the sleep. * * In non-atomic context where the exact wakeup time is flexible, use * usleep_range() or its variants instead of udelay(). The sleep improves * responsiveness by avoiding the CPU-hogging busy-wait of udelay(). */ void __sched usleep_range_state(unsigned long min, unsigned long max, unsigned int state) { ktime_t exp = ktime_add_us(ktime_get(), min); u64 delta = (u64)(max - min) * NSEC_PER_USEC; if (WARN_ON_ONCE(max < min)) delta = 0; for (;;) { __set_current_state(state); /* Do not return before the requested sleep time has elapsed */ if (!schedule_hrtimeout_range(&exp, delta, HRTIMER_MODE_ABS)) break; } } EXPORT_SYMBOL(usleep_range_state); |
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 | /* SPDX-License-Identifier: GPL-2.0 */ #ifndef _ASM_X86_LOCAL_H #define _ASM_X86_LOCAL_H #include <linux/percpu.h> #include <linux/atomic.h> #include <asm/asm.h> typedef struct { atomic_long_t a; } local_t; #define LOCAL_INIT(i) { ATOMIC_LONG_INIT(i) } #define local_read(l) atomic_long_read(&(l)->a) #define local_set(l, i) atomic_long_set(&(l)->a, (i)) static inline void local_inc(local_t *l) { asm volatile(_ASM_INC "%0" : "+m" (l->a.counter)); } static inline void local_dec(local_t *l) { asm volatile(_ASM_DEC "%0" : "+m" (l->a.counter)); } static inline void local_add(long i, local_t *l) { asm volatile(_ASM_ADD "%1,%0" : "+m" (l->a.counter) : "ir" (i)); } static inline void local_sub(long i, local_t *l) { asm volatile(_ASM_SUB "%1,%0" : "+m" (l->a.counter) : "ir" (i)); } /** * local_sub_and_test - subtract value from variable and test result * @i: integer value to subtract * @l: pointer to type local_t * * Atomically subtracts @i from @l and returns * true if the result is zero, or false for all * other cases. */ static inline bool local_sub_and_test(long i, local_t *l) { return GEN_BINARY_RMWcc(_ASM_SUB, l->a.counter, e, "er", i); } /** * local_dec_and_test - decrement and test * @l: pointer to type local_t * * Atomically decrements @l by 1 and * returns true if the result is 0, or false for all other * cases. */ static inline bool local_dec_and_test(local_t *l) { return GEN_UNARY_RMWcc(_ASM_DEC, l->a.counter, e); } /** * local_inc_and_test - increment and test * @l: pointer to type local_t * * Atomically increments @l by 1 * and returns true if the result is zero, or false for all * other cases. */ static inline bool local_inc_and_test(local_t *l) { return GEN_UNARY_RMWcc(_ASM_INC, l->a.counter, e); } /** * local_add_negative - add and test if negative * @i: integer value to add * @l: pointer to type local_t * * Atomically adds @i to @l and returns true * if the result is negative, or false when * result is greater than or equal to zero. */ static inline bool local_add_negative(long i, local_t *l) { return GEN_BINARY_RMWcc(_ASM_ADD, l->a.counter, s, "er", i); } /** * local_add_return - add and return * @i: integer value to add * @l: pointer to type local_t * * Atomically adds @i to @l and returns @i + @l */ static inline long local_add_return(long i, local_t *l) { long __i = i; asm volatile(_ASM_XADD "%0, %1;" : "+r" (i), "+m" (l->a.counter) : : "memory"); return i + __i; } static inline long local_sub_return(long i, local_t *l) { return local_add_return(-i, l); } #define local_inc_return(l) (local_add_return(1, l)) #define local_dec_return(l) (local_sub_return(1, l)) static inline long local_cmpxchg(local_t *l, long old, long new) { return cmpxchg_local(&l->a.counter, old, new); } static inline bool local_try_cmpxchg(local_t *l, long *old, long new) { return try_cmpxchg_local(&l->a.counter, (typeof(l->a.counter) *) old, new); } /* * Implement local_xchg using CMPXCHG instruction without the LOCK prefix. * XCHG is expensive due to the implied LOCK prefix. The processor * cannot prefetch cachelines if XCHG is used. */ static __always_inline long local_xchg(local_t *l, long n) { long c = local_read(l); do { } while (!local_try_cmpxchg(l, &c, n)); return c; } /** * local_add_unless - add unless the number is already a given value * @l: pointer of type local_t * @a: the amount to add to l... * @u: ...unless l is equal to u. * * Atomically adds @a to @l, if @v was not already @u. * Returns true if the addition was done. */ static __always_inline bool local_add_unless(local_t *l, long a, long u) { long c = local_read(l); do { if (unlikely(c == u)) return false; } while (!local_try_cmpxchg(l, &c, c + a)); return true; } #define local_inc_not_zero(l) local_add_unless((l), 1, 0) /* On x86_32, these are no better than the atomic variants. * On x86-64 these are better than the atomic variants on SMP kernels * because they dont use a lock prefix. */ #define __local_inc(l) local_inc(l) #define __local_dec(l) local_dec(l) #define __local_add(i, l) local_add((i), (l)) #define __local_sub(i, l) local_sub((i), (l)) #endif /* _ASM_X86_LOCAL_H */ |
| 20 8 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 | /* SPDX-License-Identifier: GPL-2.0 */ #ifndef _LINUX_BYTEORDER_GENERIC_H #define _LINUX_BYTEORDER_GENERIC_H /* * linux/byteorder/generic.h * Generic Byte-reordering support * * The "... p" macros, like le64_to_cpup, can be used with pointers * to unaligned data, but there will be a performance penalty on * some architectures. Use get_unaligned for unaligned data. * * Francois-Rene Rideau <fare@tunes.org> 19970707 * gathered all the good ideas from all asm-foo/byteorder.h into one file, * cleaned them up. * I hope it is compliant with non-GCC compilers. * I decided to put __BYTEORDER_HAS_U64__ in byteorder.h, * because I wasn't sure it would be ok to put it in types.h * Upgraded it to 2.1.43 * Francois-Rene Rideau <fare@tunes.org> 19971012 * Upgraded it to 2.1.57 * to please Linus T., replaced huge #ifdef's between little/big endian * by nestedly #include'd files. * Francois-Rene Rideau <fare@tunes.org> 19971205 * Made it to 2.1.71; now a facelift: * Put files under include/linux/byteorder/ * Split swab from generic support. * * TODO: * = Regular kernel maintainers could also replace all these manual * byteswap macros that remain, disseminated among drivers, * after some grep or the sources... * = Linus might want to rename all these macros and files to fit his taste, * to fit his personal naming scheme. * = it seems that a few drivers would also appreciate * nybble swapping support... * = every architecture could add their byteswap macro in asm/byteorder.h * see how some architectures already do (i386, alpha, ppc, etc) * = cpu_to_beXX and beXX_to_cpu might some day need to be well * distinguished throughout the kernel. This is not the case currently, * since little endian, big endian, and pdp endian machines needn't it. * But this might be the case for, say, a port of Linux to 20/21 bit * architectures (and F21 Linux addict around?). */ /* * The following macros are to be defined by <asm/byteorder.h>: * * Conversion of long and short int between network and host format * ntohl(__u32 x) * ntohs(__u16 x) * htonl(__u32 x) * htons(__u16 x) * It seems that some programs (which? where? or perhaps a standard? POSIX?) * might like the above to be functions, not macros (why?). * if that's true, then detect them, and take measures. * Anyway, the measure is: define only ___ntohl as a macro instead, * and in a separate file, have * unsigned long inline ntohl(x){return ___ntohl(x);} * * The same for constant arguments * __constant_ntohl(__u32 x) * __constant_ntohs(__u16 x) * __constant_htonl(__u32 x) * __constant_htons(__u16 x) * * Conversion of XX-bit integers (16- 32- or 64-) * between native CPU format and little/big endian format * 64-bit stuff only defined for proper architectures * cpu_to_[bl]eXX(__uXX x) * [bl]eXX_to_cpu(__uXX x) * * The same, but takes a pointer to the value to convert * cpu_to_[bl]eXXp(__uXX x) * [bl]eXX_to_cpup(__uXX x) * * The same, but change in situ * cpu_to_[bl]eXXs(__uXX x) * [bl]eXX_to_cpus(__uXX x) * * See asm-foo/byteorder.h for examples of how to provide * architecture-optimized versions * */ #define cpu_to_le64 __cpu_to_le64 #define le64_to_cpu __le64_to_cpu #define cpu_to_le32 __cpu_to_le32 #define le32_to_cpu __le32_to_cpu #define cpu_to_le16 __cpu_to_le16 #define le16_to_cpu __le16_to_cpu #define cpu_to_be64 __cpu_to_be64 #define be64_to_cpu __be64_to_cpu #define cpu_to_be32 __cpu_to_be32 #define be32_to_cpu __be32_to_cpu #define cpu_to_be16 __cpu_to_be16 #define be16_to_cpu __be16_to_cpu #define cpu_to_le64p __cpu_to_le64p #define le64_to_cpup __le64_to_cpup #define cpu_to_le32p __cpu_to_le32p #define le32_to_cpup __le32_to_cpup #define cpu_to_le16p __cpu_to_le16p #define le16_to_cpup __le16_to_cpup #define cpu_to_be64p __cpu_to_be64p #define be64_to_cpup __be64_to_cpup #define cpu_to_be32p __cpu_to_be32p #define be32_to_cpup __be32_to_cpup #define cpu_to_be16p __cpu_to_be16p #define be16_to_cpup __be16_to_cpup #define cpu_to_le64s __cpu_to_le64s #define le64_to_cpus __le64_to_cpus #define cpu_to_le32s __cpu_to_le32s #define le32_to_cpus __le32_to_cpus #define cpu_to_le16s __cpu_to_le16s #define le16_to_cpus __le16_to_cpus #define cpu_to_be64s __cpu_to_be64s #define be64_to_cpus __be64_to_cpus #define cpu_to_be32s __cpu_to_be32s #define be32_to_cpus __be32_to_cpus #define cpu_to_be16s __cpu_to_be16s #define be16_to_cpus __be16_to_cpus /* * They have to be macros in order to do the constant folding * correctly - if the argument passed into a inline function * it is no longer constant according to gcc.. */ #undef ntohl #undef ntohs #undef htonl #undef htons #define ___htonl(x) __cpu_to_be32(x) #define ___htons(x) __cpu_to_be16(x) #define ___ntohl(x) __be32_to_cpu(x) #define ___ntohs(x) __be16_to_cpu(x) #define htonl(x) ___htonl(x) #define ntohl(x) ___ntohl(x) #define htons(x) ___htons(x) #define ntohs(x) ___ntohs(x) static inline void le16_add_cpu(__le16 *var, u16 val) { *var = cpu_to_le16(le16_to_cpu(*var) + val); } static inline void le32_add_cpu(__le32 *var, u32 val) { *var = cpu_to_le32(le32_to_cpu(*var) + val); } static inline void le64_add_cpu(__le64 *var, u64 val) { *var = cpu_to_le64(le64_to_cpu(*var) + val); } /* XXX: this stuff can be optimized */ static inline void le32_to_cpu_array(u32 *buf, unsigned int words) { while (words--) { __le32_to_cpus(buf); buf++; } } static inline void cpu_to_le32_array(u32 *buf, unsigned int words) { while (words--) { __cpu_to_le32s(buf); buf++; } } static inline void le64_to_cpu_array(u64 *buf, unsigned int words) { while (words--) { __le64_to_cpus(buf); buf++; } } static inline void cpu_to_le64_array(u64 *buf, unsigned int words) { while (words--) { __cpu_to_le64s(buf); buf++; } } static inline void memcpy_from_le32(u32 *dst, const __le32 *src, size_t words) { size_t i; for (i = 0; i < words; i++) dst[i] = le32_to_cpu(src[i]); } static inline void memcpy_to_le32(__le32 *dst, const u32 *src, size_t words) { size_t i; for (i = 0; i < words; i++) dst[i] = cpu_to_le32(src[i]); } static inline void be16_add_cpu(__be16 *var, u16 val) { *var = cpu_to_be16(be16_to_cpu(*var) + val); } static inline void be32_add_cpu(__be32 *var, u32 val) { *var = cpu_to_be32(be32_to_cpu(*var) + val); } static inline void be64_add_cpu(__be64 *var, u64 val) { *var = cpu_to_be64(be64_to_cpu(*var) + val); } static inline void cpu_to_be32_array(__be32 *dst, const u32 *src, size_t len) { size_t i; for (i = 0; i < len; i++) dst[i] = cpu_to_be32(src[i]); } static inline void be32_to_cpu_array(u32 *dst, const __be32 *src, size_t len) { size_t i; for (i = 0; i < len; i++) dst[i] = be32_to_cpu(src[i]); } #endif /* _LINUX_BYTEORDER_GENERIC_H */ |
| 10 3 7 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 | // SPDX-License-Identifier: GPL-2.0-only #include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> #include <linux/netlink.h> #include <linux/netfilter.h> #include <linux/netfilter/nf_tables.h> #include <net/netfilter/nf_tables_core.h> #include <net/netfilter/nf_tables.h> #include <net/netfilter/nft_fib.h> static void nft_fib_inet_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) { const struct nft_fib *priv = nft_expr_priv(expr); switch (nft_pf(pkt)) { case NFPROTO_IPV4: switch (priv->result) { case NFT_FIB_RESULT_OIF: case NFT_FIB_RESULT_OIFNAME: return nft_fib4_eval(expr, regs, pkt); case NFT_FIB_RESULT_ADDRTYPE: return nft_fib4_eval_type(expr, regs, pkt); } break; case NFPROTO_IPV6: switch (priv->result) { case NFT_FIB_RESULT_OIF: case NFT_FIB_RESULT_OIFNAME: return nft_fib6_eval(expr, regs, pkt); case NFT_FIB_RESULT_ADDRTYPE: return nft_fib6_eval_type(expr, regs, pkt); } break; } regs->verdict.code = NF_DROP; } static struct nft_expr_type nft_fib_inet_type; static const struct nft_expr_ops nft_fib_inet_ops = { .type = &nft_fib_inet_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_fib)), .eval = nft_fib_inet_eval, .init = nft_fib_init, .dump = nft_fib_dump, .validate = nft_fib_validate, .reduce = nft_fib_reduce, }; static struct nft_expr_type nft_fib_inet_type __read_mostly = { .family = NFPROTO_INET, .name = "fib", .ops = &nft_fib_inet_ops, .policy = nft_fib_policy, .maxattr = NFTA_FIB_MAX, .owner = THIS_MODULE, }; static int __init nft_fib_inet_module_init(void) { return nft_register_expr(&nft_fib_inet_type); } static void __exit nft_fib_inet_module_exit(void) { nft_unregister_expr(&nft_fib_inet_type); } module_init(nft_fib_inet_module_init); module_exit(nft_fib_inet_module_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Florian Westphal <fw@strlen.de>"); MODULE_ALIAS_NFT_AF_EXPR(1, "fib"); MODULE_DESCRIPTION("nftables fib inet support"); |
| 28 27 27 10 10 11 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 | // SPDX-License-Identifier: GPL-2.0-or-later /* * xt_connmark - Netfilter module to operate on connection marks * * Copyright (C) 2002,2004 MARA Systems AB <https://www.marasystems.com> * by Henrik Nordstrom <hno@marasystems.com> * Copyright © CC Computer Consultants GmbH, 2007 - 2008 * Jan Engelhardt <jengelh@medozas.de> */ #include <linux/module.h> #include <linux/skbuff.h> #include <net/netfilter/nf_conntrack.h> #include <net/netfilter/nf_conntrack_ecache.h> #include <linux/netfilter/x_tables.h> #include <linux/netfilter/xt_connmark.h> MODULE_AUTHOR("Henrik Nordstrom <hno@marasystems.com>"); MODULE_DESCRIPTION("Xtables: connection mark operations"); MODULE_LICENSE("GPL"); MODULE_ALIAS("ipt_CONNMARK"); MODULE_ALIAS("ip6t_CONNMARK"); MODULE_ALIAS("ipt_connmark"); MODULE_ALIAS("ip6t_connmark"); static unsigned int connmark_tg_shift(struct sk_buff *skb, const struct xt_connmark_tginfo2 *info) { enum ip_conntrack_info ctinfo; u_int32_t new_targetmark; struct nf_conn *ct; u_int32_t newmark; u_int32_t oldmark; ct = nf_ct_get(skb, &ctinfo); if (ct == NULL) return XT_CONTINUE; switch (info->mode) { case XT_CONNMARK_SET: oldmark = READ_ONCE(ct->mark); newmark = (oldmark & ~info->ctmask) ^ info->ctmark; if (info->shift_dir == D_SHIFT_RIGHT) newmark >>= info->shift_bits; else newmark <<= info->shift_bits; if (READ_ONCE(ct->mark) != newmark) { WRITE_ONCE(ct->mark, newmark); nf_conntrack_event_cache(IPCT_MARK, ct); } break; case XT_CONNMARK_SAVE: new_targetmark = (skb->mark & info->nfmask); if (info->shift_dir == D_SHIFT_RIGHT) new_targetmark >>= info->shift_bits; else new_targetmark <<= info->shift_bits; newmark = (READ_ONCE(ct->mark) & ~info->ctmask) ^ new_targetmark; if (READ_ONCE(ct->mark) != newmark) { WRITE_ONCE(ct->mark, newmark); nf_conntrack_event_cache(IPCT_MARK, ct); } break; case XT_CONNMARK_RESTORE: new_targetmark = (READ_ONCE(ct->mark) & info->ctmask); if (info->shift_dir == D_SHIFT_RIGHT) new_targetmark >>= info->shift_bits; else new_targetmark <<= info->shift_bits; newmark = (skb->mark & ~info->nfmask) ^ new_targetmark; skb->mark = newmark; break; } return XT_CONTINUE; } static unsigned int connmark_tg(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_connmark_tginfo1 *info = par->targinfo; const struct xt_connmark_tginfo2 info2 = { .ctmark = info->ctmark, .ctmask = info->ctmask, .nfmask = info->nfmask, .mode = info->mode, }; return connmark_tg_shift(skb, &info2); } static unsigned int connmark_tg_v2(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_connmark_tginfo2 *info = par->targinfo; return connmark_tg_shift(skb, info); } static int connmark_tg_check(const struct xt_tgchk_param *par) { int ret; ret = nf_ct_netns_get(par->net, par->family); if (ret < 0) pr_info_ratelimited("cannot load conntrack support for proto=%u\n", par->family); return ret; } static void connmark_tg_destroy(const struct xt_tgdtor_param *par) { nf_ct_netns_put(par->net, par->family); } static bool connmark_mt(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_connmark_mtinfo1 *info = par->matchinfo; enum ip_conntrack_info ctinfo; const struct nf_conn *ct; ct = nf_ct_get(skb, &ctinfo); if (ct == NULL) return false; return ((READ_ONCE(ct->mark) & info->mask) == info->mark) ^ info->invert; } static int connmark_mt_check(const struct xt_mtchk_param *par) { int ret; ret = nf_ct_netns_get(par->net, par->family); if (ret < 0) pr_info_ratelimited("cannot load conntrack support for proto=%u\n", par->family); return ret; } static void connmark_mt_destroy(const struct xt_mtdtor_param *par) { nf_ct_netns_put(par->net, par->family); } static struct xt_target connmark_tg_reg[] __read_mostly = { { .name = "CONNMARK", .revision = 1, .family = NFPROTO_IPV4, .checkentry = connmark_tg_check, .target = connmark_tg, .targetsize = sizeof(struct xt_connmark_tginfo1), .destroy = connmark_tg_destroy, .me = THIS_MODULE, }, { .name = "CONNMARK", .revision = 2, .family = NFPROTO_IPV4, .checkentry = connmark_tg_check, .target = connmark_tg_v2, .targetsize = sizeof(struct xt_connmark_tginfo2), .destroy = connmark_tg_destroy, .me = THIS_MODULE, }, #if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) { .name = "CONNMARK", .revision = 1, .family = NFPROTO_IPV6, .checkentry = connmark_tg_check, .target = connmark_tg, .targetsize = sizeof(struct xt_connmark_tginfo1), .destroy = connmark_tg_destroy, .me = THIS_MODULE, }, { .name = "CONNMARK", .revision = 2, .family = NFPROTO_IPV6, .checkentry = connmark_tg_check, .target = connmark_tg_v2, .targetsize = sizeof(struct xt_connmark_tginfo2), .destroy = connmark_tg_destroy, .me = THIS_MODULE, }, #endif }; static struct xt_match connmark_mt_reg __read_mostly = { .name = "connmark", .revision = 1, .family = NFPROTO_UNSPEC, .checkentry = connmark_mt_check, .match = connmark_mt, .matchsize = sizeof(struct xt_connmark_mtinfo1), .destroy = connmark_mt_destroy, .me = THIS_MODULE, }; static int __init connmark_mt_init(void) { int ret; ret = xt_register_targets(connmark_tg_reg, ARRAY_SIZE(connmark_tg_reg)); if (ret < 0) return ret; ret = xt_register_match(&connmark_mt_reg); if (ret < 0) { xt_unregister_targets(connmark_tg_reg, ARRAY_SIZE(connmark_tg_reg)); return ret; } return 0; } static void __exit connmark_mt_exit(void) { xt_unregister_match(&connmark_mt_reg); xt_unregister_targets(connmark_tg_reg, ARRAY_SIZE(connmark_tg_reg)); } module_init(connmark_mt_init); module_exit(connmark_mt_exit); |
| 7 7 7 7 7 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 | // SPDX-License-Identifier: GPL-2.0 #include <linux/highmem.h> #include <linux/module.h> #include <linux/security.h> #include <linux/slab.h> #include <linux/types.h> #include "sysfs.h" /* * sysfs support for firmware loader */ void __fw_load_abort(struct fw_priv *fw_priv) { /* * There is a small window in which user can write to 'loading' * between loading done/aborted and disappearance of 'loading' */ if (fw_state_is_aborted(fw_priv) || fw_state_is_done(fw_priv)) return; fw_state_aborted(fw_priv); } #ifdef CONFIG_FW_LOADER_USER_HELPER static ssize_t timeout_show(const struct class *class, const struct class_attribute *attr, char *buf) { return sysfs_emit(buf, "%d\n", __firmware_loading_timeout()); } /** * timeout_store() - set number of seconds to wait for firmware * @class: device class pointer * @attr: device attribute pointer * @buf: buffer to scan for timeout value * @count: number of bytes in @buf * * Sets the number of seconds to wait for the firmware. Once * this expires an error will be returned to the driver and no * firmware will be provided. * * Note: zero means 'wait forever'. **/ static ssize_t timeout_store(const struct class *class, const struct class_attribute *attr, const char *buf, size_t count) { int tmp_loading_timeout; if (kstrtoint(buf, 10, &tmp_loading_timeout)) return -EINVAL; if (tmp_loading_timeout < 0) tmp_loading_timeout = 0; __fw_fallback_set_timeout(tmp_loading_timeout); return count; } static CLASS_ATTR_RW(timeout); static struct attribute *firmware_class_attrs[] = { &class_attr_timeout.attr, NULL, }; ATTRIBUTE_GROUPS(firmware_class); static int do_firmware_uevent(const struct fw_sysfs *fw_sysfs, struct kobj_uevent_env *env) { if (add_uevent_var(env, "FIRMWARE=%s", fw_sysfs->fw_priv->fw_name)) return -ENOMEM; if (add_uevent_var(env, "TIMEOUT=%i", __firmware_loading_timeout())) return -ENOMEM; if (add_uevent_var(env, "ASYNC=%d", fw_sysfs->nowait)) return -ENOMEM; return 0; } static int firmware_uevent(const struct device *dev, struct kobj_uevent_env *env) { const struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); int err = 0; mutex_lock(&fw_lock); if (fw_sysfs->fw_priv) err = do_firmware_uevent(fw_sysfs, env); mutex_unlock(&fw_lock); return err; } #endif /* CONFIG_FW_LOADER_USER_HELPER */ static void fw_dev_release(struct device *dev) { struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); if (fw_sysfs->fw_upload_priv) fw_upload_free(fw_sysfs); kfree(fw_sysfs); } static struct class firmware_class = { .name = "firmware", #ifdef CONFIG_FW_LOADER_USER_HELPER .class_groups = firmware_class_groups, .dev_uevent = firmware_uevent, #endif .dev_release = fw_dev_release, }; int register_sysfs_loader(void) { int ret = class_register(&firmware_class); if (ret != 0) return ret; return register_firmware_config_sysctl(); } void unregister_sysfs_loader(void) { unregister_firmware_config_sysctl(); class_unregister(&firmware_class); } static ssize_t firmware_loading_show(struct device *dev, struct device_attribute *attr, char *buf) { struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); int loading = 0; mutex_lock(&fw_lock); if (fw_sysfs->fw_priv) loading = fw_state_is_loading(fw_sysfs->fw_priv); mutex_unlock(&fw_lock); return sysfs_emit(buf, "%d\n", loading); } /** * firmware_loading_store() - set value in the 'loading' control file * @dev: device pointer * @attr: device attribute pointer * @buf: buffer to scan for loading control value * @count: number of bytes in @buf * * The relevant values are: * * 1: Start a load, discarding any previous partial load. * 0: Conclude the load and hand the data to the driver code. * -1: Conclude the load with an error and discard any written data. **/ static ssize_t firmware_loading_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); struct fw_priv *fw_priv; ssize_t written = count; int loading; if (kstrtoint(buf, 10, &loading)) return -EINVAL; mutex_lock(&fw_lock); fw_priv = fw_sysfs->fw_priv; if (fw_state_is_aborted(fw_priv) || fw_state_is_done(fw_priv)) goto out; switch (loading) { case 1: /* discarding any previous partial load */ fw_free_paged_buf(fw_priv); fw_state_start(fw_priv); break; case 0: if (fw_state_is_loading(fw_priv)) { int rc; /* * Several loading requests may be pending on * one same firmware buf, so let all requests * see the mapped 'buf->data' once the loading * is completed. */ rc = fw_map_paged_buf(fw_priv); if (rc) dev_err(dev, "%s: map pages failed\n", __func__); else rc = security_kernel_post_load_data(fw_priv->data, fw_priv->size, LOADING_FIRMWARE, "blob"); /* * Same logic as fw_load_abort, only the DONE bit * is ignored and we set ABORT only on failure. */ if (rc) { fw_state_aborted(fw_priv); written = rc; } else { fw_state_done(fw_priv); /* * If this is a user-initiated firmware upload * then start the upload in a worker thread now. */ rc = fw_upload_start(fw_sysfs); if (rc) written = rc; } break; } fallthrough; default: dev_err(dev, "%s: unexpected value (%d)\n", __func__, loading); fallthrough; case -1: fw_load_abort(fw_sysfs); if (fw_sysfs->fw_upload_priv) fw_state_init(fw_sysfs->fw_priv); break; } out: mutex_unlock(&fw_lock); return written; } DEVICE_ATTR(loading, 0644, firmware_loading_show, firmware_loading_store); static void firmware_rw_data(struct fw_priv *fw_priv, char *buffer, loff_t offset, size_t count, bool read) { if (read) memcpy(buffer, fw_priv->data + offset, count); else memcpy(fw_priv->data + offset, buffer, count); } static void firmware_rw(struct fw_priv *fw_priv, char *buffer, loff_t offset, size_t count, bool read) { while (count) { int page_nr = offset >> PAGE_SHIFT; int page_ofs = offset & (PAGE_SIZE - 1); int page_cnt = min_t(size_t, PAGE_SIZE - page_ofs, count); if (read) memcpy_from_page(buffer, fw_priv->pages[page_nr], page_ofs, page_cnt); else memcpy_to_page(fw_priv->pages[page_nr], page_ofs, buffer, page_cnt); buffer += page_cnt; offset += page_cnt; count -= page_cnt; } } static ssize_t firmware_data_read(struct file *filp, struct kobject *kobj, const struct bin_attribute *bin_attr, char *buffer, loff_t offset, size_t count) { struct device *dev = kobj_to_dev(kobj); struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); struct fw_priv *fw_priv; ssize_t ret_count; mutex_lock(&fw_lock); fw_priv = fw_sysfs->fw_priv; if (!fw_priv || fw_state_is_done(fw_priv)) { ret_count = -ENODEV; goto out; } if (offset > fw_priv->size) { ret_count = 0; goto out; } if (count > fw_priv->size - offset) count = fw_priv->size - offset; ret_count = count; if (fw_priv->data) firmware_rw_data(fw_priv, buffer, offset, count, true); else firmware_rw(fw_priv, buffer, offset, count, true); out: mutex_unlock(&fw_lock); return ret_count; } static int fw_realloc_pages(struct fw_sysfs *fw_sysfs, int min_size) { int err; err = fw_grow_paged_buf(fw_sysfs->fw_priv, PAGE_ALIGN(min_size) >> PAGE_SHIFT); if (err) fw_load_abort(fw_sysfs); return err; } /** * firmware_data_write() - write method for firmware * @filp: open sysfs file * @kobj: kobject for the device * @bin_attr: bin_attr structure * @buffer: buffer being written * @offset: buffer offset for write in total data store area * @count: buffer size * * Data written to the 'data' attribute will be later handed to * the driver as a firmware image. **/ static ssize_t firmware_data_write(struct file *filp, struct kobject *kobj, const struct bin_attribute *bin_attr, char *buffer, loff_t offset, size_t count) { struct device *dev = kobj_to_dev(kobj); struct fw_sysfs *fw_sysfs = to_fw_sysfs(dev); struct fw_priv *fw_priv; ssize_t retval; if (!capable(CAP_SYS_RAWIO)) return -EPERM; mutex_lock(&fw_lock); fw_priv = fw_sysfs->fw_priv; if (!fw_priv || fw_state_is_done(fw_priv)) { retval = -ENODEV; goto out; } if (fw_priv->data) { if (offset + count > fw_priv->allocated_size) { retval = -ENOMEM; goto out; } firmware_rw_data(fw_priv, buffer, offset, count, false); retval = count; } else { retval = fw_realloc_pages(fw_sysfs, offset + count); if (retval) goto out; retval = count; firmware_rw(fw_priv, buffer, offset, count, false); } fw_priv->size = max_t(size_t, offset + count, fw_priv->size); out: mutex_unlock(&fw_lock); return retval; } static const struct bin_attribute firmware_attr_data = { .attr = { .name = "data", .mode = 0644 }, .size = 0, .read = firmware_data_read, .write = firmware_data_write, }; static struct attribute *fw_dev_attrs[] = { &dev_attr_loading.attr, #ifdef CONFIG_FW_UPLOAD &dev_attr_cancel.attr, &dev_attr_status.attr, &dev_attr_error.attr, &dev_attr_remaining_size.attr, #endif NULL }; static const struct bin_attribute *const fw_dev_bin_attrs[] = { &firmware_attr_data, NULL }; static const struct attribute_group fw_dev_attr_group = { .attrs = fw_dev_attrs, .bin_attrs = fw_dev_bin_attrs, #ifdef CONFIG_FW_UPLOAD .is_visible = fw_upload_is_visible, #endif }; static const struct attribute_group *fw_dev_attr_groups[] = { &fw_dev_attr_group, NULL }; struct fw_sysfs * fw_create_instance(struct firmware *firmware, const char *fw_name, struct device *device, u32 opt_flags) { struct fw_sysfs *fw_sysfs; struct device *f_dev; fw_sysfs = kzalloc(sizeof(*fw_sysfs), GFP_KERNEL); if (!fw_sysfs) { fw_sysfs = ERR_PTR(-ENOMEM); goto exit; } fw_sysfs->nowait = !!(opt_flags & FW_OPT_NOWAIT); fw_sysfs->fw = firmware; f_dev = &fw_sysfs->dev; device_initialize(f_dev); dev_set_name(f_dev, "%s", fw_name); f_dev->parent = device; f_dev->class = &firmware_class; f_dev->groups = fw_dev_attr_groups; exit: return fw_sysfs; } |
| 203 201 4 1 2 14 14 1 13 4 8 10 10 4 12 12 12 1 3 1 10 10 1 10 22 22 1 21 20 20 11 10 3 12 1 14 10 8 8 8 395 19 1 2 19 19 19 18 2 1 1 306 307 5 5 306 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 | // SPDX-License-Identifier: GPL-2.0-only /* * Kernel-based Virtual Machine driver for Linux * * This module enables kernel and guest-mode vCPU access to guest physical * memory with suitable invalidation mechanisms. * * Copyright © 2021 Amazon.com, Inc. or its affiliates. * * Authors: * David Woodhouse <dwmw2@infradead.org> */ #include <linux/kvm_host.h> #include <linux/kvm.h> #include <linux/highmem.h> #include <linux/module.h> #include <linux/errno.h> #include "kvm_mm.h" /* * MMU notifier 'invalidate_range_start' hook. */ void gfn_to_pfn_cache_invalidate_start(struct kvm *kvm, unsigned long start, unsigned long end) { struct gfn_to_pfn_cache *gpc; spin_lock(&kvm->gpc_lock); list_for_each_entry(gpc, &kvm->gpc_list, list) { read_lock_irq(&gpc->lock); /* Only a single page so no need to care about length */ if (gpc->valid && !is_error_noslot_pfn(gpc->pfn) && gpc->uhva >= start && gpc->uhva < end) { read_unlock_irq(&gpc->lock); /* * There is a small window here where the cache could * be modified, and invalidation would no longer be * necessary. Hence check again whether invalidation * is still necessary once the write lock has been * acquired. */ write_lock_irq(&gpc->lock); if (gpc->valid && !is_error_noslot_pfn(gpc->pfn) && gpc->uhva >= start && gpc->uhva < end) gpc->valid = false; write_unlock_irq(&gpc->lock); continue; } read_unlock_irq(&gpc->lock); } spin_unlock(&kvm->gpc_lock); } static bool kvm_gpc_is_valid_len(gpa_t gpa, unsigned long uhva, unsigned long len) { unsigned long offset = kvm_is_error_gpa(gpa) ? offset_in_page(uhva) : offset_in_page(gpa); /* * The cached access must fit within a single page. The 'len' argument * to activate() and refresh() exists only to enforce that. */ return offset + len <= PAGE_SIZE; } bool kvm_gpc_check(struct gfn_to_pfn_cache *gpc, unsigned long len) { struct kvm_memslots *slots = kvm_memslots(gpc->kvm); if (!gpc->active) return false; /* * If the page was cached from a memslot, make sure the memslots have * not been re-configured. */ if (!kvm_is_error_gpa(gpc->gpa) && gpc->generation != slots->generation) return false; if (kvm_is_error_hva(gpc->uhva)) return false; if (!kvm_gpc_is_valid_len(gpc->gpa, gpc->uhva, len)) return false; if (!gpc->valid) return false; return true; } static void *gpc_map(kvm_pfn_t pfn) { if (pfn_valid(pfn)) return kmap(pfn_to_page(pfn)); #ifdef CONFIG_HAS_IOMEM return memremap(pfn_to_hpa(pfn), PAGE_SIZE, MEMREMAP_WB); #else return NULL; #endif } static void gpc_unmap(kvm_pfn_t pfn, void *khva) { /* Unmap the old pfn/page if it was mapped before. */ if (is_error_noslot_pfn(pfn) || !khva) return; if (pfn_valid(pfn)) { kunmap(pfn_to_page(pfn)); return; } #ifdef CONFIG_HAS_IOMEM memunmap(khva); #endif } static inline bool mmu_notifier_retry_cache(struct kvm *kvm, unsigned long mmu_seq) { /* * mn_active_invalidate_count acts for all intents and purposes * like mmu_invalidate_in_progress here; but the latter cannot * be used here because the invalidation of caches in the * mmu_notifier event occurs _before_ mmu_invalidate_in_progress * is elevated. * * Note, it does not matter that mn_active_invalidate_count * is not protected by gpc->lock. It is guaranteed to * be elevated before the mmu_notifier acquires gpc->lock, and * isn't dropped until after mmu_invalidate_seq is updated. */ if (kvm->mn_active_invalidate_count) return true; /* * Ensure mn_active_invalidate_count is read before * mmu_invalidate_seq. This pairs with the smp_wmb() in * mmu_notifier_invalidate_range_end() to guarantee either the * old (non-zero) value of mn_active_invalidate_count or the * new (incremented) value of mmu_invalidate_seq is observed. */ smp_rmb(); return kvm->mmu_invalidate_seq != mmu_seq; } static kvm_pfn_t hva_to_pfn_retry(struct gfn_to_pfn_cache *gpc) { /* Note, the new page offset may be different than the old! */ void *old_khva = (void *)PAGE_ALIGN_DOWN((uintptr_t)gpc->khva); kvm_pfn_t new_pfn = KVM_PFN_ERR_FAULT; void *new_khva = NULL; unsigned long mmu_seq; struct page *page; struct kvm_follow_pfn kfp = { .slot = gpc->memslot, .gfn = gpa_to_gfn(gpc->gpa), .flags = FOLL_WRITE, .hva = gpc->uhva, .refcounted_page = &page, }; lockdep_assert_held(&gpc->refresh_lock); lockdep_assert_held_write(&gpc->lock); /* * Invalidate the cache prior to dropping gpc->lock, the gpa=>uhva * assets have already been updated and so a concurrent check() from a * different task may not fail the gpa/uhva/generation checks. */ gpc->valid = false; do { mmu_seq = gpc->kvm->mmu_invalidate_seq; smp_rmb(); write_unlock_irq(&gpc->lock); /* * If the previous iteration "failed" due to an mmu_notifier * event, release the pfn and unmap the kernel virtual address * from the previous attempt. Unmapping might sleep, so this * needs to be done after dropping the lock. Opportunistically * check for resched while the lock isn't held. */ if (new_pfn != KVM_PFN_ERR_FAULT) { /* * Keep the mapping if the previous iteration reused * the existing mapping and didn't create a new one. */ if (new_khva != old_khva) gpc_unmap(new_pfn, new_khva); kvm_release_page_unused(page); cond_resched(); } new_pfn = hva_to_pfn(&kfp); if (is_error_noslot_pfn(new_pfn)) goto out_error; /* * Obtain a new kernel mapping if KVM itself will access the * pfn. Note, kmap() and memremap() can both sleep, so this * too must be done outside of gpc->lock! */ if (new_pfn == gpc->pfn) new_khva = old_khva; else new_khva = gpc_map(new_pfn); if (!new_khva) { kvm_release_page_unused(page); goto out_error; } write_lock_irq(&gpc->lock); /* * Other tasks must wait for _this_ refresh to complete before * attempting to refresh. */ WARN_ON_ONCE(gpc->valid); } while (mmu_notifier_retry_cache(gpc->kvm, mmu_seq)); gpc->valid = true; gpc->pfn = new_pfn; gpc->khva = new_khva + offset_in_page(gpc->uhva); /* * Put the reference to the _new_ page. The page is now tracked by the * cache and can be safely migrated, swapped, etc... as the cache will * invalidate any mappings in response to relevant mmu_notifier events. */ kvm_release_page_clean(page); return 0; out_error: write_lock_irq(&gpc->lock); return -EFAULT; } static int __kvm_gpc_refresh(struct gfn_to_pfn_cache *gpc, gpa_t gpa, unsigned long uhva) { unsigned long page_offset; bool unmap_old = false; unsigned long old_uhva; kvm_pfn_t old_pfn; bool hva_change = false; void *old_khva; int ret; /* Either gpa or uhva must be valid, but not both */ if (WARN_ON_ONCE(kvm_is_error_gpa(gpa) == kvm_is_error_hva(uhva))) return -EINVAL; lockdep_assert_held(&gpc->refresh_lock); write_lock_irq(&gpc->lock); if (!gpc->active) { ret = -EINVAL; goto out_unlock; } old_pfn = gpc->pfn; old_khva = (void *)PAGE_ALIGN_DOWN((uintptr_t)gpc->khva); old_uhva = PAGE_ALIGN_DOWN(gpc->uhva); if (kvm_is_error_gpa(gpa)) { page_offset = offset_in_page(uhva); gpc->gpa = INVALID_GPA; gpc->memslot = NULL; gpc->uhva = PAGE_ALIGN_DOWN(uhva); if (gpc->uhva != old_uhva) hva_change = true; } else { struct kvm_memslots *slots = kvm_memslots(gpc->kvm); page_offset = offset_in_page(gpa); if (gpc->gpa != gpa || gpc->generation != slots->generation || kvm_is_error_hva(gpc->uhva)) { gfn_t gfn = gpa_to_gfn(gpa); gpc->gpa = gpa; gpc->generation = slots->generation; gpc->memslot = __gfn_to_memslot(slots, gfn); gpc->uhva = gfn_to_hva_memslot(gpc->memslot, gfn); if (kvm_is_error_hva(gpc->uhva)) { ret = -EFAULT; goto out; } /* * Even if the GPA and/or the memslot generation changed, the * HVA may still be the same. */ if (gpc->uhva != old_uhva) hva_change = true; } else { gpc->uhva = old_uhva; } } /* Note: the offset must be correct before calling hva_to_pfn_retry() */ gpc->uhva += page_offset; /* * If the userspace HVA changed or the PFN was already invalid, * drop the lock and do the HVA to PFN lookup again. */ if (!gpc->valid || hva_change) { ret = hva_to_pfn_retry(gpc); } else { /* * If the HVA→PFN mapping was already valid, don't unmap it. * But do update gpc->khva because the offset within the page * may have changed. */ gpc->khva = old_khva + page_offset; ret = 0; goto out_unlock; } out: /* * Invalidate the cache and purge the pfn/khva if the refresh failed. * Some/all of the uhva, gpa, and memslot generation info may still be * valid, leave it as is. */ if (ret) { gpc->valid = false; gpc->pfn = KVM_PFN_ERR_FAULT; gpc->khva = NULL; } /* Detect a pfn change before dropping the lock! */ unmap_old = (old_pfn != gpc->pfn); out_unlock: write_unlock_irq(&gpc->lock); if (unmap_old) gpc_unmap(old_pfn, old_khva); return ret; } int kvm_gpc_refresh(struct gfn_to_pfn_cache *gpc, unsigned long len) { unsigned long uhva; guard(mutex)(&gpc->refresh_lock); if (!kvm_gpc_is_valid_len(gpc->gpa, gpc->uhva, len)) return -EINVAL; /* * If the GPA is valid then ignore the HVA, as a cache can be GPA-based * or HVA-based, not both. For GPA-based caches, the HVA will be * recomputed during refresh if necessary. */ uhva = kvm_is_error_gpa(gpc->gpa) ? gpc->uhva : KVM_HVA_ERR_BAD; return __kvm_gpc_refresh(gpc, gpc->gpa, uhva); } void kvm_gpc_init(struct gfn_to_pfn_cache *gpc, struct kvm *kvm) { rwlock_init(&gpc->lock); mutex_init(&gpc->refresh_lock); gpc->kvm = kvm; gpc->pfn = KVM_PFN_ERR_FAULT; gpc->gpa = INVALID_GPA; gpc->uhva = KVM_HVA_ERR_BAD; gpc->active = gpc->valid = false; } static int __kvm_gpc_activate(struct gfn_to_pfn_cache *gpc, gpa_t gpa, unsigned long uhva, unsigned long len) { struct kvm *kvm = gpc->kvm; if (!kvm_gpc_is_valid_len(gpa, uhva, len)) return -EINVAL; guard(mutex)(&gpc->refresh_lock); if (!gpc->active) { if (KVM_BUG_ON(gpc->valid, kvm)) return -EIO; spin_lock(&kvm->gpc_lock); list_add(&gpc->list, &kvm->gpc_list); spin_unlock(&kvm->gpc_lock); /* * Activate the cache after adding it to the list, a concurrent * refresh must not establish a mapping until the cache is * reachable by mmu_notifier events. */ write_lock_irq(&gpc->lock); gpc->active = true; write_unlock_irq(&gpc->lock); } return __kvm_gpc_refresh(gpc, gpa, uhva); } int kvm_gpc_activate(struct gfn_to_pfn_cache *gpc, gpa_t gpa, unsigned long len) { /* * Explicitly disallow INVALID_GPA so that the magic value can be used * by KVM to differentiate between GPA-based and HVA-based caches. */ if (WARN_ON_ONCE(kvm_is_error_gpa(gpa))) return -EINVAL; return __kvm_gpc_activate(gpc, gpa, KVM_HVA_ERR_BAD, len); } int kvm_gpc_activate_hva(struct gfn_to_pfn_cache *gpc, unsigned long uhva, unsigned long len) { if (!access_ok((void __user *)uhva, len)) return -EINVAL; return __kvm_gpc_activate(gpc, INVALID_GPA, uhva, len); } void kvm_gpc_deactivate(struct gfn_to_pfn_cache *gpc) { struct kvm *kvm = gpc->kvm; kvm_pfn_t old_pfn; void *old_khva; guard(mutex)(&gpc->refresh_lock); if (gpc->active) { /* * Deactivate the cache before removing it from the list, KVM * must stall mmu_notifier events until all users go away, i.e. * until gpc->lock is dropped and refresh is guaranteed to fail. */ write_lock_irq(&gpc->lock); gpc->active = false; gpc->valid = false; /* * Leave the GPA => uHVA cache intact, it's protected by the * memslot generation. The PFN lookup needs to be redone every * time as mmu_notifier protection is lost when the cache is * removed from the VM's gpc_list. */ old_khva = gpc->khva - offset_in_page(gpc->khva); gpc->khva = NULL; old_pfn = gpc->pfn; gpc->pfn = KVM_PFN_ERR_FAULT; write_unlock_irq(&gpc->lock); spin_lock(&kvm->gpc_lock); list_del(&gpc->list); spin_unlock(&kvm->gpc_lock); gpc_unmap(old_pfn, old_khva); } } |
| 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 | // SPDX-License-Identifier: GPL-2.0-only /* huawei_cdc_ncm.c - handles Huawei devices using the CDC NCM protocol as * transport layer. * Copyright (C) 2013 Enrico Mioso <mrkiko.rs@gmail.com> * * ABSTRACT: * This driver handles devices resembling the CDC NCM standard, but * encapsulating another protocol inside it. An example are some Huawei 3G * devices, exposing an embedded AT channel where you can set up the NCM * connection. * This code has been heavily inspired by the cdc_mbim.c driver, which is * Copyright (c) 2012 Smith Micro Software, Inc. * Copyright (c) 2012 Bjørn Mork <bjorn@mork.no> */ #include <linux/module.h> #include <linux/netdevice.h> #include <linux/ethtool.h> #include <linux/if_vlan.h> #include <linux/ip.h> #include <linux/mii.h> #include <linux/usb.h> #include <linux/usb/cdc.h> #include <linux/usb/usbnet.h> #include <linux/usb/cdc-wdm.h> #include <linux/usb/cdc_ncm.h> /* Driver data */ struct huawei_cdc_ncm_state { struct cdc_ncm_ctx *ctx; atomic_t pmcount; struct usb_driver *subdriver; struct usb_interface *control; struct usb_interface *data; }; static int huawei_cdc_ncm_manage_power(struct usbnet *usbnet_dev, int on) { struct huawei_cdc_ncm_state *drvstate = (void *)&usbnet_dev->data; int rv; if ((on && atomic_add_return(1, &drvstate->pmcount) == 1) || (!on && atomic_dec_and_test(&drvstate->pmcount))) { rv = usb_autopm_get_interface(usbnet_dev->intf); usbnet_dev->intf->needs_remote_wakeup = on; if (!rv) usb_autopm_put_interface(usbnet_dev->intf); } return 0; } static int huawei_cdc_ncm_wdm_manage_power(struct usb_interface *intf, int status) { struct usbnet *usbnet_dev = usb_get_intfdata(intf); /* can be called while disconnecting */ if (!usbnet_dev) return 0; return huawei_cdc_ncm_manage_power(usbnet_dev, status); } static int huawei_cdc_ncm_bind(struct usbnet *usbnet_dev, struct usb_interface *intf) { struct cdc_ncm_ctx *ctx; struct usb_driver *subdriver = ERR_PTR(-ENODEV); int ret; struct huawei_cdc_ncm_state *drvstate = (void *)&usbnet_dev->data; int drvflags = 0; /* altsetting should always be 1 for NCM devices - so we hard-coded * it here. Some huawei devices will need the NDP part of the NCM package to * be at the end of the frame. */ drvflags |= CDC_NCM_FLAG_NDP_TO_END; /* For many Huawei devices the NTB32 mode is the default and the best mode * they work with. Huawei E5785 and E5885 devices refuse to work in NTB16 mode at all. */ drvflags |= CDC_NCM_FLAG_PREFER_NTB32; ret = cdc_ncm_bind_common(usbnet_dev, intf, 1, drvflags); if (ret) goto err; ctx = drvstate->ctx; if (usbnet_dev->status) /* The wMaxCommand buffer must be big enough to hold * any message from the modem. Experience has shown * that some replies are more than 256 bytes long */ subdriver = usb_cdc_wdm_register(ctx->control, &usbnet_dev->status->desc, 1024, /* wMaxCommand */ WWAN_PORT_AT, huawei_cdc_ncm_wdm_manage_power); if (IS_ERR(subdriver)) { ret = PTR_ERR(subdriver); cdc_ncm_unbind(usbnet_dev, intf); goto err; } /* Prevent usbnet from using the status descriptor */ usbnet_dev->status = NULL; drvstate->subdriver = subdriver; err: return ret; } static void huawei_cdc_ncm_unbind(struct usbnet *usbnet_dev, struct usb_interface *intf) { struct huawei_cdc_ncm_state *drvstate = (void *)&usbnet_dev->data; struct cdc_ncm_ctx *ctx = drvstate->ctx; if (drvstate->subdriver && drvstate->subdriver->disconnect) drvstate->subdriver->disconnect(ctx->control); drvstate->subdriver = NULL; cdc_ncm_unbind(usbnet_dev, intf); } static int huawei_cdc_ncm_suspend(struct usb_interface *intf, pm_message_t message) { int ret = 0; struct usbnet *usbnet_dev = usb_get_intfdata(intf); struct huawei_cdc_ncm_state *drvstate = (void *)&usbnet_dev->data; struct cdc_ncm_ctx *ctx = drvstate->ctx; if (ctx == NULL) { ret = -ENODEV; goto error; } ret = usbnet_suspend(intf, message); if (ret < 0) goto error; if (intf == ctx->control && drvstate->subdriver && drvstate->subdriver->suspend) ret = drvstate->subdriver->suspend(intf, message); if (ret < 0) usbnet_resume(intf); error: return ret; } static int huawei_cdc_ncm_resume(struct usb_interface *intf) { int ret = 0; struct usbnet *usbnet_dev = usb_get_intfdata(intf); struct huawei_cdc_ncm_state *drvstate = (void *)&usbnet_dev->data; bool callsub; struct cdc_ncm_ctx *ctx = drvstate->ctx; /* should we call subdriver's resume function? */ callsub = (intf == ctx->control && drvstate->subdriver && drvstate->subdriver->resume); if (callsub) ret = drvstate->subdriver->resume(intf); if (ret < 0) goto err; ret = usbnet_resume(intf); if (ret < 0 && callsub) drvstate->subdriver->suspend(intf, PMSG_SUSPEND); err: return ret; } static const struct driver_info huawei_cdc_ncm_info = { .description = "Huawei CDC NCM device", .flags = FLAG_NO_SETINT | FLAG_MULTI_PACKET | FLAG_WWAN, .bind = huawei_cdc_ncm_bind, .unbind = huawei_cdc_ncm_unbind, .manage_power = huawei_cdc_ncm_manage_power, .rx_fixup = cdc_ncm_rx_fixup, .tx_fixup = cdc_ncm_tx_fixup, }; static const struct usb_device_id huawei_cdc_ncm_devs[] = { /* Huawei NCM devices disguised as vendor specific */ { USB_VENDOR_AND_INTERFACE_INFO(0x12d1, 0xff, 0x02, 0x16), .driver_info = (unsigned long)&huawei_cdc_ncm_info, }, { USB_VENDOR_AND_INTERFACE_INFO(0x12d1, 0xff, 0x02, 0x46), .driver_info = (unsigned long)&huawei_cdc_ncm_info, }, { USB_VENDOR_AND_INTERFACE_INFO(0x12d1, 0xff, 0x02, 0x76), .driver_info = (unsigned long)&huawei_cdc_ncm_info, }, { USB_VENDOR_AND_INTERFACE_INFO(0x12d1, 0xff, 0x03, 0x16), .driver_info = (unsigned long)&huawei_cdc_ncm_info, }, /* Terminating entry */ { }, }; MODULE_DEVICE_TABLE(usb, huawei_cdc_ncm_devs); static struct usb_driver huawei_cdc_ncm_driver = { .name = "huawei_cdc_ncm", .id_table = huawei_cdc_ncm_devs, .probe = usbnet_probe, .disconnect = usbnet_disconnect, .suspend = huawei_cdc_ncm_suspend, .resume = huawei_cdc_ncm_resume, .reset_resume = huawei_cdc_ncm_resume, .supports_autosuspend = 1, .disable_hub_initiated_lpm = 1, }; module_usb_driver(huawei_cdc_ncm_driver); MODULE_AUTHOR("Enrico Mioso <mrkiko.rs@gmail.com>"); MODULE_DESCRIPTION("USB CDC NCM host driver with encapsulated protocol support"); MODULE_LICENSE("GPL"); |
| 24485 2 2 723 11639 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 | /* SPDX-License-Identifier: GPL-2.0 */ #ifndef __LINUX_BIT_SPINLOCK_H #define __LINUX_BIT_SPINLOCK_H #include <linux/kernel.h> #include <linux/preempt.h> #include <linux/atomic.h> #include <linux/bug.h> #include <asm/processor.h> /* for cpu_relax() */ /* * For static context analysis, we need a unique token for each possible bit * that can be used as a bit_spinlock. The easiest way to do that is to create a * fake context that we can cast to with the __bitlock(bitnum, addr) macro * below, which will give us unique instances for each (bit, addr) pair that the * static analysis can use. */ context_lock_struct(__context_bitlock) { }; #define __bitlock(bitnum, addr) (struct __context_bitlock *)(bitnum + (addr)) /* * bit-based spin_lock() * * Don't use this unless you really need to: spin_lock() and spin_unlock() * are significantly faster. */ static __always_inline void bit_spin_lock(int bitnum, unsigned long *addr) __acquires(__bitlock(bitnum, addr)) { /* * Assuming the lock is uncontended, this never enters * the body of the outer loop. If it is contended, then * within the inner loop a non-atomic test is used to * busywait with less bus contention for a good time to * attempt to acquire the lock bit. */ preempt_disable(); #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) while (unlikely(test_and_set_bit_lock(bitnum, addr))) { preempt_enable(); do { cpu_relax(); } while (test_bit(bitnum, addr)); preempt_disable(); } #endif __acquire(__bitlock(bitnum, addr)); } /* * Return true if it was acquired */ static __always_inline int bit_spin_trylock(int bitnum, unsigned long *addr) __cond_acquires(true, __bitlock(bitnum, addr)) { preempt_disable(); #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) if (unlikely(test_and_set_bit_lock(bitnum, addr))) { preempt_enable(); return 0; } #endif __acquire(__bitlock(bitnum, addr)); return 1; } /* * bit-based spin_unlock() */ static __always_inline void bit_spin_unlock(int bitnum, unsigned long *addr) __releases(__bitlock(bitnum, addr)) { #ifdef CONFIG_DEBUG_SPINLOCK BUG_ON(!test_bit(bitnum, addr)); #endif #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) clear_bit_unlock(bitnum, addr); #endif preempt_enable(); __release(__bitlock(bitnum, addr)); } /* * bit-based spin_unlock() * non-atomic version, which can be used eg. if the bit lock itself is * protecting the rest of the flags in the word. */ static __always_inline void __bit_spin_unlock(int bitnum, unsigned long *addr) __releases(__bitlock(bitnum, addr)) { #ifdef CONFIG_DEBUG_SPINLOCK BUG_ON(!test_bit(bitnum, addr)); #endif #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) __clear_bit_unlock(bitnum, addr); #endif preempt_enable(); __release(__bitlock(bitnum, addr)); } /* * Return true if the lock is held. */ static inline int bit_spin_is_locked(int bitnum, unsigned long *addr) { #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) return test_bit(bitnum, addr); #elif defined CONFIG_PREEMPT_COUNT return preempt_count(); #else return 1; #endif } #endif /* __LINUX_BIT_SPINLOCK_H */ |
| 3 57 57 57 57 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 | // SPDX-License-Identifier: GPL-2.0 #include <linux/types.h> #include <linux/netfilter.h> #include <linux/slab.h> #include <linux/module.h> #include <linux/skbuff.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/percpu.h> #include <linux/netdevice.h> #include <linux/security.h> #include <net/net_namespace.h> #ifdef CONFIG_SYSCTL #include <linux/sysctl.h> #endif #include <net/netfilter/nf_log.h> #include <net/netfilter/nf_conntrack.h> #include <net/netfilter/nf_conntrack_core.h> #include <net/netfilter/nf_conntrack_l4proto.h> #include <net/netfilter/nf_conntrack_expect.h> #include <net/netfilter/nf_conntrack_helper.h> #include <net/netfilter/nf_conntrack_acct.h> #include <net/netfilter/nf_conntrack_zones.h> #include <net/netfilter/nf_conntrack_timestamp.h> #include <linux/rculist_nulls.h> static bool enable_hooks __read_mostly; MODULE_PARM_DESC(enable_hooks, "Always enable conntrack hooks"); module_param(enable_hooks, bool, 0000); unsigned int nf_conntrack_net_id __read_mostly; #ifdef CONFIG_NF_CONNTRACK_PROCFS void print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, const struct nf_conntrack_l4proto *l4proto) { switch (tuple->src.l3num) { case NFPROTO_IPV4: seq_printf(s, "src=%pI4 dst=%pI4 ", &tuple->src.u3.ip, &tuple->dst.u3.ip); break; case NFPROTO_IPV6: seq_printf(s, "src=%pI6 dst=%pI6 ", tuple->src.u3.ip6, tuple->dst.u3.ip6); break; default: break; } switch (l4proto->l4proto) { case IPPROTO_ICMP: seq_printf(s, "type=%u code=%u id=%u ", tuple->dst.u.icmp.type, tuple->dst.u.icmp.code, ntohs(tuple->src.u.icmp.id)); break; case IPPROTO_TCP: seq_printf(s, "sport=%hu dport=%hu ", ntohs(tuple->src.u.tcp.port), ntohs(tuple->dst.u.tcp.port)); break; case IPPROTO_UDPLITE: case IPPROTO_UDP: seq_printf(s, "sport=%hu dport=%hu ", ntohs(tuple->src.u.udp.port), ntohs(tuple->dst.u.udp.port)); break; case IPPROTO_SCTP: seq_printf(s, "sport=%hu dport=%hu ", ntohs(tuple->src.u.sctp.port), ntohs(tuple->dst.u.sctp.port)); break; case IPPROTO_ICMPV6: seq_printf(s, "type=%u code=%u id=%u ", tuple->dst.u.icmp.type, tuple->dst.u.icmp.code, ntohs(tuple->src.u.icmp.id)); break; case IPPROTO_GRE: seq_printf(s, "srckey=0x%x dstkey=0x%x ", ntohs(tuple->src.u.gre.key), ntohs(tuple->dst.u.gre.key)); break; default: break; } } EXPORT_SYMBOL_GPL(print_tuple); struct ct_iter_state { struct seq_net_private p; struct hlist_nulls_head *hash; unsigned int htable_size; unsigned int skip_elems; unsigned int bucket; u_int64_t time_now; }; static struct nf_conntrack_tuple_hash *ct_get_next(const struct net *net, struct ct_iter_state *st) { struct nf_conntrack_tuple_hash *h; struct hlist_nulls_node *n; unsigned int i; for (i = st->bucket; i < st->htable_size; i++) { unsigned int skip = 0; restart: hlist_nulls_for_each_entry_rcu(h, n, &st->hash[i], hnnode) { struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); struct hlist_nulls_node *tmp = n; if (!net_eq(net, nf_ct_net(ct))) continue; if (++skip <= st->skip_elems) continue; /* h should be returned, skip to nulls marker. */ while (!is_a_nulls(tmp)) tmp = rcu_dereference(hlist_nulls_next_rcu(tmp)); /* check if h is still linked to hash[i] */ if (get_nulls_value(tmp) != i) { skip = 0; goto restart; } st->skip_elems = skip; st->bucket = i; return h; } skip = 0; if (get_nulls_value(n) != i) goto restart; st->skip_elems = 0; } st->bucket = i; return NULL; } static void *ct_seq_start(struct seq_file *seq, loff_t *pos) __acquires(RCU) { struct ct_iter_state *st = seq->private; struct net *net = seq_file_net(seq); st->time_now = ktime_get_real_ns(); rcu_read_lock(); nf_conntrack_get_ht(&st->hash, &st->htable_size); if (*pos == 0) { st->skip_elems = 0; st->bucket = 0; } else if (st->skip_elems) { /* resume from last dumped entry */ st->skip_elems--; } return ct_get_next(net, st); } static void *ct_seq_next(struct seq_file *s, void *v, loff_t *pos) { struct ct_iter_state *st = s->private; struct net *net = seq_file_net(s); (*pos)++; return ct_get_next(net, st); } static void ct_seq_stop(struct seq_file *s, void *v) __releases(RCU) { rcu_read_unlock(); } #ifdef CONFIG_NF_CONNTRACK_SECMARK static void ct_show_secctx(struct seq_file *s, const struct nf_conn *ct) { struct lsm_context ctx; int ret; ret = security_secid_to_secctx(ct->secmark, &ctx); if (ret < 0) return; seq_printf(s, "secctx=%s ", ctx.context); security_release_secctx(&ctx); } #else static inline void ct_show_secctx(struct seq_file *s, const struct nf_conn *ct) { } #endif #ifdef CONFIG_NF_CONNTRACK_ZONES static void ct_show_zone(struct seq_file *s, const struct nf_conn *ct, int dir) { const struct nf_conntrack_zone *zone = nf_ct_zone(ct); if (zone->dir != dir) return; switch (zone->dir) { case NF_CT_DEFAULT_ZONE_DIR: seq_printf(s, "zone=%u ", zone->id); break; case NF_CT_ZONE_DIR_ORIG: seq_printf(s, "zone-orig=%u ", zone->id); break; case NF_CT_ZONE_DIR_REPL: seq_printf(s, "zone-reply=%u ", zone->id); break; default: break; } } #else static inline void ct_show_zone(struct seq_file *s, const struct nf_conn *ct, int dir) { } #endif #ifdef CONFIG_NF_CONNTRACK_TIMESTAMP static void ct_show_delta_time(struct seq_file *s, const struct nf_conn *ct) { struct ct_iter_state *st = s->private; struct nf_conn_tstamp *tstamp; s64 delta_time; tstamp = nf_conn_tstamp_find(ct); if (tstamp) { delta_time = st->time_now - tstamp->start; if (delta_time > 0) delta_time = div_s64(delta_time, NSEC_PER_SEC); else delta_time = 0; seq_printf(s, "delta-time=%llu ", (unsigned long long)delta_time); } return; } #else static inline void ct_show_delta_time(struct seq_file *s, const struct nf_conn *ct) { } #endif static const char* l3proto_name(u16 proto) { switch (proto) { case AF_INET: return "ipv4"; case AF_INET6: return "ipv6"; } return "unknown"; } static const char* l4proto_name(u16 proto) { switch (proto) { case IPPROTO_ICMP: return "icmp"; case IPPROTO_TCP: return "tcp"; case IPPROTO_UDP: return "udp"; case IPPROTO_GRE: return "gre"; case IPPROTO_SCTP: return "sctp"; case IPPROTO_UDPLITE: return "udplite"; case IPPROTO_ICMPV6: return "icmpv6"; } return "unknown"; } static void seq_print_acct(struct seq_file *s, const struct nf_conn *ct, int dir) { struct nf_conn_acct *acct; struct nf_conn_counter *counter; acct = nf_conn_acct_find(ct); if (!acct) return; counter = acct->counter; seq_printf(s, "packets=%llu bytes=%llu ", (unsigned long long)atomic64_read(&counter[dir].packets), (unsigned long long)atomic64_read(&counter[dir].bytes)); } /* return 0 on success, 1 in case of error */ static int ct_seq_show(struct seq_file *s, void *v) { struct nf_conntrack_tuple_hash *hash = v; struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(hash); const struct nf_conntrack_l4proto *l4proto; struct net *net = seq_file_net(s); int ret = 0; WARN_ON(!ct); if (unlikely(!refcount_inc_not_zero(&ct->ct_general.use))) return 0; /* load ->status after refcount increase */ smp_acquire__after_ctrl_dep(); if (nf_ct_should_gc(ct)) { struct ct_iter_state *st = s->private; st->skip_elems--; nf_ct_kill(ct); goto release; } /* we only want to print DIR_ORIGINAL */ if (NF_CT_DIRECTION(hash)) goto release; if (!net_eq(nf_ct_net(ct), net)) goto release; l4proto = nf_ct_l4proto_find(nf_ct_protonum(ct)); ret = -ENOSPC; seq_printf(s, "%-8s %u %-8s %u ", l3proto_name(nf_ct_l3num(ct)), nf_ct_l3num(ct), l4proto_name(l4proto->l4proto), nf_ct_protonum(ct)); if (!test_bit(IPS_OFFLOAD_BIT, &ct->status)) seq_printf(s, "%ld ", nf_ct_expires(ct) / HZ); if (l4proto->print_conntrack) l4proto->print_conntrack(s, ct); print_tuple(s, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, l4proto); ct_show_zone(s, ct, NF_CT_ZONE_DIR_ORIG); if (seq_has_overflowed(s)) goto release; seq_print_acct(s, ct, IP_CT_DIR_ORIGINAL); if (!(test_bit(IPS_SEEN_REPLY_BIT, &ct->status))) seq_puts(s, "[UNREPLIED] "); print_tuple(s, &ct->tuplehash[IP_CT_DIR_REPLY].tuple, l4proto); ct_show_zone(s, ct, NF_CT_ZONE_DIR_REPL); seq_print_acct(s, ct, IP_CT_DIR_REPLY); if (test_bit(IPS_HW_OFFLOAD_BIT, &ct->status)) seq_puts(s, "[HW_OFFLOAD] "); else if (test_bit(IPS_OFFLOAD_BIT, &ct->status)) seq_puts(s, "[OFFLOAD] "); else if (test_bit(IPS_ASSURED_BIT, &ct->status)) seq_puts(s, "[ASSURED] "); if (seq_has_overflowed(s)) goto release; #if defined(CONFIG_NF_CONNTRACK_MARK) seq_printf(s, "mark=%u ", READ_ONCE(ct->mark)); #endif ct_show_secctx(s, ct); ct_show_zone(s, ct, NF_CT_DEFAULT_ZONE_DIR); ct_show_delta_time(s, ct); seq_printf(s, "use=%u\n", refcount_read(&ct->ct_general.use)); if (seq_has_overflowed(s)) goto release; ret = 0; release: nf_ct_put(ct); return ret; } static const struct seq_operations ct_seq_ops = { .start = ct_seq_start, .next = ct_seq_next, .stop = ct_seq_stop, .show = ct_seq_show }; static void *ct_cpu_seq_start(struct seq_file *seq, loff_t *pos) { struct net *net = seq_file_net(seq); int cpu; if (*pos == 0) return SEQ_START_TOKEN; for (cpu = *pos-1; cpu < nr_cpu_ids; ++cpu) { if (!cpu_possible(cpu)) continue; *pos = cpu + 1; return per_cpu_ptr(net->ct.stat, cpu); } return NULL; } static void *ct_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct net *net = seq_file_net(seq); int cpu; for (cpu = *pos; cpu < nr_cpu_ids; ++cpu) { if (!cpu_possible(cpu)) continue; *pos = cpu + 1; return per_cpu_ptr(net->ct.stat, cpu); } (*pos)++; return NULL; } static void ct_cpu_seq_stop(struct seq_file *seq, void *v) { } static int ct_cpu_seq_show(struct seq_file *seq, void *v) { struct net *net = seq_file_net(seq); const struct ip_conntrack_stat *st = v; unsigned int nr_conntracks; if (v == SEQ_START_TOKEN) { seq_puts(seq, "entries clashres found new invalid ignore delete chainlength insert insert_failed drop early_drop icmp_error expect_new expect_create expect_delete search_restart\n"); return 0; } nr_conntracks = nf_conntrack_count(net); seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x " "%08x %08x %08x %08x %08x %08x %08x %08x %08x\n", nr_conntracks, st->clash_resolve, st->found, 0, st->invalid, 0, 0, st->chaintoolong, st->insert, st->insert_failed, st->drop, st->early_drop, st->error, st->expect_new, st->expect_create, st->expect_delete, st->search_restart ); return 0; } static const struct seq_operations ct_cpu_seq_ops = { .start = ct_cpu_seq_start, .next = ct_cpu_seq_next, .stop = ct_cpu_seq_stop, .show = ct_cpu_seq_show, }; static int nf_conntrack_standalone_init_proc(struct net *net) { struct proc_dir_entry *pde; kuid_t root_uid; kgid_t root_gid; pde = proc_create_net("nf_conntrack", 0440, net->proc_net, &ct_seq_ops, sizeof(struct ct_iter_state)); if (!pde) goto out_nf_conntrack; root_uid = make_kuid(net->user_ns, 0); root_gid = make_kgid(net->user_ns, 0); if (uid_valid(root_uid) && gid_valid(root_gid)) proc_set_user(pde, root_uid, root_gid); pde = proc_create_net("nf_conntrack", 0444, net->proc_net_stat, &ct_cpu_seq_ops, sizeof(struct seq_net_private)); if (!pde) goto out_stat_nf_conntrack; return 0; out_stat_nf_conntrack: remove_proc_entry("nf_conntrack", net->proc_net); out_nf_conntrack: return -ENOMEM; } static void nf_conntrack_standalone_fini_proc(struct net *net) { remove_proc_entry("nf_conntrack", net->proc_net_stat); remove_proc_entry("nf_conntrack", net->proc_net); } #else static int nf_conntrack_standalone_init_proc(struct net *net) { return 0; } static void nf_conntrack_standalone_fini_proc(struct net *net) { } #endif /* CONFIG_NF_CONNTRACK_PROCFS */ u32 nf_conntrack_count(const struct net *net) { const struct nf_conntrack_net *cnet = nf_ct_pernet(net); return atomic_read(&cnet->count); } EXPORT_SYMBOL_GPL(nf_conntrack_count); /* Sysctl support */ #ifdef CONFIG_SYSCTL /* size the user *wants to set */ static unsigned int nf_conntrack_htable_size_user __read_mostly; static int nf_conntrack_hash_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { int ret; /* module_param hashsize could have changed value */ nf_conntrack_htable_size_user = nf_conntrack_htable_size; ret = proc_dointvec(table, write, buffer, lenp, ppos); if (ret < 0 || !write) return ret; /* update ret, we might not be able to satisfy request */ ret = nf_conntrack_hash_resize(nf_conntrack_htable_size_user); /* update it to the actual value used by conntrack */ nf_conntrack_htable_size_user = nf_conntrack_htable_size; return ret; } static int nf_conntrack_log_invalid_sysctl(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { int ret, i; ret = proc_dou8vec_minmax(table, write, buffer, lenp, ppos); if (ret < 0 || !write) return ret; if (*(u8 *)table->data == 0) return 0; /* Load nf_log_syslog only if no logger is currently registered */ for (i = 0; i < NFPROTO_NUMPROTO; i++) { if (nf_log_is_registered(i)) return 0; } request_module("%s", "nf_log_syslog"); return 0; } static struct ctl_table_header *nf_ct_netfilter_header; enum nf_ct_sysctl_index { NF_SYSCTL_CT_MAX, NF_SYSCTL_CT_COUNT, NF_SYSCTL_CT_BUCKETS, NF_SYSCTL_CT_CHECKSUM, NF_SYSCTL_CT_LOG_INVALID, NF_SYSCTL_CT_EXPECT_MAX, NF_SYSCTL_CT_ACCT, #ifdef CONFIG_NF_CONNTRACK_EVENTS NF_SYSCTL_CT_EVENTS, #endif #ifdef CONFIG_NF_CONNTRACK_TIMESTAMP NF_SYSCTL_CT_TIMESTAMP, #endif NF_SYSCTL_CT_PROTO_TIMEOUT_GENERIC, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_SYN_SENT, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_SYN_RECV, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_ESTABLISHED, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_FIN_WAIT, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_CLOSE_WAIT, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_LAST_ACK, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_TIME_WAIT, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_CLOSE, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_RETRANS, NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_UNACK, #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_OFFLOAD, #endif NF_SYSCTL_CT_PROTO_TCP_LOOSE, NF_SYSCTL_CT_PROTO_TCP_LIBERAL, NF_SYSCTL_CT_PROTO_TCP_IGNORE_INVALID_RST, NF_SYSCTL_CT_PROTO_TCP_MAX_RETRANS, NF_SYSCTL_CT_PROTO_TIMEOUT_UDP, NF_SYSCTL_CT_PROTO_TIMEOUT_UDP_STREAM, #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) NF_SYSCTL_CT_PROTO_TIMEOUT_UDP_OFFLOAD, #endif NF_SYSCTL_CT_PROTO_TIMEOUT_ICMP, NF_SYSCTL_CT_PROTO_TIMEOUT_ICMPV6, #ifdef CONFIG_NF_CT_PROTO_SCTP NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_CLOSED, NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_COOKIE_WAIT, NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_COOKIE_ECHOED, NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_ESTABLISHED, NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_SHUTDOWN_SENT, NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_SHUTDOWN_RECD, NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_SHUTDOWN_ACK_SENT, NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_HEARTBEAT_SENT, #endif #ifdef CONFIG_NF_CT_PROTO_GRE NF_SYSCTL_CT_PROTO_TIMEOUT_GRE, NF_SYSCTL_CT_PROTO_TIMEOUT_GRE_STREAM, #endif NF_SYSCTL_CT_LAST_SYSCTL, }; static struct ctl_table nf_ct_sysctl_table[] = { [NF_SYSCTL_CT_MAX] = { .procname = "nf_conntrack_max", .data = &nf_conntrack_max, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ONE, .extra2 = SYSCTL_INT_MAX, }, [NF_SYSCTL_CT_COUNT] = { .procname = "nf_conntrack_count", .maxlen = sizeof(int), .mode = 0444, .proc_handler = proc_dointvec, }, [NF_SYSCTL_CT_BUCKETS] = { .procname = "nf_conntrack_buckets", .data = &nf_conntrack_htable_size_user, .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = nf_conntrack_hash_sysctl, }, [NF_SYSCTL_CT_CHECKSUM] = { .procname = "nf_conntrack_checksum", .data = &init_net.ct.sysctl_checksum, .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, [NF_SYSCTL_CT_LOG_INVALID] = { .procname = "nf_conntrack_log_invalid", .data = &init_net.ct.sysctl_log_invalid, .maxlen = sizeof(u8), .mode = 0644, .proc_handler = nf_conntrack_log_invalid_sysctl, }, [NF_SYSCTL_CT_EXPECT_MAX] = { .procname = "nf_conntrack_expect_max", .data = &nf_ct_expect_max, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ONE, .extra2 = SYSCTL_INT_MAX, }, [NF_SYSCTL_CT_ACCT] = { .procname = "nf_conntrack_acct", .data = &init_net.ct.sysctl_acct, .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, #ifdef CONFIG_NF_CONNTRACK_EVENTS [NF_SYSCTL_CT_EVENTS] = { .procname = "nf_conntrack_events", .data = &init_net.ct.sysctl_events, .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_TWO, }, #endif #ifdef CONFIG_NF_CONNTRACK_TIMESTAMP [NF_SYSCTL_CT_TIMESTAMP] = { .procname = "nf_conntrack_timestamp", .data = &init_net.ct.sysctl_tstamp, .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, #endif [NF_SYSCTL_CT_PROTO_TIMEOUT_GENERIC] = { .procname = "nf_conntrack_generic_timeout", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_SYN_SENT] = { .procname = "nf_conntrack_tcp_timeout_syn_sent", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_SYN_RECV] = { .procname = "nf_conntrack_tcp_timeout_syn_recv", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_ESTABLISHED] = { .procname = "nf_conntrack_tcp_timeout_established", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_FIN_WAIT] = { .procname = "nf_conntrack_tcp_timeout_fin_wait", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_CLOSE_WAIT] = { .procname = "nf_conntrack_tcp_timeout_close_wait", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_LAST_ACK] = { .procname = "nf_conntrack_tcp_timeout_last_ack", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_TIME_WAIT] = { .procname = "nf_conntrack_tcp_timeout_time_wait", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_CLOSE] = { .procname = "nf_conntrack_tcp_timeout_close", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_RETRANS] = { .procname = "nf_conntrack_tcp_timeout_max_retrans", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_UNACK] = { .procname = "nf_conntrack_tcp_timeout_unacknowledged", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) [NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_OFFLOAD] = { .procname = "nf_flowtable_tcp_timeout", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, #endif [NF_SYSCTL_CT_PROTO_TCP_LOOSE] = { .procname = "nf_conntrack_tcp_loose", .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, [NF_SYSCTL_CT_PROTO_TCP_LIBERAL] = { .procname = "nf_conntrack_tcp_be_liberal", .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, [NF_SYSCTL_CT_PROTO_TCP_IGNORE_INVALID_RST] = { .procname = "nf_conntrack_tcp_ignore_invalid_rst", .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, [NF_SYSCTL_CT_PROTO_TCP_MAX_RETRANS] = { .procname = "nf_conntrack_tcp_max_retrans", .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_UDP] = { .procname = "nf_conntrack_udp_timeout", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_UDP_STREAM] = { .procname = "nf_conntrack_udp_timeout_stream", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) [NF_SYSCTL_CT_PROTO_TIMEOUT_UDP_OFFLOAD] = { .procname = "nf_flowtable_udp_timeout", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, #endif [NF_SYSCTL_CT_PROTO_TIMEOUT_ICMP] = { .procname = "nf_conntrack_icmp_timeout", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_ICMPV6] = { .procname = "nf_conntrack_icmpv6_timeout", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, #ifdef CONFIG_NF_CT_PROTO_SCTP [NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_CLOSED] = { .procname = "nf_conntrack_sctp_timeout_closed", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_COOKIE_WAIT] = { .procname = "nf_conntrack_sctp_timeout_cookie_wait", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_COOKIE_ECHOED] = { .procname = "nf_conntrack_sctp_timeout_cookie_echoed", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_ESTABLISHED] = { .procname = "nf_conntrack_sctp_timeout_established", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_SHUTDOWN_SENT] = { .procname = "nf_conntrack_sctp_timeout_shutdown_sent", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_SHUTDOWN_RECD] = { .procname = "nf_conntrack_sctp_timeout_shutdown_recd", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_SHUTDOWN_ACK_SENT] = { .procname = "nf_conntrack_sctp_timeout_shutdown_ack_sent", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_HEARTBEAT_SENT] = { .procname = "nf_conntrack_sctp_timeout_heartbeat_sent", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, #endif #ifdef CONFIG_NF_CT_PROTO_GRE [NF_SYSCTL_CT_PROTO_TIMEOUT_GRE] = { .procname = "nf_conntrack_gre_timeout", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, [NF_SYSCTL_CT_PROTO_TIMEOUT_GRE_STREAM] = { .procname = "nf_conntrack_gre_timeout_stream", .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, #endif }; static struct ctl_table nf_ct_netfilter_table[] = { { .procname = "nf_conntrack_max", .data = &nf_conntrack_max, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ONE, .extra2 = SYSCTL_INT_MAX, }, }; static void nf_conntrack_standalone_init_tcp_sysctl(struct net *net, struct ctl_table *table) { struct nf_tcp_net *tn = nf_tcp_pernet(net); #define XASSIGN(XNAME, tn) \ table[NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_ ## XNAME].data = \ &(tn)->timeouts[TCP_CONNTRACK_ ## XNAME] XASSIGN(SYN_SENT, tn); XASSIGN(SYN_RECV, tn); XASSIGN(ESTABLISHED, tn); XASSIGN(FIN_WAIT, tn); XASSIGN(CLOSE_WAIT, tn); XASSIGN(LAST_ACK, tn); XASSIGN(TIME_WAIT, tn); XASSIGN(CLOSE, tn); XASSIGN(RETRANS, tn); XASSIGN(UNACK, tn); #undef XASSIGN #define XASSIGN(XNAME, rval) \ table[NF_SYSCTL_CT_PROTO_TCP_ ## XNAME].data = (rval) XASSIGN(LOOSE, &tn->tcp_loose); XASSIGN(LIBERAL, &tn->tcp_be_liberal); XASSIGN(MAX_RETRANS, &tn->tcp_max_retrans); XASSIGN(IGNORE_INVALID_RST, &tn->tcp_ignore_invalid_rst); #undef XASSIGN #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) table[NF_SYSCTL_CT_PROTO_TIMEOUT_TCP_OFFLOAD].data = &tn->offload_timeout; #endif } static void nf_conntrack_standalone_init_sctp_sysctl(struct net *net, struct ctl_table *table) { #ifdef CONFIG_NF_CT_PROTO_SCTP struct nf_sctp_net *sn = nf_sctp_pernet(net); #define XASSIGN(XNAME, sn) \ table[NF_SYSCTL_CT_PROTO_TIMEOUT_SCTP_ ## XNAME].data = \ &(sn)->timeouts[SCTP_CONNTRACK_ ## XNAME] XASSIGN(CLOSED, sn); XASSIGN(COOKIE_WAIT, sn); XASSIGN(COOKIE_ECHOED, sn); XASSIGN(ESTABLISHED, sn); XASSIGN(SHUTDOWN_SENT, sn); XASSIGN(SHUTDOWN_RECD, sn); XASSIGN(SHUTDOWN_ACK_SENT, sn); XASSIGN(HEARTBEAT_SENT, sn); #undef XASSIGN #endif } static void nf_conntrack_standalone_init_gre_sysctl(struct net *net, struct ctl_table *table) { #ifdef CONFIG_NF_CT_PROTO_GRE struct nf_gre_net *gn = nf_gre_pernet(net); table[NF_SYSCTL_CT_PROTO_TIMEOUT_GRE].data = &gn->timeouts[GRE_CT_UNREPLIED]; table[NF_SYSCTL_CT_PROTO_TIMEOUT_GRE_STREAM].data = &gn->timeouts[GRE_CT_REPLIED]; #endif } static int nf_conntrack_standalone_init_sysctl(struct net *net) { struct nf_conntrack_net *cnet = nf_ct_pernet(net); struct nf_udp_net *un = nf_udp_pernet(net); struct ctl_table *table; BUILD_BUG_ON(ARRAY_SIZE(nf_ct_sysctl_table) != NF_SYSCTL_CT_LAST_SYSCTL); table = kmemdup(nf_ct_sysctl_table, sizeof(nf_ct_sysctl_table), GFP_KERNEL); if (!table) return -ENOMEM; table[NF_SYSCTL_CT_COUNT].data = &cnet->count; table[NF_SYSCTL_CT_CHECKSUM].data = &net->ct.sysctl_checksum; table[NF_SYSCTL_CT_LOG_INVALID].data = &net->ct.sysctl_log_invalid; table[NF_SYSCTL_CT_ACCT].data = &net->ct.sysctl_acct; #ifdef CONFIG_NF_CONNTRACK_EVENTS table[NF_SYSCTL_CT_EVENTS].data = &net->ct.sysctl_events; #endif #ifdef CONFIG_NF_CONNTRACK_TIMESTAMP table[NF_SYSCTL_CT_TIMESTAMP].data = &net->ct.sysctl_tstamp; #endif table[NF_SYSCTL_CT_PROTO_TIMEOUT_GENERIC].data = &nf_generic_pernet(net)->timeout; table[NF_SYSCTL_CT_PROTO_TIMEOUT_ICMP].data = &nf_icmp_pernet(net)->timeout; table[NF_SYSCTL_CT_PROTO_TIMEOUT_ICMPV6].data = &nf_icmpv6_pernet(net)->timeout; table[NF_SYSCTL_CT_PROTO_TIMEOUT_UDP].data = &un->timeouts[UDP_CT_UNREPLIED]; table[NF_SYSCTL_CT_PROTO_TIMEOUT_UDP_STREAM].data = &un->timeouts[UDP_CT_REPLIED]; #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) table[NF_SYSCTL_CT_PROTO_TIMEOUT_UDP_OFFLOAD].data = &un->offload_timeout; #endif nf_conntrack_standalone_init_tcp_sysctl(net, table); nf_conntrack_standalone_init_sctp_sysctl(net, table); nf_conntrack_standalone_init_gre_sysctl(net, table); /* Don't allow non-init_net ns to alter global sysctls */ if (!net_eq(&init_net, net)) { table[NF_SYSCTL_CT_MAX].mode = 0444; table[NF_SYSCTL_CT_EXPECT_MAX].mode = 0444; table[NF_SYSCTL_CT_BUCKETS].mode = 0444; } cnet->sysctl_header = register_net_sysctl_sz(net, "net/netfilter", table, ARRAY_SIZE(nf_ct_sysctl_table)); if (!cnet->sysctl_header) goto out_unregister_netfilter; return 0; out_unregister_netfilter: kfree(table); return -ENOMEM; } static void nf_conntrack_standalone_fini_sysctl(struct net *net) { struct nf_conntrack_net *cnet = nf_ct_pernet(net); const struct ctl_table *table; table = cnet->sysctl_header->ctl_table_arg; unregister_net_sysctl_table(cnet->sysctl_header); kfree(table); } #else static int nf_conntrack_standalone_init_sysctl(struct net *net) { return 0; } static void nf_conntrack_standalone_fini_sysctl(struct net *net) { } #endif /* CONFIG_SYSCTL */ static void nf_conntrack_fini_net(struct net *net) { if (enable_hooks) nf_ct_netns_put(net, NFPROTO_INET); nf_conntrack_standalone_fini_proc(net); nf_conntrack_standalone_fini_sysctl(net); } static int nf_conntrack_pernet_init(struct net *net) { int ret; net->ct.sysctl_checksum = 1; ret = nf_conntrack_standalone_init_sysctl(net); if (ret < 0) return ret; ret = nf_conntrack_standalone_init_proc(net); if (ret < 0) goto out_proc; ret = nf_conntrack_init_net(net); if (ret < 0) goto out_init_net; if (enable_hooks) { ret = nf_ct_netns_get(net, NFPROTO_INET); if (ret < 0) goto out_hooks; } return 0; out_hooks: nf_conntrack_cleanup_net(net); out_init_net: nf_conntrack_standalone_fini_proc(net); out_proc: nf_conntrack_standalone_fini_sysctl(net); return ret; } static void nf_conntrack_pernet_exit(struct list_head *net_exit_list) { struct net *net; list_for_each_entry(net, net_exit_list, exit_list) nf_conntrack_fini_net(net); nf_conntrack_cleanup_net_list(net_exit_list); } static struct pernet_operations nf_conntrack_net_ops = { .init = nf_conntrack_pernet_init, .exit_batch = nf_conntrack_pernet_exit, .id = &nf_conntrack_net_id, .size = sizeof(struct nf_conntrack_net), }; static int __init nf_conntrack_standalone_init(void) { int ret = nf_conntrack_init_start(); if (ret < 0) goto out_start; BUILD_BUG_ON(NFCT_INFOMASK <= IP_CT_NUMBER); #ifdef CONFIG_SYSCTL nf_ct_netfilter_header = register_net_sysctl(&init_net, "net", nf_ct_netfilter_table); if (!nf_ct_netfilter_header) { pr_err("nf_conntrack: can't register to sysctl.\n"); ret = -ENOMEM; goto out_sysctl; } nf_conntrack_htable_size_user = nf_conntrack_htable_size; #endif nf_conntrack_init_end(); ret = register_pernet_subsys(&nf_conntrack_net_ops); if (ret < 0) goto out_pernet; return 0; out_pernet: #ifdef CONFIG_SYSCTL unregister_net_sysctl_table(nf_ct_netfilter_header); out_sysctl: #endif nf_conntrack_cleanup_end(); out_start: return ret; } static void __exit nf_conntrack_standalone_fini(void) { nf_conntrack_cleanup_start(); unregister_pernet_subsys(&nf_conntrack_net_ops); #ifdef CONFIG_SYSCTL unregister_net_sysctl_table(nf_ct_netfilter_header); #endif nf_conntrack_cleanup_end(); } module_init(nf_conntrack_standalone_init); module_exit(nf_conntrack_standalone_fini); |
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 | // SPDX-License-Identifier: GPL-2.0+ /* * Copyright (C) 2016 Oracle. All Rights Reserved. * Author: Darrick J. Wong <darrick.wong@oracle.com> */ #ifndef __XFS_REFCOUNT_ITEM_H__ #define __XFS_REFCOUNT_ITEM_H__ /* * There are (currently) two pairs of refcount btree redo item types: * increase and decrease. The log items for these are CUI (refcount * update intent) and CUD (refcount update done). The redo item type * is encoded in the flags field of each xfs_map_extent. * * *I items should be recorded in the *first* of a series of rolled * transactions, and the *D items should be recorded in the same * transaction that records the associated refcountbt updates. * * Should the system crash after the commit of the first transaction * but before the commit of the final transaction in a series, log * recovery will use the redo information recorded by the intent items * to replay the refcountbt metadata updates. */ /* kernel only CUI/CUD definitions */ struct xfs_mount; struct kmem_cache; /* * Max number of extents in fast allocation path. */ #define XFS_CUI_MAX_FAST_EXTENTS 16 /* * This is the "refcount update intent" log item. It is used to log * the fact that some reverse mappings need to change. It is used in * conjunction with the "refcount update done" log item described * below. * * These log items follow the same rules as struct xfs_efi_log_item; * see the comments about that structure (in xfs_extfree_item.h) for * more details. */ struct xfs_cui_log_item { struct xfs_log_item cui_item; atomic_t cui_refcount; atomic_t cui_next_extent; struct xfs_cui_log_format cui_format; }; static inline size_t xfs_cui_log_item_sizeof( unsigned int nr) { return offsetof(struct xfs_cui_log_item, cui_format) + xfs_cui_log_format_sizeof(nr); } /* * This is the "refcount update done" log item. It is used to log the * fact that some refcountbt updates mentioned in an earlier cui item * have been performed. */ struct xfs_cud_log_item { struct xfs_log_item cud_item; struct xfs_cui_log_item *cud_cuip; struct xfs_cud_log_format cud_format; }; extern struct kmem_cache *xfs_cui_cache; extern struct kmem_cache *xfs_cud_cache; struct xfs_refcount_intent; void xfs_refcount_defer_add(struct xfs_trans *tp, struct xfs_refcount_intent *ri); unsigned int xfs_cui_log_space(unsigned int nr); unsigned int xfs_cud_log_space(void); #endif /* __XFS_REFCOUNT_ITEM_H__ */ |
| 12 2 5 5 4 5 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 | /* SPDX-License-Identifier: GPL-2.0 */ #include <linux/kernel.h> #include <linux/init.h> #include <linux/module.h> #include <linux/seqlock.h> #include <linux/netlink.h> #include <linux/netfilter.h> #include <linux/netfilter/nf_tables.h> #include <net/netfilter/nf_tables.h> #include <net/dst_metadata.h> #include <net/ip_tunnels.h> #include <net/vxlan.h> #include <net/erspan.h> #include <net/geneve.h> struct nft_tunnel { enum nft_tunnel_keys key:8; u8 dreg; enum nft_tunnel_mode mode:8; u8 len; }; static void nft_tunnel_get_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) { const struct nft_tunnel *priv = nft_expr_priv(expr); u32 *dest = ®s->data[priv->dreg]; struct ip_tunnel_info *tun_info; tun_info = skb_tunnel_info(pkt->skb); switch (priv->key) { case NFT_TUNNEL_PATH: if (!tun_info) { nft_reg_store8(dest, false); return; } if (priv->mode == NFT_TUNNEL_MODE_NONE || (priv->mode == NFT_TUNNEL_MODE_RX && !(tun_info->mode & IP_TUNNEL_INFO_TX)) || (priv->mode == NFT_TUNNEL_MODE_TX && (tun_info->mode & IP_TUNNEL_INFO_TX))) nft_reg_store8(dest, true); else nft_reg_store8(dest, false); break; case NFT_TUNNEL_ID: if (!tun_info) { regs->verdict.code = NFT_BREAK; return; } if (priv->mode == NFT_TUNNEL_MODE_NONE || (priv->mode == NFT_TUNNEL_MODE_RX && !(tun_info->mode & IP_TUNNEL_INFO_TX)) || (priv->mode == NFT_TUNNEL_MODE_TX && (tun_info->mode & IP_TUNNEL_INFO_TX))) *dest = ntohl(tunnel_id_to_key32(tun_info->key.tun_id)); else regs->verdict.code = NFT_BREAK; break; default: WARN_ON(1); regs->verdict.code = NFT_BREAK; } } static const struct nla_policy nft_tunnel_policy[NFTA_TUNNEL_MAX + 1] = { [NFTA_TUNNEL_KEY] = NLA_POLICY_MAX(NLA_BE32, 255), [NFTA_TUNNEL_DREG] = { .type = NLA_U32 }, [NFTA_TUNNEL_MODE] = NLA_POLICY_MAX(NLA_BE32, 255), }; static int nft_tunnel_get_init(const struct nft_ctx *ctx, const struct nft_expr *expr, const struct nlattr * const tb[]) { struct nft_tunnel *priv = nft_expr_priv(expr); u32 len; if (!tb[NFTA_TUNNEL_KEY] || !tb[NFTA_TUNNEL_DREG]) return -EINVAL; priv->key = ntohl(nla_get_be32(tb[NFTA_TUNNEL_KEY])); switch (priv->key) { case NFT_TUNNEL_PATH: len = sizeof(u8); break; case NFT_TUNNEL_ID: len = sizeof(u32); break; default: return -EOPNOTSUPP; } if (tb[NFTA_TUNNEL_MODE]) { priv->mode = ntohl(nla_get_be32(tb[NFTA_TUNNEL_MODE])); if (priv->mode > NFT_TUNNEL_MODE_MAX) return -EOPNOTSUPP; } else { priv->mode = NFT_TUNNEL_MODE_NONE; } priv->len = len; return nft_parse_register_store(ctx, tb[NFTA_TUNNEL_DREG], &priv->dreg, NULL, NFT_DATA_VALUE, len); } static int nft_tunnel_get_dump(struct sk_buff *skb, const struct nft_expr *expr, bool reset) { const struct nft_tunnel *priv = nft_expr_priv(expr); if (nla_put_be32(skb, NFTA_TUNNEL_KEY, htonl(priv->key))) goto nla_put_failure; if (nft_dump_register(skb, NFTA_TUNNEL_DREG, priv->dreg)) goto nla_put_failure; if (nla_put_be32(skb, NFTA_TUNNEL_MODE, htonl(priv->mode))) goto nla_put_failure; return 0; nla_put_failure: return -1; } static bool nft_tunnel_get_reduce(struct nft_regs_track *track, const struct nft_expr *expr) { const struct nft_tunnel *priv = nft_expr_priv(expr); const struct nft_tunnel *tunnel; if (!nft_reg_track_cmp(track, expr, priv->dreg)) { nft_reg_track_update(track, expr, priv->dreg, priv->len); return false; } tunnel = nft_expr_priv(track->regs[priv->dreg].selector); if (priv->key != tunnel->key || priv->dreg != tunnel->dreg || priv->mode != tunnel->mode) { nft_reg_track_update(track, expr, priv->dreg, priv->len); return false; } if (!track->regs[priv->dreg].bitwise) return true; return false; } static struct nft_expr_type nft_tunnel_type; static const struct nft_expr_ops nft_tunnel_get_ops = { .type = &nft_tunnel_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_tunnel)), .eval = nft_tunnel_get_eval, .init = nft_tunnel_get_init, .dump = nft_tunnel_get_dump, .reduce = nft_tunnel_get_reduce, }; static struct nft_expr_type nft_tunnel_type __read_mostly = { .name = "tunnel", .family = NFPROTO_NETDEV, .ops = &nft_tunnel_get_ops, .policy = nft_tunnel_policy, .maxattr = NFTA_TUNNEL_MAX, .owner = THIS_MODULE, }; struct nft_tunnel_opts { union { struct vxlan_metadata vxlan; struct erspan_metadata erspan; u8 data[IP_TUNNEL_OPTS_MAX]; } u; IP_TUNNEL_DECLARE_FLAGS(flags); u32 len; }; struct nft_tunnel_obj { struct metadata_dst *md; struct nft_tunnel_opts opts; }; static const struct nla_policy nft_tunnel_ip_policy[NFTA_TUNNEL_KEY_IP_MAX + 1] = { [NFTA_TUNNEL_KEY_IP_SRC] = { .type = NLA_U32 }, [NFTA_TUNNEL_KEY_IP_DST] = { .type = NLA_U32 }, }; static int nft_tunnel_obj_ip_init(const struct nft_ctx *ctx, const struct nlattr *attr, struct ip_tunnel_info *info) { struct nlattr *tb[NFTA_TUNNEL_KEY_IP_MAX + 1]; int err; err = nla_parse_nested_deprecated(tb, NFTA_TUNNEL_KEY_IP_MAX, attr, nft_tunnel_ip_policy, NULL); if (err < 0) return err; if (!tb[NFTA_TUNNEL_KEY_IP_DST]) return -EINVAL; if (tb[NFTA_TUNNEL_KEY_IP_SRC]) info->key.u.ipv4.src = nla_get_be32(tb[NFTA_TUNNEL_KEY_IP_SRC]); if (tb[NFTA_TUNNEL_KEY_IP_DST]) info->key.u.ipv4.dst = nla_get_be32(tb[NFTA_TUNNEL_KEY_IP_DST]); return 0; } static const struct nla_policy nft_tunnel_ip6_policy[NFTA_TUNNEL_KEY_IP6_MAX + 1] = { [NFTA_TUNNEL_KEY_IP6_SRC] = { .len = sizeof(struct in6_addr), }, [NFTA_TUNNEL_KEY_IP6_DST] = { .len = sizeof(struct in6_addr), }, [NFTA_TUNNEL_KEY_IP6_FLOWLABEL] = { .type = NLA_U32, } }; static int nft_tunnel_obj_ip6_init(const struct nft_ctx *ctx, const struct nlattr *attr, struct ip_tunnel_info *info) { struct nlattr *tb[NFTA_TUNNEL_KEY_IP6_MAX + 1]; int err; err = nla_parse_nested_deprecated(tb, NFTA_TUNNEL_KEY_IP6_MAX, attr, nft_tunnel_ip6_policy, NULL); if (err < 0) return err; if (!tb[NFTA_TUNNEL_KEY_IP6_DST]) return -EINVAL; if (tb[NFTA_TUNNEL_KEY_IP6_SRC]) { memcpy(&info->key.u.ipv6.src, nla_data(tb[NFTA_TUNNEL_KEY_IP6_SRC]), sizeof(struct in6_addr)); } if (tb[NFTA_TUNNEL_KEY_IP6_DST]) { memcpy(&info->key.u.ipv6.dst, nla_data(tb[NFTA_TUNNEL_KEY_IP6_DST]), sizeof(struct in6_addr)); } if (tb[NFTA_TUNNEL_KEY_IP6_FLOWLABEL]) info->key.label = nla_get_be32(tb[NFTA_TUNNEL_KEY_IP6_FLOWLABEL]); info->mode |= IP_TUNNEL_INFO_IPV6; return 0; } static const struct nla_policy nft_tunnel_opts_vxlan_policy[NFTA_TUNNEL_KEY_VXLAN_MAX + 1] = { [NFTA_TUNNEL_KEY_VXLAN_GBP] = { .type = NLA_U32 }, }; static int nft_tunnel_obj_vxlan_init(const struct nlattr *attr, struct nft_tunnel_opts *opts) { struct nlattr *tb[NFTA_TUNNEL_KEY_VXLAN_MAX + 1]; int err; err = nla_parse_nested_deprecated(tb, NFTA_TUNNEL_KEY_VXLAN_MAX, attr, nft_tunnel_opts_vxlan_policy, NULL); if (err < 0) return err; if (!tb[NFTA_TUNNEL_KEY_VXLAN_GBP]) return -EINVAL; opts->u.vxlan.gbp = ntohl(nla_get_be32(tb[NFTA_TUNNEL_KEY_VXLAN_GBP])); opts->len = sizeof(struct vxlan_metadata); ip_tunnel_flags_zero(opts->flags); __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, opts->flags); return 0; } static const struct nla_policy nft_tunnel_opts_erspan_policy[NFTA_TUNNEL_KEY_ERSPAN_MAX + 1] = { [NFTA_TUNNEL_KEY_ERSPAN_VERSION] = { .type = NLA_U32 }, [NFTA_TUNNEL_KEY_ERSPAN_V1_INDEX] = { .type = NLA_U32 }, [NFTA_TUNNEL_KEY_ERSPAN_V2_DIR] = { .type = NLA_U8 }, [NFTA_TUNNEL_KEY_ERSPAN_V2_HWID] = { .type = NLA_U8 }, }; static int nft_tunnel_obj_erspan_init(const struct nlattr *attr, struct nft_tunnel_opts *opts) { struct nlattr *tb[NFTA_TUNNEL_KEY_ERSPAN_MAX + 1]; uint8_t hwid, dir; int err, version; err = nla_parse_nested_deprecated(tb, NFTA_TUNNEL_KEY_ERSPAN_MAX, attr, nft_tunnel_opts_erspan_policy, NULL); if (err < 0) return err; if (!tb[NFTA_TUNNEL_KEY_ERSPAN_VERSION]) return -EINVAL; version = ntohl(nla_get_be32(tb[NFTA_TUNNEL_KEY_ERSPAN_VERSION])); switch (version) { case ERSPAN_VERSION: if (!tb[NFTA_TUNNEL_KEY_ERSPAN_V1_INDEX]) return -EINVAL; opts->u.erspan.u.index = nla_get_be32(tb[NFTA_TUNNEL_KEY_ERSPAN_V1_INDEX]); break; case ERSPAN_VERSION2: if (!tb[NFTA_TUNNEL_KEY_ERSPAN_V2_DIR] || !tb[NFTA_TUNNEL_KEY_ERSPAN_V2_HWID]) return -EINVAL; hwid = nla_get_u8(tb[NFTA_TUNNEL_KEY_ERSPAN_V2_HWID]); dir = nla_get_u8(tb[NFTA_TUNNEL_KEY_ERSPAN_V2_DIR]); set_hwid(&opts->u.erspan.u.md2, hwid); opts->u.erspan.u.md2.dir = dir; break; default: return -EOPNOTSUPP; } opts->u.erspan.version = version; opts->len = sizeof(struct erspan_metadata); ip_tunnel_flags_zero(opts->flags); __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, opts->flags); return 0; } static const struct nla_policy nft_tunnel_opts_geneve_policy[NFTA_TUNNEL_KEY_GENEVE_MAX + 1] = { [NFTA_TUNNEL_KEY_GENEVE_CLASS] = { .type = NLA_U16 }, [NFTA_TUNNEL_KEY_GENEVE_TYPE] = { .type = NLA_U8 }, [NFTA_TUNNEL_KEY_GENEVE_DATA] = { .type = NLA_BINARY, .len = 127 }, }; static int nft_tunnel_obj_geneve_init(const struct nlattr *attr, struct nft_tunnel_opts *opts) { struct geneve_opt *opt = (struct geneve_opt *)(opts->u.data + opts->len); struct nlattr *tb[NFTA_TUNNEL_KEY_GENEVE_MAX + 1]; int err, data_len; err = nla_parse_nested(tb, NFTA_TUNNEL_KEY_GENEVE_MAX, attr, nft_tunnel_opts_geneve_policy, NULL); if (err < 0) return err; if (!tb[NFTA_TUNNEL_KEY_GENEVE_CLASS] || !tb[NFTA_TUNNEL_KEY_GENEVE_TYPE] || !tb[NFTA_TUNNEL_KEY_GENEVE_DATA]) return -EINVAL; attr = tb[NFTA_TUNNEL_KEY_GENEVE_DATA]; data_len = nla_len(attr); if (data_len % 4) return -EINVAL; opts->len += sizeof(*opt) + data_len; if (opts->len > IP_TUNNEL_OPTS_MAX) return -EINVAL; memcpy(opt->opt_data, nla_data(attr), data_len); opt->length = data_len / 4; opt->opt_class = nla_get_be16(tb[NFTA_TUNNEL_KEY_GENEVE_CLASS]); opt->type = nla_get_u8(tb[NFTA_TUNNEL_KEY_GENEVE_TYPE]); ip_tunnel_flags_zero(opts->flags); __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, opts->flags); return 0; } static const struct nla_policy nft_tunnel_opts_policy[NFTA_TUNNEL_KEY_OPTS_MAX + 1] = { [NFTA_TUNNEL_KEY_OPTS_UNSPEC] = { .strict_start_type = NFTA_TUNNEL_KEY_OPTS_GENEVE }, [NFTA_TUNNEL_KEY_OPTS_VXLAN] = { .type = NLA_NESTED, }, [NFTA_TUNNEL_KEY_OPTS_ERSPAN] = { .type = NLA_NESTED, }, [NFTA_TUNNEL_KEY_OPTS_GENEVE] = { .type = NLA_NESTED, }, }; static int nft_tunnel_obj_opts_init(const struct nft_ctx *ctx, const struct nlattr *attr, struct ip_tunnel_info *info, struct nft_tunnel_opts *opts) { struct nlattr *nla; int err, rem; u32 type = 0; err = nla_validate_nested_deprecated(attr, NFTA_TUNNEL_KEY_OPTS_MAX, nft_tunnel_opts_policy, NULL); if (err < 0) return err; nla_for_each_attr(nla, nla_data(attr), nla_len(attr), rem) { switch (nla_type(nla)) { case NFTA_TUNNEL_KEY_OPTS_VXLAN: if (type) return -EINVAL; err = nft_tunnel_obj_vxlan_init(nla, opts); if (err) return err; type = IP_TUNNEL_VXLAN_OPT_BIT; break; case NFTA_TUNNEL_KEY_OPTS_ERSPAN: if (type) return -EINVAL; err = nft_tunnel_obj_erspan_init(nla, opts); if (err) return err; type = IP_TUNNEL_ERSPAN_OPT_BIT; break; case NFTA_TUNNEL_KEY_OPTS_GENEVE: if (type && type != IP_TUNNEL_GENEVE_OPT_BIT) return -EINVAL; err = nft_tunnel_obj_geneve_init(nla, opts); if (err) return err; type = IP_TUNNEL_GENEVE_OPT_BIT; break; default: return -EOPNOTSUPP; } } return err; } static const struct nla_policy nft_tunnel_key_policy[NFTA_TUNNEL_KEY_MAX + 1] = { [NFTA_TUNNEL_KEY_IP] = { .type = NLA_NESTED, }, [NFTA_TUNNEL_KEY_IP6] = { .type = NLA_NESTED, }, [NFTA_TUNNEL_KEY_ID] = { .type = NLA_U32, }, [NFTA_TUNNEL_KEY_FLAGS] = { .type = NLA_U32, }, [NFTA_TUNNEL_KEY_TOS] = { .type = NLA_U8, }, [NFTA_TUNNEL_KEY_TTL] = { .type = NLA_U8, }, [NFTA_TUNNEL_KEY_SPORT] = { .type = NLA_U16, }, [NFTA_TUNNEL_KEY_DPORT] = { .type = NLA_U16, }, [NFTA_TUNNEL_KEY_OPTS] = { .type = NLA_NESTED, }, }; static int nft_tunnel_obj_init(const struct nft_ctx *ctx, const struct nlattr * const tb[], struct nft_object *obj) { struct nft_tunnel_obj *priv = nft_obj_data(obj); struct ip_tunnel_info info; struct metadata_dst *md; int err; if (!tb[NFTA_TUNNEL_KEY_ID]) return -EINVAL; memset(&info, 0, sizeof(info)); info.mode = IP_TUNNEL_INFO_TX; info.key.tun_id = key32_to_tunnel_id(nla_get_be32(tb[NFTA_TUNNEL_KEY_ID])); __set_bit(IP_TUNNEL_KEY_BIT, info.key.tun_flags); __set_bit(IP_TUNNEL_CSUM_BIT, info.key.tun_flags); __set_bit(IP_TUNNEL_NOCACHE_BIT, info.key.tun_flags); if (tb[NFTA_TUNNEL_KEY_IP]) { err = nft_tunnel_obj_ip_init(ctx, tb[NFTA_TUNNEL_KEY_IP], &info); if (err < 0) return err; } else if (tb[NFTA_TUNNEL_KEY_IP6]) { err = nft_tunnel_obj_ip6_init(ctx, tb[NFTA_TUNNEL_KEY_IP6], &info); if (err < 0) return err; } else { return -EINVAL; } if (tb[NFTA_TUNNEL_KEY_SPORT]) { info.key.tp_src = nla_get_be16(tb[NFTA_TUNNEL_KEY_SPORT]); } if (tb[NFTA_TUNNEL_KEY_DPORT]) { info.key.tp_dst = nla_get_be16(tb[NFTA_TUNNEL_KEY_DPORT]); } if (tb[NFTA_TUNNEL_KEY_FLAGS]) { u32 tun_flags; tun_flags = ntohl(nla_get_be32(tb[NFTA_TUNNEL_KEY_FLAGS])); if (tun_flags & ~NFT_TUNNEL_F_MASK) return -EOPNOTSUPP; if (tun_flags & NFT_TUNNEL_F_ZERO_CSUM_TX) __clear_bit(IP_TUNNEL_CSUM_BIT, info.key.tun_flags); if (tun_flags & NFT_TUNNEL_F_DONT_FRAGMENT) __set_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, info.key.tun_flags); if (tun_flags & NFT_TUNNEL_F_SEQ_NUMBER) __set_bit(IP_TUNNEL_SEQ_BIT, info.key.tun_flags); } if (tb[NFTA_TUNNEL_KEY_TOS]) info.key.tos = nla_get_u8(tb[NFTA_TUNNEL_KEY_TOS]); info.key.ttl = nla_get_u8_default(tb[NFTA_TUNNEL_KEY_TTL], U8_MAX); if (tb[NFTA_TUNNEL_KEY_OPTS]) { err = nft_tunnel_obj_opts_init(ctx, tb[NFTA_TUNNEL_KEY_OPTS], &info, &priv->opts); if (err < 0) return err; } md = metadata_dst_alloc(priv->opts.len, METADATA_IP_TUNNEL, GFP_KERNEL_ACCOUNT); if (!md) return -ENOMEM; memcpy(&md->u.tun_info, &info, sizeof(info)); #ifdef CONFIG_DST_CACHE err = dst_cache_init(&md->u.tun_info.dst_cache, GFP_KERNEL_ACCOUNT); if (err < 0) { metadata_dst_free(md); return err; } #endif ip_tunnel_info_opts_set(&md->u.tun_info, &priv->opts.u, priv->opts.len, priv->opts.flags); priv->md = md; return 0; } static inline void nft_tunnel_obj_eval(struct nft_object *obj, struct nft_regs *regs, const struct nft_pktinfo *pkt) { struct nft_tunnel_obj *priv = nft_obj_data(obj); struct sk_buff *skb = pkt->skb; skb_dst_drop(skb); dst_hold((struct dst_entry *) priv->md); skb_dst_set(skb, (struct dst_entry *) priv->md); } static int nft_tunnel_ip_dump(struct sk_buff *skb, struct ip_tunnel_info *info) { struct nlattr *nest; if (info->mode & IP_TUNNEL_INFO_IPV6) { nest = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_IP6); if (!nest) return -1; if (nla_put_in6_addr(skb, NFTA_TUNNEL_KEY_IP6_SRC, &info->key.u.ipv6.src) < 0 || nla_put_in6_addr(skb, NFTA_TUNNEL_KEY_IP6_DST, &info->key.u.ipv6.dst) < 0 || nla_put_be32(skb, NFTA_TUNNEL_KEY_IP6_FLOWLABEL, info->key.label)) { nla_nest_cancel(skb, nest); return -1; } nla_nest_end(skb, nest); } else { nest = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_IP); if (!nest) return -1; if (nla_put_in_addr(skb, NFTA_TUNNEL_KEY_IP_SRC, info->key.u.ipv4.src) < 0 || nla_put_in_addr(skb, NFTA_TUNNEL_KEY_IP_DST, info->key.u.ipv4.dst) < 0) { nla_nest_cancel(skb, nest); return -1; } nla_nest_end(skb, nest); } return 0; } static int nft_tunnel_opts_dump(struct sk_buff *skb, struct nft_tunnel_obj *priv) { struct nft_tunnel_opts *opts = &priv->opts; struct nlattr *nest, *inner; nest = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS); if (!nest) return -1; if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, opts->flags)) { inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_VXLAN); if (!inner) goto failure; if (nla_put_be32(skb, NFTA_TUNNEL_KEY_VXLAN_GBP, htonl(opts->u.vxlan.gbp))) goto inner_failure; nla_nest_end(skb, inner); } else if (test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, opts->flags)) { inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_ERSPAN); if (!inner) goto failure; if (nla_put_be32(skb, NFTA_TUNNEL_KEY_ERSPAN_VERSION, htonl(opts->u.erspan.version))) goto inner_failure; switch (opts->u.erspan.version) { case ERSPAN_VERSION: if (nla_put_be32(skb, NFTA_TUNNEL_KEY_ERSPAN_V1_INDEX, opts->u.erspan.u.index)) goto inner_failure; break; case ERSPAN_VERSION2: if (nla_put_u8(skb, NFTA_TUNNEL_KEY_ERSPAN_V2_HWID, get_hwid(&opts->u.erspan.u.md2)) || nla_put_u8(skb, NFTA_TUNNEL_KEY_ERSPAN_V2_DIR, opts->u.erspan.u.md2.dir)) goto inner_failure; break; } nla_nest_end(skb, inner); } else if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, opts->flags)) { struct geneve_opt *opt; int offset = 0; while (opts->len > offset) { inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_GENEVE); if (!inner) goto failure; opt = (struct geneve_opt *)(opts->u.data + offset); if (nla_put_be16(skb, NFTA_TUNNEL_KEY_GENEVE_CLASS, opt->opt_class) || nla_put_u8(skb, NFTA_TUNNEL_KEY_GENEVE_TYPE, opt->type) || nla_put(skb, NFTA_TUNNEL_KEY_GENEVE_DATA, opt->length * 4, opt->opt_data)) goto inner_failure; offset += sizeof(*opt) + opt->length * 4; nla_nest_end(skb, inner); } } nla_nest_end(skb, nest); return 0; inner_failure: nla_nest_cancel(skb, inner); failure: nla_nest_cancel(skb, nest); return -1; } static int nft_tunnel_ports_dump(struct sk_buff *skb, struct ip_tunnel_info *info) { if (nla_put_be16(skb, NFTA_TUNNEL_KEY_SPORT, info->key.tp_src) < 0 || nla_put_be16(skb, NFTA_TUNNEL_KEY_DPORT, info->key.tp_dst) < 0) return -1; return 0; } static int nft_tunnel_flags_dump(struct sk_buff *skb, struct ip_tunnel_info *info) { u32 flags = 0; if (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, info->key.tun_flags)) flags |= NFT_TUNNEL_F_DONT_FRAGMENT; if (!test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags)) flags |= NFT_TUNNEL_F_ZERO_CSUM_TX; if (test_bit(IP_TUNNEL_SEQ_BIT, info->key.tun_flags)) flags |= NFT_TUNNEL_F_SEQ_NUMBER; if (nla_put_be32(skb, NFTA_TUNNEL_KEY_FLAGS, htonl(flags)) < 0) return -1; return 0; } static int nft_tunnel_obj_dump(struct sk_buff *skb, struct nft_object *obj, bool reset) { struct nft_tunnel_obj *priv = nft_obj_data(obj); struct ip_tunnel_info *info = &priv->md->u.tun_info; if (nla_put_be32(skb, NFTA_TUNNEL_KEY_ID, tunnel_id_to_key32(info->key.tun_id)) || nft_tunnel_ip_dump(skb, info) < 0 || nft_tunnel_ports_dump(skb, info) < 0 || nft_tunnel_flags_dump(skb, info) < 0 || nla_put_u8(skb, NFTA_TUNNEL_KEY_TOS, info->key.tos) || nla_put_u8(skb, NFTA_TUNNEL_KEY_TTL, info->key.ttl) || nft_tunnel_opts_dump(skb, priv) < 0) goto nla_put_failure; return 0; nla_put_failure: return -1; } static void nft_tunnel_obj_destroy(const struct nft_ctx *ctx, struct nft_object *obj) { struct nft_tunnel_obj *priv = nft_obj_data(obj); metadata_dst_free(priv->md); } static struct nft_object_type nft_tunnel_obj_type; static const struct nft_object_ops nft_tunnel_obj_ops = { .type = &nft_tunnel_obj_type, .size = sizeof(struct nft_tunnel_obj), .eval = nft_tunnel_obj_eval, .init = nft_tunnel_obj_init, .destroy = nft_tunnel_obj_destroy, .dump = nft_tunnel_obj_dump, }; static struct nft_object_type nft_tunnel_obj_type __read_mostly = { .type = NFT_OBJECT_TUNNEL, .family = NFPROTO_NETDEV, .ops = &nft_tunnel_obj_ops, .maxattr = NFTA_TUNNEL_KEY_MAX, .policy = nft_tunnel_key_policy, .owner = THIS_MODULE, }; static int __init nft_tunnel_module_init(void) { int err; err = nft_register_expr(&nft_tunnel_type); if (err < 0) return err; err = nft_register_obj(&nft_tunnel_obj_type); if (err < 0) nft_unregister_expr(&nft_tunnel_type); return err; } static void __exit nft_tunnel_module_exit(void) { nft_unregister_obj(&nft_tunnel_obj_type); nft_unregister_expr(&nft_tunnel_type); } module_init(nft_tunnel_module_init); module_exit(nft_tunnel_module_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>"); MODULE_ALIAS_NFT_EXPR("tunnel"); MODULE_ALIAS_NFT_OBJ(NFT_OBJECT_TUNNEL); MODULE_DESCRIPTION("nftables tunnel expression support"); |
| 32 40 42 4 11 42 34 32 8 32 4 4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 | // SPDX-License-Identifier: GPL-2.0-or-later /* * Cryptographic API. * * RNG operations. * * Copyright (c) 2008 Neil Horman <nhorman@tuxdriver.com> * Copyright (c) 2015 Herbert Xu <herbert@gondor.apana.org.au> */ #include <crypto/internal/rng.h> #include <linux/atomic.h> #include <linux/cryptouser.h> #include <linux/err.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/mutex.h> #include <linux/random.h> #include <linux/seq_file.h> #include <linux/slab.h> #include <linux/string.h> #include <net/netlink.h> #include "internal.h" static DEFINE_MUTEX(crypto_default_rng_lock); struct crypto_rng *crypto_default_rng; EXPORT_SYMBOL_GPL(crypto_default_rng); static int crypto_default_rng_refcnt; int crypto_rng_reset(struct crypto_rng *tfm, const u8 *seed, unsigned int slen) { u8 *buf = NULL; int err; if (!seed && slen) { buf = kmalloc(slen, GFP_KERNEL); if (!buf) return -ENOMEM; err = get_random_bytes_wait(buf, slen); if (err) goto out; seed = buf; } err = crypto_rng_alg(tfm)->seed(tfm, seed, slen); out: kfree_sensitive(buf); return err; } EXPORT_SYMBOL_GPL(crypto_rng_reset); static int crypto_rng_init_tfm(struct crypto_tfm *tfm) { return 0; } static unsigned int seedsize(struct crypto_alg *alg) { struct rng_alg *ralg = container_of(alg, struct rng_alg, base); return ralg->seedsize; } static int __maybe_unused crypto_rng_report( struct sk_buff *skb, struct crypto_alg *alg) { struct crypto_report_rng rrng; memset(&rrng, 0, sizeof(rrng)); strscpy(rrng.type, "rng", sizeof(rrng.type)); rrng.seedsize = seedsize(alg); return nla_put(skb, CRYPTOCFGA_REPORT_RNG, sizeof(rrng), &rrng); } static void __maybe_unused crypto_rng_show(struct seq_file *m, struct crypto_alg *alg) { seq_printf(m, "type : rng\n"); seq_printf(m, "seedsize : %u\n", seedsize(alg)); } static const struct crypto_type crypto_rng_type = { .extsize = crypto_alg_extsize, .init_tfm = crypto_rng_init_tfm, #ifdef CONFIG_PROC_FS .show = crypto_rng_show, #endif #if IS_ENABLED(CONFIG_CRYPTO_USER) .report = crypto_rng_report, #endif .maskclear = ~CRYPTO_ALG_TYPE_MASK, .maskset = CRYPTO_ALG_TYPE_MASK, .type = CRYPTO_ALG_TYPE_RNG, .tfmsize = offsetof(struct crypto_rng, base), .algsize = offsetof(struct rng_alg, base), }; struct crypto_rng *crypto_alloc_rng(const char *alg_name, u32 type, u32 mask) { return crypto_alloc_tfm(alg_name, &crypto_rng_type, type, mask); } EXPORT_SYMBOL_GPL(crypto_alloc_rng); int crypto_get_default_rng(void) { struct crypto_rng *rng; int err; mutex_lock(&crypto_default_rng_lock); if (!crypto_default_rng) { rng = crypto_alloc_rng("stdrng", 0, 0); err = PTR_ERR(rng); if (IS_ERR(rng)) goto unlock; err = crypto_rng_reset(rng, NULL, crypto_rng_seedsize(rng)); if (err) { crypto_free_rng(rng); goto unlock; } crypto_default_rng = rng; } crypto_default_rng_refcnt++; err = 0; unlock: mutex_unlock(&crypto_default_rng_lock); return err; } EXPORT_SYMBOL_GPL(crypto_get_default_rng); void crypto_put_default_rng(void) { mutex_lock(&crypto_default_rng_lock); crypto_default_rng_refcnt--; mutex_unlock(&crypto_default_rng_lock); } EXPORT_SYMBOL_GPL(crypto_put_default_rng); #if defined(CONFIG_CRYPTO_RNG) || defined(CONFIG_CRYPTO_RNG_MODULE) int crypto_del_default_rng(void) { int err = -EBUSY; mutex_lock(&crypto_default_rng_lock); if (crypto_default_rng_refcnt) goto out; crypto_free_rng(crypto_default_rng); crypto_default_rng = NULL; err = 0; out: mutex_unlock(&crypto_default_rng_lock); return err; } EXPORT_SYMBOL_GPL(crypto_del_default_rng); #endif static void rng_default_set_ent(struct crypto_rng *tfm, const u8 *data, unsigned int len) { } int crypto_register_rng(struct rng_alg *alg) { struct crypto_alg *base = &alg->base; if (alg->seedsize > PAGE_SIZE / 8) return -EINVAL; base->cra_type = &crypto_rng_type; base->cra_flags &= ~CRYPTO_ALG_TYPE_MASK; base->cra_flags |= CRYPTO_ALG_TYPE_RNG; if (!alg->set_ent) alg->set_ent = rng_default_set_ent; return crypto_register_alg(base); } EXPORT_SYMBOL_GPL(crypto_register_rng); void crypto_unregister_rng(struct rng_alg *alg) { crypto_unregister_alg(&alg->base); } EXPORT_SYMBOL_GPL(crypto_unregister_rng); int crypto_register_rngs(struct rng_alg *algs, int count) { int i, ret; for (i = 0; i < count; i++) { ret = crypto_register_rng(algs + i); if (ret) goto err; } return 0; err: for (--i; i >= 0; --i) crypto_unregister_rng(algs + i); return ret; } EXPORT_SYMBOL_GPL(crypto_register_rngs); void crypto_unregister_rngs(struct rng_alg *algs, int count) { int i; for (i = count - 1; i >= 0; --i) crypto_unregister_rng(algs + i); } EXPORT_SYMBOL_GPL(crypto_unregister_rngs); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("Random Number Generator"); |
| 20 1 1 9 5 5 18 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 | // SPDX-License-Identifier: GPL-2.0-only /* iptables module for the packet checksum mangling * * (C) 2002 by Harald Welte <laforge@netfilter.org> * (C) 2010 Red Hat, Inc. * * Author: Michael S. Tsirkin <mst@redhat.com> */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/module.h> #include <linux/skbuff.h> #include <linux/netfilter/x_tables.h> #include <linux/netfilter/xt_CHECKSUM.h> #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv6/ip6_tables.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Michael S. Tsirkin <mst@redhat.com>"); MODULE_DESCRIPTION("Xtables: checksum modification"); MODULE_ALIAS("ipt_CHECKSUM"); MODULE_ALIAS("ip6t_CHECKSUM"); static unsigned int checksum_tg(struct sk_buff *skb, const struct xt_action_param *par) { if (skb->ip_summed == CHECKSUM_PARTIAL && !skb_is_gso(skb)) skb_checksum_help(skb); return XT_CONTINUE; } static int checksum_tg_check(const struct xt_tgchk_param *par) { const struct xt_CHECKSUM_info *einfo = par->targinfo; const struct ip6t_ip6 *i6 = par->entryinfo; const struct ipt_ip *i4 = par->entryinfo; if (einfo->operation & ~XT_CHECKSUM_OP_FILL) { pr_info_ratelimited("unsupported CHECKSUM operation %x\n", einfo->operation); return -EINVAL; } if (!einfo->operation) return -EINVAL; switch (par->family) { case NFPROTO_IPV4: if (i4->proto == IPPROTO_UDP && (i4->invflags & XT_INV_PROTO) == 0) return 0; break; case NFPROTO_IPV6: if ((i6->flags & IP6T_F_PROTO) && i6->proto == IPPROTO_UDP && (i6->invflags & XT_INV_PROTO) == 0) return 0; break; } pr_warn_once("CHECKSUM should be avoided. If really needed, restrict with \"-p udp\" and only use in OUTPUT\n"); return 0; } static struct xt_target checksum_tg_reg[] __read_mostly = { { .name = "CHECKSUM", .family = NFPROTO_IPV4, .target = checksum_tg, .targetsize = sizeof(struct xt_CHECKSUM_info), .table = "mangle", .checkentry = checksum_tg_check, .me = THIS_MODULE, }, #if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) { .name = "CHECKSUM", .family = NFPROTO_IPV6, .target = checksum_tg, .targetsize = sizeof(struct xt_CHECKSUM_info), .table = "mangle", .checkentry = checksum_tg_check, .me = THIS_MODULE, }, #endif }; static int __init checksum_tg_init(void) { return xt_register_targets(checksum_tg_reg, ARRAY_SIZE(checksum_tg_reg)); } static void __exit checksum_tg_exit(void) { xt_unregister_targets(checksum_tg_reg, ARRAY_SIZE(checksum_tg_reg)); } module_init(checksum_tg_init); module_exit(checksum_tg_exit); |
| 210 210 210 209 162 56 42 23 23 23 23 23 23 43 43 57 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 | // SPDX-License-Identifier: GPL-2.0-or-later /* * ip_vs_proto_udp.c: UDP load balancing support for IPVS * * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> * Julian Anastasov <ja@ssi.bg> * * Changes: Hans Schillstrom <hans.schillstrom@ericsson.com> * Network name space (netns) aware. */ #define pr_fmt(fmt) "IPVS: " fmt #include <linux/in.h> #include <linux/ip.h> #include <linux/kernel.h> #include <linux/netfilter.h> #include <linux/netfilter_ipv4.h> #include <linux/udp.h> #include <linux/indirect_call_wrapper.h> #include <net/ip_vs.h> #include <net/ip.h> #include <net/ip6_checksum.h> static int udp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp); static int udp_conn_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, int *verdict, struct ip_vs_conn **cpp, struct ip_vs_iphdr *iph) { struct ip_vs_service *svc; struct udphdr _udph, *uh; __be16 _ports[2], *ports = NULL; if (likely(!ip_vs_iph_icmp(iph))) { /* IPv6 fragments, only first fragment will hit this */ uh = skb_header_pointer(skb, iph->len, sizeof(_udph), &_udph); if (uh) ports = &uh->source; } else { ports = skb_header_pointer( skb, iph->len, sizeof(_ports), &_ports); } if (!ports) { *verdict = NF_DROP; return 0; } if (likely(!ip_vs_iph_inverse(iph))) svc = ip_vs_service_find(ipvs, af, skb->mark, iph->protocol, &iph->daddr, ports[1]); else svc = ip_vs_service_find(ipvs, af, skb->mark, iph->protocol, &iph->saddr, ports[0]); if (svc) { int ignored; if (ip_vs_todrop(ipvs)) { /* * It seems that we are very loaded. * We have to drop this packet :( */ *verdict = NF_DROP; return 0; } /* * Let the virtual server select a real server for the * incoming connection, and create a connection entry. */ *cpp = ip_vs_schedule(svc, skb, pd, &ignored, iph); if (!*cpp && ignored <= 0) { if (!ignored) *verdict = ip_vs_leave(svc, skb, pd, iph); else *verdict = NF_DROP; return 0; } } /* NF_ACCEPT */ return 1; } static inline void udp_fast_csum_update(int af, struct udphdr *uhdr, const union nf_inet_addr *oldip, const union nf_inet_addr *newip, __be16 oldport, __be16 newport) { #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) uhdr->check = csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, ip_vs_check_diff2(oldport, newport, ~csum_unfold(uhdr->check)))); else #endif uhdr->check = csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, ip_vs_check_diff2(oldport, newport, ~csum_unfold(uhdr->check)))); if (!uhdr->check) uhdr->check = CSUM_MANGLED_0; } static inline void udp_partial_csum_update(int af, struct udphdr *uhdr, const union nf_inet_addr *oldip, const union nf_inet_addr *newip, __be16 oldlen, __be16 newlen) { #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) uhdr->check = ~csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, ip_vs_check_diff2(oldlen, newlen, csum_unfold(uhdr->check)))); else #endif uhdr->check = ~csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, ip_vs_check_diff2(oldlen, newlen, csum_unfold(uhdr->check)))); } INDIRECT_CALLABLE_SCOPE int udp_snat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp, struct ip_vs_conn *cp, struct ip_vs_iphdr *iph) { struct udphdr *udph; unsigned int udphoff = iph->len; bool payload_csum = false; int oldlen; #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6 && iph->fragoffs) return 1; #endif oldlen = skb->len - udphoff; /* csum_check requires unshared skb */ if (skb_ensure_writable(skb, udphoff + sizeof(*udph))) return 0; if (unlikely(cp->app != NULL)) { int ret; /* Some checks before mangling */ if (!udp_csum_check(cp->af, skb, pp)) return 0; /* * Call application helper if needed */ if (!(ret = ip_vs_app_pkt_out(cp, skb, iph))) return 0; /* ret=2: csum update is needed after payload mangling */ if (ret == 1) oldlen = skb->len - udphoff; else payload_csum = true; } udph = (void *)skb_network_header(skb) + udphoff; udph->source = cp->vport; /* * Adjust UDP checksums */ if (skb->ip_summed == CHECKSUM_PARTIAL) { udp_partial_csum_update(cp->af, udph, &cp->daddr, &cp->vaddr, htons(oldlen), htons(skb->len - udphoff)); } else if (!payload_csum && (udph->check != 0)) { /* Only port and addr are changed, do fast csum update */ udp_fast_csum_update(cp->af, udph, &cp->daddr, &cp->vaddr, cp->dport, cp->vport); if (skb->ip_summed == CHECKSUM_COMPLETE) skb->ip_summed = cp->app ? CHECKSUM_UNNECESSARY : CHECKSUM_NONE; } else { /* full checksum calculation */ udph->check = 0; skb->csum = skb_checksum(skb, udphoff, skb->len - udphoff, 0); #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) udph->check = csum_ipv6_magic(&cp->vaddr.in6, &cp->caddr.in6, skb->len - udphoff, cp->protocol, skb->csum); else #endif udph->check = csum_tcpudp_magic(cp->vaddr.ip, cp->caddr.ip, skb->len - udphoff, cp->protocol, skb->csum); if (udph->check == 0) udph->check = CSUM_MANGLED_0; skb->ip_summed = CHECKSUM_UNNECESSARY; IP_VS_DBG(11, "O-pkt: %s O-csum=%d (+%zd)\n", pp->name, udph->check, (char*)&(udph->check) - (char*)udph); } return 1; } static int udp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp, struct ip_vs_conn *cp, struct ip_vs_iphdr *iph) { struct udphdr *udph; unsigned int udphoff = iph->len; bool payload_csum = false; int oldlen; #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6 && iph->fragoffs) return 1; #endif oldlen = skb->len - udphoff; /* csum_check requires unshared skb */ if (skb_ensure_writable(skb, udphoff + sizeof(*udph))) return 0; if (unlikely(cp->app != NULL)) { int ret; /* Some checks before mangling */ if (!udp_csum_check(cp->af, skb, pp)) return 0; /* * Attempt ip_vs_app call. * It will fix ip_vs_conn */ if (!(ret = ip_vs_app_pkt_in(cp, skb, iph))) return 0; /* ret=2: csum update is needed after payload mangling */ if (ret == 1) oldlen = skb->len - udphoff; else payload_csum = true; } udph = (void *)skb_network_header(skb) + udphoff; udph->dest = cp->dport; /* * Adjust UDP checksums */ if (skb->ip_summed == CHECKSUM_PARTIAL) { udp_partial_csum_update(cp->af, udph, &cp->vaddr, &cp->daddr, htons(oldlen), htons(skb->len - udphoff)); } else if (!payload_csum && (udph->check != 0)) { /* Only port and addr are changed, do fast csum update */ udp_fast_csum_update(cp->af, udph, &cp->vaddr, &cp->daddr, cp->vport, cp->dport); if (skb->ip_summed == CHECKSUM_COMPLETE) skb->ip_summed = cp->app ? CHECKSUM_UNNECESSARY : CHECKSUM_NONE; } else { /* full checksum calculation */ udph->check = 0; skb->csum = skb_checksum(skb, udphoff, skb->len - udphoff, 0); #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) udph->check = csum_ipv6_magic(&cp->caddr.in6, &cp->daddr.in6, skb->len - udphoff, cp->protocol, skb->csum); else #endif udph->check = csum_tcpudp_magic(cp->caddr.ip, cp->daddr.ip, skb->len - udphoff, cp->protocol, skb->csum); if (udph->check == 0) udph->check = CSUM_MANGLED_0; skb->ip_summed = CHECKSUM_UNNECESSARY; } return 1; } static int udp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) { struct udphdr _udph, *uh; unsigned int udphoff; #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) udphoff = sizeof(struct ipv6hdr); else #endif udphoff = ip_hdrlen(skb); uh = skb_header_pointer(skb, udphoff, sizeof(_udph), &_udph); if (uh == NULL) return 0; if (uh->check != 0) { switch (skb->ip_summed) { case CHECKSUM_NONE: skb->csum = skb_checksum(skb, udphoff, skb->len - udphoff, 0); fallthrough; case CHECKSUM_COMPLETE: #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { if (csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->len - udphoff, ipv6_hdr(skb)->nexthdr, skb->csum)) { IP_VS_DBG_RL_PKT(0, af, pp, skb, 0, "Failed checksum for"); return 0; } } else #endif if (csum_tcpudp_magic(ip_hdr(skb)->saddr, ip_hdr(skb)->daddr, skb->len - udphoff, ip_hdr(skb)->protocol, skb->csum)) { IP_VS_DBG_RL_PKT(0, af, pp, skb, 0, "Failed checksum for"); return 0; } break; default: /* No need to checksum. */ break; } } return 1; } static inline __u16 udp_app_hashkey(__be16 port) { return (((__force u16)port >> UDP_APP_TAB_BITS) ^ (__force u16)port) & UDP_APP_TAB_MASK; } static int udp_register_app(struct netns_ipvs *ipvs, struct ip_vs_app *inc) { struct ip_vs_app *i; __u16 hash; __be16 port = inc->port; int ret = 0; struct ip_vs_proto_data *pd = ip_vs_proto_data_get(ipvs, IPPROTO_UDP); hash = udp_app_hashkey(port); list_for_each_entry(i, &ipvs->udp_apps[hash], p_list) { if (i->port == port) { ret = -EEXIST; goto out; } } list_add_rcu(&inc->p_list, &ipvs->udp_apps[hash]); atomic_inc(&pd->appcnt); out: return ret; } static void udp_unregister_app(struct netns_ipvs *ipvs, struct ip_vs_app *inc) { struct ip_vs_proto_data *pd = ip_vs_proto_data_get(ipvs, IPPROTO_UDP); atomic_dec(&pd->appcnt); list_del_rcu(&inc->p_list); } static int udp_app_conn_bind(struct ip_vs_conn *cp) { struct netns_ipvs *ipvs = cp->ipvs; int hash; struct ip_vs_app *inc; int result = 0; /* Default binding: bind app only for NAT */ if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ) return 0; /* Lookup application incarnations and bind the right one */ hash = udp_app_hashkey(cp->vport); list_for_each_entry_rcu(inc, &ipvs->udp_apps[hash], p_list) { if (inc->port == cp->vport) { if (unlikely(!ip_vs_app_inc_get(inc))) break; IP_VS_DBG_BUF(9, "%s(): Binding conn %s:%u->" "%s:%u to app %s on port %u\n", __func__, IP_VS_DBG_ADDR(cp->af, &cp->caddr), ntohs(cp->cport), IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport), inc->name, ntohs(inc->port)); cp->app = inc; if (inc->init_conn) result = inc->init_conn(inc, cp); break; } } return result; } static const int udp_timeouts[IP_VS_UDP_S_LAST+1] = { [IP_VS_UDP_S_NORMAL] = 5*60*HZ, [IP_VS_UDP_S_LAST] = 2*HZ, }; static const char *const udp_state_name_table[IP_VS_UDP_S_LAST+1] = { [IP_VS_UDP_S_NORMAL] = "UDP", [IP_VS_UDP_S_LAST] = "BUG!", }; static const char * udp_state_name(int state) { if (state >= IP_VS_UDP_S_LAST) return "ERR!"; return udp_state_name_table[state] ? udp_state_name_table[state] : "?"; } static void udp_state_transition(struct ip_vs_conn *cp, int direction, const struct sk_buff *skb, struct ip_vs_proto_data *pd) { if (unlikely(!pd)) { pr_err("UDP no ns data\n"); return; } cp->timeout = pd->timeout_table[IP_VS_UDP_S_NORMAL]; if (direction == IP_VS_DIR_OUTPUT) ip_vs_control_assure_ct(cp); } static int __udp_init(struct netns_ipvs *ipvs, struct ip_vs_proto_data *pd) { ip_vs_init_hash_table(ipvs->udp_apps, UDP_APP_TAB_SIZE); pd->timeout_table = ip_vs_create_timeout_table((int *)udp_timeouts, sizeof(udp_timeouts)); if (!pd->timeout_table) return -ENOMEM; return 0; } static void __udp_exit(struct netns_ipvs *ipvs, struct ip_vs_proto_data *pd) { kfree(pd->timeout_table); } struct ip_vs_protocol ip_vs_protocol_udp = { .name = "UDP", .protocol = IPPROTO_UDP, .num_states = IP_VS_UDP_S_LAST, .dont_defrag = 0, .init = NULL, .exit = NULL, .init_netns = __udp_init, .exit_netns = __udp_exit, .conn_schedule = udp_conn_schedule, .conn_in_get = ip_vs_conn_in_get_proto, .conn_out_get = ip_vs_conn_out_get_proto, .snat_handler = udp_snat_handler, .dnat_handler = udp_dnat_handler, .state_transition = udp_state_transition, .state_name = udp_state_name, .register_app = udp_register_app, .unregister_app = udp_unregister_app, .app_conn_bind = udp_app_conn_bind, .debug_packet = ip_vs_tcpudp_debug_packet, .timeout_change = NULL, }; |
| 122 123 121 122 3 122 20 20 143 123 20 122 122 122 20 20 20 142 122 20 142 1 123 143 21 122 1 1 122 112 112 112 20 1 49 2 47 158 157 1 1 159 159 156 1 158 1 160 160 2 3 47 47 143 143 123 121 20 142 142 3 3 3 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 | // SPDX-License-Identifier: GPL-2.0-or-later /* * slot_map.c * * Copyright (C) 2002, 2004 Oracle. All rights reserved. */ #include <linux/types.h> #include <linux/slab.h> #include <linux/highmem.h> #include <cluster/masklog.h> #include "ocfs2.h" #include "dlmglue.h" #include "extent_map.h" #include "heartbeat.h" #include "inode.h" #include "slot_map.h" #include "super.h" #include "sysfile.h" #include "ocfs2_trace.h" #include "buffer_head_io.h" struct ocfs2_slot { int sl_valid; unsigned int sl_node_num; }; struct ocfs2_slot_info { int si_extended; int si_slots_per_block; struct inode *si_inode; unsigned int si_blocks; struct buffer_head **si_bh; unsigned int si_num_slots; struct ocfs2_slot si_slots[] __counted_by(si_num_slots); }; static int __ocfs2_node_num_to_slot(struct ocfs2_slot_info *si, unsigned int node_num); static int ocfs2_validate_slot_map_block(struct super_block *sb, struct buffer_head *bh); static void ocfs2_invalidate_slot(struct ocfs2_slot_info *si, int slot_num) { BUG_ON((slot_num < 0) || (slot_num >= si->si_num_slots)); si->si_slots[slot_num].sl_valid = 0; } static void ocfs2_set_slot(struct ocfs2_slot_info *si, int slot_num, unsigned int node_num) { BUG_ON((slot_num < 0) || (slot_num >= si->si_num_slots)); si->si_slots[slot_num].sl_valid = 1; si->si_slots[slot_num].sl_node_num = node_num; } /* This version is for the extended slot map */ static void ocfs2_update_slot_info_extended(struct ocfs2_slot_info *si) { int b, i, slotno; struct ocfs2_slot_map_extended *se; slotno = 0; for (b = 0; b < si->si_blocks; b++) { se = (struct ocfs2_slot_map_extended *)si->si_bh[b]->b_data; for (i = 0; (i < si->si_slots_per_block) && (slotno < si->si_num_slots); i++, slotno++) { if (se->se_slots[i].es_valid) ocfs2_set_slot(si, slotno, le32_to_cpu(se->se_slots[i].es_node_num)); else ocfs2_invalidate_slot(si, slotno); } } } /* * Post the slot information on disk into our slot_info struct. * Must be protected by osb_lock. */ static void ocfs2_update_slot_info_old(struct ocfs2_slot_info *si) { int i; struct ocfs2_slot_map *sm; sm = (struct ocfs2_slot_map *)si->si_bh[0]->b_data; for (i = 0; i < si->si_num_slots; i++) { if (le16_to_cpu(sm->sm_slots[i]) == (u16)OCFS2_INVALID_SLOT) ocfs2_invalidate_slot(si, i); else ocfs2_set_slot(si, i, le16_to_cpu(sm->sm_slots[i])); } } static void ocfs2_update_slot_info(struct ocfs2_slot_info *si) { /* * The slot data will have been refreshed when ocfs2_super_lock * was taken. */ if (si->si_extended) ocfs2_update_slot_info_extended(si); else ocfs2_update_slot_info_old(si); } int ocfs2_refresh_slot_info(struct ocfs2_super *osb) { int ret; struct ocfs2_slot_info *si = osb->slot_info; if (si == NULL) return 0; BUG_ON(si->si_blocks == 0); BUG_ON(si->si_bh == NULL); trace_ocfs2_refresh_slot_info(si->si_blocks); /* * We pass -1 as blocknr because we expect all of si->si_bh to * be !NULL. Thus, ocfs2_read_blocks() will ignore blocknr. If * this is not true, the read of -1 (UINT64_MAX) will fail. */ ret = ocfs2_read_blocks(INODE_CACHE(si->si_inode), -1, si->si_blocks, si->si_bh, OCFS2_BH_IGNORE_CACHE, ocfs2_validate_slot_map_block); if (ret == 0) { spin_lock(&osb->osb_lock); ocfs2_update_slot_info(si); spin_unlock(&osb->osb_lock); } return ret; } /* post the our slot info stuff into it's destination bh and write it * out. */ static void ocfs2_update_disk_slot_extended(struct ocfs2_slot_info *si, int slot_num, struct buffer_head **bh) { int blkind = slot_num / si->si_slots_per_block; int slotno = slot_num % si->si_slots_per_block; struct ocfs2_slot_map_extended *se; BUG_ON(blkind >= si->si_blocks); se = (struct ocfs2_slot_map_extended *)si->si_bh[blkind]->b_data; se->se_slots[slotno].es_valid = si->si_slots[slot_num].sl_valid; if (si->si_slots[slot_num].sl_valid) se->se_slots[slotno].es_node_num = cpu_to_le32(si->si_slots[slot_num].sl_node_num); *bh = si->si_bh[blkind]; } static void ocfs2_update_disk_slot_old(struct ocfs2_slot_info *si, int slot_num, struct buffer_head **bh) { int i; struct ocfs2_slot_map *sm; sm = (struct ocfs2_slot_map *)si->si_bh[0]->b_data; for (i = 0; i < si->si_num_slots; i++) { if (si->si_slots[i].sl_valid) sm->sm_slots[i] = cpu_to_le16(si->si_slots[i].sl_node_num); else sm->sm_slots[i] = cpu_to_le16(OCFS2_INVALID_SLOT); } *bh = si->si_bh[0]; } static int ocfs2_update_disk_slot(struct ocfs2_super *osb, struct ocfs2_slot_info *si, int slot_num) { int status; struct buffer_head *bh; spin_lock(&osb->osb_lock); if (si->si_extended) ocfs2_update_disk_slot_extended(si, slot_num, &bh); else ocfs2_update_disk_slot_old(si, slot_num, &bh); spin_unlock(&osb->osb_lock); status = ocfs2_write_block(osb, bh, INODE_CACHE(si->si_inode)); if (status < 0) mlog_errno(status); return status; } /* * Calculate how many bytes are needed by the slot map. Returns * an error if the slot map file is too small. */ static int ocfs2_slot_map_physical_size(struct ocfs2_super *osb, struct inode *inode, unsigned long long *bytes) { unsigned long long bytes_needed; if (ocfs2_uses_extended_slot_map(osb)) { bytes_needed = osb->max_slots * sizeof(struct ocfs2_extended_slot); } else { bytes_needed = osb->max_slots * sizeof(__le16); } if (bytes_needed > i_size_read(inode)) { mlog(ML_ERROR, "Slot map file is too small! (size %llu, needed %llu)\n", i_size_read(inode), bytes_needed); return -ENOSPC; } *bytes = bytes_needed; return 0; } /* try to find global node in the slot info. Returns -ENOENT * if nothing is found. */ static int __ocfs2_node_num_to_slot(struct ocfs2_slot_info *si, unsigned int node_num) { int i, ret = -ENOENT; for(i = 0; i < si->si_num_slots; i++) { if (si->si_slots[i].sl_valid && (node_num == si->si_slots[i].sl_node_num)) { ret = i; break; } } return ret; } static int __ocfs2_find_empty_slot(struct ocfs2_slot_info *si, int preferred) { int i, ret = -ENOSPC; if ((preferred >= 0) && (preferred < si->si_num_slots)) { if (!si->si_slots[preferred].sl_valid) { ret = preferred; goto out; } } for(i = 0; i < si->si_num_slots; i++) { if (!si->si_slots[i].sl_valid) { ret = i; break; } } out: return ret; } int ocfs2_node_num_to_slot(struct ocfs2_super *osb, unsigned int node_num) { int slot; struct ocfs2_slot_info *si = osb->slot_info; spin_lock(&osb->osb_lock); slot = __ocfs2_node_num_to_slot(si, node_num); spin_unlock(&osb->osb_lock); return slot; } int ocfs2_slot_to_node_num_locked(struct ocfs2_super *osb, int slot_num, unsigned int *node_num) { struct ocfs2_slot_info *si = osb->slot_info; assert_spin_locked(&osb->osb_lock); BUG_ON(slot_num < 0); BUG_ON(slot_num >= osb->max_slots); if (!si->si_slots[slot_num].sl_valid) return -ENOENT; *node_num = si->si_slots[slot_num].sl_node_num; return 0; } static void __ocfs2_free_slot_info(struct ocfs2_slot_info *si) { unsigned int i; if (si == NULL) return; iput(si->si_inode); if (si->si_bh) { for (i = 0; i < si->si_blocks; i++) { if (si->si_bh[i]) { brelse(si->si_bh[i]); si->si_bh[i] = NULL; } } kfree(si->si_bh); } kfree(si); } int ocfs2_clear_slot(struct ocfs2_super *osb, int slot_num) { struct ocfs2_slot_info *si = osb->slot_info; if (si == NULL) return 0; spin_lock(&osb->osb_lock); ocfs2_invalidate_slot(si, slot_num); spin_unlock(&osb->osb_lock); return ocfs2_update_disk_slot(osb, osb->slot_info, slot_num); } static int ocfs2_validate_slot_map_block(struct super_block *sb, struct buffer_head *bh) { int rc; BUG_ON(!buffer_uptodate(bh)); if (bh->b_blocknr < OCFS2_SUPER_BLOCK_BLKNO) { rc = ocfs2_error(sb, "Invalid Slot Map Buffer Head " "Block Number : %llu, Should be >= %d", (unsigned long long)bh->b_blocknr, OCFS2_SUPER_BLOCK_BLKNO); return rc; } return 0; } static int ocfs2_map_slot_buffers(struct ocfs2_super *osb, struct ocfs2_slot_info *si) { int status = 0; u64 blkno; unsigned long long blocks, bytes = 0; unsigned int i; struct buffer_head *bh; status = ocfs2_slot_map_physical_size(osb, si->si_inode, &bytes); if (status) goto bail; blocks = ocfs2_blocks_for_bytes(si->si_inode->i_sb, bytes); BUG_ON(blocks > UINT_MAX); si->si_blocks = blocks; if (!si->si_blocks) goto bail; if (si->si_extended) si->si_slots_per_block = (osb->sb->s_blocksize / sizeof(struct ocfs2_extended_slot)); else si->si_slots_per_block = osb->sb->s_blocksize / sizeof(__le16); /* The size checks above should ensure this */ BUG_ON((osb->max_slots / si->si_slots_per_block) > blocks); trace_ocfs2_map_slot_buffers(bytes, si->si_blocks); si->si_bh = kcalloc(si->si_blocks, sizeof(struct buffer_head *), GFP_KERNEL); if (!si->si_bh) { status = -ENOMEM; mlog_errno(status); goto bail; } for (i = 0; i < si->si_blocks; i++) { status = ocfs2_extent_map_get_blocks(si->si_inode, i, &blkno, NULL, NULL); if (status < 0) { mlog_errno(status); goto bail; } trace_ocfs2_map_slot_buffers_block((unsigned long long)blkno, i); bh = NULL; /* Acquire a fresh bh */ status = ocfs2_read_blocks(INODE_CACHE(si->si_inode), blkno, 1, &bh, OCFS2_BH_IGNORE_CACHE, ocfs2_validate_slot_map_block); if (status < 0) { mlog_errno(status); goto bail; } si->si_bh[i] = bh; } bail: return status; } int ocfs2_init_slot_info(struct ocfs2_super *osb) { int status; struct inode *inode = NULL; struct ocfs2_slot_info *si; si = kzalloc(struct_size(si, si_slots, osb->max_slots), GFP_KERNEL); if (!si) { status = -ENOMEM; mlog_errno(status); return status; } si->si_extended = ocfs2_uses_extended_slot_map(osb); si->si_num_slots = osb->max_slots; inode = ocfs2_get_system_file_inode(osb, SLOT_MAP_SYSTEM_INODE, OCFS2_INVALID_SLOT); if (!inode) { status = -EINVAL; mlog_errno(status); goto bail; } si->si_inode = inode; status = ocfs2_map_slot_buffers(osb, si); if (status < 0) { mlog_errno(status); goto bail; } osb->slot_info = (struct ocfs2_slot_info *)si; bail: if (status < 0) __ocfs2_free_slot_info(si); return status; } void ocfs2_free_slot_info(struct ocfs2_super *osb) { struct ocfs2_slot_info *si = osb->slot_info; osb->slot_info = NULL; __ocfs2_free_slot_info(si); } int ocfs2_find_slot(struct ocfs2_super *osb) { int status; int slot; struct ocfs2_slot_info *si; si = osb->slot_info; spin_lock(&osb->osb_lock); ocfs2_update_slot_info(si); /* search for ourselves first and take the slot if it already * exists. Perhaps we need to mark this in a variable for our * own journal recovery? Possibly not, though we certainly * need to warn to the user */ slot = __ocfs2_node_num_to_slot(si, osb->node_num); if (slot < 0) { /* if no slot yet, then just take 1st available * one. */ slot = __ocfs2_find_empty_slot(si, osb->preferred_slot); if (slot < 0) { spin_unlock(&osb->osb_lock); mlog(ML_ERROR, "no free slots available!\n"); status = -EINVAL; goto bail; } } else printk(KERN_INFO "ocfs2: Slot %d on device (%s) was already " "allocated to this node!\n", slot, osb->dev_str); ocfs2_set_slot(si, slot, osb->node_num); osb->slot_num = slot; spin_unlock(&osb->osb_lock); trace_ocfs2_find_slot(osb->slot_num); status = ocfs2_update_disk_slot(osb, si, osb->slot_num); if (status < 0) { mlog_errno(status); /* * if write block failed, invalidate slot to avoid overwrite * slot during dismount in case another node rightly has mounted */ spin_lock(&osb->osb_lock); ocfs2_invalidate_slot(si, osb->slot_num); osb->slot_num = OCFS2_INVALID_SLOT; spin_unlock(&osb->osb_lock); } bail: return status; } void ocfs2_put_slot(struct ocfs2_super *osb) { int status, slot_num; struct ocfs2_slot_info *si = osb->slot_info; if (!si) return; spin_lock(&osb->osb_lock); ocfs2_update_slot_info(si); slot_num = osb->slot_num; ocfs2_invalidate_slot(si, osb->slot_num); osb->slot_num = OCFS2_INVALID_SLOT; spin_unlock(&osb->osb_lock); status = ocfs2_update_disk_slot(osb, si, slot_num); if (status < 0) mlog_errno(status); ocfs2_free_slot_info(osb); } |
| 307 307 211 307 277 1 278 278 4 277 1 278 4 4 4 307 307 211 211 267 237 237 237 248 11 237 237 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 | // SPDX-License-Identifier: GPL-2.0 /* * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. * All Rights Reserved. */ #include "xfs_platform.h" #include "xfs_fs.h" #include "xfs_shared.h" #include "xfs_format.h" #include "xfs_log_format.h" #include "xfs_trans_resv.h" #include "xfs_bit.h" #include "xfs_mount.h" #include "xfs_btree.h" #include "xfs_btree_staging.h" #include "xfs_ialloc.h" #include "xfs_ialloc_btree.h" #include "xfs_alloc.h" #include "xfs_error.h" #include "xfs_health.h" #include "xfs_trace.h" #include "xfs_trans.h" #include "xfs_rmap.h" #include "xfs_ag.h" static struct kmem_cache *xfs_inobt_cur_cache; STATIC int xfs_inobt_get_minrecs( struct xfs_btree_cur *cur, int level) { return M_IGEO(cur->bc_mp)->inobt_mnr[level != 0]; } STATIC struct xfs_btree_cur * xfs_inobt_dup_cursor( struct xfs_btree_cur *cur) { return xfs_inobt_init_cursor(to_perag(cur->bc_group), cur->bc_tp, cur->bc_ag.agbp); } STATIC struct xfs_btree_cur * xfs_finobt_dup_cursor( struct xfs_btree_cur *cur) { return xfs_finobt_init_cursor(to_perag(cur->bc_group), cur->bc_tp, cur->bc_ag.agbp); } STATIC void xfs_inobt_set_root( struct xfs_btree_cur *cur, const union xfs_btree_ptr *nptr, int inc) /* level change */ { struct xfs_buf *agbp = cur->bc_ag.agbp; struct xfs_agi *agi = agbp->b_addr; agi->agi_root = nptr->s; be32_add_cpu(&agi->agi_level, inc); xfs_ialloc_log_agi(cur->bc_tp, agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL); } STATIC void xfs_finobt_set_root( struct xfs_btree_cur *cur, const union xfs_btree_ptr *nptr, int inc) /* level change */ { struct xfs_buf *agbp = cur->bc_ag.agbp; struct xfs_agi *agi = agbp->b_addr; agi->agi_free_root = nptr->s; be32_add_cpu(&agi->agi_free_level, inc); xfs_ialloc_log_agi(cur->bc_tp, agbp, XFS_AGI_FREE_ROOT | XFS_AGI_FREE_LEVEL); } /* Update the inode btree block counter for this btree. */ static inline void xfs_inobt_mod_blockcount( struct xfs_btree_cur *cur, int howmuch) { struct xfs_buf *agbp = cur->bc_ag.agbp; struct xfs_agi *agi = agbp->b_addr; if (!xfs_has_inobtcounts(cur->bc_mp)) return; if (xfs_btree_is_fino(cur->bc_ops)) be32_add_cpu(&agi->agi_fblocks, howmuch); else be32_add_cpu(&agi->agi_iblocks, howmuch); xfs_ialloc_log_agi(cur->bc_tp, agbp, XFS_AGI_IBLOCKS); } STATIC int __xfs_inobt_alloc_block( struct xfs_btree_cur *cur, const union xfs_btree_ptr *start, union xfs_btree_ptr *new, int *stat, enum xfs_ag_resv_type resv) { xfs_alloc_arg_t args; /* block allocation args */ int error; /* error return value */ xfs_agblock_t sbno = be32_to_cpu(start->s); memset(&args, 0, sizeof(args)); args.tp = cur->bc_tp; args.mp = cur->bc_mp; args.pag = to_perag(cur->bc_group); args.oinfo = XFS_RMAP_OINFO_INOBT; args.minlen = 1; args.maxlen = 1; args.prod = 1; args.resv = resv; error = xfs_alloc_vextent_near_bno(&args, xfs_agbno_to_fsb(args.pag, sbno)); if (error) return error; if (args.fsbno == NULLFSBLOCK) { *stat = 0; return 0; } ASSERT(args.len == 1); new->s = cpu_to_be32(XFS_FSB_TO_AGBNO(args.mp, args.fsbno)); *stat = 1; xfs_inobt_mod_blockcount(cur, 1); return 0; } STATIC int xfs_inobt_alloc_block( struct xfs_btree_cur *cur, const union xfs_btree_ptr *start, union xfs_btree_ptr *new, int *stat) { return __xfs_inobt_alloc_block(cur, start, new, stat, XFS_AG_RESV_NONE); } STATIC int xfs_finobt_alloc_block( struct xfs_btree_cur *cur, const union xfs_btree_ptr *start, union xfs_btree_ptr *new, int *stat) { if (cur->bc_mp->m_finobt_nores) return xfs_inobt_alloc_block(cur, start, new, stat); return __xfs_inobt_alloc_block(cur, start, new, stat, XFS_AG_RESV_METADATA); } STATIC int __xfs_inobt_free_block( struct xfs_btree_cur *cur, struct xfs_buf *bp, enum xfs_ag_resv_type resv) { xfs_fsblock_t fsbno; xfs_inobt_mod_blockcount(cur, -1); fsbno = XFS_DADDR_TO_FSB(cur->bc_mp, xfs_buf_daddr(bp)); return xfs_free_extent_later(cur->bc_tp, fsbno, 1, &XFS_RMAP_OINFO_INOBT, resv, 0); } STATIC int xfs_inobt_free_block( struct xfs_btree_cur *cur, struct xfs_buf *bp) { return __xfs_inobt_free_block(cur, bp, XFS_AG_RESV_NONE); } STATIC int xfs_finobt_free_block( struct xfs_btree_cur *cur, struct xfs_buf *bp) { if (cur->bc_mp->m_finobt_nores) return xfs_inobt_free_block(cur, bp); return __xfs_inobt_free_block(cur, bp, XFS_AG_RESV_METADATA); } STATIC int xfs_inobt_get_maxrecs( struct xfs_btree_cur *cur, int level) { return M_IGEO(cur->bc_mp)->inobt_mxr[level != 0]; } STATIC void xfs_inobt_init_key_from_rec( union xfs_btree_key *key, const union xfs_btree_rec *rec) { key->inobt.ir_startino = rec->inobt.ir_startino; } STATIC void xfs_inobt_init_high_key_from_rec( union xfs_btree_key *key, const union xfs_btree_rec *rec) { __u32 x; x = be32_to_cpu(rec->inobt.ir_startino); x += XFS_INODES_PER_CHUNK - 1; key->inobt.ir_startino = cpu_to_be32(x); } STATIC void xfs_inobt_init_rec_from_cur( struct xfs_btree_cur *cur, union xfs_btree_rec *rec) { rec->inobt.ir_startino = cpu_to_be32(cur->bc_rec.i.ir_startino); if (xfs_has_sparseinodes(cur->bc_mp)) { rec->inobt.ir_u.sp.ir_holemask = cpu_to_be16(cur->bc_rec.i.ir_holemask); rec->inobt.ir_u.sp.ir_count = cur->bc_rec.i.ir_count; rec->inobt.ir_u.sp.ir_freecount = cur->bc_rec.i.ir_freecount; } else { /* ir_holemask/ir_count not supported on-disk */ rec->inobt.ir_u.f.ir_freecount = cpu_to_be32(cur->bc_rec.i.ir_freecount); } rec->inobt.ir_free = cpu_to_be64(cur->bc_rec.i.ir_free); } /* * initial value of ptr for lookup */ STATIC void xfs_inobt_init_ptr_from_cur( struct xfs_btree_cur *cur, union xfs_btree_ptr *ptr) { struct xfs_agi *agi = cur->bc_ag.agbp->b_addr; ASSERT(cur->bc_group->xg_gno == be32_to_cpu(agi->agi_seqno)); ptr->s = agi->agi_root; } STATIC void xfs_finobt_init_ptr_from_cur( struct xfs_btree_cur *cur, union xfs_btree_ptr *ptr) { struct xfs_agi *agi = cur->bc_ag.agbp->b_addr; ASSERT(cur->bc_group->xg_gno == be32_to_cpu(agi->agi_seqno)); ptr->s = agi->agi_free_root; } STATIC int xfs_inobt_cmp_key_with_cur( struct xfs_btree_cur *cur, const union xfs_btree_key *key) { return cmp_int(be32_to_cpu(key->inobt.ir_startino), cur->bc_rec.i.ir_startino); } STATIC int xfs_inobt_cmp_two_keys( struct xfs_btree_cur *cur, const union xfs_btree_key *k1, const union xfs_btree_key *k2, const union xfs_btree_key *mask) { ASSERT(!mask || mask->inobt.ir_startino); return cmp_int(be32_to_cpu(k1->inobt.ir_startino), be32_to_cpu(k2->inobt.ir_startino)); } static xfs_failaddr_t xfs_inobt_verify( struct xfs_buf *bp) { struct xfs_mount *mp = bp->b_mount; struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); xfs_failaddr_t fa; unsigned int level; if (!xfs_verify_magic(bp, block->bb_magic)) return __this_address; /* * During growfs operations, we can't verify the exact owner as the * perag is not fully initialised and hence not attached to the buffer. * * Similarly, during log recovery we will have a perag structure * attached, but the agi information will not yet have been initialised * from the on disk AGI. We don't currently use any of this information, * but beware of the landmine (i.e. need to check * xfs_perag_initialised_agi(pag)) if we ever do. */ if (xfs_has_crc(mp)) { fa = xfs_btree_agblock_v5hdr_verify(bp); if (fa) return fa; } /* level verification */ level = be16_to_cpu(block->bb_level); if (level >= M_IGEO(mp)->inobt_maxlevels) return __this_address; return xfs_btree_agblock_verify(bp, M_IGEO(mp)->inobt_mxr[level != 0]); } static void xfs_inobt_read_verify( struct xfs_buf *bp) { xfs_failaddr_t fa; if (!xfs_btree_agblock_verify_crc(bp)) xfs_verifier_error(bp, -EFSBADCRC, __this_address); else { fa = xfs_inobt_verify(bp); if (fa) xfs_verifier_error(bp, -EFSCORRUPTED, fa); } if (bp->b_error) trace_xfs_btree_corrupt(bp, _RET_IP_); } static void xfs_inobt_write_verify( struct xfs_buf *bp) { xfs_failaddr_t fa; fa = xfs_inobt_verify(bp); if (fa) { trace_xfs_btree_corrupt(bp, _RET_IP_); xfs_verifier_error(bp, -EFSCORRUPTED, fa); return; } xfs_btree_agblock_calc_crc(bp); } const struct xfs_buf_ops xfs_inobt_buf_ops = { .name = "xfs_inobt", .magic = { cpu_to_be32(XFS_IBT_MAGIC), cpu_to_be32(XFS_IBT_CRC_MAGIC) }, .verify_read = xfs_inobt_read_verify, .verify_write = xfs_inobt_write_verify, .verify_struct = xfs_inobt_verify, }; const struct xfs_buf_ops xfs_finobt_buf_ops = { .name = "xfs_finobt", .magic = { cpu_to_be32(XFS_FIBT_MAGIC), cpu_to_be32(XFS_FIBT_CRC_MAGIC) }, .verify_read = xfs_inobt_read_verify, .verify_write = xfs_inobt_write_verify, .verify_struct = xfs_inobt_verify, }; STATIC int xfs_inobt_keys_inorder( struct xfs_btree_cur *cur, const union xfs_btree_key *k1, const union xfs_btree_key *k2) { return be32_to_cpu(k1->inobt.ir_startino) < be32_to_cpu(k2->inobt.ir_startino); } STATIC int xfs_inobt_recs_inorder( struct xfs_btree_cur *cur, const union xfs_btree_rec *r1, const union xfs_btree_rec *r2) { return be32_to_cpu(r1->inobt.ir_startino) + XFS_INODES_PER_CHUNK <= be32_to_cpu(r2->inobt.ir_startino); } STATIC enum xbtree_key_contig xfs_inobt_keys_contiguous( struct xfs_btree_cur *cur, const union xfs_btree_key *key1, const union xfs_btree_key *key2, const union xfs_btree_key *mask) { ASSERT(!mask || mask->inobt.ir_startino); return xbtree_key_contig(be32_to_cpu(key1->inobt.ir_startino), be32_to_cpu(key2->inobt.ir_startino)); } const struct xfs_btree_ops xfs_inobt_ops = { .name = "ino", .type = XFS_BTREE_TYPE_AG, .rec_len = sizeof(xfs_inobt_rec_t), .key_len = sizeof(xfs_inobt_key_t), .ptr_len = XFS_BTREE_SHORT_PTR_LEN, .lru_refs = XFS_INO_BTREE_REF, .statoff = XFS_STATS_CALC_INDEX(xs_ibt_2), .sick_mask = XFS_SICK_AG_INOBT, .dup_cursor = xfs_inobt_dup_cursor, .set_root = xfs_inobt_set_root, .alloc_block = xfs_inobt_alloc_block, .free_block = xfs_inobt_free_block, .get_minrecs = xfs_inobt_get_minrecs, .get_maxrecs = xfs_inobt_get_maxrecs, .init_key_from_rec = xfs_inobt_init_key_from_rec, .init_high_key_from_rec = xfs_inobt_init_high_key_from_rec, .init_rec_from_cur = xfs_inobt_init_rec_from_cur, .init_ptr_from_cur = xfs_inobt_init_ptr_from_cur, .cmp_key_with_cur = xfs_inobt_cmp_key_with_cur, .buf_ops = &xfs_inobt_buf_ops, .cmp_two_keys = xfs_inobt_cmp_two_keys, .keys_inorder = xfs_inobt_keys_inorder, .recs_inorder = xfs_inobt_recs_inorder, .keys_contiguous = xfs_inobt_keys_contiguous, }; const struct xfs_btree_ops xfs_finobt_ops = { .name = "fino", .type = XFS_BTREE_TYPE_AG, .rec_len = sizeof(xfs_inobt_rec_t), .key_len = sizeof(xfs_inobt_key_t), .ptr_len = XFS_BTREE_SHORT_PTR_LEN, .lru_refs = XFS_INO_BTREE_REF, .statoff = XFS_STATS_CALC_INDEX(xs_fibt_2), .sick_mask = XFS_SICK_AG_FINOBT, .dup_cursor = xfs_finobt_dup_cursor, .set_root = xfs_finobt_set_root, .alloc_block = xfs_finobt_alloc_block, .free_block = xfs_finobt_free_block, .get_minrecs = xfs_inobt_get_minrecs, .get_maxrecs = xfs_inobt_get_maxrecs, .init_key_from_rec = xfs_inobt_init_key_from_rec, .init_high_key_from_rec = xfs_inobt_init_high_key_from_rec, .init_rec_from_cur = xfs_inobt_init_rec_from_cur, .init_ptr_from_cur = xfs_finobt_init_ptr_from_cur, .cmp_key_with_cur = xfs_inobt_cmp_key_with_cur, .buf_ops = &xfs_finobt_buf_ops, .cmp_two_keys = xfs_inobt_cmp_two_keys, .keys_inorder = xfs_inobt_keys_inorder, .recs_inorder = xfs_inobt_recs_inorder, .keys_contiguous = xfs_inobt_keys_contiguous, }; /* * Create an inode btree cursor. * * For staging cursors tp and agbp are NULL. */ struct xfs_btree_cur * xfs_inobt_init_cursor( struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp) { struct xfs_mount *mp = pag_mount(pag); struct xfs_btree_cur *cur; cur = xfs_btree_alloc_cursor(mp, tp, &xfs_inobt_ops, M_IGEO(mp)->inobt_maxlevels, xfs_inobt_cur_cache); cur->bc_group = xfs_group_hold(pag_group(pag)); cur->bc_ag.agbp = agbp; if (agbp) { struct xfs_agi *agi = agbp->b_addr; cur->bc_nlevels = be32_to_cpu(agi->agi_level); } return cur; } /* * Create a free inode btree cursor. * * For staging cursors tp and agbp are NULL. */ struct xfs_btree_cur * xfs_finobt_init_cursor( struct xfs_perag *pag, struct xfs_trans *tp, struct xfs_buf *agbp) { struct xfs_mount *mp = pag_mount(pag); struct xfs_btree_cur *cur; cur = xfs_btree_alloc_cursor(mp, tp, &xfs_finobt_ops, M_IGEO(mp)->inobt_maxlevels, xfs_inobt_cur_cache); cur->bc_group = xfs_group_hold(pag_group(pag)); cur->bc_ag.agbp = agbp; if (agbp) { struct xfs_agi *agi = agbp->b_addr; cur->bc_nlevels = be32_to_cpu(agi->agi_free_level); } return cur; } /* * Install a new inobt btree root. Caller is responsible for invalidating * and freeing the old btree blocks. */ void xfs_inobt_commit_staged_btree( struct xfs_btree_cur *cur, struct xfs_trans *tp, struct xfs_buf *agbp) { struct xfs_agi *agi = agbp->b_addr; struct xbtree_afakeroot *afake = cur->bc_ag.afake; int fields; ASSERT(cur->bc_flags & XFS_BTREE_STAGING); if (xfs_btree_is_ino(cur->bc_ops)) { fields = XFS_AGI_ROOT | XFS_AGI_LEVEL; agi->agi_root = cpu_to_be32(afake->af_root); agi->agi_level = cpu_to_be32(afake->af_levels); if (xfs_has_inobtcounts(cur->bc_mp)) { agi->agi_iblocks = cpu_to_be32(afake->af_blocks); fields |= XFS_AGI_IBLOCKS; } xfs_ialloc_log_agi(tp, agbp, fields); xfs_btree_commit_afakeroot(cur, tp, agbp); } else { fields = XFS_AGI_FREE_ROOT | XFS_AGI_FREE_LEVEL; agi->agi_free_root = cpu_to_be32(afake->af_root); agi->agi_free_level = cpu_to_be32(afake->af_levels); if (xfs_has_inobtcounts(cur->bc_mp)) { agi->agi_fblocks = cpu_to_be32(afake->af_blocks); fields |= XFS_AGI_IBLOCKS; } xfs_ialloc_log_agi(tp, agbp, fields); xfs_btree_commit_afakeroot(cur, tp, agbp); } } /* Calculate number of records in an inode btree block. */ static inline unsigned int xfs_inobt_block_maxrecs( unsigned int blocklen, bool leaf) { if (leaf) return blocklen / sizeof(xfs_inobt_rec_t); return blocklen / (sizeof(xfs_inobt_key_t) + sizeof(xfs_inobt_ptr_t)); } /* * Calculate number of records in an inobt btree block. */ unsigned int xfs_inobt_maxrecs( struct xfs_mount *mp, unsigned int blocklen, bool leaf) { blocklen -= XFS_INOBT_BLOCK_LEN(mp); return xfs_inobt_block_maxrecs(blocklen, leaf); } /* * Maximum number of inode btree records per AG. Pretend that we can fill an * entire AG completely full of inodes except for the AG headers. */ #define XFS_MAX_INODE_RECORDS \ ((XFS_MAX_AG_BYTES - (4 * BBSIZE)) / XFS_DINODE_MIN_SIZE) / \ XFS_INODES_PER_CHUNK /* Compute the max possible height for the inode btree. */ static inline unsigned int xfs_inobt_maxlevels_ondisk(void) { unsigned int minrecs[2]; unsigned int blocklen; blocklen = min(XFS_MIN_BLOCKSIZE - XFS_BTREE_SBLOCK_LEN, XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN); minrecs[0] = xfs_inobt_block_maxrecs(blocklen, true) / 2; minrecs[1] = xfs_inobt_block_maxrecs(blocklen, false) / 2; return xfs_btree_compute_maxlevels(minrecs, XFS_MAX_INODE_RECORDS); } /* Compute the max possible height for the free inode btree. */ static inline unsigned int xfs_finobt_maxlevels_ondisk(void) { unsigned int minrecs[2]; unsigned int blocklen; blocklen = XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN; minrecs[0] = xfs_inobt_block_maxrecs(blocklen, true) / 2; minrecs[1] = xfs_inobt_block_maxrecs(blocklen, false) / 2; return xfs_btree_compute_maxlevels(minrecs, XFS_MAX_INODE_RECORDS); } /* Compute the max possible height for either inode btree. */ unsigned int xfs_iallocbt_maxlevels_ondisk(void) { return max(xfs_inobt_maxlevels_ondisk(), xfs_finobt_maxlevels_ondisk()); } /* * Convert the inode record holemask to an inode allocation bitmap. The inode * allocation bitmap is inode granularity and specifies whether an inode is * physically allocated on disk (not whether the inode is considered allocated * or free by the fs). * * A bit value of 1 means the inode is allocated, a value of 0 means it is free. */ uint64_t xfs_inobt_irec_to_allocmask( const struct xfs_inobt_rec_incore *rec) { uint64_t bitmap = 0; uint64_t inodespbit; int nextbit; uint allocbitmap; /* * The holemask has 16-bits for a 64 inode record. Therefore each * holemask bit represents multiple inodes. Create a mask of bits to set * in the allocmask for each holemask bit. */ inodespbit = (1 << XFS_INODES_PER_HOLEMASK_BIT) - 1; /* * Allocated inodes are represented by 0 bits in holemask. Invert the 0 * bits to 1 and convert to a uint so we can use xfs_next_bit(). Mask * anything beyond the 16 holemask bits since this casts to a larger * type. */ allocbitmap = ~rec->ir_holemask & ((1 << XFS_INOBT_HOLEMASK_BITS) - 1); /* * allocbitmap is the inverted holemask so every set bit represents * allocated inodes. To expand from 16-bit holemask granularity to * 64-bit (e.g., bit-per-inode), set inodespbit bits in the target * bitmap for every holemask bit. */ nextbit = xfs_next_bit(&allocbitmap, 1, 0); while (nextbit != -1) { ASSERT(nextbit < (sizeof(rec->ir_holemask) * NBBY)); bitmap |= (inodespbit << (nextbit * XFS_INODES_PER_HOLEMASK_BIT)); nextbit = xfs_next_bit(&allocbitmap, 1, nextbit + 1); } return bitmap; } #if defined(DEBUG) || defined(XFS_WARN) /* * Verify that an in-core inode record has a valid inode count. */ int xfs_inobt_rec_check_count( struct xfs_mount *mp, struct xfs_inobt_rec_incore *rec) { int inocount = 0; int nextbit = 0; uint64_t allocbmap; int wordsz; wordsz = sizeof(allocbmap) / sizeof(unsigned int); allocbmap = xfs_inobt_irec_to_allocmask(rec); nextbit = xfs_next_bit((uint *) &allocbmap, wordsz, nextbit); while (nextbit != -1) { inocount++; nextbit = xfs_next_bit((uint *) &allocbmap, wordsz, nextbit + 1); } if (inocount != rec->ir_count) return -EFSCORRUPTED; return 0; } #endif /* DEBUG */ static xfs_extlen_t xfs_inobt_max_size( struct xfs_perag *pag) { struct xfs_mount *mp = pag_mount(pag); xfs_agblock_t agblocks = pag_group(pag)->xg_block_count; /* Bail out if we're uninitialized, which can happen in mkfs. */ if (M_IGEO(mp)->inobt_mxr[0] == 0) return 0; /* * The log is permanently allocated, so the space it occupies will * never be available for the kinds of things that would require btree * expansion. We therefore can pretend the space isn't there. */ if (xfs_ag_contains_log(mp, pag_agno(pag))) agblocks -= mp->m_sb.sb_logblocks; return xfs_btree_calc_size(M_IGEO(mp)->inobt_mnr, (uint64_t)agblocks * mp->m_sb.sb_inopblock / XFS_INODES_PER_CHUNK); } static int xfs_finobt_count_blocks( struct xfs_perag *pag, struct xfs_trans *tp, xfs_extlen_t *tree_blocks) { struct xfs_buf *agbp = NULL; struct xfs_btree_cur *cur; xfs_filblks_t blocks; int error; error = xfs_ialloc_read_agi(pag, tp, 0, &agbp); if (error) return error; cur = xfs_finobt_init_cursor(pag, tp, agbp); error = xfs_btree_count_blocks(cur, &blocks); xfs_btree_del_cursor(cur, error); xfs_trans_brelse(tp, agbp); *tree_blocks = blocks; return error; } /* Read finobt block count from AGI header. */ static int xfs_finobt_read_blocks( struct xfs_perag *pag, struct xfs_trans *tp, xfs_extlen_t *tree_blocks) { struct xfs_buf *agbp; struct xfs_agi *agi; int error; error = xfs_ialloc_read_agi(pag, tp, 0, &agbp); if (error) return error; agi = agbp->b_addr; *tree_blocks = be32_to_cpu(agi->agi_fblocks); xfs_trans_brelse(tp, agbp); return 0; } /* * Figure out how many blocks to reserve and how many are used by this btree. */ int xfs_finobt_calc_reserves( struct xfs_perag *pag, struct xfs_trans *tp, xfs_extlen_t *ask, xfs_extlen_t *used) { xfs_extlen_t tree_len = 0; int error; if (!xfs_has_finobt(pag_mount(pag))) return 0; if (xfs_has_inobtcounts(pag_mount(pag))) error = xfs_finobt_read_blocks(pag, tp, &tree_len); else error = xfs_finobt_count_blocks(pag, tp, &tree_len); if (error) return error; *ask += xfs_inobt_max_size(pag); *used += tree_len; return 0; } /* Calculate the inobt btree size for some records. */ xfs_extlen_t xfs_iallocbt_calc_size( struct xfs_mount *mp, unsigned long long len) { return xfs_btree_calc_size(M_IGEO(mp)->inobt_mnr, len); } int __init xfs_inobt_init_cur_cache(void) { xfs_inobt_cur_cache = kmem_cache_create("xfs_inobt_cur", xfs_btree_cur_sizeof(xfs_inobt_maxlevels_ondisk()), 0, 0, NULL); if (!xfs_inobt_cur_cache) return -ENOMEM; return 0; } void xfs_inobt_destroy_cur_cache(void) { kmem_cache_destroy(xfs_inobt_cur_cache); xfs_inobt_cur_cache = NULL; } |
| 22 22 1 21 20 1 21 2 2 2 2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 | // SPDX-License-Identifier: GPL-2.0 /* * linux/fs/hpfs/inode.c * * Mikulas Patocka (mikulas@artax.karlin.mff.cuni.cz), 1998-1999 * * inode VFS functions */ #include <linux/slab.h> #include <linux/user_namespace.h> #include "hpfs_fn.h" void hpfs_init_inode(struct inode *i) { struct super_block *sb = i->i_sb; struct hpfs_inode_info *hpfs_inode = hpfs_i(i); i->i_uid = hpfs_sb(sb)->sb_uid; i->i_gid = hpfs_sb(sb)->sb_gid; i->i_mode = hpfs_sb(sb)->sb_mode; i->i_size = -1; i->i_blocks = -1; hpfs_inode->i_dno = 0; hpfs_inode->i_n_secs = 0; hpfs_inode->i_file_sec = 0; hpfs_inode->i_disk_sec = 0; hpfs_inode->i_dpos = 0; hpfs_inode->i_dsubdno = 0; hpfs_inode->i_ea_mode = 0; hpfs_inode->i_ea_uid = 0; hpfs_inode->i_ea_gid = 0; hpfs_inode->i_ea_size = 0; hpfs_inode->i_rddir_off = NULL; hpfs_inode->i_dirty = 0; inode_set_ctime(i, 0, 0); inode_set_mtime(i, 0, 0); inode_set_atime(i, 0, 0); } void hpfs_read_inode(struct inode *i) { struct buffer_head *bh; struct fnode *fnode; struct super_block *sb = i->i_sb; struct hpfs_inode_info *hpfs_inode = hpfs_i(i); void *ea; int ea_size; if (!(fnode = hpfs_map_fnode(sb, i->i_ino, &bh))) { /*i->i_mode |= S_IFREG; i->i_mode &= ~0111; i->i_op = &hpfs_file_iops; i->i_fop = &hpfs_file_ops; clear_nlink(i);*/ make_bad_inode(i); return; } if (hpfs_sb(i->i_sb)->sb_eas) { if ((ea = hpfs_get_ea(i->i_sb, fnode, "UID", &ea_size))) { if (ea_size == 2) { i_uid_write(i, le16_to_cpu(*(__le16*)ea)); hpfs_inode->i_ea_uid = 1; } kfree(ea); } if ((ea = hpfs_get_ea(i->i_sb, fnode, "GID", &ea_size))) { if (ea_size == 2) { i_gid_write(i, le16_to_cpu(*(__le16*)ea)); hpfs_inode->i_ea_gid = 1; } kfree(ea); } if ((ea = hpfs_get_ea(i->i_sb, fnode, "SYMLINK", &ea_size))) { kfree(ea); i->i_mode = S_IFLNK | 0777; i->i_op = &page_symlink_inode_operations; inode_nohighmem(i); i->i_data.a_ops = &hpfs_symlink_aops; set_nlink(i, 1); i->i_size = ea_size; i->i_blocks = 1; brelse(bh); return; } if ((ea = hpfs_get_ea(i->i_sb, fnode, "MODE", &ea_size))) { int rdev = 0; umode_t mode = hpfs_sb(sb)->sb_mode; if (ea_size == 2) { mode = le16_to_cpu(*(__le16*)ea); hpfs_inode->i_ea_mode = 1; } kfree(ea); i->i_mode = mode; if (S_ISBLK(mode) || S_ISCHR(mode)) { if ((ea = hpfs_get_ea(i->i_sb, fnode, "DEV", &ea_size))) { if (ea_size == 4) rdev = le32_to_cpu(*(__le32*)ea); kfree(ea); } } if (S_ISBLK(mode) || S_ISCHR(mode) || S_ISFIFO(mode) || S_ISSOCK(mode)) { brelse(bh); set_nlink(i, 1); i->i_size = 0; i->i_blocks = 1; init_special_inode(i, mode, new_decode_dev(rdev)); return; } } } if (fnode_is_dir(fnode)) { int n_dnodes, n_subdirs; i->i_mode |= S_IFDIR; i->i_op = &hpfs_dir_iops; i->i_fop = &hpfs_dir_ops; hpfs_inode->i_parent_dir = le32_to_cpu(fnode->up); hpfs_inode->i_dno = le32_to_cpu(fnode->u.external[0].disk_secno); if (hpfs_sb(sb)->sb_chk >= 2) { struct buffer_head *bh0; if (hpfs_map_fnode(sb, hpfs_inode->i_parent_dir, &bh0)) brelse(bh0); } n_dnodes = 0; n_subdirs = 0; hpfs_count_dnodes(i->i_sb, hpfs_inode->i_dno, &n_dnodes, &n_subdirs, NULL); i->i_blocks = 4 * n_dnodes; i->i_size = 2048 * n_dnodes; set_nlink(i, 2 + n_subdirs); } else { i->i_mode |= S_IFREG; if (!hpfs_inode->i_ea_mode) i->i_mode &= ~0111; i->i_op = &hpfs_file_iops; i->i_fop = &hpfs_file_ops; set_nlink(i, 1); i->i_size = le32_to_cpu(fnode->file_size); i->i_blocks = ((i->i_size + 511) >> 9) + 1; i->i_data.a_ops = &hpfs_aops; hpfs_i(i)->mmu_private = i->i_size; } brelse(bh); } static void hpfs_write_inode_ea(struct inode *i, struct fnode *fnode) { struct hpfs_inode_info *hpfs_inode = hpfs_i(i); /*if (le32_to_cpu(fnode->acl_size_l) || le16_to_cpu(fnode->acl_size_s)) { Some unknown structures like ACL may be in fnode, we'd better not overwrite them hpfs_error(i->i_sb, "fnode %08x has some unknown HPFS386 structures", i->i_ino); } else*/ if (hpfs_sb(i->i_sb)->sb_eas >= 2) { __le32 ea; if (!uid_eq(i->i_uid, hpfs_sb(i->i_sb)->sb_uid) || hpfs_inode->i_ea_uid) { ea = cpu_to_le32(i_uid_read(i)); hpfs_set_ea(i, fnode, "UID", (char*)&ea, 2); hpfs_inode->i_ea_uid = 1; } if (!gid_eq(i->i_gid, hpfs_sb(i->i_sb)->sb_gid) || hpfs_inode->i_ea_gid) { ea = cpu_to_le32(i_gid_read(i)); hpfs_set_ea(i, fnode, "GID", (char *)&ea, 2); hpfs_inode->i_ea_gid = 1; } if (!S_ISLNK(i->i_mode)) if ((i->i_mode != ((hpfs_sb(i->i_sb)->sb_mode & ~(S_ISDIR(i->i_mode) ? 0 : 0111)) | (S_ISDIR(i->i_mode) ? S_IFDIR : S_IFREG)) && i->i_mode != ((hpfs_sb(i->i_sb)->sb_mode & ~(S_ISDIR(i->i_mode) ? 0222 : 0333)) | (S_ISDIR(i->i_mode) ? S_IFDIR : S_IFREG))) || hpfs_inode->i_ea_mode) { ea = cpu_to_le32(i->i_mode); /* sick, but legal */ hpfs_set_ea(i, fnode, "MODE", (char *)&ea, 2); hpfs_inode->i_ea_mode = 1; } if (S_ISBLK(i->i_mode) || S_ISCHR(i->i_mode)) { ea = cpu_to_le32(new_encode_dev(i->i_rdev)); hpfs_set_ea(i, fnode, "DEV", (char *)&ea, 4); } } } void hpfs_write_inode(struct inode *i) { struct hpfs_inode_info *hpfs_inode = hpfs_i(i); struct inode *parent; if (i->i_ino == hpfs_sb(i->i_sb)->sb_root) return; if (hpfs_inode->i_rddir_off && !icount_read(i)) { if (*hpfs_inode->i_rddir_off) pr_err("write_inode: some position still there\n"); kfree(hpfs_inode->i_rddir_off); hpfs_inode->i_rddir_off = NULL; } if (!i->i_nlink) { return; } parent = iget_locked(i->i_sb, hpfs_inode->i_parent_dir); if (parent) { hpfs_inode->i_dirty = 0; if (inode_state_read_once(parent) & I_NEW) { hpfs_init_inode(parent); hpfs_read_inode(parent); unlock_new_inode(parent); } hpfs_write_inode_nolock(i); iput(parent); } } void hpfs_write_inode_nolock(struct inode *i) { struct hpfs_inode_info *hpfs_inode = hpfs_i(i); struct buffer_head *bh; struct fnode *fnode; struct quad_buffer_head qbh; struct hpfs_dirent *de; if (i->i_ino == hpfs_sb(i->i_sb)->sb_root) return; if (!(fnode = hpfs_map_fnode(i->i_sb, i->i_ino, &bh))) return; if (i->i_ino != hpfs_sb(i->i_sb)->sb_root && i->i_nlink) { if (!(de = map_fnode_dirent(i->i_sb, i->i_ino, fnode, &qbh))) { brelse(bh); return; } } else de = NULL; if (S_ISREG(i->i_mode)) { fnode->file_size = cpu_to_le32(i->i_size); if (de) de->file_size = cpu_to_le32(i->i_size); } else if (S_ISDIR(i->i_mode)) { fnode->file_size = cpu_to_le32(0); if (de) de->file_size = cpu_to_le32(0); } hpfs_write_inode_ea(i, fnode); if (de) { de->write_date = cpu_to_le32(gmt_to_local(i->i_sb, inode_get_mtime_sec(i))); de->read_date = cpu_to_le32(gmt_to_local(i->i_sb, inode_get_atime_sec(i))); de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, inode_get_ctime_sec(i))); de->read_only = !(i->i_mode & 0222); de->ea_size = cpu_to_le32(hpfs_inode->i_ea_size); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); } if (S_ISDIR(i->i_mode)) { if ((de = map_dirent(i, hpfs_inode->i_dno, "\001\001", 2, NULL, &qbh))) { de->write_date = cpu_to_le32(gmt_to_local(i->i_sb, inode_get_mtime_sec(i))); de->read_date = cpu_to_le32(gmt_to_local(i->i_sb, inode_get_atime_sec(i))); de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, inode_get_ctime_sec(i))); de->read_only = !(i->i_mode & 0222); de->ea_size = cpu_to_le32(/*hpfs_inode->i_ea_size*/0); de->file_size = cpu_to_le32(0); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); } else hpfs_error(i->i_sb, "directory %08lx doesn't have '.' entry", (unsigned long)i->i_ino); } mark_buffer_dirty(bh); brelse(bh); } int hpfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); int error = -EINVAL; hpfs_lock(inode->i_sb); if (inode->i_ino == hpfs_sb(inode->i_sb)->sb_root) goto out_unlock; if ((attr->ia_valid & ATTR_UID) && from_kuid(&init_user_ns, attr->ia_uid) >= 0x10000) goto out_unlock; if ((attr->ia_valid & ATTR_GID) && from_kgid(&init_user_ns, attr->ia_gid) >= 0x10000) goto out_unlock; if ((attr->ia_valid & ATTR_SIZE) && attr->ia_size > inode->i_size) goto out_unlock; error = setattr_prepare(&nop_mnt_idmap, dentry, attr); if (error) goto out_unlock; if ((attr->ia_valid & ATTR_SIZE) && attr->ia_size != i_size_read(inode)) { error = inode_newsize_ok(inode, attr->ia_size); if (error) goto out_unlock; truncate_setsize(inode, attr->ia_size); hpfs_truncate(inode); } setattr_copy(&nop_mnt_idmap, inode, attr); hpfs_write_inode(inode); out_unlock: hpfs_unlock(inode->i_sb); return error; } void hpfs_write_if_changed(struct inode *inode) { struct hpfs_inode_info *hpfs_inode = hpfs_i(inode); if (hpfs_inode->i_dirty) hpfs_write_inode(inode); } void hpfs_evict_inode(struct inode *inode) { truncate_inode_pages_final(&inode->i_data); clear_inode(inode); if (!inode->i_nlink) { hpfs_lock(inode->i_sb); hpfs_remove_fnode(inode->i_sb, inode->i_ino); hpfs_unlock(inode->i_sb); } } |
| 27 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 | /* SPDX-License-Identifier: GPL-2.0-or-later */ /* * Copyright (C) International Business Machines Corp., 2000-2004 */ #ifndef _H_JFS_TYPES #define _H_JFS_TYPES /* * jfs_types.h: * * basic type/utility definitions * * note: this header file must be the 1st include file * of JFS include list in all JFS .c file. */ #include <linux/types.h> #include <linux/nls.h> /* * transaction and lock id's * * Don't change these without carefully considering the impact on the * size and alignment of all of the linelock variants */ typedef u16 tid_t; typedef u16 lid_t; /* * Almost identical to Linux's timespec, but not quite */ struct timestruc_t { __le32 tv_sec; __le32 tv_nsec; }; /* * handy */ #define LEFTMOSTONE 0x80000000 #define HIGHORDER 0x80000000u /* high order bit on */ #define ONES 0xffffffffu /* all bit on */ /* * physical xd (pxd) * * The leftmost 24 bits of len_addr are the extent length. * The rightmost 8 bits of len_addr are the most signficant bits of * the extent address */ typedef struct { __le32 len_addr; __le32 addr2; } pxd_t; /* xd_t field construction */ static inline void PXDlength(pxd_t *pxd, __u32 len) { pxd->len_addr = (pxd->len_addr & cpu_to_le32(~0xffffff)) | cpu_to_le32(len & 0xffffff); } static inline void PXDaddress(pxd_t *pxd, __u64 addr) { pxd->len_addr = (pxd->len_addr & cpu_to_le32(0xffffff)) | cpu_to_le32((addr >> 32)<<24); pxd->addr2 = cpu_to_le32(addr & 0xffffffff); } /* xd_t field extraction */ static inline __u32 lengthPXD(pxd_t *pxd) { return le32_to_cpu((pxd)->len_addr) & 0xffffff; } static inline __u64 addressPXD(pxd_t *pxd) { __u64 n = le32_to_cpu(pxd->len_addr) & ~0xffffff; return (n << 8) + le32_to_cpu(pxd->addr2); } #define MAXTREEHEIGHT 8 /* pxd list */ struct pxdlist { s16 maxnpxd; s16 npxd; pxd_t pxd[MAXTREEHEIGHT]; }; /* * data extent descriptor (dxd) */ typedef struct { __u8 flag; /* 1: flags */ __u8 rsrvd[3]; __le32 size; /* 4: size in byte */ pxd_t loc; /* 8: address and length in unit of fsblksize */ } dxd_t; /* - 16 - */ /* dxd_t flags */ #define DXD_INDEX 0x80 /* B+-tree index */ #define DXD_INLINE 0x40 /* in-line data extent */ #define DXD_EXTENT 0x20 /* out-of-line single extent */ #define DXD_FILE 0x10 /* out-of-line file (inode) */ #define DXD_CORRUPT 0x08 /* Inconsistency detected */ /* dxd_t field construction */ #define DXDlength(dxd, len) PXDlength(&(dxd)->loc, len) #define DXDaddress(dxd, addr) PXDaddress(&(dxd)->loc, addr) #define lengthDXD(dxd) lengthPXD(&(dxd)->loc) #define addressDXD(dxd) addressPXD(&(dxd)->loc) #define DXDsize(dxd, size32) ((dxd)->size = cpu_to_le32(size32)) #define sizeDXD(dxd) le32_to_cpu((dxd)->size) /* * directory entry argument */ struct component_name { int namlen; wchar_t *name; }; /* * DASD limit information - stored in directory inode */ struct dasd { u8 thresh; /* Alert Threshold (in percent) */ u8 delta; /* Alert Threshold delta (in percent) */ u8 rsrvd1; u8 limit_hi; /* DASD limit (in logical blocks) */ __le32 limit_lo; /* DASD limit (in logical blocks) */ u8 rsrvd2[3]; u8 used_hi; /* DASD usage (in logical blocks) */ __le32 used_lo; /* DASD usage (in logical blocks) */ }; #define DASDLIMIT(dasdp) \ (((u64)((dasdp)->limit_hi) << 32) + __le32_to_cpu((dasdp)->limit_lo)) #define setDASDLIMIT(dasdp, limit)\ {\ (dasdp)->limit_hi = ((u64)limit) >> 32;\ (dasdp)->limit_lo = __cpu_to_le32(limit);\ } #define DASDUSED(dasdp) \ (((u64)((dasdp)->used_hi) << 32) + __le32_to_cpu((dasdp)->used_lo)) #define setDASDUSED(dasdp, used)\ {\ (dasdp)->used_hi = ((u64)used) >> 32;\ (dasdp)->used_lo = __cpu_to_le32(used);\ } #endif /* !_H_JFS_TYPES */ |
| 1 1 2 2 1 1 1 1 1 1 1 2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 | // SPDX-License-Identifier: GPL-2.0+ /****************************************************************************** * speedtch.c - Alcatel SpeedTouch USB xDSL modem driver * * Copyright (C) 2001, Alcatel * Copyright (C) 2003, Duncan Sands * Copyright (C) 2004, David Woodhouse * * Based on "modem_run.c", copyright (C) 2001, Benoit Papillault ******************************************************************************/ #include <asm/page.h> #include <linux/device.h> #include <linux/errno.h> #include <linux/firmware.h> #include <linux/hex.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/moduleparam.h> #include <linux/slab.h> #include <linux/stat.h> #include <linux/timer.h> #include <linux/types.h> #include <linux/usb/ch9.h> #include <linux/workqueue.h> #include "usbatm.h" #define DRIVER_AUTHOR "Johan Verrept, Duncan Sands <duncan.sands@free.fr>" #define DRIVER_DESC "Alcatel SpeedTouch USB driver" static const char speedtch_driver_name[] = "speedtch"; #define CTRL_TIMEOUT 2000 /* milliseconds */ #define DATA_TIMEOUT 2000 /* milliseconds */ #define OFFSET_7 0 /* size 1 */ #define OFFSET_b 1 /* size 8 */ #define OFFSET_d 9 /* size 4 */ #define OFFSET_e 13 /* size 1 */ #define OFFSET_f 14 /* size 1 */ #define SIZE_7 1 #define SIZE_b 8 #define SIZE_d 4 #define SIZE_e 1 #define SIZE_f 1 #define MIN_POLL_DELAY 5000 /* milliseconds */ #define MAX_POLL_DELAY 60000 /* milliseconds */ #define RESUBMIT_DELAY 1000 /* milliseconds */ #define DEFAULT_BULK_ALTSETTING 1 #define DEFAULT_ISOC_ALTSETTING 3 #define DEFAULT_DL_512_FIRST 0 #define DEFAULT_ENABLE_ISOC 0 #define DEFAULT_SW_BUFFERING 0 static unsigned int altsetting = 0; /* zero means: use the default */ static bool dl_512_first = DEFAULT_DL_512_FIRST; static bool enable_isoc = DEFAULT_ENABLE_ISOC; static bool sw_buffering = DEFAULT_SW_BUFFERING; #define DEFAULT_B_MAX_DSL 8128 #define DEFAULT_MODEM_MODE 11 #define MODEM_OPTION_LENGTH 16 static const unsigned char DEFAULT_MODEM_OPTION[MODEM_OPTION_LENGTH] = { 0x10, 0x00, 0x00, 0x00, 0x20, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }; static unsigned int BMaxDSL = DEFAULT_B_MAX_DSL; static unsigned char ModemMode = DEFAULT_MODEM_MODE; static unsigned char ModemOption[MODEM_OPTION_LENGTH]; static unsigned int num_ModemOption; module_param(altsetting, uint, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(altsetting, "Alternative setting for data interface (bulk_default: " __MODULE_STRING(DEFAULT_BULK_ALTSETTING) "; isoc_default: " __MODULE_STRING(DEFAULT_ISOC_ALTSETTING) ")"); module_param(dl_512_first, bool, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(dl_512_first, "Read 512 bytes before sending firmware (default: " __MODULE_STRING(DEFAULT_DL_512_FIRST) ")"); module_param(enable_isoc, bool, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(enable_isoc, "Use isochronous transfers if available (default: " __MODULE_STRING(DEFAULT_ENABLE_ISOC) ")"); module_param(sw_buffering, bool, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(sw_buffering, "Enable software buffering (default: " __MODULE_STRING(DEFAULT_SW_BUFFERING) ")"); module_param(BMaxDSL, uint, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(BMaxDSL, "default: " __MODULE_STRING(DEFAULT_B_MAX_DSL)); module_param(ModemMode, byte, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(ModemMode, "default: " __MODULE_STRING(DEFAULT_MODEM_MODE)); module_param_array(ModemOption, byte, &num_ModemOption, S_IRUGO); MODULE_PARM_DESC(ModemOption, "default: 0x10,0x00,0x00,0x00,0x20"); #define INTERFACE_DATA 1 #define ENDPOINT_INT 0x81 #define ENDPOINT_BULK_DATA 0x07 #define ENDPOINT_ISOC_DATA 0x07 #define ENDPOINT_FIRMWARE 0x05 struct speedtch_params { unsigned int altsetting; unsigned int BMaxDSL; unsigned char ModemMode; unsigned char ModemOption[MODEM_OPTION_LENGTH]; }; struct speedtch_instance_data { struct usbatm_data *usbatm; struct speedtch_params params; /* set in probe, constant afterwards */ struct timer_list status_check_timer; struct work_struct status_check_work; unsigned char last_status; int poll_delay; /* milliseconds */ struct timer_list resubmit_timer; struct urb *int_urb; unsigned char int_data[16]; unsigned char scratch_buffer[16]; }; /*************** ** firmware ** ***************/ static void speedtch_set_swbuff(struct speedtch_instance_data *instance, int state) { struct usbatm_data *usbatm = instance->usbatm; struct usb_device *usb_dev = usbatm->usb_dev; int ret; ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), 0x32, 0x40, state ? 0x01 : 0x00, 0x00, NULL, 0, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%sabling SW buffering: usb_control_msg returned %d\n", state ? "En" : "Dis", ret); else usb_dbg(usbatm, "speedtch_set_swbuff: %sbled SW buffering\n", state ? "En" : "Dis"); } static void speedtch_test_sequence(struct speedtch_instance_data *instance) { struct usbatm_data *usbatm = instance->usbatm; struct usb_device *usb_dev = usbatm->usb_dev; unsigned char *buf = instance->scratch_buffer; int ret; /* URB 147 */ buf[0] = 0x1c; buf[1] = 0x50; ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), 0x01, 0x40, 0x0b, 0x00, buf, 2, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%s failed on URB147: %d\n", __func__, ret); /* URB 148 */ buf[0] = 0x32; buf[1] = 0x00; ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), 0x01, 0x40, 0x02, 0x00, buf, 2, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%s failed on URB148: %d\n", __func__, ret); /* URB 149 */ buf[0] = 0x01; buf[1] = 0x00; buf[2] = 0x01; ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), 0x01, 0x40, 0x03, 0x00, buf, 3, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%s failed on URB149: %d\n", __func__, ret); /* URB 150 */ buf[0] = 0x01; buf[1] = 0x00; buf[2] = 0x01; ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), 0x01, 0x40, 0x04, 0x00, buf, 3, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%s failed on URB150: %d\n", __func__, ret); /* Extra initialisation in recent drivers - gives higher speeds */ /* URBext1 */ buf[0] = instance->params.ModemMode; ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), 0x01, 0x40, 0x11, 0x00, buf, 1, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%s failed on URBext1: %d\n", __func__, ret); /* URBext2 */ /* This seems to be the one which actually triggers the higher sync rate -- it does require the new firmware too, although it works OK with older firmware */ ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), 0x01, 0x40, 0x14, 0x00, instance->params.ModemOption, MODEM_OPTION_LENGTH, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%s failed on URBext2: %d\n", __func__, ret); /* URBext3 */ buf[0] = instance->params.BMaxDSL & 0xff; buf[1] = instance->params.BMaxDSL >> 8; ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), 0x01, 0x40, 0x12, 0x00, buf, 2, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%s failed on URBext3: %d\n", __func__, ret); } static int speedtch_upload_firmware(struct speedtch_instance_data *instance, const struct firmware *fw1, const struct firmware *fw2) { unsigned char *buffer; struct usbatm_data *usbatm = instance->usbatm; struct usb_device *usb_dev = usbatm->usb_dev; int actual_length; int ret = 0; int offset; usb_dbg(usbatm, "%s entered\n", __func__); buffer = (unsigned char *)__get_free_page(GFP_KERNEL); if (!buffer) { ret = -ENOMEM; usb_dbg(usbatm, "%s: no memory for buffer!\n", __func__); goto out; } if (!usb_ifnum_to_if(usb_dev, 2)) { ret = -ENODEV; usb_dbg(usbatm, "%s: interface not found!\n", __func__); goto out_free; } /* URB 7 */ if (dl_512_first) { /* some modems need a read before writing the firmware */ ret = usb_bulk_msg(usb_dev, usb_rcvbulkpipe(usb_dev, ENDPOINT_FIRMWARE), buffer, 0x200, &actual_length, 2000); if (ret < 0 && ret != -ETIMEDOUT) usb_warn(usbatm, "%s: read BLOCK0 from modem failed (%d)!\n", __func__, ret); else usb_dbg(usbatm, "%s: BLOCK0 downloaded (%d bytes)\n", __func__, ret); } /* URB 8 : both leds are static green */ for (offset = 0; offset < fw1->size; offset += PAGE_SIZE) { int thislen = min_t(int, PAGE_SIZE, fw1->size - offset); memcpy(buffer, fw1->data + offset, thislen); ret = usb_bulk_msg(usb_dev, usb_sndbulkpipe(usb_dev, ENDPOINT_FIRMWARE), buffer, thislen, &actual_length, DATA_TIMEOUT); if (ret < 0) { usb_err(usbatm, "%s: write BLOCK1 to modem failed (%d)!\n", __func__, ret); goto out_free; } usb_dbg(usbatm, "%s: BLOCK1 uploaded (%zu bytes)\n", __func__, fw1->size); } /* USB led blinking green, ADSL led off */ /* URB 11 */ ret = usb_bulk_msg(usb_dev, usb_rcvbulkpipe(usb_dev, ENDPOINT_FIRMWARE), buffer, 0x200, &actual_length, DATA_TIMEOUT); if (ret < 0) { usb_err(usbatm, "%s: read BLOCK2 from modem failed (%d)!\n", __func__, ret); goto out_free; } usb_dbg(usbatm, "%s: BLOCK2 downloaded (%d bytes)\n", __func__, actual_length); /* URBs 12 to 139 - USB led blinking green, ADSL led off */ for (offset = 0; offset < fw2->size; offset += PAGE_SIZE) { int thislen = min_t(int, PAGE_SIZE, fw2->size - offset); memcpy(buffer, fw2->data + offset, thislen); ret = usb_bulk_msg(usb_dev, usb_sndbulkpipe(usb_dev, ENDPOINT_FIRMWARE), buffer, thislen, &actual_length, DATA_TIMEOUT); if (ret < 0) { usb_err(usbatm, "%s: write BLOCK3 to modem failed (%d)!\n", __func__, ret); goto out_free; } } usb_dbg(usbatm, "%s: BLOCK3 uploaded (%zu bytes)\n", __func__, fw2->size); /* USB led static green, ADSL led static red */ /* URB 142 */ ret = usb_bulk_msg(usb_dev, usb_rcvbulkpipe(usb_dev, ENDPOINT_FIRMWARE), buffer, 0x200, &actual_length, DATA_TIMEOUT); if (ret < 0) { usb_err(usbatm, "%s: read BLOCK4 from modem failed (%d)!\n", __func__, ret); goto out_free; } /* success */ usb_dbg(usbatm, "%s: BLOCK4 downloaded (%d bytes)\n", __func__, actual_length); /* Delay to allow firmware to start up. We can do this here because we're in our own kernel thread anyway. */ msleep_interruptible(1000); if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, instance->params.altsetting)) < 0) { usb_err(usbatm, "%s: setting interface to %d failed (%d)!\n", __func__, instance->params.altsetting, ret); goto out_free; } /* Enable software buffering, if requested */ if (sw_buffering) speedtch_set_swbuff(instance, 1); /* Magic spell; don't ask us what this does */ speedtch_test_sequence(instance); ret = 0; out_free: free_page((unsigned long)buffer); out: return ret; } static int speedtch_find_firmware(struct usbatm_data *usbatm, struct usb_interface *intf, int phase, const struct firmware **fw_p) { struct device *dev = &intf->dev; const u16 bcdDevice = le16_to_cpu(interface_to_usbdev(intf)->descriptor.bcdDevice); const u8 major_revision = bcdDevice >> 8; const u8 minor_revision = bcdDevice & 0xff; char buf[24]; sprintf(buf, "speedtch-%d.bin.%x.%02x", phase, major_revision, minor_revision); usb_dbg(usbatm, "%s: looking for %s\n", __func__, buf); if (request_firmware(fw_p, buf, dev)) { sprintf(buf, "speedtch-%d.bin.%x", phase, major_revision); usb_dbg(usbatm, "%s: looking for %s\n", __func__, buf); if (request_firmware(fw_p, buf, dev)) { sprintf(buf, "speedtch-%d.bin", phase); usb_dbg(usbatm, "%s: looking for %s\n", __func__, buf); if (request_firmware(fw_p, buf, dev)) { usb_err(usbatm, "%s: no stage %d firmware found!\n", __func__, phase); return -ENOENT; } } } usb_info(usbatm, "found stage %d firmware %s\n", phase, buf); return 0; } static int speedtch_heavy_init(struct usbatm_data *usbatm, struct usb_interface *intf) { const struct firmware *fw1, *fw2; struct speedtch_instance_data *instance = usbatm->driver_data; int ret; if ((ret = speedtch_find_firmware(usbatm, intf, 1, &fw1)) < 0) return ret; if ((ret = speedtch_find_firmware(usbatm, intf, 2, &fw2)) < 0) { release_firmware(fw1); return ret; } if ((ret = speedtch_upload_firmware(instance, fw1, fw2)) < 0) usb_err(usbatm, "%s: firmware upload failed (%d)!\n", __func__, ret); release_firmware(fw2); release_firmware(fw1); return ret; } /********** ** ATM ** **********/ static int speedtch_read_status(struct speedtch_instance_data *instance) { struct usbatm_data *usbatm = instance->usbatm; struct usb_device *usb_dev = usbatm->usb_dev; unsigned char *buf = instance->scratch_buffer; int ret; memset(buf, 0, 16); ret = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0), 0x12, 0xc0, 0x07, 0x00, buf + OFFSET_7, SIZE_7, CTRL_TIMEOUT); if (ret < 0) { atm_dbg(usbatm, "%s: MSG 7 failed\n", __func__); return ret; } ret = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0), 0x12, 0xc0, 0x0b, 0x00, buf + OFFSET_b, SIZE_b, CTRL_TIMEOUT); if (ret < 0) { atm_dbg(usbatm, "%s: MSG B failed\n", __func__); return ret; } ret = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0), 0x12, 0xc0, 0x0d, 0x00, buf + OFFSET_d, SIZE_d, CTRL_TIMEOUT); if (ret < 0) { atm_dbg(usbatm, "%s: MSG D failed\n", __func__); return ret; } ret = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0), 0x01, 0xc0, 0x0e, 0x00, buf + OFFSET_e, SIZE_e, CTRL_TIMEOUT); if (ret < 0) { atm_dbg(usbatm, "%s: MSG E failed\n", __func__); return ret; } ret = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0), 0x01, 0xc0, 0x0f, 0x00, buf + OFFSET_f, SIZE_f, CTRL_TIMEOUT); if (ret < 0) { atm_dbg(usbatm, "%s: MSG F failed\n", __func__); return ret; } return 0; } static int speedtch_start_synchro(struct speedtch_instance_data *instance) { struct usbatm_data *usbatm = instance->usbatm; struct usb_device *usb_dev = usbatm->usb_dev; unsigned char *buf = instance->scratch_buffer; int ret; atm_dbg(usbatm, "%s entered\n", __func__); memset(buf, 0, 2); ret = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0), 0x12, 0xc0, 0x04, 0x00, buf, 2, CTRL_TIMEOUT); if (ret < 0) atm_warn(usbatm, "failed to start ADSL synchronisation: %d\n", ret); else atm_dbg(usbatm, "%s: modem prodded. %d bytes returned: %02x %02x\n", __func__, ret, buf[0], buf[1]); return ret; } static void speedtch_check_status(struct work_struct *work) { struct speedtch_instance_data *instance = container_of(work, struct speedtch_instance_data, status_check_work); struct usbatm_data *usbatm = instance->usbatm; struct atm_dev *atm_dev = usbatm->atm_dev; unsigned char *buf = instance->scratch_buffer; int down_speed, up_speed, ret; unsigned char status; #ifdef VERBOSE_DEBUG atm_dbg(usbatm, "%s entered\n", __func__); #endif ret = speedtch_read_status(instance); if (ret < 0) { atm_warn(usbatm, "error %d fetching device status\n", ret); instance->poll_delay = min(2 * instance->poll_delay, MAX_POLL_DELAY); return; } instance->poll_delay = max(instance->poll_delay / 2, MIN_POLL_DELAY); status = buf[OFFSET_7]; if ((status != instance->last_status) || !status) { atm_dbg(usbatm, "%s: line state 0x%02x\n", __func__, status); switch (status) { case 0: atm_dev_signal_change(atm_dev, ATM_PHY_SIG_LOST); if (instance->last_status) atm_info(usbatm, "ADSL line is down\n"); /* It may never resync again unless we ask it to... */ ret = speedtch_start_synchro(instance); break; case 0x08: atm_dev_signal_change(atm_dev, ATM_PHY_SIG_UNKNOWN); atm_info(usbatm, "ADSL line is blocked?\n"); break; case 0x10: atm_dev_signal_change(atm_dev, ATM_PHY_SIG_LOST); atm_info(usbatm, "ADSL line is synchronising\n"); break; case 0x20: down_speed = buf[OFFSET_b] | (buf[OFFSET_b + 1] << 8) | (buf[OFFSET_b + 2] << 16) | (buf[OFFSET_b + 3] << 24); up_speed = buf[OFFSET_b + 4] | (buf[OFFSET_b + 5] << 8) | (buf[OFFSET_b + 6] << 16) | (buf[OFFSET_b + 7] << 24); if (!(down_speed & 0x0000ffff) && !(up_speed & 0x0000ffff)) { down_speed >>= 16; up_speed >>= 16; } atm_dev->link_rate = down_speed * 1000 / 424; atm_dev_signal_change(atm_dev, ATM_PHY_SIG_FOUND); atm_info(usbatm, "ADSL line is up (%d kb/s down | %d kb/s up)\n", down_speed, up_speed); break; default: atm_dev_signal_change(atm_dev, ATM_PHY_SIG_UNKNOWN); atm_info(usbatm, "unknown line state %02x\n", status); break; } instance->last_status = status; } } static void speedtch_status_poll(struct timer_list *t) { struct speedtch_instance_data *instance = timer_container_of(instance, t, status_check_timer); schedule_work(&instance->status_check_work); /* The following check is racy, but the race is harmless */ if (instance->poll_delay < MAX_POLL_DELAY) mod_timer(&instance->status_check_timer, jiffies + msecs_to_jiffies(instance->poll_delay)); else atm_warn(instance->usbatm, "Too many failures - disabling line status polling\n"); } static void speedtch_resubmit_int(struct timer_list *t) { struct speedtch_instance_data *instance = timer_container_of(instance, t, resubmit_timer); struct urb *int_urb = instance->int_urb; int ret; atm_dbg(instance->usbatm, "%s entered\n", __func__); if (int_urb) { ret = usb_submit_urb(int_urb, GFP_ATOMIC); if (!ret) schedule_work(&instance->status_check_work); else { atm_dbg(instance->usbatm, "%s: usb_submit_urb failed with result %d\n", __func__, ret); mod_timer(&instance->resubmit_timer, jiffies + msecs_to_jiffies(RESUBMIT_DELAY)); } } } static void speedtch_handle_int(struct urb *int_urb) { struct speedtch_instance_data *instance = int_urb->context; struct usbatm_data *usbatm = instance->usbatm; unsigned int count = int_urb->actual_length; int status = int_urb->status; int ret; /* The magic interrupt for "up state" */ static const unsigned char up_int[6] = { 0xa1, 0x00, 0x01, 0x00, 0x00, 0x00 }; /* The magic interrupt for "down state" */ static const unsigned char down_int[6] = { 0xa1, 0x00, 0x00, 0x00, 0x00, 0x00 }; atm_dbg(usbatm, "%s entered\n", __func__); if (status < 0) { atm_dbg(usbatm, "%s: nonzero urb status %d!\n", __func__, status); goto fail; } if ((count == 6) && !memcmp(up_int, instance->int_data, 6)) { timer_delete(&instance->status_check_timer); atm_info(usbatm, "DSL line goes up\n"); } else if ((count == 6) && !memcmp(down_int, instance->int_data, 6)) { atm_info(usbatm, "DSL line goes down\n"); } else { int i; atm_dbg(usbatm, "%s: unknown interrupt packet of length %d:", __func__, count); for (i = 0; i < count; i++) printk(" %02x", instance->int_data[i]); printk("\n"); goto fail; } int_urb = instance->int_urb; if (int_urb) { ret = usb_submit_urb(int_urb, GFP_ATOMIC); schedule_work(&instance->status_check_work); if (ret < 0) { atm_dbg(usbatm, "%s: usb_submit_urb failed with result %d\n", __func__, ret); goto fail; } } return; fail: int_urb = instance->int_urb; if (int_urb) mod_timer(&instance->resubmit_timer, jiffies + msecs_to_jiffies(RESUBMIT_DELAY)); } static int speedtch_atm_start(struct usbatm_data *usbatm, struct atm_dev *atm_dev) { struct usb_device *usb_dev = usbatm->usb_dev; struct speedtch_instance_data *instance = usbatm->driver_data; int i, ret; unsigned char mac_str[13]; atm_dbg(usbatm, "%s entered\n", __func__); /* Set MAC address, it is stored in the serial number */ memset(atm_dev->esi, 0, sizeof(atm_dev->esi)); if (usb_string(usb_dev, usb_dev->descriptor.iSerialNumber, mac_str, sizeof(mac_str)) == 12) { for (i = 0; i < 6; i++) atm_dev->esi[i] = (hex_to_bin(mac_str[i * 2]) << 4) + hex_to_bin(mac_str[i * 2 + 1]); } /* Start modem synchronisation */ ret = speedtch_start_synchro(instance); /* Set up interrupt endpoint */ if (instance->int_urb) { ret = usb_submit_urb(instance->int_urb, GFP_KERNEL); if (ret < 0) { /* Doesn't matter; we'll poll anyway */ atm_dbg(usbatm, "%s: submission of interrupt URB failed (%d)!\n", __func__, ret); usb_free_urb(instance->int_urb); instance->int_urb = NULL; } } /* Start status polling */ mod_timer(&instance->status_check_timer, jiffies + msecs_to_jiffies(1000)); return 0; } static void speedtch_atm_stop(struct usbatm_data *usbatm, struct atm_dev *atm_dev) { struct speedtch_instance_data *instance = usbatm->driver_data; struct urb *int_urb = instance->int_urb; atm_dbg(usbatm, "%s entered\n", __func__); timer_delete_sync(&instance->status_check_timer); /* * Since resubmit_timer and int_urb can schedule themselves and * each other, shutting them down correctly takes some care */ instance->int_urb = NULL; /* signal shutdown */ mb(); usb_kill_urb(int_urb); timer_delete_sync(&instance->resubmit_timer); /* * At this point, speedtch_handle_int and speedtch_resubmit_int * can run or be running, but instance->int_urb == NULL means that * they will not reschedule */ usb_kill_urb(int_urb); timer_delete_sync(&instance->resubmit_timer); usb_free_urb(int_urb); flush_work(&instance->status_check_work); } static int speedtch_pre_reset(struct usb_interface *intf) { return 0; } static int speedtch_post_reset(struct usb_interface *intf) { return 0; } /********** ** USB ** **********/ static const struct usb_device_id speedtch_usb_ids[] = { {USB_DEVICE(0x06b9, 0x4061)}, {} }; MODULE_DEVICE_TABLE(usb, speedtch_usb_ids); static int speedtch_usb_probe(struct usb_interface *, const struct usb_device_id *); static struct usb_driver speedtch_usb_driver = { .name = speedtch_driver_name, .probe = speedtch_usb_probe, .disconnect = usbatm_usb_disconnect, .pre_reset = speedtch_pre_reset, .post_reset = speedtch_post_reset, .id_table = speedtch_usb_ids }; static void speedtch_release_interfaces(struct usb_device *usb_dev, int num_interfaces) { struct usb_interface *cur_intf; int i; for (i = 0; i < num_interfaces; i++) { cur_intf = usb_ifnum_to_if(usb_dev, i); if (cur_intf) { usb_set_intfdata(cur_intf, NULL); usb_driver_release_interface(&speedtch_usb_driver, cur_intf); } } } static int speedtch_bind(struct usbatm_data *usbatm, struct usb_interface *intf, const struct usb_device_id *id) { struct usb_device *usb_dev = interface_to_usbdev(intf); struct usb_interface *cur_intf, *data_intf; struct speedtch_instance_data *instance; int ifnum = intf->altsetting->desc.bInterfaceNumber; int num_interfaces = usb_dev->actconfig->desc.bNumInterfaces; int i, ret; int use_isoc; usb_dbg(usbatm, "%s entered\n", __func__); /* sanity checks */ if (usb_dev->descriptor.bDeviceClass != USB_CLASS_VENDOR_SPEC) { usb_err(usbatm, "%s: wrong device class %d\n", __func__, usb_dev->descriptor.bDeviceClass); return -ENODEV; } data_intf = usb_ifnum_to_if(usb_dev, INTERFACE_DATA); if (!data_intf) { usb_err(usbatm, "%s: data interface not found!\n", __func__); return -ENODEV; } /* claim all interfaces */ for (i = 0; i < num_interfaces; i++) { cur_intf = usb_ifnum_to_if(usb_dev, i); if ((i != ifnum) && cur_intf) { ret = usb_driver_claim_interface(&speedtch_usb_driver, cur_intf, usbatm); if (ret < 0) { usb_err(usbatm, "%s: failed to claim interface %2d (%d)!\n", __func__, i, ret); speedtch_release_interfaces(usb_dev, i); return ret; } } } instance = kzalloc(sizeof(*instance), GFP_KERNEL); if (!instance) { ret = -ENOMEM; goto fail_release; } instance->usbatm = usbatm; /* module parameters may change at any moment, so take a snapshot */ instance->params.altsetting = altsetting; instance->params.BMaxDSL = BMaxDSL; instance->params.ModemMode = ModemMode; memcpy(instance->params.ModemOption, DEFAULT_MODEM_OPTION, MODEM_OPTION_LENGTH); memcpy(instance->params.ModemOption, ModemOption, num_ModemOption); use_isoc = enable_isoc; if (instance->params.altsetting) if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, instance->params.altsetting)) < 0) { usb_err(usbatm, "%s: setting interface to %2d failed (%d)!\n", __func__, instance->params.altsetting, ret); instance->params.altsetting = 0; /* fall back to default */ } if (!instance->params.altsetting && use_isoc) if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, DEFAULT_ISOC_ALTSETTING)) < 0) { usb_dbg(usbatm, "%s: setting interface to %2d failed (%d)!\n", __func__, DEFAULT_ISOC_ALTSETTING, ret); use_isoc = 0; /* fall back to bulk */ } if (use_isoc) { const struct usb_host_interface *desc = data_intf->cur_altsetting; const __u8 target_address = USB_DIR_IN | usbatm->driver->isoc_in; use_isoc = 0; /* fall back to bulk if endpoint not found */ for (i = 0; i < desc->desc.bNumEndpoints; i++) { const struct usb_endpoint_descriptor *endpoint_desc = &desc->endpoint[i].desc; if ((endpoint_desc->bEndpointAddress == target_address)) { use_isoc = usb_endpoint_xfer_isoc(endpoint_desc); break; } } if (!use_isoc) usb_info(usbatm, "isochronous transfer not supported - using bulk\n"); } if (!use_isoc && !instance->params.altsetting) if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, DEFAULT_BULK_ALTSETTING)) < 0) { usb_err(usbatm, "%s: setting interface to %2d failed (%d)!\n", __func__, DEFAULT_BULK_ALTSETTING, ret); goto fail_free; } if (!instance->params.altsetting) instance->params.altsetting = use_isoc ? DEFAULT_ISOC_ALTSETTING : DEFAULT_BULK_ALTSETTING; usbatm->flags |= (use_isoc ? UDSL_USE_ISOC : 0); INIT_WORK(&instance->status_check_work, speedtch_check_status); timer_setup(&instance->status_check_timer, speedtch_status_poll, 0); instance->last_status = 0xff; instance->poll_delay = MIN_POLL_DELAY; timer_setup(&instance->resubmit_timer, speedtch_resubmit_int, 0); instance->int_urb = usb_alloc_urb(0, GFP_KERNEL); if (instance->int_urb) usb_fill_int_urb(instance->int_urb, usb_dev, usb_rcvintpipe(usb_dev, ENDPOINT_INT), instance->int_data, sizeof(instance->int_data), speedtch_handle_int, instance, 16); else usb_dbg(usbatm, "%s: no memory for interrupt urb!\n", __func__); /* check whether the modem already seems to be alive */ ret = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0), 0x12, 0xc0, 0x07, 0x00, instance->scratch_buffer + OFFSET_7, SIZE_7, 500); usbatm->flags |= (ret == SIZE_7 ? UDSL_SKIP_HEAVY_INIT : 0); usb_dbg(usbatm, "%s: firmware %s loaded\n", __func__, usbatm->flags & UDSL_SKIP_HEAVY_INIT ? "already" : "not"); if (!(usbatm->flags & UDSL_SKIP_HEAVY_INIT)) if ((ret = usb_reset_device(usb_dev)) < 0) { usb_err(usbatm, "%s: device reset failed (%d)!\n", __func__, ret); goto fail_free; } usbatm->driver_data = instance; return 0; fail_free: usb_free_urb(instance->int_urb); kfree(instance); fail_release: speedtch_release_interfaces(usb_dev, num_interfaces); return ret; } static void speedtch_unbind(struct usbatm_data *usbatm, struct usb_interface *intf) { struct usb_device *usb_dev = interface_to_usbdev(intf); struct speedtch_instance_data *instance = usbatm->driver_data; usb_dbg(usbatm, "%s entered\n", __func__); speedtch_release_interfaces(usb_dev, usb_dev->actconfig->desc.bNumInterfaces); usb_free_urb(instance->int_urb); kfree(instance); } /*********** ** init ** ***********/ static struct usbatm_driver speedtch_usbatm_driver = { .driver_name = speedtch_driver_name, .bind = speedtch_bind, .heavy_init = speedtch_heavy_init, .unbind = speedtch_unbind, .atm_start = speedtch_atm_start, .atm_stop = speedtch_atm_stop, .bulk_in = ENDPOINT_BULK_DATA, .bulk_out = ENDPOINT_BULK_DATA, .isoc_in = ENDPOINT_ISOC_DATA }; static int speedtch_usb_probe(struct usb_interface *intf, const struct usb_device_id *id) { return usbatm_usb_probe(intf, id, &speedtch_usbatm_driver); } module_usb_driver(speedtch_usb_driver); MODULE_AUTHOR(DRIVER_AUTHOR); MODULE_DESCRIPTION(DRIVER_DESC); MODULE_LICENSE("GPL"); |
| 5 5 2 1 1 1 63 63 1 1 19 64 64 1 20 60 5 60 60 60 29 45 33 33 1 64 64 63 39 26 62 59 5 62 72 72 2 2 2 5 34 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 | // SPDX-License-Identifier: GPL-2.0 /* * Copyright (c) 2000-2005 Silicon Graphics, Inc. * Copyright (c) 2016-2025 Christoph Hellwig. * All Rights Reserved. */ #include "xfs_platform.h" #include "xfs_shared.h" #include "xfs_format.h" #include "xfs_log_format.h" #include "xfs_trans_resv.h" #include "xfs_mount.h" #include "xfs_inode.h" #include "xfs_trans.h" #include "xfs_iomap.h" #include "xfs_trace.h" #include "xfs_bmap.h" #include "xfs_bmap_util.h" #include "xfs_reflink.h" #include "xfs_errortag.h" #include "xfs_error.h" #include "xfs_icache.h" #include "xfs_zone_alloc.h" #include "xfs_rtgroup.h" struct xfs_writepage_ctx { struct iomap_writepage_ctx ctx; unsigned int data_seq; unsigned int cow_seq; }; static inline struct xfs_writepage_ctx * XFS_WPC(struct iomap_writepage_ctx *ctx) { return container_of(ctx, struct xfs_writepage_ctx, ctx); } /* * Fast and loose check if this write could update the on-disk inode size. */ static inline bool xfs_ioend_is_append(struct iomap_ioend *ioend) { return ioend->io_offset + ioend->io_size > XFS_I(ioend->io_inode)->i_disk_size; } /* * Update on-disk file size now that data has been written to disk. */ int xfs_setfilesize( struct xfs_inode *ip, xfs_off_t offset, size_t size) { struct xfs_mount *mp = ip->i_mount; struct xfs_trans *tp; xfs_fsize_t isize; int error; error = xfs_trans_alloc(mp, &M_RES(mp)->tr_fsyncts, 0, 0, 0, &tp); if (error) return error; xfs_ilock(ip, XFS_ILOCK_EXCL); isize = xfs_new_eof(ip, offset + size); if (!isize) { xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_trans_cancel(tp); return 0; } trace_xfs_setfilesize(ip, offset, size); ip->i_disk_size = isize; xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); return xfs_trans_commit(tp); } static void xfs_ioend_put_open_zones( struct iomap_ioend *ioend) { struct iomap_ioend *tmp; /* * Put the open zone for all ioends merged into this one (if any). */ list_for_each_entry(tmp, &ioend->io_list, io_list) xfs_open_zone_put(tmp->io_private); /* * The main ioend might not have an open zone if the submission failed * before xfs_zone_alloc_and_submit got called. */ if (ioend->io_private) xfs_open_zone_put(ioend->io_private); } /* * IO write completion. */ STATIC void xfs_end_ioend_write( struct iomap_ioend *ioend) { struct xfs_inode *ip = XFS_I(ioend->io_inode); struct xfs_mount *mp = ip->i_mount; bool is_zoned = xfs_is_zoned_inode(ip); xfs_off_t offset = ioend->io_offset; size_t size = ioend->io_size; unsigned int nofs_flag; int error; /* * We can allocate memory here while doing writeback on behalf of * memory reclaim. To avoid memory allocation deadlocks set the * task-wide nofs context for the following operations. */ nofs_flag = memalloc_nofs_save(); /* * Just clean up the in-memory structures if the fs has been shut down. */ if (xfs_is_shutdown(mp)) { error = -EIO; goto done; } /* * Clean up all COW blocks and underlying data fork delalloc blocks on * I/O error. The delalloc punch is required because this ioend was * mapped to blocks in the COW fork and the associated pages are no * longer dirty. If we don't remove delalloc blocks here, they become * stale and can corrupt free space accounting on unmount. */ error = blk_status_to_errno(ioend->io_bio.bi_status); if (unlikely(error)) { if (ioend->io_flags & IOMAP_IOEND_SHARED) { ASSERT(!is_zoned); xfs_reflink_cancel_cow_range(ip, offset, size, true); xfs_bmap_punch_delalloc_range(ip, XFS_DATA_FORK, offset, offset + size, NULL); } goto done; } /* * Success: commit the COW or unwritten blocks if needed. */ if (is_zoned) error = xfs_zoned_end_io(ip, offset, size, ioend->io_sector, ioend->io_private, NULLFSBLOCK); else if (ioend->io_flags & IOMAP_IOEND_SHARED) error = xfs_reflink_end_cow(ip, offset, size); else if (ioend->io_flags & IOMAP_IOEND_UNWRITTEN) error = xfs_iomap_write_unwritten(ip, offset, size, false); if (!error && !(ioend->io_flags & IOMAP_IOEND_DIRECT) && xfs_ioend_is_append(ioend)) error = xfs_setfilesize(ip, offset, size); done: if (is_zoned) xfs_ioend_put_open_zones(ioend); iomap_finish_ioends(ioend, error); memalloc_nofs_restore(nofs_flag); } /* * Finish all pending IO completions that require transactional modifications. * * We try to merge physical and logically contiguous ioends before completion to * minimise the number of transactions we need to perform during IO completion. * Both unwritten extent conversion and COW remapping need to iterate and modify * one physical extent at a time, so we gain nothing by merging physically * discontiguous extents here. * * The ioend chain length that we can be processing here is largely unbound in * length and we may have to perform significant amounts of work on each ioend * to complete it. Hence we have to be careful about holding the CPU for too * long in this loop. */ void xfs_end_io( struct work_struct *work) { struct xfs_inode *ip = container_of(work, struct xfs_inode, i_ioend_work); struct iomap_ioend *ioend; struct list_head tmp; unsigned long flags; spin_lock_irqsave(&ip->i_ioend_lock, flags); list_replace_init(&ip->i_ioend_list, &tmp); spin_unlock_irqrestore(&ip->i_ioend_lock, flags); iomap_sort_ioends(&tmp); while ((ioend = list_first_entry_or_null(&tmp, struct iomap_ioend, io_list))) { list_del_init(&ioend->io_list); iomap_ioend_try_merge(ioend, &tmp); if (bio_op(&ioend->io_bio) == REQ_OP_READ) iomap_finish_ioends(ioend, blk_status_to_errno(ioend->io_bio.bi_status)); else xfs_end_ioend_write(ioend); cond_resched(); } } void xfs_end_bio( struct bio *bio) { struct iomap_ioend *ioend = iomap_ioend_from_bio(bio); struct xfs_inode *ip = XFS_I(ioend->io_inode); struct xfs_mount *mp = ip->i_mount; unsigned long flags; /* * For Appends record the actually written block number and set the * boundary flag if needed. */ if (IS_ENABLED(CONFIG_XFS_RT) && bio_is_zone_append(bio)) { ioend->io_sector = bio->bi_iter.bi_sector; xfs_mark_rtg_boundary(ioend); } spin_lock_irqsave(&ip->i_ioend_lock, flags); if (list_empty(&ip->i_ioend_list)) WARN_ON_ONCE(!queue_work(mp->m_unwritten_workqueue, &ip->i_ioend_work)); list_add_tail(&ioend->io_list, &ip->i_ioend_list); spin_unlock_irqrestore(&ip->i_ioend_lock, flags); } /* * We cannot cancel the ioend directly on error. We may have already set other * pages under writeback and hence we have to run I/O completion to mark the * error state of the pages under writeback appropriately. * * If the folio has delalloc blocks on it, the caller is asking us to punch them * out. If we don't, we can leave a stale delalloc mapping covered by a clean * page that needs to be dirtied again before the delalloc mapping can be * converted. This stale delalloc mapping can trip up a later direct I/O read * operation on the same region. * * We prevent this by truncating away the delalloc regions on the folio. Because * they are delalloc, we can do this without needing a transaction. Indeed - if * we get ENOSPC errors, we have to be able to do this truncation without a * transaction as there is no space left for block reservation (typically why * we see a ENOSPC in writeback). */ static void xfs_discard_folio( struct folio *folio, loff_t pos) { struct xfs_inode *ip = XFS_I(folio->mapping->host); struct xfs_mount *mp = ip->i_mount; if (xfs_is_shutdown(mp)) return; xfs_alert_ratelimited(mp, "page discard on page "PTR_FMT", inode 0x%llx, pos %llu.", folio, ip->i_ino, pos); /* * The end of the punch range is always the offset of the first * byte of the next folio. Hence the end offset is only dependent on the * folio itself and not the start offset that is passed in. */ xfs_bmap_punch_delalloc_range(ip, XFS_DATA_FORK, pos, folio_next_pos(folio), NULL); } /* * Fast revalidation of the cached writeback mapping. Return true if the current * mapping is valid, false otherwise. */ static bool xfs_imap_valid( struct iomap_writepage_ctx *wpc, struct xfs_inode *ip, loff_t offset) { if (offset < wpc->iomap.offset || offset >= wpc->iomap.offset + wpc->iomap.length) return false; /* * If this is a COW mapping, it is sufficient to check that the mapping * covers the offset. Be careful to check this first because the caller * can revalidate a COW mapping without updating the data seqno. */ if (wpc->iomap.flags & IOMAP_F_SHARED) return true; /* * This is not a COW mapping. Check the sequence number of the data fork * because concurrent changes could have invalidated the extent. Check * the COW fork because concurrent changes since the last time we * checked (and found nothing at this offset) could have added * overlapping blocks. */ if (XFS_WPC(wpc)->data_seq != READ_ONCE(ip->i_df.if_seq)) { trace_xfs_wb_data_iomap_invalid(ip, &wpc->iomap, XFS_WPC(wpc)->data_seq, XFS_DATA_FORK); return false; } if (xfs_inode_has_cow_data(ip) && XFS_WPC(wpc)->cow_seq != READ_ONCE(ip->i_cowfp->if_seq)) { trace_xfs_wb_cow_iomap_invalid(ip, &wpc->iomap, XFS_WPC(wpc)->cow_seq, XFS_COW_FORK); return false; } return true; } static int xfs_map_blocks( struct iomap_writepage_ctx *wpc, loff_t offset, unsigned int len) { struct xfs_inode *ip = XFS_I(wpc->inode); struct xfs_mount *mp = ip->i_mount; ssize_t count = i_blocksize(wpc->inode); xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset); xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, offset + count); xfs_fileoff_t cow_fsb; int whichfork; struct xfs_bmbt_irec imap; struct xfs_iext_cursor icur; int retries = 0; int error = 0; unsigned int *seq; if (xfs_is_shutdown(mp)) return -EIO; XFS_ERRORTAG_DELAY(mp, XFS_ERRTAG_WB_DELAY_MS); /* * COW fork blocks can overlap data fork blocks even if the blocks * aren't shared. COW I/O always takes precedent, so we must always * check for overlap on reflink inodes unless the mapping is already a * COW one, or the COW fork hasn't changed from the last time we looked * at it. * * It's safe to check the COW fork if_seq here without the ILOCK because * we've indirectly protected against concurrent updates: writeback has * the page locked, which prevents concurrent invalidations by reflink * and directio and prevents concurrent buffered writes to the same * page. Changes to if_seq always happen under i_lock, which protects * against concurrent updates and provides a memory barrier on the way * out that ensures that we always see the current value. */ if (xfs_imap_valid(wpc, ip, offset)) return 0; /* * If we don't have a valid map, now it's time to get a new one for this * offset. This will convert delayed allocations (including COW ones) * into real extents. If we return without a valid map, it means we * landed in a hole and we skip the block. */ retry: cow_fsb = NULLFILEOFF; whichfork = XFS_DATA_FORK; xfs_ilock(ip, XFS_ILOCK_SHARED); ASSERT(!xfs_need_iread_extents(&ip->i_df)); /* * Check if this is offset is covered by a COW extents, and if yes use * it directly instead of looking up anything in the data fork. */ if (xfs_inode_has_cow_data(ip) && xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &imap)) cow_fsb = imap.br_startoff; if (cow_fsb != NULLFILEOFF && cow_fsb <= offset_fsb) { XFS_WPC(wpc)->cow_seq = READ_ONCE(ip->i_cowfp->if_seq); xfs_iunlock(ip, XFS_ILOCK_SHARED); whichfork = XFS_COW_FORK; goto allocate_blocks; } /* * No COW extent overlap. Revalidate now that we may have updated * ->cow_seq. If the data mapping is still valid, we're done. */ if (xfs_imap_valid(wpc, ip, offset)) { xfs_iunlock(ip, XFS_ILOCK_SHARED); return 0; } /* * If we don't have a valid map, now it's time to get a new one for this * offset. This will convert delayed allocations (including COW ones) * into real extents. */ if (!xfs_iext_lookup_extent(ip, &ip->i_df, offset_fsb, &icur, &imap)) imap.br_startoff = end_fsb; /* fake a hole past EOF */ XFS_WPC(wpc)->data_seq = READ_ONCE(ip->i_df.if_seq); xfs_iunlock(ip, XFS_ILOCK_SHARED); /* landed in a hole or beyond EOF? */ if (imap.br_startoff > offset_fsb) { imap.br_blockcount = imap.br_startoff - offset_fsb; imap.br_startoff = offset_fsb; imap.br_startblock = HOLESTARTBLOCK; imap.br_state = XFS_EXT_NORM; } /* * Truncate to the next COW extent if there is one. This is the only * opportunity to do this because we can skip COW fork lookups for the * subsequent blocks in the mapping; however, the requirement to treat * the COW range separately remains. */ if (cow_fsb != NULLFILEOFF && cow_fsb < imap.br_startoff + imap.br_blockcount) imap.br_blockcount = cow_fsb - imap.br_startoff; /* got a delalloc extent? */ if (imap.br_startblock != HOLESTARTBLOCK && isnullstartblock(imap.br_startblock)) goto allocate_blocks; xfs_bmbt_to_iomap(ip, &wpc->iomap, &imap, 0, 0, XFS_WPC(wpc)->data_seq); trace_xfs_map_blocks_found(ip, offset, count, whichfork, &imap); return 0; allocate_blocks: /* * Convert a dellalloc extent to a real one. The current page is held * locked so nothing could have removed the block backing offset_fsb, * although it could have moved from the COW to the data fork by another * thread. */ if (whichfork == XFS_COW_FORK) seq = &XFS_WPC(wpc)->cow_seq; else seq = &XFS_WPC(wpc)->data_seq; error = xfs_bmapi_convert_delalloc(ip, whichfork, offset, &wpc->iomap, seq); if (error) { /* * If we failed to find the extent in the COW fork we might have * raced with a COW to data fork conversion or truncate. * Restart the lookup to catch the extent in the data fork for * the former case, but prevent additional retries to avoid * looping forever for the latter case. */ if (error == -EAGAIN && whichfork == XFS_COW_FORK && !retries++) goto retry; ASSERT(error != -EAGAIN); return error; } /* * Due to merging the return real extent might be larger than the * original delalloc one. Trim the return extent to the next COW * boundary again to force a re-lookup. */ if (whichfork != XFS_COW_FORK && cow_fsb != NULLFILEOFF) { loff_t cow_offset = XFS_FSB_TO_B(mp, cow_fsb); if (cow_offset < wpc->iomap.offset + wpc->iomap.length) wpc->iomap.length = cow_offset - wpc->iomap.offset; } ASSERT(wpc->iomap.offset <= offset); ASSERT(wpc->iomap.offset + wpc->iomap.length > offset); trace_xfs_map_blocks_alloc(ip, offset, count, whichfork, &imap); return 0; } static ssize_t xfs_writeback_range( struct iomap_writepage_ctx *wpc, struct folio *folio, u64 offset, unsigned int len, u64 end_pos) { ssize_t ret; ret = xfs_map_blocks(wpc, offset, len); if (!ret) ret = iomap_add_to_ioend(wpc, folio, offset, end_pos, len); if (ret < 0) xfs_discard_folio(folio, offset); return ret; } static bool xfs_ioend_needs_wq_completion( struct iomap_ioend *ioend) { /* Changing inode size requires a transaction. */ if (xfs_ioend_is_append(ioend)) return true; /* Extent manipulation requires a transaction. */ if (ioend->io_flags & (IOMAP_IOEND_UNWRITTEN | IOMAP_IOEND_SHARED)) return true; /* Page cache invalidation cannot be done in irq context. */ if (ioend->io_flags & IOMAP_IOEND_DONTCACHE) return true; return false; } static int xfs_writeback_submit( struct iomap_writepage_ctx *wpc, int error) { struct iomap_ioend *ioend = wpc->wb_ctx; /* * Convert CoW extents to regular. * * We can allocate memory here while doing writeback on behalf of memory * reclaim. To avoid memory allocation deadlocks, set the task-wide * nofs context. */ if (!error && (ioend->io_flags & IOMAP_IOEND_SHARED)) { unsigned int nofs_flag; nofs_flag = memalloc_nofs_save(); error = xfs_reflink_convert_cow(XFS_I(ioend->io_inode), ioend->io_offset, ioend->io_size); memalloc_nofs_restore(nofs_flag); } /* * Send ioends that might require a transaction to the completion wq. */ if (xfs_ioend_needs_wq_completion(ioend)) ioend->io_bio.bi_end_io = xfs_end_bio; return iomap_ioend_writeback_submit(wpc, error); } static const struct iomap_writeback_ops xfs_writeback_ops = { .writeback_range = xfs_writeback_range, .writeback_submit = xfs_writeback_submit, }; struct xfs_zoned_writepage_ctx { struct iomap_writepage_ctx ctx; struct xfs_open_zone *open_zone; }; static inline struct xfs_zoned_writepage_ctx * XFS_ZWPC(struct iomap_writepage_ctx *ctx) { return container_of(ctx, struct xfs_zoned_writepage_ctx, ctx); } static int xfs_zoned_map_blocks( struct iomap_writepage_ctx *wpc, loff_t offset, unsigned int len) { struct xfs_inode *ip = XFS_I(wpc->inode); struct xfs_mount *mp = ip->i_mount; xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset); xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, offset + len); xfs_filblks_t count_fsb; struct xfs_bmbt_irec imap, del; struct xfs_iext_cursor icur; if (xfs_is_shutdown(mp)) return -EIO; XFS_ERRORTAG_DELAY(mp, XFS_ERRTAG_WB_DELAY_MS); /* * All dirty data must be covered by delalloc extents. But truncate can * remove delalloc extents underneath us or reduce their size. * Returning a hole tells iomap to not write back any data from this * range, which is the right thing to do in that case. * * Otherwise just tell iomap to treat ranges previously covered by a * delalloc extent as mapped. The actual block allocation will be done * just before submitting the bio. * * This implies we never map outside folios that are locked or marked * as under writeback, and thus there is no need check the fork sequence * count here. */ xfs_ilock(ip, XFS_ILOCK_EXCL); if (!xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &imap)) imap.br_startoff = end_fsb; /* fake a hole past EOF */ if (imap.br_startoff > offset_fsb) { imap.br_blockcount = imap.br_startoff - offset_fsb; imap.br_startoff = offset_fsb; imap.br_startblock = HOLESTARTBLOCK; imap.br_state = XFS_EXT_NORM; xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_bmbt_to_iomap(ip, &wpc->iomap, &imap, 0, 0, 0); return 0; } end_fsb = min(end_fsb, imap.br_startoff + imap.br_blockcount); count_fsb = end_fsb - offset_fsb; del = imap; xfs_trim_extent(&del, offset_fsb, count_fsb); xfs_bmap_del_extent_delay(ip, XFS_COW_FORK, &icur, &imap, &del, XFS_BMAPI_REMAP); xfs_iunlock(ip, XFS_ILOCK_EXCL); wpc->iomap.type = IOMAP_MAPPED; wpc->iomap.flags = IOMAP_F_DIRTY; wpc->iomap.bdev = mp->m_rtdev_targp->bt_bdev; wpc->iomap.offset = offset; wpc->iomap.length = XFS_FSB_TO_B(mp, count_fsb); wpc->iomap.flags = IOMAP_F_ANON_WRITE; trace_xfs_zoned_map_blocks(ip, offset, wpc->iomap.length); return 0; } static ssize_t xfs_zoned_writeback_range( struct iomap_writepage_ctx *wpc, struct folio *folio, u64 offset, unsigned int len, u64 end_pos) { ssize_t ret; ret = xfs_zoned_map_blocks(wpc, offset, len); if (!ret) ret = iomap_add_to_ioend(wpc, folio, offset, end_pos, len); if (ret < 0) xfs_discard_folio(folio, offset); return ret; } static int xfs_zoned_writeback_submit( struct iomap_writepage_ctx *wpc, int error) { struct iomap_ioend *ioend = wpc->wb_ctx; ioend->io_bio.bi_end_io = xfs_end_bio; if (error) { ioend->io_bio.bi_status = errno_to_blk_status(error); bio_endio(&ioend->io_bio); return error; } xfs_zone_alloc_and_submit(ioend, &XFS_ZWPC(wpc)->open_zone); return 0; } static const struct iomap_writeback_ops xfs_zoned_writeback_ops = { .writeback_range = xfs_zoned_writeback_range, .writeback_submit = xfs_zoned_writeback_submit, }; STATIC int xfs_vm_writepages( struct address_space *mapping, struct writeback_control *wbc) { struct xfs_inode *ip = XFS_I(mapping->host); xfs_iflags_clear(ip, XFS_ITRUNCATED); if (xfs_is_zoned_inode(ip)) { struct xfs_zoned_writepage_ctx xc = { .ctx = { .inode = mapping->host, .wbc = wbc, .ops = &xfs_zoned_writeback_ops }, }; int error; error = iomap_writepages(&xc.ctx); if (xc.open_zone) xfs_open_zone_put(xc.open_zone); return error; } else { struct xfs_writepage_ctx wpc = { .ctx = { .inode = mapping->host, .wbc = wbc, .ops = &xfs_writeback_ops }, }; return iomap_writepages(&wpc.ctx); } } STATIC int xfs_dax_writepages( struct address_space *mapping, struct writeback_control *wbc) { struct xfs_inode *ip = XFS_I(mapping->host); xfs_iflags_clear(ip, XFS_ITRUNCATED); return dax_writeback_mapping_range(mapping, xfs_inode_buftarg(ip)->bt_daxdev, wbc); } STATIC sector_t xfs_vm_bmap( struct address_space *mapping, sector_t block) { struct xfs_inode *ip = XFS_I(mapping->host); trace_xfs_vm_bmap(ip); /* * The swap code (ab-)uses ->bmap to get a block mapping and then * bypasses the file system for actual I/O. We really can't allow * that on reflinks inodes, so we have to skip out here. And yes, * 0 is the magic code for a bmap error. * * Since we don't pass back blockdev info, we can't return bmap * information for rt files either. */ if (xfs_is_cow_inode(ip) || XFS_IS_REALTIME_INODE(ip)) return 0; return iomap_bmap(mapping, block, &xfs_read_iomap_ops); } STATIC int xfs_vm_read_folio( struct file *unused, struct folio *folio) { iomap_bio_read_folio(folio, &xfs_read_iomap_ops); return 0; } STATIC void xfs_vm_readahead( struct readahead_control *rac) { iomap_bio_readahead(rac, &xfs_read_iomap_ops); } static int xfs_vm_swap_activate( struct swap_info_struct *sis, struct file *swap_file, sector_t *span) { struct xfs_inode *ip = XFS_I(file_inode(swap_file)); if (xfs_is_zoned_inode(ip)) return -EINVAL; /* * Swap file activation can race against concurrent shared extent * removal in files that have been cloned. If this happens, * iomap_swapfile_iter() can fail because it encountered a shared * extent even though an operation is in progress to remove those * shared extents. * * This race becomes problematic when we defer extent removal * operations beyond the end of a syscall (i.e. use async background * processing algorithms). Users think the extents are no longer * shared, but iomap_swapfile_iter() still sees them as shared * because the refcountbt entries for the extents being removed have * not yet been updated. Hence the swapon call fails unexpectedly. * * The race condition is currently most obvious from the unlink() * operation as extent removal is deferred until after the last * reference to the inode goes away. We then process the extent * removal asynchronously, hence triggers the "syscall completed but * work not done" condition mentioned above. To close this race * window, we need to flush any pending inodegc operations to ensure * they have updated the refcountbt records before we try to map the * swapfile. */ xfs_inodegc_flush(ip->i_mount); /* * Direct the swap code to the correct block device when this file * sits on the RT device. */ sis->bdev = xfs_inode_buftarg(ip)->bt_bdev; return iomap_swapfile_activate(sis, swap_file, span, &xfs_read_iomap_ops); } const struct address_space_operations xfs_address_space_operations = { .read_folio = xfs_vm_read_folio, .readahead = xfs_vm_readahead, .writepages = xfs_vm_writepages, .dirty_folio = iomap_dirty_folio, .release_folio = iomap_release_folio, .invalidate_folio = iomap_invalidate_folio, .bmap = xfs_vm_bmap, .migrate_folio = filemap_migrate_folio, .is_partially_uptodate = iomap_is_partially_uptodate, .error_remove_folio = generic_error_remove_folio, .swap_activate = xfs_vm_swap_activate, }; const struct address_space_operations xfs_dax_aops = { .writepages = xfs_dax_writepages, .dirty_folio = noop_dirty_folio, .swap_activate = xfs_vm_swap_activate, }; |
| 13 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 | /* * Copyright (c) 2005 Topspin Communications. All rights reserved. * Copyright (c) 2005, 2006 Cisco Systems. All rights reserved. * Copyright (c) 2005-2017 Mellanox Technologies. All rights reserved. * Copyright (c) 2005 Voltaire, Inc. All rights reserved. * Copyright (c) 2005 PathScale, Inc. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU * General Public License (GPL) Version 2, available from the file * COPYING in the main directory of this source tree, or the * OpenIB.org BSD license below: * * Redistribution and use in source and binary forms, with or * without modification, are permitted provided that the following * conditions are met: * * - Redistributions of source code must retain the above * copyright notice, this list of conditions and the following * disclaimer. * * - Redistributions in binary form must reproduce the above * copyright notice, this list of conditions and the following * disclaimer in the documentation and/or other materials * provided with the distribution. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ #ifndef RDMA_CORE_H #define RDMA_CORE_H #include <linux/idr.h> #include <rdma/uverbs_types.h> #include <rdma/uverbs_ioctl.h> #include <rdma/ib_verbs.h> #include <linux/mutex.h> struct ib_uverbs_device; void uverbs_destroy_ufile_hw(struct ib_uverbs_file *ufile, enum rdma_remove_reason reason); int uobj_destroy(struct ib_uobject *uobj, struct uverbs_attr_bundle *attrs); /* * Get an ib_uobject that corresponds to the given id from ufile, assuming * the object is from the given type. Lock it to the required access when * applicable. * This function could create (access == NEW), destroy (access == DESTROY) * or unlock (access == READ || access == WRITE) objects if required. * The action will be finalized only when uverbs_finalize_object or * uverbs_finalize_objects are called. */ struct ib_uobject * uverbs_get_uobject_from_file(u16 object_id, enum uverbs_obj_access access, s64 id, struct uverbs_attr_bundle *attrs); void uverbs_finalize_object(struct ib_uobject *uobj, enum uverbs_obj_access access, bool hw_obj_valid, bool commit, struct uverbs_attr_bundle *attrs); int uverbs_output_written(const struct uverbs_attr_bundle *bundle, size_t idx); void setup_ufile_idr_uobject(struct ib_uverbs_file *ufile); void release_ufile_idr_uobject(struct ib_uverbs_file *ufile); struct ib_udata *uverbs_get_cleared_udata(struct uverbs_attr_bundle *attrs); /* * This is the runtime description of the uverbs API, used by the syscall * machinery to validate and dispatch calls. */ /* * Depending on ID the slot pointer in the radix tree points at one of these * structs. */ struct uverbs_api_ioctl_method { int(__rcu *handler)(struct uverbs_attr_bundle *attrs); DECLARE_BITMAP(attr_mandatory, UVERBS_API_ATTR_BKEY_LEN); u16 bundle_size; u8 use_stack:1; u8 driver_method:1; u8 disabled:1; u8 has_udata:1; u8 key_bitmap_len; u8 destroy_bkey; }; struct uverbs_api_write_method { int (*handler)(struct uverbs_attr_bundle *attrs); u8 disabled:1; u8 is_ex:1; u8 has_udata:1; u8 has_resp:1; u8 req_size; u8 resp_size; }; struct uverbs_api_attr { struct uverbs_attr_spec spec; }; struct uverbs_api { /* radix tree contains struct uverbs_api_* pointers */ struct radix_tree_root radix; enum rdma_driver_id driver_id; unsigned int num_write; unsigned int num_write_ex; struct uverbs_api_write_method notsupp_method; const struct uverbs_api_write_method **write_methods; const struct uverbs_api_write_method **write_ex_methods; }; /* * Get an uverbs_api_object that corresponds to the given object_id. * Note: * -ENOMSG means that any object is allowed to match during lookup. */ static inline const struct uverbs_api_object * uapi_get_object(struct uverbs_api *uapi, u16 object_id) { const struct uverbs_api_object *res; if (object_id == UVERBS_IDR_ANY_OBJECT) return ERR_PTR(-ENOMSG); res = radix_tree_lookup(&uapi->radix, uapi_key_obj(object_id)); if (!res) return ERR_PTR(-ENOENT); return res; } char *uapi_key_format(char *S, unsigned int key); struct uverbs_api *uverbs_alloc_api(struct ib_device *ibdev); void uverbs_disassociate_api_pre(struct ib_uverbs_device *uverbs_dev); void uverbs_disassociate_api(struct uverbs_api *uapi); void uverbs_destroy_api(struct uverbs_api *uapi); void uapi_compute_bundle_size(struct uverbs_api_ioctl_method *method_elm, unsigned int num_attrs); void uverbs_user_mmap_disassociate(struct ib_uverbs_file *ufile); extern const struct uapi_definition uverbs_def_obj_async_fd[]; extern const struct uapi_definition uverbs_def_obj_counters[]; extern const struct uapi_definition uverbs_def_obj_cq[]; extern const struct uapi_definition uverbs_def_obj_device[]; extern const struct uapi_definition uverbs_def_obj_dm[]; extern const struct uapi_definition uverbs_def_obj_dmah[]; extern const struct uapi_definition uverbs_def_obj_flow_action[]; extern const struct uapi_definition uverbs_def_obj_intf[]; extern const struct uapi_definition uverbs_def_obj_mr[]; extern const struct uapi_definition uverbs_def_obj_qp[]; extern const struct uapi_definition uverbs_def_obj_srq[]; extern const struct uapi_definition uverbs_def_obj_wq[]; extern const struct uapi_definition uverbs_def_write_intf[]; static inline const struct uverbs_api_write_method * uapi_get_method(const struct uverbs_api *uapi, u32 command) { u32 cmd_idx = command & IB_USER_VERBS_CMD_COMMAND_MASK; if (command & ~(u32)(IB_USER_VERBS_CMD_FLAG_EXTENDED | IB_USER_VERBS_CMD_COMMAND_MASK)) return ERR_PTR(-EINVAL); if (command & IB_USER_VERBS_CMD_FLAG_EXTENDED) { if (cmd_idx >= uapi->num_write_ex) return ERR_PTR(-EOPNOTSUPP); return uapi->write_ex_methods[cmd_idx]; } if (cmd_idx >= uapi->num_write) return ERR_PTR(-EOPNOTSUPP); return uapi->write_methods[cmd_idx]; } void uverbs_fill_udata(struct uverbs_attr_bundle *bundle, struct ib_udata *udata, unsigned int attr_in, unsigned int attr_out); #endif /* RDMA_CORE_H */ |
| 88 583 565 839 778 149 149 68 87 564 524 133 133 665 6 661 117 586 64 64 93 1 93 72 25 11 17 2 3 3 1 2 594 598 28 28 14 14 8 5 4 58 25 2 26 37 22 3 15 28 62 35 37 37 1 1 28 1 6 1 33 23 5 2 2 26 5 31 563 759 4 24 46 34 12 12 12 12 39 9 41 172 262 264 264 263 70 182 60 40 39 60 60 60 39 39 201 202 200 28 17 202 53 53 14 335 34 35 4 323 315 327 208 4 3 197 13 201 7 206 208 207 208 1 208 195 28 5 9 2 207 11 5 2 3 2 3 358 358 356 35 175 3 192 193 3 6 326 8 347 327 3 13 13 13 128 286 327 327 297 295 2 106 13 125 312 17 294 314 291 18 125 285 21 8 28 29 561 1 9 559 1 1 7 572 574 2 308 22 1 248 217 341 565 9 571 530 6 30 2 555 341 1 1 13 5 219 224 233 233 4 1 1 1 48 24 4 28 28 26 2 2 4 3 1 11 1 3 11 3 1 2 11 1 1 2 4 5 26 1 1 4 23 23 23 3 4 31 24 556 521 27 522 107 2 324 54 125 2 2 26 53 2 53 574 2 23 23 27 4 529 343 235 342 39 599 597 1 1 640 1 1 639 82 611 611 607 11 1 11 11 11 1 11 12 12 1 11 5 11 11 11 11 11 3 11 14 15 13 13 3 34 34 34 5 1 1 7 3 4 1 1 3 2 2 2 2 33 13 20 2 35 6 20 32 4 15 9 18 52 47 51 18 7 7 1 1 1 1 5 18 11 28 79 1 1 31 76 76 76 7 16 7 9 9 9 2 10 3 2 6 1 8 16 16 16 1 18 1 4 16 16 16 16 1 57 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 | // SPDX-License-Identifier: GPL-2.0-or-later /* * net/sched/sch_api.c Packet scheduler API. * * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> * * Fixes: * * Rani Assaf <rani@magic.metawire.com> :980802: JIFFIES and CPU clock sources are repaired. * Eduardo J. Blanco <ejbs@netlabs.com.uy> :990222: kmod support * Jamal Hadi Salim <hadi@nortelnetworks.com>: 990601: ingress support */ #include <linux/module.h> #include <linux/types.h> #include <linux/kernel.h> #include <linux/string.h> #include <linux/errno.h> #include <linux/skbuff.h> #include <linux/init.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/kmod.h> #include <linux/list.h> #include <linux/hrtimer.h> #include <linux/slab.h> #include <linux/hashtable.h> #include <linux/bpf.h> #include <net/netdev_lock.h> #include <net/net_namespace.h> #include <net/sock.h> #include <net/netlink.h> #include <net/pkt_sched.h> #include <net/pkt_cls.h> #include <net/tc_wrapper.h> #include <trace/events/qdisc.h> /* Short review. ------------- This file consists of two interrelated parts: 1. queueing disciplines manager frontend. 2. traffic classes manager frontend. Generally, queueing discipline ("qdisc") is a black box, which is able to enqueue packets and to dequeue them (when device is ready to send something) in order and at times determined by algorithm hidden in it. qdisc's are divided to two categories: - "queues", which have no internal structure visible from outside. - "schedulers", which split all the packets to "traffic classes", using "packet classifiers" (look at cls_api.c) In turn, classes may have child qdiscs (as rule, queues) attached to them etc. etc. etc. The goal of the routines in this file is to translate information supplied by user in the form of handles to more intelligible for kernel form, to make some sanity checks and part of work, which is common to all qdiscs and to provide rtnetlink notifications. All real intelligent work is done inside qdisc modules. Every discipline has two major routines: enqueue and dequeue. ---dequeue dequeue usually returns a skb to send. It is allowed to return NULL, but it does not mean that queue is empty, it just means that discipline does not want to send anything this time. Queue is really empty if q->q.qlen == 0. For complicated disciplines with multiple queues q->q is not real packet queue, but however q->q.qlen must be valid. ---enqueue enqueue returns 0, if packet was enqueued successfully. If packet (this one or another one) was dropped, it returns not zero error code. NET_XMIT_DROP - this packet dropped Expected action: do not backoff, but wait until queue will clear. NET_XMIT_CN - probably this packet enqueued, but another one dropped. Expected action: backoff or ignore Auxiliary routines: ---peek like dequeue but without removing a packet from the queue ---reset returns qdisc to initial state: purge all buffers, clear all timers, counters (except for statistics) etc. ---init initializes newly created qdisc. ---destroy destroys resources allocated by init and during lifetime of qdisc. ---change changes qdisc parameters. */ /* Protects list of registered TC modules. It is pure SMP lock. */ static DEFINE_RWLOCK(qdisc_mod_lock); /************************************************ * Queueing disciplines manipulation. * ************************************************/ /* The list of all installed queueing disciplines. */ static struct Qdisc_ops *qdisc_base; /* Register/unregister queueing discipline */ int register_qdisc(struct Qdisc_ops *qops) { struct Qdisc_ops *q, **qp; int rc = -EEXIST; write_lock(&qdisc_mod_lock); for (qp = &qdisc_base; (q = *qp) != NULL; qp = &q->next) if (!strcmp(qops->id, q->id)) goto out; if (qops->enqueue == NULL) qops->enqueue = noop_qdisc_ops.enqueue; if (qops->peek == NULL) { if (qops->dequeue == NULL) qops->peek = noop_qdisc_ops.peek; else goto out_einval; } if (qops->dequeue == NULL) qops->dequeue = noop_qdisc_ops.dequeue; if (qops->cl_ops) { const struct Qdisc_class_ops *cops = qops->cl_ops; if (!(cops->find && cops->walk && cops->leaf)) goto out_einval; if (cops->tcf_block && !(cops->bind_tcf && cops->unbind_tcf)) goto out_einval; } qops->next = NULL; *qp = qops; rc = 0; out: write_unlock(&qdisc_mod_lock); return rc; out_einval: rc = -EINVAL; goto out; } EXPORT_SYMBOL(register_qdisc); void unregister_qdisc(struct Qdisc_ops *qops) { struct Qdisc_ops *q, **qp; int err = -ENOENT; write_lock(&qdisc_mod_lock); for (qp = &qdisc_base; (q = *qp) != NULL; qp = &q->next) if (q == qops) break; if (q) { *qp = q->next; q->next = NULL; err = 0; } write_unlock(&qdisc_mod_lock); WARN(err, "unregister qdisc(%s) failed\n", qops->id); } EXPORT_SYMBOL(unregister_qdisc); /* Get default qdisc if not otherwise specified */ void qdisc_get_default(char *name, size_t len) { read_lock(&qdisc_mod_lock); strscpy(name, default_qdisc_ops->id, len); read_unlock(&qdisc_mod_lock); } static struct Qdisc_ops *qdisc_lookup_default(const char *name) { struct Qdisc_ops *q = NULL; for (q = qdisc_base; q; q = q->next) { if (!strcmp(name, q->id)) { if (!bpf_try_module_get(q, q->owner)) q = NULL; break; } } return q; } /* Set new default qdisc to use */ int qdisc_set_default(const char *name) { const struct Qdisc_ops *ops; if (!capable(CAP_NET_ADMIN)) return -EPERM; write_lock(&qdisc_mod_lock); ops = qdisc_lookup_default(name); if (!ops) { /* Not found, drop lock and try to load module */ write_unlock(&qdisc_mod_lock); request_module(NET_SCH_ALIAS_PREFIX "%s", name); write_lock(&qdisc_mod_lock); ops = qdisc_lookup_default(name); } if (ops) { /* Set new default */ bpf_module_put(default_qdisc_ops, default_qdisc_ops->owner); default_qdisc_ops = ops; } write_unlock(&qdisc_mod_lock); return ops ? 0 : -ENOENT; } #ifdef CONFIG_NET_SCH_DEFAULT /* Set default value from kernel config */ static int __init sch_default_qdisc(void) { return qdisc_set_default(CONFIG_DEFAULT_NET_SCH); } late_initcall(sch_default_qdisc); #endif /* We know handle. Find qdisc among all qdisc's attached to device * (root qdisc, all its children, children of children etc.) * Note: caller either uses rtnl or rcu_read_lock() */ static struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle) { struct Qdisc *q; if (!qdisc_dev(root)) return (root->handle == handle ? root : NULL); if (!(root->flags & TCQ_F_BUILTIN) && root->handle == handle) return root; hash_for_each_possible_rcu(qdisc_dev(root)->qdisc_hash, q, hash, handle, lockdep_rtnl_is_held()) { if (q->handle == handle) return q; } return NULL; } void qdisc_hash_add(struct Qdisc *q, bool invisible) { if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { ASSERT_RTNL(); hash_add_rcu(qdisc_dev(q)->qdisc_hash, &q->hash, q->handle); if (invisible) q->flags |= TCQ_F_INVISIBLE; } } EXPORT_SYMBOL(qdisc_hash_add); void qdisc_hash_del(struct Qdisc *q) { if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { ASSERT_RTNL(); hash_del_rcu(&q->hash); } } EXPORT_SYMBOL(qdisc_hash_del); struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) { struct Qdisc *q; if (!handle) return NULL; q = qdisc_match_from_root(rtnl_dereference(dev->qdisc), handle); if (q) goto out; if (dev_ingress_queue(dev)) q = qdisc_match_from_root( rtnl_dereference(dev_ingress_queue(dev)->qdisc_sleeping), handle); out: return q; } struct Qdisc *qdisc_lookup_rcu(struct net_device *dev, u32 handle) { struct netdev_queue *nq; struct Qdisc *q; if (!handle) return NULL; q = qdisc_match_from_root(rcu_dereference(dev->qdisc), handle); if (q) goto out; nq = dev_ingress_queue_rcu(dev); if (nq) q = qdisc_match_from_root(rcu_dereference(nq->qdisc_sleeping), handle); out: return q; } static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid, struct netlink_ext_ack *extack) { unsigned long cl; const struct Qdisc_class_ops *cops = p->ops->cl_ops; if (cops == NULL) { NL_SET_ERR_MSG(extack, "Parent qdisc is not classful"); return ERR_PTR(-EOPNOTSUPP); } cl = cops->find(p, classid); if (cl == 0) { NL_SET_ERR_MSG(extack, "Specified class not found"); return ERR_PTR(-ENOENT); } return cops->leaf(p, cl); } /* Find queueing discipline by name */ static struct Qdisc_ops *qdisc_lookup_ops(struct nlattr *kind) { struct Qdisc_ops *q = NULL; if (kind) { read_lock(&qdisc_mod_lock); for (q = qdisc_base; q; q = q->next) { if (nla_strcmp(kind, q->id) == 0) { if (!bpf_try_module_get(q, q->owner)) q = NULL; break; } } read_unlock(&qdisc_mod_lock); } return q; } /* The linklayer setting were not transferred from iproute2, in older * versions, and the rate tables lookup systems have been dropped in * the kernel. To keep backward compatible with older iproute2 tc * utils, we detect the linklayer setting by detecting if the rate * table were modified. * * For linklayer ATM table entries, the rate table will be aligned to * 48 bytes, thus some table entries will contain the same value. The * mpu (min packet unit) is also encoded into the old rate table, thus * starting from the mpu, we find low and high table entries for * mapping this cell. If these entries contain the same value, when * the rate tables have been modified for linklayer ATM. * * This is done by rounding mpu to the nearest 48 bytes cell/entry, * and then roundup to the next cell, calc the table entry one below, * and compare. */ static __u8 __detect_linklayer(struct tc_ratespec *r, __u32 *rtab) { int low = roundup(r->mpu, 48); int high = roundup(low+1, 48); int cell_low = low >> r->cell_log; int cell_high = (high >> r->cell_log) - 1; /* rtab is too inaccurate at rates > 100Mbit/s */ if ((r->rate > (100000000/8)) || (rtab[0] == 0)) { pr_debug("TC linklayer: Giving up ATM detection\n"); return TC_LINKLAYER_ETHERNET; } if ((cell_high > cell_low) && (cell_high < 256) && (rtab[cell_low] == rtab[cell_high])) { pr_debug("TC linklayer: Detected ATM, low(%d)=high(%d)=%u\n", cell_low, cell_high, rtab[cell_high]); return TC_LINKLAYER_ATM; } return TC_LINKLAYER_ETHERNET; } static struct qdisc_rate_table *qdisc_rtab_list; struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct nlattr *tab, struct netlink_ext_ack *extack) { struct qdisc_rate_table *rtab; if (tab == NULL || r->rate == 0 || r->cell_log == 0 || r->cell_log >= 32 || nla_len(tab) != TC_RTAB_SIZE) { NL_SET_ERR_MSG(extack, "Invalid rate table parameters for searching"); return NULL; } for (rtab = qdisc_rtab_list; rtab; rtab = rtab->next) { if (!memcmp(&rtab->rate, r, sizeof(struct tc_ratespec)) && !memcmp(&rtab->data, nla_data(tab), TC_RTAB_SIZE)) { rtab->refcnt++; return rtab; } } rtab = kmalloc(sizeof(*rtab), GFP_KERNEL); if (rtab) { rtab->rate = *r; rtab->refcnt = 1; memcpy(rtab->data, nla_data(tab), TC_RTAB_SIZE); if (r->linklayer == TC_LINKLAYER_UNAWARE) r->linklayer = __detect_linklayer(r, rtab->data); rtab->next = qdisc_rtab_list; qdisc_rtab_list = rtab; } else { NL_SET_ERR_MSG(extack, "Failed to allocate new qdisc rate table"); } return rtab; } EXPORT_SYMBOL(qdisc_get_rtab); void qdisc_put_rtab(struct qdisc_rate_table *tab) { struct qdisc_rate_table *rtab, **rtabp; if (!tab || --tab->refcnt) return; for (rtabp = &qdisc_rtab_list; (rtab = *rtabp) != NULL; rtabp = &rtab->next) { if (rtab == tab) { *rtabp = rtab->next; kfree(rtab); return; } } } EXPORT_SYMBOL(qdisc_put_rtab); static LIST_HEAD(qdisc_stab_list); static const struct nla_policy stab_policy[TCA_STAB_MAX + 1] = { [TCA_STAB_BASE] = { .len = sizeof(struct tc_sizespec) }, [TCA_STAB_DATA] = { .type = NLA_BINARY }, }; static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt, struct netlink_ext_ack *extack) { struct nlattr *tb[TCA_STAB_MAX + 1]; struct qdisc_size_table *stab; struct tc_sizespec *s; unsigned int tsize = 0; u16 *tab = NULL; int err; err = nla_parse_nested_deprecated(tb, TCA_STAB_MAX, opt, stab_policy, extack); if (err < 0) return ERR_PTR(err); if (!tb[TCA_STAB_BASE]) { NL_SET_ERR_MSG(extack, "Size table base attribute is missing"); return ERR_PTR(-EINVAL); } s = nla_data(tb[TCA_STAB_BASE]); if (s->tsize > 0) { if (!tb[TCA_STAB_DATA]) { NL_SET_ERR_MSG(extack, "Size table data attribute is missing"); return ERR_PTR(-EINVAL); } tab = nla_data(tb[TCA_STAB_DATA]); tsize = nla_len(tb[TCA_STAB_DATA]) / sizeof(u16); } if (tsize != s->tsize || (!tab && tsize > 0)) { NL_SET_ERR_MSG(extack, "Invalid size of size table"); return ERR_PTR(-EINVAL); } list_for_each_entry(stab, &qdisc_stab_list, list) { if (memcmp(&stab->szopts, s, sizeof(*s))) continue; if (tsize > 0 && memcmp(stab->data, tab, flex_array_size(stab, data, tsize))) continue; stab->refcnt++; return stab; } if (s->size_log > STAB_SIZE_LOG_MAX || s->cell_log > STAB_SIZE_LOG_MAX) { NL_SET_ERR_MSG(extack, "Invalid logarithmic size of size table"); return ERR_PTR(-EINVAL); } stab = kmalloc(struct_size(stab, data, tsize), GFP_KERNEL); if (!stab) return ERR_PTR(-ENOMEM); stab->refcnt = 1; stab->szopts = *s; if (tsize > 0) memcpy(stab->data, tab, flex_array_size(stab, data, tsize)); list_add_tail(&stab->list, &qdisc_stab_list); return stab; } void qdisc_put_stab(struct qdisc_size_table *tab) { if (!tab) return; if (--tab->refcnt == 0) { list_del(&tab->list); kfree_rcu(tab, rcu); } } EXPORT_SYMBOL(qdisc_put_stab); static int qdisc_dump_stab(struct sk_buff *skb, struct qdisc_size_table *stab) { struct nlattr *nest; nest = nla_nest_start_noflag(skb, TCA_STAB); if (nest == NULL) goto nla_put_failure; if (nla_put(skb, TCA_STAB_BASE, sizeof(stab->szopts), &stab->szopts)) goto nla_put_failure; nla_nest_end(skb, nest); return skb->len; nla_put_failure: return -1; } void __qdisc_calculate_pkt_len(struct sk_buff *skb, const struct qdisc_size_table *stab) { int pkt_len, slot; pkt_len = skb->len + stab->szopts.overhead; if (unlikely(!stab->szopts.tsize)) goto out; slot = pkt_len + stab->szopts.cell_align; if (unlikely(slot < 0)) slot = 0; slot >>= stab->szopts.cell_log; if (likely(slot < stab->szopts.tsize)) pkt_len = stab->data[slot]; else pkt_len = stab->data[stab->szopts.tsize - 1] * (slot / stab->szopts.tsize) + stab->data[slot % stab->szopts.tsize]; pkt_len <<= stab->szopts.size_log; out: if (unlikely(pkt_len < 1)) pkt_len = 1; qdisc_skb_cb(skb)->pkt_len = pkt_len; } static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer) { struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog, timer); rcu_read_lock(); __netif_schedule(qdisc_root(wd->qdisc)); rcu_read_unlock(); return HRTIMER_NORESTART; } void qdisc_watchdog_init_clockid(struct qdisc_watchdog *wd, struct Qdisc *qdisc, clockid_t clockid) { hrtimer_setup(&wd->timer, qdisc_watchdog, clockid, HRTIMER_MODE_ABS_PINNED); wd->qdisc = qdisc; } EXPORT_SYMBOL(qdisc_watchdog_init_clockid); void qdisc_watchdog_init(struct qdisc_watchdog *wd, struct Qdisc *qdisc) { qdisc_watchdog_init_clockid(wd, qdisc, CLOCK_MONOTONIC); } EXPORT_SYMBOL(qdisc_watchdog_init); void qdisc_watchdog_schedule_range_ns(struct qdisc_watchdog *wd, u64 expires, u64 delta_ns) { bool deactivated; rcu_read_lock(); deactivated = test_bit(__QDISC_STATE_DEACTIVATED, &qdisc_root_sleeping(wd->qdisc)->state); rcu_read_unlock(); if (deactivated) return; if (hrtimer_is_queued(&wd->timer)) { u64 softexpires; softexpires = ktime_to_ns(hrtimer_get_softexpires(&wd->timer)); /* If timer is already set in [expires, expires + delta_ns], * do not reprogram it. */ if (softexpires - expires <= delta_ns) return; } hrtimer_start_range_ns(&wd->timer, ns_to_ktime(expires), delta_ns, HRTIMER_MODE_ABS_PINNED); } EXPORT_SYMBOL(qdisc_watchdog_schedule_range_ns); void qdisc_watchdog_cancel(struct qdisc_watchdog *wd) { hrtimer_cancel(&wd->timer); } EXPORT_SYMBOL(qdisc_watchdog_cancel); static struct hlist_head *qdisc_class_hash_alloc(unsigned int n) { struct hlist_head *h; unsigned int i; h = kvmalloc_array(n, sizeof(struct hlist_head), GFP_KERNEL); if (h != NULL) { for (i = 0; i < n; i++) INIT_HLIST_HEAD(&h[i]); } return h; } void qdisc_class_hash_grow(struct Qdisc *sch, struct Qdisc_class_hash *clhash) { struct Qdisc_class_common *cl; struct hlist_node *next; struct hlist_head *nhash, *ohash; unsigned int nsize, nmask, osize; unsigned int i, h; /* Rehash when load factor exceeds 0.75 */ if (clhash->hashelems * 4 <= clhash->hashsize * 3) return; nsize = clhash->hashsize * 2; nmask = nsize - 1; nhash = qdisc_class_hash_alloc(nsize); if (nhash == NULL) return; ohash = clhash->hash; osize = clhash->hashsize; sch_tree_lock(sch); for (i = 0; i < osize; i++) { hlist_for_each_entry_safe(cl, next, &ohash[i], hnode) { h = qdisc_class_hash(cl->classid, nmask); hlist_add_head(&cl->hnode, &nhash[h]); } } clhash->hash = nhash; clhash->hashsize = nsize; clhash->hashmask = nmask; sch_tree_unlock(sch); kvfree(ohash); } EXPORT_SYMBOL(qdisc_class_hash_grow); int qdisc_class_hash_init(struct Qdisc_class_hash *clhash) { unsigned int size = 4; clhash->hash = qdisc_class_hash_alloc(size); if (!clhash->hash) return -ENOMEM; clhash->hashsize = size; clhash->hashmask = size - 1; clhash->hashelems = 0; return 0; } EXPORT_SYMBOL(qdisc_class_hash_init); void qdisc_class_hash_destroy(struct Qdisc_class_hash *clhash) { kvfree(clhash->hash); } EXPORT_SYMBOL(qdisc_class_hash_destroy); void qdisc_class_hash_insert(struct Qdisc_class_hash *clhash, struct Qdisc_class_common *cl) { unsigned int h; INIT_HLIST_NODE(&cl->hnode); h = qdisc_class_hash(cl->classid, clhash->hashmask); hlist_add_head(&cl->hnode, &clhash->hash[h]); clhash->hashelems++; } EXPORT_SYMBOL(qdisc_class_hash_insert); void qdisc_class_hash_remove(struct Qdisc_class_hash *clhash, struct Qdisc_class_common *cl) { hlist_del(&cl->hnode); clhash->hashelems--; } EXPORT_SYMBOL(qdisc_class_hash_remove); /* Allocate an unique handle from space managed by kernel * Possible range is [8000-FFFF]:0000 (0x8000 values) */ static u32 qdisc_alloc_handle(struct net_device *dev) { int i = 0x8000; static u32 autohandle = TC_H_MAKE(0x80000000U, 0); do { autohandle += TC_H_MAKE(0x10000U, 0); if (autohandle == TC_H_MAKE(TC_H_ROOT, 0)) autohandle = TC_H_MAKE(0x80000000U, 0); if (!qdisc_lookup(dev, autohandle)) return autohandle; cond_resched(); } while (--i > 0); return 0; } void qdisc_tree_reduce_backlog(struct Qdisc *sch, int n, int len) { const struct Qdisc_class_ops *cops; unsigned long cl; u32 parentid; bool notify; int drops; drops = max_t(int, n, 0); rcu_read_lock(); while ((parentid = sch->parent)) { if (parentid == TC_H_ROOT) break; if (sch->flags & TCQ_F_NOPARENT) break; /* Notify parent qdisc only if child qdisc becomes empty. */ notify = !sch->q.qlen; /* TODO: perform the search on a per txq basis */ sch = qdisc_lookup_rcu(qdisc_dev(sch), TC_H_MAJ(parentid)); if (sch == NULL) { WARN_ON_ONCE(parentid != TC_H_ROOT); break; } cops = sch->ops->cl_ops; if (notify && cops->qlen_notify) { /* Note that qlen_notify must be idempotent as it may get called * multiple times. */ cl = cops->find(sch, parentid); cops->qlen_notify(sch, cl); } sch->q.qlen -= n; sch->qstats.backlog -= len; __qdisc_qstats_drop(sch, drops); } rcu_read_unlock(); } EXPORT_SYMBOL(qdisc_tree_reduce_backlog); int qdisc_offload_dump_helper(struct Qdisc *sch, enum tc_setup_type type, void *type_data) { struct net_device *dev = qdisc_dev(sch); int err; sch->flags &= ~TCQ_F_OFFLOADED; if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) return 0; err = dev->netdev_ops->ndo_setup_tc(dev, type, type_data); if (err == -EOPNOTSUPP) return 0; if (!err) sch->flags |= TCQ_F_OFFLOADED; return err; } EXPORT_SYMBOL(qdisc_offload_dump_helper); void qdisc_offload_graft_helper(struct net_device *dev, struct Qdisc *sch, struct Qdisc *new, struct Qdisc *old, enum tc_setup_type type, void *type_data, struct netlink_ext_ack *extack) { bool any_qdisc_is_offloaded; int err; if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) return; err = dev->netdev_ops->ndo_setup_tc(dev, type, type_data); /* Don't report error if the graft is part of destroy operation. */ if (!err || !new || new == &noop_qdisc) return; /* Don't report error if the parent, the old child and the new * one are not offloaded. */ any_qdisc_is_offloaded = new->flags & TCQ_F_OFFLOADED; any_qdisc_is_offloaded |= sch && sch->flags & TCQ_F_OFFLOADED; any_qdisc_is_offloaded |= old && old->flags & TCQ_F_OFFLOADED; if (any_qdisc_is_offloaded) NL_SET_ERR_MSG(extack, "Offloading graft operation failed."); } EXPORT_SYMBOL(qdisc_offload_graft_helper); void qdisc_offload_query_caps(struct net_device *dev, enum tc_setup_type type, void *caps, size_t caps_len) { const struct net_device_ops *ops = dev->netdev_ops; struct tc_query_caps_base base = { .type = type, .caps = caps, }; memset(caps, 0, caps_len); if (ops->ndo_setup_tc) ops->ndo_setup_tc(dev, TC_QUERY_CAPS, &base); } EXPORT_SYMBOL(qdisc_offload_query_caps); static void qdisc_offload_graft_root(struct net_device *dev, struct Qdisc *new, struct Qdisc *old, struct netlink_ext_ack *extack) { struct tc_root_qopt_offload graft_offload = { .command = TC_ROOT_GRAFT, .handle = new ? new->handle : 0, .ingress = (new && new->flags & TCQ_F_INGRESS) || (old && old->flags & TCQ_F_INGRESS), }; qdisc_offload_graft_helper(dev, NULL, new, old, TC_SETUP_ROOT_QDISC, &graft_offload, extack); } static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid, u32 portid, u32 seq, u16 flags, int event, struct netlink_ext_ack *extack) { struct gnet_stats_basic_sync __percpu *cpu_bstats = NULL; struct gnet_stats_queue __percpu *cpu_qstats = NULL; struct tcmsg *tcm; struct nlmsghdr *nlh; unsigned char *b = skb_tail_pointer(skb); struct gnet_dump d; struct qdisc_size_table *stab; u32 block_index; __u32 qlen; cond_resched(); nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); if (!nlh) goto out_nlmsg_trim; tcm = nlmsg_data(nlh); tcm->tcm_family = AF_UNSPEC; tcm->tcm__pad1 = 0; tcm->tcm__pad2 = 0; tcm->tcm_ifindex = qdisc_dev(q)->ifindex; tcm->tcm_parent = clid; tcm->tcm_handle = q->handle; tcm->tcm_info = refcount_read(&q->refcnt); if (nla_put_string(skb, TCA_KIND, q->ops->id)) goto nla_put_failure; if (q->ops->ingress_block_get) { block_index = q->ops->ingress_block_get(q); if (block_index && nla_put_u32(skb, TCA_INGRESS_BLOCK, block_index)) goto nla_put_failure; } if (q->ops->egress_block_get) { block_index = q->ops->egress_block_get(q); if (block_index && nla_put_u32(skb, TCA_EGRESS_BLOCK, block_index)) goto nla_put_failure; } if (q->ops->dump && q->ops->dump(q, skb) < 0) goto nla_put_failure; if (nla_put_u8(skb, TCA_HW_OFFLOAD, !!(q->flags & TCQ_F_OFFLOADED))) goto nla_put_failure; qlen = qdisc_qlen_sum(q); stab = rtnl_dereference(q->stab); if (stab && qdisc_dump_stab(skb, stab) < 0) goto nla_put_failure; if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, NULL, &d, TCA_PAD) < 0) goto nla_put_failure; if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0) goto nla_put_failure; if (qdisc_is_percpu_stats(q)) { cpu_bstats = q->cpu_bstats; cpu_qstats = q->cpu_qstats; } if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats, true) < 0 || gnet_stats_copy_rate_est(&d, &q->rate_est) < 0 || gnet_stats_copy_queue(&d, cpu_qstats, &q->qstats, qlen) < 0) goto nla_put_failure; if (gnet_stats_finish_copy(&d) < 0) goto nla_put_failure; if (extack && extack->_msg && nla_put_string(skb, TCA_EXT_WARN_MSG, extack->_msg)) goto out_nlmsg_trim; nlh->nlmsg_len = skb_tail_pointer(skb) - b; return skb->len; out_nlmsg_trim: nla_put_failure: nlmsg_trim(skb, b); return -1; } static bool tc_qdisc_dump_ignore(struct Qdisc *q, bool dump_invisible) { if (q->flags & TCQ_F_BUILTIN) return true; if ((q->flags & TCQ_F_INVISIBLE) && !dump_invisible) return true; return false; } static int qdisc_get_notify(struct net *net, struct sk_buff *oskb, struct nlmsghdr *n, u32 clid, struct Qdisc *q, struct netlink_ext_ack *extack) { struct sk_buff *skb; u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb) return -ENOBUFS; if (!tc_qdisc_dump_ignore(q, false)) { if (tc_fill_qdisc(skb, q, clid, portid, n->nlmsg_seq, 0, RTM_NEWQDISC, extack) < 0) goto err_out; } if (skb->len) return rtnetlink_send(skb, net, portid, RTNLGRP_TC, n->nlmsg_flags & NLM_F_ECHO); err_out: kfree_skb(skb); return -EINVAL; } static int qdisc_notify(struct net *net, struct sk_buff *oskb, struct nlmsghdr *n, u32 clid, struct Qdisc *old, struct Qdisc *new, struct netlink_ext_ack *extack) { struct sk_buff *skb; u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; if (!rtnl_notify_needed(net, n->nlmsg_flags, RTNLGRP_TC)) return 0; skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb) return -ENOBUFS; if (old && !tc_qdisc_dump_ignore(old, false)) { if (tc_fill_qdisc(skb, old, clid, portid, n->nlmsg_seq, 0, RTM_DELQDISC, extack) < 0) goto err_out; } if (new && !tc_qdisc_dump_ignore(new, false)) { if (tc_fill_qdisc(skb, new, clid, portid, n->nlmsg_seq, old ? NLM_F_REPLACE : 0, RTM_NEWQDISC, extack) < 0) goto err_out; } if (skb->len) return rtnetlink_send(skb, net, portid, RTNLGRP_TC, n->nlmsg_flags & NLM_F_ECHO); err_out: kfree_skb(skb); return -EINVAL; } static void notify_and_destroy(struct net *net, struct sk_buff *skb, struct nlmsghdr *n, u32 clid, struct Qdisc *old, struct Qdisc *new, struct netlink_ext_ack *extack) { if (new || old) qdisc_notify(net, skb, n, clid, old, new, extack); if (old) qdisc_put(old); } static void qdisc_clear_nolock(struct Qdisc *sch) { sch->flags &= ~TCQ_F_NOLOCK; if (!(sch->flags & TCQ_F_CPUSTATS)) return; free_percpu(sch->cpu_bstats); free_percpu(sch->cpu_qstats); sch->cpu_bstats = NULL; sch->cpu_qstats = NULL; sch->flags &= ~TCQ_F_CPUSTATS; } /* Graft qdisc "new" to class "classid" of qdisc "parent" or * to device "dev". * * When appropriate send a netlink notification using 'skb' * and "n". * * On success, destroy old qdisc. */ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, struct sk_buff *skb, struct nlmsghdr *n, u32 classid, struct Qdisc *new, struct Qdisc *old, struct netlink_ext_ack *extack) { struct Qdisc *q = old; struct net *net = dev_net(dev); if (parent == NULL) { unsigned int i, num_q, ingress; struct netdev_queue *dev_queue; ingress = 0; num_q = dev->num_tx_queues; if ((q && q->flags & TCQ_F_INGRESS) || (new && new->flags & TCQ_F_INGRESS)) { ingress = 1; dev_queue = dev_ingress_queue(dev); if (!dev_queue) { NL_SET_ERR_MSG(extack, "Device does not have an ingress queue"); return -ENOENT; } q = rtnl_dereference(dev_queue->qdisc_sleeping); /* This is the counterpart of that qdisc_refcount_inc_nz() call in * __tcf_qdisc_find() for filter requests. */ if (!qdisc_refcount_dec_if_one(q)) { NL_SET_ERR_MSG(extack, "Current ingress or clsact Qdisc has ongoing filter requests"); return -EBUSY; } } if (dev->flags & IFF_UP) dev_deactivate(dev); qdisc_offload_graft_root(dev, new, old, extack); if (new && new->ops->attach && !ingress) goto skip; if (!ingress) { for (i = 0; i < num_q; i++) { dev_queue = netdev_get_tx_queue(dev, i); old = dev_graft_qdisc(dev_queue, new); if (new && i > 0) qdisc_refcount_inc(new); qdisc_put(old); } } else { old = dev_graft_qdisc(dev_queue, NULL); /* {ingress,clsact}_destroy() @old before grafting @new to avoid * unprotected concurrent accesses to net_device::miniq_{in,e}gress * pointer(s) in mini_qdisc_pair_swap(). */ qdisc_notify(net, skb, n, classid, old, new, extack); qdisc_destroy(old); dev_graft_qdisc(dev_queue, new); } skip: if (!ingress) { old = rtnl_dereference(dev->qdisc); if (new && !new->ops->attach) qdisc_refcount_inc(new); rcu_assign_pointer(dev->qdisc, new ? : &noop_qdisc); notify_and_destroy(net, skb, n, classid, old, new, extack); if (new && new->ops->attach) new->ops->attach(new); } if (dev->flags & IFF_UP) dev_activate(dev); } else { const struct Qdisc_class_ops *cops = parent->ops->cl_ops; unsigned long cl; int err; /* Only support running class lockless if parent is lockless */ if (new && (new->flags & TCQ_F_NOLOCK) && !(parent->flags & TCQ_F_NOLOCK)) qdisc_clear_nolock(new); if (!cops || !cops->graft) return -EOPNOTSUPP; cl = cops->find(parent, classid); if (!cl) { NL_SET_ERR_MSG(extack, "Specified class not found"); return -ENOENT; } if (new && new->ops == &noqueue_qdisc_ops) { NL_SET_ERR_MSG(extack, "Cannot assign noqueue to a class"); return -EINVAL; } if (new && !(parent->flags & TCQ_F_MQROOT) && rcu_access_pointer(new->stab)) { NL_SET_ERR_MSG(extack, "STAB not supported on a non root"); return -EINVAL; } err = cops->graft(parent, cl, new, &old, extack); if (err) return err; notify_and_destroy(net, skb, n, classid, old, new, extack); } return 0; } static int qdisc_block_indexes_set(struct Qdisc *sch, struct nlattr **tca, struct netlink_ext_ack *extack) { u32 block_index; if (tca[TCA_INGRESS_BLOCK]) { block_index = nla_get_u32(tca[TCA_INGRESS_BLOCK]); if (!block_index) { NL_SET_ERR_MSG(extack, "Ingress block index cannot be 0"); return -EINVAL; } if (!sch->ops->ingress_block_set) { NL_SET_ERR_MSG(extack, "Ingress block sharing is not supported"); return -EOPNOTSUPP; } sch->ops->ingress_block_set(sch, block_index); } if (tca[TCA_EGRESS_BLOCK]) { block_index = nla_get_u32(tca[TCA_EGRESS_BLOCK]); if (!block_index) { NL_SET_ERR_MSG(extack, "Egress block index cannot be 0"); return -EINVAL; } if (!sch->ops->egress_block_set) { NL_SET_ERR_MSG(extack, "Egress block sharing is not supported"); return -EOPNOTSUPP; } sch->ops->egress_block_set(sch, block_index); } return 0; } /* Allocate and initialize new qdisc. Parameters are passed via opt. */ static struct Qdisc *qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, u32 parent, u32 handle, struct nlattr **tca, int *errp, struct netlink_ext_ack *extack) { int err; struct nlattr *kind = tca[TCA_KIND]; struct Qdisc *sch; struct Qdisc_ops *ops; struct qdisc_size_table *stab; ops = qdisc_lookup_ops(kind); if (!ops) { err = -ENOENT; NL_SET_ERR_MSG(extack, "Specified qdisc kind is unknown"); goto err_out; } sch = qdisc_alloc(dev_queue, ops, extack); if (IS_ERR(sch)) { err = PTR_ERR(sch); goto err_out2; } sch->parent = parent; if (handle == TC_H_INGRESS) { if (!(sch->flags & TCQ_F_INGRESS)) { NL_SET_ERR_MSG(extack, "Specified parent ID is reserved for ingress and clsact Qdiscs"); err = -EINVAL; goto err_out3; } handle = TC_H_MAKE(TC_H_INGRESS, 0); } else { if (handle == 0) { handle = qdisc_alloc_handle(dev); if (handle == 0) { NL_SET_ERR_MSG(extack, "Maximum number of qdisc handles was exceeded"); err = -ENOSPC; goto err_out3; } } if (!netif_is_multiqueue(dev)) sch->flags |= TCQ_F_ONETXQUEUE; } sch->handle = handle; /* This exist to keep backward compatible with a userspace * loophole, what allowed userspace to get IFF_NO_QUEUE * facility on older kernels by setting tx_queue_len=0 (prior * to qdisc init), and then forgot to reinit tx_queue_len * before again attaching a qdisc. */ if ((dev->priv_flags & IFF_NO_QUEUE) && (dev->tx_queue_len == 0)) { WRITE_ONCE(dev->tx_queue_len, DEFAULT_TX_QUEUE_LEN); netdev_info(dev, "Caught tx_queue_len zero misconfig\n"); } err = qdisc_block_indexes_set(sch, tca, extack); if (err) goto err_out3; if (tca[TCA_STAB]) { stab = qdisc_get_stab(tca[TCA_STAB], extack); if (IS_ERR(stab)) { err = PTR_ERR(stab); goto err_out3; } rcu_assign_pointer(sch->stab, stab); } if (ops->init) { err = ops->init(sch, tca[TCA_OPTIONS], extack); if (err != 0) goto err_out4; } if (tca[TCA_RATE]) { err = -EOPNOTSUPP; if (sch->flags & TCQ_F_MQROOT) { NL_SET_ERR_MSG(extack, "Cannot attach rate estimator to a multi-queue root qdisc"); goto err_out4; } err = gen_new_estimator(&sch->bstats, sch->cpu_bstats, &sch->rate_est, NULL, true, tca[TCA_RATE]); if (err) { NL_SET_ERR_MSG(extack, "Failed to generate new estimator"); goto err_out4; } } qdisc_hash_add(sch, false); trace_qdisc_create(ops, dev, parent); return sch; err_out4: /* Even if ops->init() failed, we call ops->destroy() * like qdisc_create_dflt(). */ if (ops->destroy) ops->destroy(sch); qdisc_put_stab(rtnl_dereference(sch->stab)); err_out3: lockdep_unregister_key(&sch->root_lock_key); netdev_put(dev, &sch->dev_tracker); qdisc_free(sch); err_out2: bpf_module_put(ops, ops->owner); err_out: *errp = err; return NULL; } static int qdisc_change(struct Qdisc *sch, struct nlattr **tca, struct netlink_ext_ack *extack) { struct qdisc_size_table *ostab, *stab = NULL; int err = 0; if (tca[TCA_OPTIONS]) { if (!sch->ops->change) { NL_SET_ERR_MSG(extack, "Change operation not supported by specified qdisc"); return -EINVAL; } if (tca[TCA_INGRESS_BLOCK] || tca[TCA_EGRESS_BLOCK]) { NL_SET_ERR_MSG(extack, "Change of blocks is not supported"); return -EOPNOTSUPP; } err = sch->ops->change(sch, tca[TCA_OPTIONS], extack); if (err) return err; } if (tca[TCA_STAB]) { stab = qdisc_get_stab(tca[TCA_STAB], extack); if (IS_ERR(stab)) return PTR_ERR(stab); } ostab = rtnl_dereference(sch->stab); rcu_assign_pointer(sch->stab, stab); qdisc_put_stab(ostab); if (tca[TCA_RATE]) { /* NB: ignores errors from replace_estimator because change can't be undone. */ if (sch->flags & TCQ_F_MQROOT) goto out; gen_replace_estimator(&sch->bstats, sch->cpu_bstats, &sch->rate_est, NULL, true, tca[TCA_RATE]); } out: return 0; } struct check_loop_arg { struct qdisc_walker w; struct Qdisc *p; int depth; }; static int check_loop_fn(struct Qdisc *q, unsigned long cl, struct qdisc_walker *w); static int check_loop(struct Qdisc *q, struct Qdisc *p, int depth) { struct check_loop_arg arg; if (q->ops->cl_ops == NULL) return 0; arg.w.stop = arg.w.skip = arg.w.count = 0; arg.w.fn = check_loop_fn; arg.depth = depth; arg.p = p; q->ops->cl_ops->walk(q, &arg.w); return arg.w.stop ? -ELOOP : 0; } static int check_loop_fn(struct Qdisc *q, unsigned long cl, struct qdisc_walker *w) { struct Qdisc *leaf; const struct Qdisc_class_ops *cops = q->ops->cl_ops; struct check_loop_arg *arg = (struct check_loop_arg *)w; leaf = cops->leaf(q, cl); if (leaf) { if (leaf == arg->p || arg->depth > 7) return -ELOOP; return check_loop(leaf, arg->p, arg->depth + 1); } return 0; } const struct nla_policy rtm_tca_policy[TCA_MAX + 1] = { [TCA_KIND] = { .type = NLA_STRING }, [TCA_RATE] = { .type = NLA_BINARY, .len = sizeof(struct tc_estimator) }, [TCA_STAB] = { .type = NLA_NESTED }, [TCA_DUMP_INVISIBLE] = { .type = NLA_FLAG }, [TCA_CHAIN] = { .type = NLA_U32 }, [TCA_INGRESS_BLOCK] = { .type = NLA_U32 }, [TCA_EGRESS_BLOCK] = { .type = NLA_U32 }, }; /* * Delete/get qdisc. */ static int __tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack, struct net_device *dev, struct nlattr *tca[TCA_MAX + 1], struct tcmsg *tcm) { struct net *net = sock_net(skb->sk); struct Qdisc *q = NULL; struct Qdisc *p = NULL; u32 clid; int err; clid = tcm->tcm_parent; if (clid) { if (clid != TC_H_ROOT) { if (TC_H_MAJ(clid) != TC_H_MAJ(TC_H_INGRESS)) { p = qdisc_lookup(dev, TC_H_MAJ(clid)); if (!p) { NL_SET_ERR_MSG(extack, "Failed to find qdisc with specified classid"); return -ENOENT; } q = qdisc_leaf(p, clid, extack); } else if (dev_ingress_queue(dev)) { q = rtnl_dereference(dev_ingress_queue(dev)->qdisc_sleeping); } } else { q = rtnl_dereference(dev->qdisc); } if (!q) { NL_SET_ERR_MSG(extack, "Cannot find specified qdisc on specified device"); return -ENOENT; } if (IS_ERR(q)) return PTR_ERR(q); if (tcm->tcm_handle && q->handle != tcm->tcm_handle) { NL_SET_ERR_MSG(extack, "Invalid handle"); return -EINVAL; } } else { q = qdisc_lookup(dev, tcm->tcm_handle); if (!q) { NL_SET_ERR_MSG(extack, "Failed to find qdisc with specified handle"); return -ENOENT; } } if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id)) { NL_SET_ERR_MSG(extack, "Invalid qdisc name: must match existing qdisc"); return -EINVAL; } if (n->nlmsg_type == RTM_DELQDISC) { if (!clid) { NL_SET_ERR_MSG(extack, "Classid cannot be zero"); return -EINVAL; } if (q->handle == 0) { NL_SET_ERR_MSG(extack, "Cannot delete qdisc with handle of zero"); return -ENOENT; } err = qdisc_graft(dev, p, skb, n, clid, NULL, q, extack); if (err != 0) return err; } else { qdisc_get_notify(net, skb, n, clid, q, NULL); } return 0; } static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack) { struct net *net = sock_net(skb->sk); struct tcmsg *tcm = nlmsg_data(n); struct nlattr *tca[TCA_MAX + 1]; struct net_device *dev; int err; err = nlmsg_parse_deprecated(n, sizeof(*tcm), tca, TCA_MAX, rtm_tca_policy, extack); if (err < 0) return err; dev = __dev_get_by_index(net, tcm->tcm_ifindex); if (!dev) return -ENODEV; netdev_lock_ops(dev); err = __tc_get_qdisc(skb, n, extack, dev, tca, tcm); netdev_unlock_ops(dev); return err; } static bool req_create_or_replace(struct nlmsghdr *n) { return (n->nlmsg_flags & NLM_F_CREATE && n->nlmsg_flags & NLM_F_REPLACE); } static bool req_create_exclusive(struct nlmsghdr *n) { return (n->nlmsg_flags & NLM_F_CREATE && n->nlmsg_flags & NLM_F_EXCL); } static bool req_change(struct nlmsghdr *n) { return (!(n->nlmsg_flags & NLM_F_CREATE) && !(n->nlmsg_flags & NLM_F_REPLACE) && !(n->nlmsg_flags & NLM_F_EXCL)); } static int __tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack, struct net_device *dev, struct nlattr *tca[TCA_MAX + 1], struct tcmsg *tcm) { struct Qdisc *q = NULL; struct Qdisc *p = NULL; u32 clid; int err; clid = tcm->tcm_parent; if (clid) { if (clid != TC_H_ROOT) { if (clid != TC_H_INGRESS) { p = qdisc_lookup(dev, TC_H_MAJ(clid)); if (!p) { NL_SET_ERR_MSG(extack, "Failed to find specified qdisc"); return -ENOENT; } if (p->flags & TCQ_F_INGRESS) { NL_SET_ERR_MSG(extack, "Cannot add children to ingress/clsact qdisc"); return -EOPNOTSUPP; } q = qdisc_leaf(p, clid, extack); if (IS_ERR(q)) return PTR_ERR(q); } else if (dev_ingress_queue_create(dev)) { q = rtnl_dereference(dev_ingress_queue(dev)->qdisc_sleeping); } } else { q = rtnl_dereference(dev->qdisc); } /* It may be default qdisc, ignore it */ if (q && q->handle == 0) q = NULL; if (!q || !tcm->tcm_handle || q->handle != tcm->tcm_handle) { if (tcm->tcm_handle) { if (q && !(n->nlmsg_flags & NLM_F_REPLACE)) { NL_SET_ERR_MSG(extack, "NLM_F_REPLACE needed to override"); return -EEXIST; } if (TC_H_MIN(tcm->tcm_handle)) { NL_SET_ERR_MSG(extack, "Invalid minor handle"); return -EINVAL; } q = qdisc_lookup(dev, tcm->tcm_handle); if (!q) goto create_n_graft; if (q->parent != tcm->tcm_parent) { NL_SET_ERR_MSG(extack, "Cannot move an existing qdisc to a different parent"); return -EINVAL; } if (n->nlmsg_flags & NLM_F_EXCL) { NL_SET_ERR_MSG(extack, "Exclusivity flag on, cannot override"); return -EEXIST; } if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id)) { NL_SET_ERR_MSG(extack, "Invalid qdisc name: must match existing qdisc"); return -EINVAL; } if (q->flags & TCQ_F_INGRESS) { NL_SET_ERR_MSG(extack, "Cannot regraft ingress or clsact Qdiscs"); return -EINVAL; } if (q == p || (p && check_loop(q, p, 0))) { NL_SET_ERR_MSG(extack, "Qdisc parent/child loop detected"); return -ELOOP; } if (clid == TC_H_INGRESS) { NL_SET_ERR_MSG(extack, "Ingress cannot graft directly"); return -EINVAL; } qdisc_refcount_inc(q); goto graft; } else { if (!q) goto create_n_graft; /* This magic test requires explanation. * * We know, that some child q is already * attached to this parent and have choice: * 1) change it or 2) create/graft new one. * If the requested qdisc kind is different * than the existing one, then we choose graft. * If they are the same then this is "change" * operation - just let it fallthrough.. * * 1. We are allowed to create/graft only * if the request is explicitly stating * "please create if it doesn't exist". * * 2. If the request is to exclusive create * then the qdisc tcm_handle is not expected * to exist, so that we choose create/graft too. * * 3. The last case is when no flags are set. * This will happen when for example tc * utility issues a "change" command. * Alas, it is sort of hole in API, we * cannot decide what to do unambiguously. * For now we select create/graft. */ if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id)) { if (req_create_or_replace(n) || req_create_exclusive(n)) goto create_n_graft; else if (req_change(n)) goto create_n_graft2; } } } } else { if (!tcm->tcm_handle) { NL_SET_ERR_MSG(extack, "Handle cannot be zero"); return -EINVAL; } q = qdisc_lookup(dev, tcm->tcm_handle); } /* Change qdisc parameters */ if (!q) { NL_SET_ERR_MSG(extack, "Specified qdisc not found"); return -ENOENT; } if (n->nlmsg_flags & NLM_F_EXCL) { NL_SET_ERR_MSG(extack, "Exclusivity flag on, cannot modify"); return -EEXIST; } if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id)) { NL_SET_ERR_MSG(extack, "Invalid qdisc name: must match existing qdisc"); return -EINVAL; } err = qdisc_change(q, tca, extack); if (err == 0) qdisc_notify(sock_net(skb->sk), skb, n, clid, NULL, q, extack); return err; create_n_graft: if (!(n->nlmsg_flags & NLM_F_CREATE)) { NL_SET_ERR_MSG(extack, "Qdisc not found. To create specify NLM_F_CREATE flag"); return -ENOENT; } create_n_graft2: if (clid == TC_H_INGRESS) { if (dev_ingress_queue(dev)) { q = qdisc_create(dev, dev_ingress_queue(dev), tcm->tcm_parent, tcm->tcm_parent, tca, &err, extack); } else { NL_SET_ERR_MSG(extack, "Cannot find ingress queue for specified device"); err = -ENOENT; } } else { struct netdev_queue *dev_queue; if (p && p->ops->cl_ops && p->ops->cl_ops->select_queue) dev_queue = p->ops->cl_ops->select_queue(p, tcm); else if (p) dev_queue = p->dev_queue; else dev_queue = netdev_get_tx_queue(dev, 0); q = qdisc_create(dev, dev_queue, tcm->tcm_parent, tcm->tcm_handle, tca, &err, extack); } if (!q) return err; graft: err = qdisc_graft(dev, p, skb, n, clid, q, NULL, extack); if (err) { if (q) qdisc_put(q); return err; } return 0; } static void request_qdisc_module(struct nlattr *kind) { struct Qdisc_ops *ops; char name[IFNAMSIZ]; if (!kind) return; ops = qdisc_lookup_ops(kind); if (ops) { bpf_module_put(ops, ops->owner); return; } if (nla_strscpy(name, kind, IFNAMSIZ) >= 0) { rtnl_unlock(); request_module(NET_SCH_ALIAS_PREFIX "%s", name); rtnl_lock(); } } /* * Create/change qdisc. */ static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack) { struct net *net = sock_net(skb->sk); struct nlattr *tca[TCA_MAX + 1]; struct net_device *dev; struct tcmsg *tcm; int err; err = nlmsg_parse_deprecated(n, sizeof(*tcm), tca, TCA_MAX, rtm_tca_policy, extack); if (err < 0) return err; request_qdisc_module(tca[TCA_KIND]); tcm = nlmsg_data(n); dev = __dev_get_by_index(net, tcm->tcm_ifindex); if (!dev) return -ENODEV; netdev_lock_ops(dev); err = __tc_modify_qdisc(skb, n, extack, dev, tca, tcm); netdev_unlock_ops(dev); return err; } static int tc_dump_qdisc_root(struct Qdisc *root, struct sk_buff *skb, struct netlink_callback *cb, int *q_idx_p, int s_q_idx, bool recur, bool dump_invisible) { int ret = 0, q_idx = *q_idx_p; struct Qdisc *q; int b; if (!root) return 0; q = root; if (q_idx < s_q_idx) { q_idx++; } else { if (!tc_qdisc_dump_ignore(q, dump_invisible) && tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC, NULL) <= 0) goto done; q_idx++; } /* If dumping singletons, there is no qdisc_dev(root) and the singleton * itself has already been dumped. * * If we've already dumped the top-level (ingress) qdisc above and the global * qdisc hashtable, we don't want to hit it again */ if (!qdisc_dev(root) || !recur) goto out; hash_for_each(qdisc_dev(root)->qdisc_hash, b, q, hash) { if (q_idx < s_q_idx) { q_idx++; continue; } if (!tc_qdisc_dump_ignore(q, dump_invisible) && tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC, NULL) <= 0) goto done; q_idx++; } out: *q_idx_p = q_idx; return ret; done: ret = -1; goto out; } static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb) { struct net *net = sock_net(skb->sk); int idx, q_idx; int s_idx, s_q_idx; struct net_device *dev; const struct nlmsghdr *nlh = cb->nlh; struct nlattr *tca[TCA_MAX + 1]; int err; s_idx = cb->args[0]; s_q_idx = q_idx = cb->args[1]; idx = 0; ASSERT_RTNL(); err = nlmsg_parse_deprecated(nlh, sizeof(struct tcmsg), tca, TCA_MAX, rtm_tca_policy, cb->extack); if (err < 0) return err; for_each_netdev(net, dev) { struct netdev_queue *dev_queue; if (idx < s_idx) goto cont; if (idx > s_idx) s_q_idx = 0; q_idx = 0; netdev_lock_ops(dev); if (tc_dump_qdisc_root(rtnl_dereference(dev->qdisc), skb, cb, &q_idx, s_q_idx, true, tca[TCA_DUMP_INVISIBLE]) < 0) { netdev_unlock_ops(dev); goto done; } dev_queue = dev_ingress_queue(dev); if (dev_queue && tc_dump_qdisc_root(rtnl_dereference(dev_queue->qdisc_sleeping), skb, cb, &q_idx, s_q_idx, false, tca[TCA_DUMP_INVISIBLE]) < 0) { netdev_unlock_ops(dev); goto done; } netdev_unlock_ops(dev); cont: idx++; } done: cb->args[0] = idx; cb->args[1] = q_idx; return skb->len; } /************************************************ * Traffic classes manipulation. * ************************************************/ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q, unsigned long cl, u32 portid, u32 seq, u16 flags, int event, struct netlink_ext_ack *extack) { struct tcmsg *tcm; struct nlmsghdr *nlh; unsigned char *b = skb_tail_pointer(skb); struct gnet_dump d; const struct Qdisc_class_ops *cl_ops = q->ops->cl_ops; cond_resched(); nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); if (!nlh) goto out_nlmsg_trim; tcm = nlmsg_data(nlh); tcm->tcm_family = AF_UNSPEC; tcm->tcm__pad1 = 0; tcm->tcm__pad2 = 0; tcm->tcm_ifindex = qdisc_dev(q)->ifindex; tcm->tcm_parent = q->handle; tcm->tcm_handle = q->handle; tcm->tcm_info = 0; if (nla_put_string(skb, TCA_KIND, q->ops->id)) goto nla_put_failure; if (cl_ops->dump && cl_ops->dump(q, cl, skb, tcm) < 0) goto nla_put_failure; if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, NULL, &d, TCA_PAD) < 0) goto nla_put_failure; if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0) goto nla_put_failure; if (gnet_stats_finish_copy(&d) < 0) goto nla_put_failure; if (extack && extack->_msg && nla_put_string(skb, TCA_EXT_WARN_MSG, extack->_msg)) goto out_nlmsg_trim; nlh->nlmsg_len = skb_tail_pointer(skb) - b; return skb->len; out_nlmsg_trim: nla_put_failure: nlmsg_trim(skb, b); return -1; } static int tclass_notify(struct net *net, struct sk_buff *oskb, struct nlmsghdr *n, struct Qdisc *q, unsigned long cl, int event, struct netlink_ext_ack *extack) { struct sk_buff *skb; u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; if (!rtnl_notify_needed(net, n->nlmsg_flags, RTNLGRP_TC)) return 0; skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb) return -ENOBUFS; if (tc_fill_tclass(skb, q, cl, portid, n->nlmsg_seq, 0, event, extack) < 0) { kfree_skb(skb); return -EINVAL; } return rtnetlink_send(skb, net, portid, RTNLGRP_TC, n->nlmsg_flags & NLM_F_ECHO); } static int tclass_get_notify(struct net *net, struct sk_buff *oskb, struct nlmsghdr *n, struct Qdisc *q, unsigned long cl, struct netlink_ext_ack *extack) { struct sk_buff *skb; u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb) return -ENOBUFS; if (tc_fill_tclass(skb, q, cl, portid, n->nlmsg_seq, 0, RTM_NEWTCLASS, extack) < 0) { kfree_skb(skb); return -EINVAL; } return rtnetlink_send(skb, net, portid, RTNLGRP_TC, n->nlmsg_flags & NLM_F_ECHO); } static int tclass_del_notify(struct net *net, const struct Qdisc_class_ops *cops, struct sk_buff *oskb, struct nlmsghdr *n, struct Qdisc *q, unsigned long cl, struct netlink_ext_ack *extack) { u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; struct sk_buff *skb; int err = 0; if (!cops->delete) return -EOPNOTSUPP; if (rtnl_notify_needed(net, n->nlmsg_flags, RTNLGRP_TC)) { skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb) return -ENOBUFS; if (tc_fill_tclass(skb, q, cl, portid, n->nlmsg_seq, 0, RTM_DELTCLASS, extack) < 0) { kfree_skb(skb); return -EINVAL; } } else { skb = NULL; } err = cops->delete(q, cl, extack); if (err) { kfree_skb(skb); return err; } err = rtnetlink_maybe_send(skb, net, portid, RTNLGRP_TC, n->nlmsg_flags & NLM_F_ECHO); return err; } #ifdef CONFIG_NET_CLS struct tcf_bind_args { struct tcf_walker w; unsigned long base; unsigned long cl; u32 classid; }; static int tcf_node_bind(struct tcf_proto *tp, void *n, struct tcf_walker *arg) { struct tcf_bind_args *a = (void *)arg; if (n && tp->ops->bind_class) { struct Qdisc *q = tcf_block_q(tp->chain->block); sch_tree_lock(q); tp->ops->bind_class(n, a->classid, a->cl, q, a->base); sch_tree_unlock(q); } return 0; } struct tc_bind_class_args { struct qdisc_walker w; unsigned long new_cl; u32 portid; u32 clid; }; static int tc_bind_class_walker(struct Qdisc *q, unsigned long cl, struct qdisc_walker *w) { struct tc_bind_class_args *a = (struct tc_bind_class_args *)w; const struct Qdisc_class_ops *cops = q->ops->cl_ops; struct tcf_block *block; struct tcf_chain *chain; block = cops->tcf_block(q, cl, NULL); if (!block) return 0; for (chain = tcf_get_next_chain(block, NULL); chain; chain = tcf_get_next_chain(block, chain)) { struct tcf_proto *tp; for (tp = tcf_get_next_proto(chain, NULL); tp; tp = tcf_get_next_proto(chain, tp)) { struct tcf_bind_args arg = {}; arg.w.fn = tcf_node_bind; arg.classid = a->clid; arg.base = cl; arg.cl = a->new_cl; tp->ops->walk(tp, &arg.w, true); } } return 0; } static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid, unsigned long new_cl) { const struct Qdisc_class_ops *cops = q->ops->cl_ops; struct tc_bind_class_args args = {}; if (!cops->tcf_block) return; args.portid = portid; args.clid = clid; args.new_cl = new_cl; args.w.fn = tc_bind_class_walker; q->ops->cl_ops->walk(q, &args.w); } #else static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid, unsigned long new_cl) { } #endif static int __tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack, struct net_device *dev, struct nlattr *tca[TCA_MAX + 1], struct tcmsg *tcm) { struct net *net = sock_net(skb->sk); const struct Qdisc_class_ops *cops; struct Qdisc *q = NULL; unsigned long cl = 0; unsigned long new_cl; u32 portid; u32 clid; u32 qid; int err; /* parent == TC_H_UNSPEC - unspecified parent. parent == TC_H_ROOT - class is root, which has no parent. parent == X:0 - parent is root class. parent == X:Y - parent is a node in hierarchy. parent == 0:Y - parent is X:Y, where X:0 is qdisc. handle == 0:0 - generate handle from kernel pool. handle == 0:Y - class is X:Y, where X:0 is qdisc. handle == X:Y - clear. handle == X:0 - root class. */ /* Step 1. Determine qdisc handle X:0 */ portid = tcm->tcm_parent; clid = tcm->tcm_handle; qid = TC_H_MAJ(clid); if (portid != TC_H_ROOT) { u32 qid1 = TC_H_MAJ(portid); if (qid && qid1) { /* If both majors are known, they must be identical. */ if (qid != qid1) return -EINVAL; } else if (qid1) { qid = qid1; } else if (qid == 0) qid = rtnl_dereference(dev->qdisc)->handle; /* Now qid is genuine qdisc handle consistent * both with parent and child. * * TC_H_MAJ(portid) still may be unspecified, complete it now. */ if (portid) portid = TC_H_MAKE(qid, portid); } else { if (qid == 0) qid = rtnl_dereference(dev->qdisc)->handle; } /* OK. Locate qdisc */ q = qdisc_lookup(dev, qid); if (!q) return -ENOENT; /* An check that it supports classes */ cops = q->ops->cl_ops; if (cops == NULL) return -EINVAL; /* Now try to get class */ if (clid == 0) { if (portid == TC_H_ROOT) clid = qid; } else clid = TC_H_MAKE(qid, clid); if (clid) cl = cops->find(q, clid); if (cl == 0) { err = -ENOENT; if (n->nlmsg_type != RTM_NEWTCLASS || !(n->nlmsg_flags & NLM_F_CREATE)) goto out; } else { switch (n->nlmsg_type) { case RTM_NEWTCLASS: err = -EEXIST; if (n->nlmsg_flags & NLM_F_EXCL) goto out; break; case RTM_DELTCLASS: err = tclass_del_notify(net, cops, skb, n, q, cl, extack); /* Unbind the class with flilters with 0 */ tc_bind_tclass(q, portid, clid, 0); goto out; case RTM_GETTCLASS: err = tclass_get_notify(net, skb, n, q, cl, extack); goto out; default: err = -EINVAL; goto out; } } if (tca[TCA_INGRESS_BLOCK] || tca[TCA_EGRESS_BLOCK]) { NL_SET_ERR_MSG(extack, "Shared blocks are not supported for classes"); return -EOPNOTSUPP; } /* Prevent creation of traffic classes with classid TC_H_ROOT */ if (clid == TC_H_ROOT) { NL_SET_ERR_MSG(extack, "Cannot create traffic class with classid TC_H_ROOT"); return -EINVAL; } new_cl = cl; err = -EOPNOTSUPP; if (cops->change) err = cops->change(q, clid, portid, tca, &new_cl, extack); if (err == 0) { tclass_notify(net, skb, n, q, new_cl, RTM_NEWTCLASS, extack); /* We just create a new class, need to do reverse binding. */ if (cl != new_cl) tc_bind_tclass(q, portid, clid, new_cl); } out: return err; } static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack) { struct net *net = sock_net(skb->sk); struct tcmsg *tcm = nlmsg_data(n); struct nlattr *tca[TCA_MAX + 1]; struct net_device *dev; int err; err = nlmsg_parse_deprecated(n, sizeof(*tcm), tca, TCA_MAX, rtm_tca_policy, extack); if (err < 0) return err; dev = __dev_get_by_index(net, tcm->tcm_ifindex); if (!dev) return -ENODEV; netdev_lock_ops(dev); err = __tc_ctl_tclass(skb, n, extack, dev, tca, tcm); netdev_unlock_ops(dev); return err; } struct qdisc_dump_args { struct qdisc_walker w; struct sk_buff *skb; struct netlink_callback *cb; }; static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walker *arg) { struct qdisc_dump_args *a = (struct qdisc_dump_args *)arg; return tc_fill_tclass(a->skb, q, cl, NETLINK_CB(a->cb->skb).portid, a->cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWTCLASS, NULL); } static int tc_dump_tclass_qdisc(struct Qdisc *q, struct sk_buff *skb, struct tcmsg *tcm, struct netlink_callback *cb, int *t_p, int s_t) { struct qdisc_dump_args arg; if (tc_qdisc_dump_ignore(q, false) || *t_p < s_t || !q->ops->cl_ops || (tcm->tcm_parent && TC_H_MAJ(tcm->tcm_parent) != q->handle)) { (*t_p)++; return 0; } if (*t_p > s_t) memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0])); arg.w.fn = qdisc_class_dump; arg.skb = skb; arg.cb = cb; arg.w.stop = 0; arg.w.skip = cb->args[1]; arg.w.count = 0; q->ops->cl_ops->walk(q, &arg.w); cb->args[1] = arg.w.count; if (arg.w.stop) return -1; (*t_p)++; return 0; } static int tc_dump_tclass_root(struct Qdisc *root, struct sk_buff *skb, struct tcmsg *tcm, struct netlink_callback *cb, int *t_p, int s_t, bool recur) { struct Qdisc *q; int b; if (!root) return 0; if (tc_dump_tclass_qdisc(root, skb, tcm, cb, t_p, s_t) < 0) return -1; if (!qdisc_dev(root) || !recur) return 0; if (tcm->tcm_parent) { q = qdisc_match_from_root(root, TC_H_MAJ(tcm->tcm_parent)); if (q && q != root && tc_dump_tclass_qdisc(q, skb, tcm, cb, t_p, s_t) < 0) return -1; return 0; } hash_for_each(qdisc_dev(root)->qdisc_hash, b, q, hash) { if (tc_dump_tclass_qdisc(q, skb, tcm, cb, t_p, s_t) < 0) return -1; } return 0; } static int __tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb, struct tcmsg *tcm, struct net_device *dev) { struct netdev_queue *dev_queue; int t, s_t; s_t = cb->args[0]; t = 0; if (tc_dump_tclass_root(rtnl_dereference(dev->qdisc), skb, tcm, cb, &t, s_t, true) < 0) goto done; dev_queue = dev_ingress_queue(dev); if (dev_queue && tc_dump_tclass_root(rtnl_dereference(dev_queue->qdisc_sleeping), skb, tcm, cb, &t, s_t, false) < 0) goto done; done: cb->args[0] = t; return skb->len; } static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb) { struct tcmsg *tcm = nlmsg_data(cb->nlh); struct net *net = sock_net(skb->sk); struct net_device *dev; int err; if (nlmsg_len(cb->nlh) < sizeof(*tcm)) return 0; dev = dev_get_by_index(net, tcm->tcm_ifindex); if (!dev) return 0; netdev_lock_ops(dev); err = __tc_dump_tclass(skb, cb, tcm, dev); netdev_unlock_ops(dev); dev_put(dev); return err; } #ifdef CONFIG_PROC_FS static int psched_show(struct seq_file *seq, void *v) { seq_printf(seq, "%08x %08x %08x %08x\n", (u32)NSEC_PER_USEC, (u32)PSCHED_TICKS2NS(1), 1000000, (u32)NSEC_PER_SEC / hrtimer_resolution); return 0; } static int __net_init psched_net_init(struct net *net) { struct proc_dir_entry *e; e = proc_create_single("psched", 0, net->proc_net, psched_show); if (e == NULL) return -ENOMEM; return 0; } static void __net_exit psched_net_exit(struct net *net) { remove_proc_entry("psched", net->proc_net); } #else static int __net_init psched_net_init(struct net *net) { return 0; } static void __net_exit psched_net_exit(struct net *net) { } #endif static struct pernet_operations psched_net_ops = { .init = psched_net_init, .exit = psched_net_exit, }; #if IS_ENABLED(CONFIG_MITIGATION_RETPOLINE) DEFINE_STATIC_KEY_FALSE(tc_skip_wrapper); #endif static const struct rtnl_msg_handler psched_rtnl_msg_handlers[] __initconst = { {.msgtype = RTM_NEWQDISC, .doit = tc_modify_qdisc}, {.msgtype = RTM_DELQDISC, .doit = tc_get_qdisc}, {.msgtype = RTM_GETQDISC, .doit = tc_get_qdisc, .dumpit = tc_dump_qdisc}, {.msgtype = RTM_NEWTCLASS, .doit = tc_ctl_tclass}, {.msgtype = RTM_DELTCLASS, .doit = tc_ctl_tclass}, {.msgtype = RTM_GETTCLASS, .doit = tc_ctl_tclass, .dumpit = tc_dump_tclass}, }; static int __init pktsched_init(void) { int err; err = register_pernet_subsys(&psched_net_ops); if (err) { pr_err("pktsched_init: " "cannot initialize per netns operations\n"); return err; } register_qdisc(&pfifo_fast_ops); register_qdisc(&pfifo_qdisc_ops); register_qdisc(&bfifo_qdisc_ops); register_qdisc(&pfifo_head_drop_qdisc_ops); register_qdisc(&mq_qdisc_ops); register_qdisc(&noqueue_qdisc_ops); rtnl_register_many(psched_rtnl_msg_handlers); tc_wrapper_init(); return 0; } subsys_initcall(pktsched_init); |
| 57 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 | // SPDX-License-Identifier: GPL-2.0-only /* * Common code for control of lockd and nfsv4 grace periods. * * Transplanted from lockd code */ #include <linux/module.h> #include <net/net_namespace.h> #include <net/netns/generic.h> #include <linux/fs.h> #include <linux/filelock.h> static unsigned int grace_net_id; static DEFINE_SPINLOCK(grace_lock); /** * locks_start_grace * @net: net namespace that this lock manager belongs to * @lm: who this grace period is for * * A grace period is a period during which locks should not be given * out. Currently grace periods are only enforced by the two lock * managers (lockd and nfsd), using the locks_in_grace() function to * check when they are in a grace period. * * This function is called to start a grace period. */ void locks_start_grace(struct net *net, struct lock_manager *lm) { struct list_head *grace_list = net_generic(net, grace_net_id); spin_lock(&grace_lock); if (list_empty(&lm->list)) list_add(&lm->list, grace_list); else WARN(1, "double list_add attempt detected in net %x %s\n", net->ns.inum, (net == &init_net) ? "(init_net)" : ""); spin_unlock(&grace_lock); } EXPORT_SYMBOL_GPL(locks_start_grace); /** * locks_end_grace * @lm: who this grace period is for * * Call this function to state that the given lock manager is ready to * resume regular locking. The grace period will not end until all lock * managers that called locks_start_grace() also call locks_end_grace(). * Note that callers count on it being safe to call this more than once, * and the second call should be a no-op. */ void locks_end_grace(struct lock_manager *lm) { spin_lock(&grace_lock); list_del_init(&lm->list); spin_unlock(&grace_lock); } EXPORT_SYMBOL_GPL(locks_end_grace); static bool __state_in_grace(struct net *net, bool open) { struct list_head *grace_list = net_generic(net, grace_net_id); struct lock_manager *lm; if (!open) return !list_empty(grace_list); spin_lock(&grace_lock); list_for_each_entry(lm, grace_list, list) { if (lm->block_opens) { spin_unlock(&grace_lock); return true; } } spin_unlock(&grace_lock); return false; } /** * locks_in_grace * @net: network namespace * * Lock managers call this function to determine when it is OK for them * to answer ordinary lock requests, and when they should accept only * lock reclaims. */ bool locks_in_grace(struct net *net) { return __state_in_grace(net, false); } EXPORT_SYMBOL_GPL(locks_in_grace); bool opens_in_grace(struct net *net) { return __state_in_grace(net, true); } EXPORT_SYMBOL_GPL(opens_in_grace); static int __net_init grace_init_net(struct net *net) { struct list_head *grace_list = net_generic(net, grace_net_id); INIT_LIST_HEAD(grace_list); return 0; } static void __net_exit grace_exit_net(struct net *net) { struct list_head *grace_list = net_generic(net, grace_net_id); WARN_ONCE(!list_empty(grace_list), "net %x %s: grace_list is not empty\n", net->ns.inum, __func__); } static struct pernet_operations grace_net_ops = { .init = grace_init_net, .exit = grace_exit_net, .id = &grace_net_id, .size = sizeof(struct list_head), }; static int __init init_grace(void) { return register_pernet_subsys(&grace_net_ops); } static void __exit exit_grace(void) { unregister_pernet_subsys(&grace_net_ops); } MODULE_AUTHOR("Jeff Layton <jlayton@primarydata.com>"); MODULE_DESCRIPTION("NFS client and server infrastructure"); MODULE_LICENSE("GPL"); module_init(init_grace) module_exit(exit_grace) |
| 213 171 219 219 216 219 219 219 218 6 6 6 4 6 6 2 2 2 2 2 2 2 15 3 17 3 3 1 3 3 3 3 3 1 2 2 2 2 3 76 41 76 76 9 7 2 8 1 7 7 5 5 3 1 1 1 3 3 3 3 5 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 | // SPDX-License-Identifier: GPL-2.0+ /* * Buffer/page management specific to NILFS * * Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation. * * Written by Ryusuke Konishi and Seiji Kihara. */ #include <linux/pagemap.h> #include <linux/writeback.h> #include <linux/swap.h> #include <linux/bitops.h> #include <linux/page-flags.h> #include <linux/list.h> #include <linux/highmem.h> #include <linux/pagevec.h> #include <linux/gfp.h> #include "nilfs.h" #include "page.h" #include "mdt.h" #define NILFS_BUFFER_INHERENT_BITS \ (BIT(BH_Uptodate) | BIT(BH_Mapped) | BIT(BH_NILFS_Node) | \ BIT(BH_NILFS_Volatile) | BIT(BH_NILFS_Checked)) static struct buffer_head *__nilfs_get_folio_block(struct folio *folio, unsigned long block, pgoff_t index, int blkbits, unsigned long b_state) { unsigned long first_block; struct buffer_head *bh = folio_buffers(folio); if (!bh) bh = create_empty_buffers(folio, 1 << blkbits, b_state); first_block = (unsigned long)index << (PAGE_SHIFT - blkbits); bh = get_nth_bh(bh, block - first_block); wait_on_buffer(bh); return bh; } struct buffer_head *nilfs_grab_buffer(struct inode *inode, struct address_space *mapping, unsigned long blkoff, unsigned long b_state) { int blkbits = inode->i_blkbits; pgoff_t index = blkoff >> (PAGE_SHIFT - blkbits); struct folio *folio; struct buffer_head *bh; folio = filemap_grab_folio(mapping, index); if (IS_ERR(folio)) return NULL; bh = __nilfs_get_folio_block(folio, blkoff, index, blkbits, b_state); if (unlikely(!bh)) { folio_unlock(folio); folio_put(folio); return NULL; } bh->b_bdev = inode->i_sb->s_bdev; return bh; } /** * nilfs_forget_buffer - discard dirty state * @bh: buffer head of the buffer to be discarded */ void nilfs_forget_buffer(struct buffer_head *bh) { struct folio *folio = bh->b_folio; const unsigned long clear_bits = (BIT(BH_Uptodate) | BIT(BH_Dirty) | BIT(BH_Mapped) | BIT(BH_Async_Write) | BIT(BH_NILFS_Volatile) | BIT(BH_NILFS_Checked) | BIT(BH_NILFS_Redirected) | BIT(BH_Delay)); lock_buffer(bh); set_mask_bits(&bh->b_state, clear_bits, 0); if (nilfs_folio_buffers_clean(folio)) __nilfs_clear_folio_dirty(folio); bh->b_blocknr = -1; folio_clear_uptodate(folio); folio_clear_mappedtodisk(folio); unlock_buffer(bh); brelse(bh); } /** * nilfs_copy_buffer -- copy buffer data and flags * @dbh: destination buffer * @sbh: source buffer */ void nilfs_copy_buffer(struct buffer_head *dbh, struct buffer_head *sbh) { void *saddr, *daddr; unsigned long bits; struct folio *sfolio = sbh->b_folio, *dfolio = dbh->b_folio; struct buffer_head *bh; saddr = kmap_local_folio(sfolio, bh_offset(sbh)); daddr = kmap_local_folio(dfolio, bh_offset(dbh)); memcpy(daddr, saddr, sbh->b_size); kunmap_local(daddr); kunmap_local(saddr); dbh->b_state = sbh->b_state & NILFS_BUFFER_INHERENT_BITS; dbh->b_blocknr = sbh->b_blocknr; dbh->b_bdev = sbh->b_bdev; bh = dbh; bits = sbh->b_state & (BIT(BH_Uptodate) | BIT(BH_Mapped)); while ((bh = bh->b_this_page) != dbh) { lock_buffer(bh); bits &= bh->b_state; unlock_buffer(bh); } if (bits & BIT(BH_Uptodate)) folio_mark_uptodate(dfolio); else folio_clear_uptodate(dfolio); if (bits & BIT(BH_Mapped)) folio_set_mappedtodisk(dfolio); else folio_clear_mappedtodisk(dfolio); } /** * nilfs_folio_buffers_clean - Check if a folio has dirty buffers or not. * @folio: Folio to be checked. * * Return: false if the folio has dirty buffers, true otherwise. */ bool nilfs_folio_buffers_clean(struct folio *folio) { struct buffer_head *bh, *head; bh = head = folio_buffers(folio); do { if (buffer_dirty(bh)) return false; bh = bh->b_this_page; } while (bh != head); return true; } void nilfs_folio_bug(struct folio *folio) { struct buffer_head *bh, *head; struct address_space *m; unsigned long ino; if (unlikely(!folio)) { printk(KERN_CRIT "NILFS_FOLIO_BUG(NULL)\n"); return; } m = folio->mapping; ino = m ? m->host->i_ino : 0; printk(KERN_CRIT "NILFS_FOLIO_BUG(%p): cnt=%d index#=%llu flags=0x%lx " "mapping=%p ino=%lu\n", folio, folio_ref_count(folio), (unsigned long long)folio->index, folio->flags.f, m, ino); head = folio_buffers(folio); if (head) { int i = 0; bh = head; do { printk(KERN_CRIT " BH[%d] %p: cnt=%d block#=%llu state=0x%lx\n", i++, bh, atomic_read(&bh->b_count), (unsigned long long)bh->b_blocknr, bh->b_state); bh = bh->b_this_page; } while (bh != head); } } /** * nilfs_copy_folio -- copy the folio with buffers * @dst: destination folio * @src: source folio * @copy_dirty: flag whether to copy dirty states on the folio's buffer heads. * * This function is for both data folios and btnode folios. The dirty flag * should be treated by caller. The folio must not be under i/o. * Both src and dst folio must be locked */ static void nilfs_copy_folio(struct folio *dst, struct folio *src, bool copy_dirty) { struct buffer_head *dbh, *dbufs, *sbh; unsigned long mask = NILFS_BUFFER_INHERENT_BITS; BUG_ON(folio_test_writeback(dst)); sbh = folio_buffers(src); dbh = folio_buffers(dst); if (!dbh) dbh = create_empty_buffers(dst, sbh->b_size, 0); if (copy_dirty) mask |= BIT(BH_Dirty); dbufs = dbh; do { lock_buffer(sbh); lock_buffer(dbh); dbh->b_state = sbh->b_state & mask; dbh->b_blocknr = sbh->b_blocknr; dbh->b_bdev = sbh->b_bdev; sbh = sbh->b_this_page; dbh = dbh->b_this_page; } while (dbh != dbufs); folio_copy(dst, src); if (folio_test_uptodate(src) && !folio_test_uptodate(dst)) folio_mark_uptodate(dst); else if (!folio_test_uptodate(src) && folio_test_uptodate(dst)) folio_clear_uptodate(dst); if (folio_test_mappedtodisk(src) && !folio_test_mappedtodisk(dst)) folio_set_mappedtodisk(dst); else if (!folio_test_mappedtodisk(src) && folio_test_mappedtodisk(dst)) folio_clear_mappedtodisk(dst); do { unlock_buffer(sbh); unlock_buffer(dbh); sbh = sbh->b_this_page; dbh = dbh->b_this_page; } while (dbh != dbufs); } int nilfs_copy_dirty_pages(struct address_space *dmap, struct address_space *smap) { struct folio_batch fbatch; unsigned int i; pgoff_t index = 0; int err = 0; folio_batch_init(&fbatch); repeat: if (!filemap_get_folios_tag(smap, &index, (pgoff_t)-1, PAGECACHE_TAG_DIRTY, &fbatch)) return 0; for (i = 0; i < folio_batch_count(&fbatch); i++) { struct folio *folio = fbatch.folios[i], *dfolio; folio_lock(folio); if (unlikely(!folio_test_dirty(folio))) NILFS_FOLIO_BUG(folio, "inconsistent dirty state"); dfolio = filemap_grab_folio(dmap, folio->index); if (IS_ERR(dfolio)) { /* No empty page is added to the page cache */ folio_unlock(folio); err = PTR_ERR(dfolio); break; } if (unlikely(!folio_buffers(folio))) NILFS_FOLIO_BUG(folio, "found empty page in dat page cache"); nilfs_copy_folio(dfolio, folio, true); filemap_dirty_folio(folio_mapping(dfolio), dfolio); folio_unlock(dfolio); folio_put(dfolio); folio_unlock(folio); } folio_batch_release(&fbatch); cond_resched(); if (likely(!err)) goto repeat; return err; } /** * nilfs_copy_back_pages -- copy back pages to original cache from shadow cache * @dmap: destination page cache * @smap: source page cache * * No pages must be added to the cache during this process. * This must be ensured by the caller. */ void nilfs_copy_back_pages(struct address_space *dmap, struct address_space *smap) { struct folio_batch fbatch; unsigned int i, n; pgoff_t start = 0; folio_batch_init(&fbatch); repeat: n = filemap_get_folios(smap, &start, ~0UL, &fbatch); if (!n) return; for (i = 0; i < folio_batch_count(&fbatch); i++) { struct folio *folio = fbatch.folios[i], *dfolio; pgoff_t index = folio->index; folio_lock(folio); dfolio = filemap_lock_folio(dmap, index); if (!IS_ERR(dfolio)) { /* overwrite existing folio in the destination cache */ WARN_ON(folio_test_dirty(dfolio)); nilfs_copy_folio(dfolio, folio, false); folio_unlock(dfolio); folio_put(dfolio); /* Do we not need to remove folio from smap here? */ } else { struct folio *f; /* move the folio to the destination cache */ xa_lock_irq(&smap->i_pages); f = __xa_erase(&smap->i_pages, index); WARN_ON(folio != f); smap->nrpages--; xa_unlock_irq(&smap->i_pages); xa_lock_irq(&dmap->i_pages); f = __xa_store(&dmap->i_pages, index, folio, GFP_NOFS); if (unlikely(f)) { /* Probably -ENOMEM */ folio->mapping = NULL; folio_put(folio); } else { folio->mapping = dmap; dmap->nrpages++; if (folio_test_dirty(folio)) __xa_set_mark(&dmap->i_pages, index, PAGECACHE_TAG_DIRTY); } xa_unlock_irq(&dmap->i_pages); } folio_unlock(folio); } folio_batch_release(&fbatch); cond_resched(); goto repeat; } /** * nilfs_clear_dirty_pages - discard dirty pages in address space * @mapping: address space with dirty pages for discarding */ void nilfs_clear_dirty_pages(struct address_space *mapping) { struct folio_batch fbatch; unsigned int i; pgoff_t index = 0; folio_batch_init(&fbatch); while (filemap_get_folios_tag(mapping, &index, (pgoff_t)-1, PAGECACHE_TAG_DIRTY, &fbatch)) { for (i = 0; i < folio_batch_count(&fbatch); i++) { struct folio *folio = fbatch.folios[i]; folio_lock(folio); /* * This folio may have been removed from the address * space by truncation or invalidation when the lock * was acquired. Skip processing in that case. */ if (likely(folio->mapping == mapping)) nilfs_clear_folio_dirty(folio); folio_unlock(folio); } folio_batch_release(&fbatch); cond_resched(); } } /** * nilfs_clear_folio_dirty - discard dirty folio * @folio: dirty folio that will be discarded * * nilfs_clear_folio_dirty() clears working states including dirty state for * the folio and its buffers. If the folio has buffers, clear only if it is * confirmed that none of the buffer heads are busy (none have valid * references and none are locked). */ void nilfs_clear_folio_dirty(struct folio *folio) { struct buffer_head *bh, *head; BUG_ON(!folio_test_locked(folio)); head = folio_buffers(folio); if (head) { const unsigned long clear_bits = (BIT(BH_Uptodate) | BIT(BH_Dirty) | BIT(BH_Mapped) | BIT(BH_Async_Write) | BIT(BH_NILFS_Volatile) | BIT(BH_NILFS_Checked) | BIT(BH_NILFS_Redirected) | BIT(BH_Delay)); bool busy, invalidated = false; recheck_buffers: busy = false; bh = head; do { if (atomic_read(&bh->b_count) | buffer_locked(bh)) { busy = true; break; } } while (bh = bh->b_this_page, bh != head); if (busy) { if (invalidated) return; invalidate_bh_lrus(); invalidated = true; goto recheck_buffers; } bh = head; do { lock_buffer(bh); set_mask_bits(&bh->b_state, clear_bits, 0); unlock_buffer(bh); } while (bh = bh->b_this_page, bh != head); } folio_clear_uptodate(folio); folio_clear_mappedtodisk(folio); folio_clear_checked(folio); __nilfs_clear_folio_dirty(folio); } unsigned int nilfs_page_count_clean_buffers(struct folio *folio, unsigned int from, unsigned int to) { unsigned int block_start, block_end; struct buffer_head *bh, *head; unsigned int nc = 0; for (bh = head = folio_buffers(folio), block_start = 0; bh != head || !block_start; block_start = block_end, bh = bh->b_this_page) { block_end = block_start + bh->b_size; if (block_end > from && block_start < to && !buffer_dirty(bh)) nc++; } return nc; } /* * NILFS2 needs clear_page_dirty() in the following two cases: * * 1) For B-tree node pages and data pages of DAT file, NILFS2 clears dirty * flag of pages when it copies back pages from shadow cache to the * original cache. * * 2) Some B-tree operations like insertion or deletion may dispose buffers * in dirty state, and this needs to cancel the dirty state of their pages. */ void __nilfs_clear_folio_dirty(struct folio *folio) { struct address_space *mapping = folio->mapping; if (mapping) { xa_lock_irq(&mapping->i_pages); if (folio_test_dirty(folio)) { __xa_clear_mark(&mapping->i_pages, folio->index, PAGECACHE_TAG_DIRTY); xa_unlock_irq(&mapping->i_pages); folio_clear_dirty_for_io(folio); return; } xa_unlock_irq(&mapping->i_pages); return; } folio_clear_dirty(folio); } /** * nilfs_find_uncommitted_extent - find extent of uncommitted data * @inode: inode * @start_blk: start block offset (in) * @blkoff: start offset of the found extent (out) * * This function searches an extent of buffers marked "delayed" which * starts from a block offset equal to or larger than @start_blk. If * such an extent was found, this will store the start offset in * @blkoff and return its length in blocks. * * Return: Length in blocks of found extent, 0 otherwise. */ unsigned long nilfs_find_uncommitted_extent(struct inode *inode, sector_t start_blk, sector_t *blkoff) { unsigned int i, nr_folios; pgoff_t index; unsigned long length = 0; struct folio_batch fbatch; struct folio *folio; if (inode->i_mapping->nrpages == 0) return 0; index = start_blk >> (PAGE_SHIFT - inode->i_blkbits); folio_batch_init(&fbatch); repeat: nr_folios = filemap_get_folios_contig(inode->i_mapping, &index, ULONG_MAX, &fbatch); if (nr_folios == 0) return length; i = 0; do { folio = fbatch.folios[i]; folio_lock(folio); if (folio_buffers(folio)) { struct buffer_head *bh, *head; sector_t b; b = folio->index << (PAGE_SHIFT - inode->i_blkbits); bh = head = folio_buffers(folio); do { if (b < start_blk) continue; if (buffer_delay(bh)) { if (length == 0) *blkoff = b; length++; } else if (length > 0) { goto out_locked; } } while (++b, bh = bh->b_this_page, bh != head); } else { if (length > 0) goto out_locked; } folio_unlock(folio); } while (++i < nr_folios); folio_batch_release(&fbatch); cond_resched(); goto repeat; out_locked: folio_unlock(folio); folio_batch_release(&fbatch); return length; } |
| 3162 5413 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 | /* SPDX-License-Identifier: GPL-2.0 */ #ifndef _LINUX_SCHED_USER_H #define _LINUX_SCHED_USER_H #include <linux/uidgid.h> #include <linux/atomic.h> #include <linux/percpu_counter.h> #include <linux/refcount.h> #include <linux/ratelimit.h> /* * Some day this will be a full-fledged user tracking system.. */ struct user_struct { refcount_t __count; /* reference count */ #ifdef CONFIG_EPOLL struct percpu_counter epoll_watches; /* The number of file descriptors currently watched */ #endif unsigned long unix_inflight; /* How many files in flight in unix sockets */ atomic_long_t pipe_bufs; /* how many pages are allocated in pipe buffers */ /* Hash table maintenance information */ struct hlist_node uidhash_node; kuid_t uid; #if defined(CONFIG_PERF_EVENTS) || defined(CONFIG_BPF_SYSCALL) || \ defined(CONFIG_NET) || defined(CONFIG_IO_URING) || \ defined(CONFIG_VFIO_PCI_ZDEV_KVM) || IS_ENABLED(CONFIG_IOMMUFD) atomic_long_t locked_vm; #endif #ifdef CONFIG_WATCH_QUEUE atomic_t nr_watches; /* The number of watches this user currently has */ #endif /* Miscellaneous per-user rate limit */ struct ratelimit_state ratelimit; }; extern int uids_sysfs_init(void); extern struct user_struct *find_user(kuid_t); extern struct user_struct root_user; #define INIT_USER (&root_user) /* per-UID process charging. */ extern struct user_struct * alloc_uid(kuid_t); static inline struct user_struct *get_uid(struct user_struct *u) { refcount_inc(&u->__count); return u; } extern void free_uid(struct user_struct *); #endif /* _LINUX_SCHED_USER_H */ |
| 6 5 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 | /* SPDX-License-Identifier: GPL-2.0 */ #ifndef __NET_TC_GACT_H #define __NET_TC_GACT_H #include <net/act_api.h> #include <linux/tc_act/tc_gact.h> struct tcf_gact { struct tc_action common; #ifdef CONFIG_GACT_PROB u16 tcfg_ptype; u16 tcfg_pval; int tcfg_paction; atomic_t packets; #endif }; #define to_gact(a) ((struct tcf_gact *)a) static inline bool __is_tcf_gact_act(const struct tc_action *a, int act, bool is_ext) { #ifdef CONFIG_NET_CLS_ACT struct tcf_gact *gact; if (a->ops && a->ops->id != TCA_ID_GACT) return false; gact = to_gact(a); if ((!is_ext && gact->tcf_action == act) || (is_ext && TC_ACT_EXT_CMP(gact->tcf_action, act))) return true; #endif return false; } static inline bool is_tcf_gact_ok(const struct tc_action *a) { return __is_tcf_gact_act(a, TC_ACT_OK, false); } static inline bool is_tcf_gact_shot(const struct tc_action *a) { return __is_tcf_gact_act(a, TC_ACT_SHOT, false); } static inline bool is_tcf_gact_trap(const struct tc_action *a) { return __is_tcf_gact_act(a, TC_ACT_TRAP, false); } static inline bool is_tcf_gact_goto_chain(const struct tc_action *a) { return __is_tcf_gact_act(a, TC_ACT_GOTO_CHAIN, true); } static inline u32 tcf_gact_goto_chain_index(const struct tc_action *a) { return READ_ONCE(a->tcfa_action) & TC_ACT_EXT_VAL_MASK; } static inline bool is_tcf_gact_continue(const struct tc_action *a) { return __is_tcf_gact_act(a, TC_ACT_UNSPEC, false); } static inline bool is_tcf_gact_reclassify(const struct tc_action *a) { return __is_tcf_gact_act(a, TC_ACT_RECLASSIFY, false); } static inline bool is_tcf_gact_pipe(const struct tc_action *a) { return __is_tcf_gact_act(a, TC_ACT_PIPE, false); } #endif /* __NET_TC_GACT_H */ |
| 8 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 | /* SPDX-License-Identifier: GPL-2.0 */ #ifndef _ASM_X86_SPECCTRL_H_ #define _ASM_X86_SPECCTRL_H_ #include <linux/thread_info.h> #include <asm/nospec-branch.h> #include <asm/msr.h> /* * On VMENTER we must preserve whatever view of the SPEC_CTRL MSR * the guest has, while on VMEXIT we restore the host view. This * would be easier if SPEC_CTRL were architecturally maskable or * shadowable for guests but this is not (currently) the case. * Takes the guest view of SPEC_CTRL MSR as a parameter and also * the guest's version of VIRT_SPEC_CTRL, if emulated. */ extern void x86_virt_spec_ctrl(u64 guest_virt_spec_ctrl, bool guest); /** * x86_spec_ctrl_set_guest - Set speculation control registers for the guest * @guest_spec_ctrl: The guest content of MSR_SPEC_CTRL * @guest_virt_spec_ctrl: The guest controlled bits of MSR_VIRT_SPEC_CTRL * (may get translated to MSR_AMD64_LS_CFG bits) * * Avoids writing to the MSR if the content/bits are the same */ static inline void x86_spec_ctrl_set_guest(u64 guest_virt_spec_ctrl) { x86_virt_spec_ctrl(guest_virt_spec_ctrl, true); } /** * x86_spec_ctrl_restore_host - Restore host speculation control registers * @guest_spec_ctrl: The guest content of MSR_SPEC_CTRL * @guest_virt_spec_ctrl: The guest controlled bits of MSR_VIRT_SPEC_CTRL * (may get translated to MSR_AMD64_LS_CFG bits) * * Avoids writing to the MSR if the content/bits are the same */ static inline void x86_spec_ctrl_restore_host(u64 guest_virt_spec_ctrl) { x86_virt_spec_ctrl(guest_virt_spec_ctrl, false); } /* AMD specific Speculative Store Bypass MSR data */ extern u64 x86_amd_ls_cfg_base; extern u64 x86_amd_ls_cfg_ssbd_mask; static inline u64 ssbd_tif_to_spec_ctrl(u64 tifn) { BUILD_BUG_ON(TIF_SSBD < SPEC_CTRL_SSBD_SHIFT); return (tifn & _TIF_SSBD) >> (TIF_SSBD - SPEC_CTRL_SSBD_SHIFT); } static inline u64 stibp_tif_to_spec_ctrl(u64 tifn) { BUILD_BUG_ON(TIF_SPEC_IB < SPEC_CTRL_STIBP_SHIFT); return (tifn & _TIF_SPEC_IB) >> (TIF_SPEC_IB - SPEC_CTRL_STIBP_SHIFT); } static inline unsigned long ssbd_spec_ctrl_to_tif(u64 spec_ctrl) { BUILD_BUG_ON(TIF_SSBD < SPEC_CTRL_SSBD_SHIFT); return (spec_ctrl & SPEC_CTRL_SSBD) << (TIF_SSBD - SPEC_CTRL_SSBD_SHIFT); } static inline unsigned long stibp_spec_ctrl_to_tif(u64 spec_ctrl) { BUILD_BUG_ON(TIF_SPEC_IB < SPEC_CTRL_STIBP_SHIFT); return (spec_ctrl & SPEC_CTRL_STIBP) << (TIF_SPEC_IB - SPEC_CTRL_STIBP_SHIFT); } static inline u64 ssbd_tif_to_amd_ls_cfg(u64 tifn) { return (tifn & _TIF_SSBD) ? x86_amd_ls_cfg_ssbd_mask : 0ULL; } /* * This can be used in noinstr functions & should only be called in bare * metal context. */ static __always_inline void __update_spec_ctrl(u64 val) { __this_cpu_write(x86_spec_ctrl_current, val); native_wrmsrq(MSR_IA32_SPEC_CTRL, val); } #ifdef CONFIG_SMP extern void speculative_store_bypass_ht_init(void); #else static inline void speculative_store_bypass_ht_init(void) { } #endif extern void speculation_ctrl_update(unsigned long tif); extern void speculation_ctrl_update_current(void); extern bool itlb_multihit_kvm_mitigation; #endif |
| 18 18 15 18 18 1 14 2 15 15 15 15 15 15 15 15 15 15 15 14 15 15 15 15 15 15 14 15 15 15 15 15 15 18 18 280 18 17 18 16 18 18 2 1 17 15 15 15 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 | // SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause /* * Copyright (c) Meta Platforms, Inc. and affiliates. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the * LICENSE file in the root directory of this source tree) and the GPLv2 (found * in the COPYING file in the root directory of this source tree). * You may select, at your option, one of the above-listed licenses. */ /* *************************************************************** * Tuning parameters *****************************************************************/ /*! * HEAPMODE : * Select how default decompression function ZSTD_decompress() allocates its context, * on stack (0), or into heap (1, default; requires malloc()). * Note that functions with explicit context such as ZSTD_decompressDCtx() are unaffected. */ #ifndef ZSTD_HEAPMODE # define ZSTD_HEAPMODE 1 #endif /*! * LEGACY_SUPPORT : * if set to 1+, ZSTD_decompress() can decode older formats (v0.1+) */ /*! * MAXWINDOWSIZE_DEFAULT : * maximum window size accepted by DStream __by default__. * Frames requiring more memory will be rejected. * It's possible to set a different limit using ZSTD_DCtx_setMaxWindowSize(). */ #ifndef ZSTD_MAXWINDOWSIZE_DEFAULT # define ZSTD_MAXWINDOWSIZE_DEFAULT (((U32)1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT) + 1) #endif /*! * NO_FORWARD_PROGRESS_MAX : * maximum allowed nb of calls to ZSTD_decompressStream() * without any forward progress * (defined as: no byte read from input, and no byte flushed to output) * before triggering an error. */ #ifndef ZSTD_NO_FORWARD_PROGRESS_MAX # define ZSTD_NO_FORWARD_PROGRESS_MAX 16 #endif /*-******************************************************* * Dependencies *********************************************************/ #include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ #include "../common/allocations.h" /* ZSTD_customMalloc, ZSTD_customCalloc, ZSTD_customFree */ #include "../common/error_private.h" #include "../common/zstd_internal.h" /* blockProperties_t */ #include "../common/mem.h" /* low level memory routines */ #include "../common/bits.h" /* ZSTD_highbit32 */ #define FSE_STATIC_LINKING_ONLY #include "../common/fse.h" #include "../common/huf.h" #include <linux/xxhash.h> /* xxh64_reset, xxh64_update, xxh64_digest, XXH64 */ #include "zstd_decompress_internal.h" /* ZSTD_DCtx */ #include "zstd_ddict.h" /* ZSTD_DDictDictContent */ #include "zstd_decompress_block.h" /* ZSTD_decompressBlock_internal */ /* *********************************** * Multiple DDicts Hashset internals * *************************************/ #define DDICT_HASHSET_MAX_LOAD_FACTOR_COUNT_MULT 4 #define DDICT_HASHSET_MAX_LOAD_FACTOR_SIZE_MULT 3 /* These two constants represent SIZE_MULT/COUNT_MULT load factor without using a float. * Currently, that means a 0.75 load factor. * So, if count * COUNT_MULT / size * SIZE_MULT != 0, then we've exceeded * the load factor of the ddict hash set. */ #define DDICT_HASHSET_TABLE_BASE_SIZE 64 #define DDICT_HASHSET_RESIZE_FACTOR 2 /* Hash function to determine starting position of dict insertion within the table * Returns an index between [0, hashSet->ddictPtrTableSize] */ static size_t ZSTD_DDictHashSet_getIndex(const ZSTD_DDictHashSet* hashSet, U32 dictID) { const U64 hash = xxh64(&dictID, sizeof(U32), 0); /* DDict ptr table size is a multiple of 2, use size - 1 as mask to get index within [0, hashSet->ddictPtrTableSize) */ return hash & (hashSet->ddictPtrTableSize - 1); } /* Adds DDict to a hashset without resizing it. * If inserting a DDict with a dictID that already exists in the set, replaces the one in the set. * Returns 0 if successful, or a zstd error code if something went wrong. */ static size_t ZSTD_DDictHashSet_emplaceDDict(ZSTD_DDictHashSet* hashSet, const ZSTD_DDict* ddict) { const U32 dictID = ZSTD_getDictID_fromDDict(ddict); size_t idx = ZSTD_DDictHashSet_getIndex(hashSet, dictID); const size_t idxRangeMask = hashSet->ddictPtrTableSize - 1; RETURN_ERROR_IF(hashSet->ddictPtrCount == hashSet->ddictPtrTableSize, GENERIC, "Hash set is full!"); DEBUGLOG(4, "Hashed index: for dictID: %u is %zu", dictID, idx); while (hashSet->ddictPtrTable[idx] != NULL) { /* Replace existing ddict if inserting ddict with same dictID */ if (ZSTD_getDictID_fromDDict(hashSet->ddictPtrTable[idx]) == dictID) { DEBUGLOG(4, "DictID already exists, replacing rather than adding"); hashSet->ddictPtrTable[idx] = ddict; return 0; } idx &= idxRangeMask; idx++; } DEBUGLOG(4, "Final idx after probing for dictID %u is: %zu", dictID, idx); hashSet->ddictPtrTable[idx] = ddict; hashSet->ddictPtrCount++; return 0; } /* Expands hash table by factor of DDICT_HASHSET_RESIZE_FACTOR and * rehashes all values, allocates new table, frees old table. * Returns 0 on success, otherwise a zstd error code. */ static size_t ZSTD_DDictHashSet_expand(ZSTD_DDictHashSet* hashSet, ZSTD_customMem customMem) { size_t newTableSize = hashSet->ddictPtrTableSize * DDICT_HASHSET_RESIZE_FACTOR; const ZSTD_DDict** newTable = (const ZSTD_DDict**)ZSTD_customCalloc(sizeof(ZSTD_DDict*) * newTableSize, customMem); const ZSTD_DDict** oldTable = hashSet->ddictPtrTable; size_t oldTableSize = hashSet->ddictPtrTableSize; size_t i; DEBUGLOG(4, "Expanding DDict hash table! Old size: %zu new size: %zu", oldTableSize, newTableSize); RETURN_ERROR_IF(!newTable, memory_allocation, "Expanded hashset allocation failed!"); hashSet->ddictPtrTable = newTable; hashSet->ddictPtrTableSize = newTableSize; hashSet->ddictPtrCount = 0; for (i = 0; i < oldTableSize; ++i) { if (oldTable[i] != NULL) { FORWARD_IF_ERROR(ZSTD_DDictHashSet_emplaceDDict(hashSet, oldTable[i]), ""); } } ZSTD_customFree((void*)oldTable, customMem); DEBUGLOG(4, "Finished re-hash"); return 0; } /* Fetches a DDict with the given dictID * Returns the ZSTD_DDict* with the requested dictID. If it doesn't exist, then returns NULL. */ static const ZSTD_DDict* ZSTD_DDictHashSet_getDDict(ZSTD_DDictHashSet* hashSet, U32 dictID) { size_t idx = ZSTD_DDictHashSet_getIndex(hashSet, dictID); const size_t idxRangeMask = hashSet->ddictPtrTableSize - 1; DEBUGLOG(4, "Hashed index: for dictID: %u is %zu", dictID, idx); for (;;) { size_t currDictID = ZSTD_getDictID_fromDDict(hashSet->ddictPtrTable[idx]); if (currDictID == dictID || currDictID == 0) { /* currDictID == 0 implies a NULL ddict entry */ break; } else { idx &= idxRangeMask; /* Goes to start of table when we reach the end */ idx++; } } DEBUGLOG(4, "Final idx after probing for dictID %u is: %zu", dictID, idx); return hashSet->ddictPtrTable[idx]; } /* Allocates space for and returns a ddict hash set * The hash set's ZSTD_DDict* table has all values automatically set to NULL to begin with. * Returns NULL if allocation failed. */ static ZSTD_DDictHashSet* ZSTD_createDDictHashSet(ZSTD_customMem customMem) { ZSTD_DDictHashSet* ret = (ZSTD_DDictHashSet*)ZSTD_customMalloc(sizeof(ZSTD_DDictHashSet), customMem); DEBUGLOG(4, "Allocating new hash set"); if (!ret) return NULL; ret->ddictPtrTable = (const ZSTD_DDict**)ZSTD_customCalloc(DDICT_HASHSET_TABLE_BASE_SIZE * sizeof(ZSTD_DDict*), customMem); if (!ret->ddictPtrTable) { ZSTD_customFree(ret, customMem); return NULL; } ret->ddictPtrTableSize = DDICT_HASHSET_TABLE_BASE_SIZE; ret->ddictPtrCount = 0; return ret; } /* Frees the table of ZSTD_DDict* within a hashset, then frees the hashset itself. * Note: The ZSTD_DDict* within the table are NOT freed. */ static void ZSTD_freeDDictHashSet(ZSTD_DDictHashSet* hashSet, ZSTD_customMem customMem) { DEBUGLOG(4, "Freeing ddict hash set"); if (hashSet && hashSet->ddictPtrTable) { ZSTD_customFree((void*)hashSet->ddictPtrTable, customMem); } if (hashSet) { ZSTD_customFree(hashSet, customMem); } } /* Public function: Adds a DDict into the ZSTD_DDictHashSet, possibly triggering a resize of the hash set. * Returns 0 on success, or a ZSTD error. */ static size_t ZSTD_DDictHashSet_addDDict(ZSTD_DDictHashSet* hashSet, const ZSTD_DDict* ddict, ZSTD_customMem customMem) { DEBUGLOG(4, "Adding dict ID: %u to hashset with - Count: %zu Tablesize: %zu", ZSTD_getDictID_fromDDict(ddict), hashSet->ddictPtrCount, hashSet->ddictPtrTableSize); if (hashSet->ddictPtrCount * DDICT_HASHSET_MAX_LOAD_FACTOR_COUNT_MULT / hashSet->ddictPtrTableSize * DDICT_HASHSET_MAX_LOAD_FACTOR_SIZE_MULT != 0) { FORWARD_IF_ERROR(ZSTD_DDictHashSet_expand(hashSet, customMem), ""); } FORWARD_IF_ERROR(ZSTD_DDictHashSet_emplaceDDict(hashSet, ddict), ""); return 0; } /*-************************************************************* * Context management ***************************************************************/ size_t ZSTD_sizeof_DCtx (const ZSTD_DCtx* dctx) { if (dctx==NULL) return 0; /* support sizeof NULL */ return sizeof(*dctx) + ZSTD_sizeof_DDict(dctx->ddictLocal) + dctx->inBuffSize + dctx->outBuffSize; } size_t ZSTD_estimateDCtxSize(void) { return sizeof(ZSTD_DCtx); } static size_t ZSTD_startingInputLength(ZSTD_format_e format) { size_t const startingInputLength = ZSTD_FRAMEHEADERSIZE_PREFIX(format); /* only supports formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless */ assert( (format == ZSTD_f_zstd1) || (format == ZSTD_f_zstd1_magicless) ); return startingInputLength; } static void ZSTD_DCtx_resetParameters(ZSTD_DCtx* dctx) { assert(dctx->streamStage == zdss_init); dctx->format = ZSTD_f_zstd1; dctx->maxWindowSize = ZSTD_MAXWINDOWSIZE_DEFAULT; dctx->outBufferMode = ZSTD_bm_buffered; dctx->forceIgnoreChecksum = ZSTD_d_validateChecksum; dctx->refMultipleDDicts = ZSTD_rmd_refSingleDDict; dctx->disableHufAsm = 0; dctx->maxBlockSizeParam = 0; } static void ZSTD_initDCtx_internal(ZSTD_DCtx* dctx) { dctx->staticSize = 0; dctx->ddict = NULL; dctx->ddictLocal = NULL; dctx->dictEnd = NULL; dctx->ddictIsCold = 0; dctx->dictUses = ZSTD_dont_use; dctx->inBuff = NULL; dctx->inBuffSize = 0; dctx->outBuffSize = 0; dctx->streamStage = zdss_init; dctx->noForwardProgress = 0; dctx->oversizedDuration = 0; dctx->isFrameDecompression = 1; #if DYNAMIC_BMI2 dctx->bmi2 = ZSTD_cpuSupportsBmi2(); #endif dctx->ddictSet = NULL; ZSTD_DCtx_resetParameters(dctx); #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION dctx->dictContentEndForFuzzing = NULL; #endif } ZSTD_DCtx* ZSTD_initStaticDCtx(void *workspace, size_t workspaceSize) { ZSTD_DCtx* const dctx = (ZSTD_DCtx*) workspace; if ((size_t)workspace & 7) return NULL; /* 8-aligned */ if (workspaceSize < sizeof(ZSTD_DCtx)) return NULL; /* minimum size */ ZSTD_initDCtx_internal(dctx); dctx->staticSize = workspaceSize; dctx->inBuff = (char*)(dctx+1); return dctx; } static ZSTD_DCtx* ZSTD_createDCtx_internal(ZSTD_customMem customMem) { if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; { ZSTD_DCtx* const dctx = (ZSTD_DCtx*)ZSTD_customMalloc(sizeof(*dctx), customMem); if (!dctx) return NULL; dctx->customMem = customMem; ZSTD_initDCtx_internal(dctx); return dctx; } } ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem) { return ZSTD_createDCtx_internal(customMem); } ZSTD_DCtx* ZSTD_createDCtx(void) { DEBUGLOG(3, "ZSTD_createDCtx"); return ZSTD_createDCtx_internal(ZSTD_defaultCMem); } static void ZSTD_clearDict(ZSTD_DCtx* dctx) { ZSTD_freeDDict(dctx->ddictLocal); dctx->ddictLocal = NULL; dctx->ddict = NULL; dctx->dictUses = ZSTD_dont_use; } size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx) { if (dctx==NULL) return 0; /* support free on NULL */ RETURN_ERROR_IF(dctx->staticSize, memory_allocation, "not compatible with static DCtx"); { ZSTD_customMem const cMem = dctx->customMem; ZSTD_clearDict(dctx); ZSTD_customFree(dctx->inBuff, cMem); dctx->inBuff = NULL; if (dctx->ddictSet) { ZSTD_freeDDictHashSet(dctx->ddictSet, cMem); dctx->ddictSet = NULL; } ZSTD_customFree(dctx, cMem); return 0; } } /* no longer useful */ void ZSTD_copyDCtx(ZSTD_DCtx* dstDCtx, const ZSTD_DCtx* srcDCtx) { size_t const toCopy = (size_t)((char*)(&dstDCtx->inBuff) - (char*)dstDCtx); ZSTD_memcpy(dstDCtx, srcDCtx, toCopy); /* no need to copy workspace */ } /* Given a dctx with a digested frame params, re-selects the correct ZSTD_DDict based on * the requested dict ID from the frame. If there exists a reference to the correct ZSTD_DDict, then * accordingly sets the ddict to be used to decompress the frame. * * If no DDict is found, then no action is taken, and the ZSTD_DCtx::ddict remains as-is. * * ZSTD_d_refMultipleDDicts must be enabled for this function to be called. */ static void ZSTD_DCtx_selectFrameDDict(ZSTD_DCtx* dctx) { assert(dctx->refMultipleDDicts && dctx->ddictSet); DEBUGLOG(4, "Adjusting DDict based on requested dict ID from frame"); if (dctx->ddict) { const ZSTD_DDict* frameDDict = ZSTD_DDictHashSet_getDDict(dctx->ddictSet, dctx->fParams.dictID); if (frameDDict) { DEBUGLOG(4, "DDict found!"); ZSTD_clearDict(dctx); dctx->dictID = dctx->fParams.dictID; dctx->ddict = frameDDict; dctx->dictUses = ZSTD_use_indefinitely; } } } /*-************************************************************* * Frame header decoding ***************************************************************/ /*! ZSTD_isFrame() : * Tells if the content of `buffer` starts with a valid Frame Identifier. * Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0. * Note 2 : Legacy Frame Identifiers are considered valid only if Legacy Support is enabled. * Note 3 : Skippable Frame Identifiers are considered valid. */ unsigned ZSTD_isFrame(const void* buffer, size_t size) { if (size < ZSTD_FRAMEIDSIZE) return 0; { U32 const magic = MEM_readLE32(buffer); if (magic == ZSTD_MAGICNUMBER) return 1; if ((magic & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) return 1; } return 0; } /*! ZSTD_isSkippableFrame() : * Tells if the content of `buffer` starts with a valid Frame Identifier for a skippable frame. * Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0. */ unsigned ZSTD_isSkippableFrame(const void* buffer, size_t size) { if (size < ZSTD_FRAMEIDSIZE) return 0; { U32 const magic = MEM_readLE32(buffer); if ((magic & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) return 1; } return 0; } /* ZSTD_frameHeaderSize_internal() : * srcSize must be large enough to reach header size fields. * note : only works for formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless. * @return : size of the Frame Header * or an error code, which can be tested with ZSTD_isError() */ static size_t ZSTD_frameHeaderSize_internal(const void* src, size_t srcSize, ZSTD_format_e format) { size_t const minInputSize = ZSTD_startingInputLength(format); RETURN_ERROR_IF(srcSize < minInputSize, srcSize_wrong, ""); { BYTE const fhd = ((const BYTE*)src)[minInputSize-1]; U32 const dictID= fhd & 3; U32 const singleSegment = (fhd >> 5) & 1; U32 const fcsId = fhd >> 6; return minInputSize + !singleSegment + ZSTD_did_fieldSize[dictID] + ZSTD_fcs_fieldSize[fcsId] + (singleSegment && !fcsId); } } /* ZSTD_frameHeaderSize() : * srcSize must be >= ZSTD_frameHeaderSize_prefix. * @return : size of the Frame Header, * or an error code (if srcSize is too small) */ size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize) { return ZSTD_frameHeaderSize_internal(src, srcSize, ZSTD_f_zstd1); } /* ZSTD_getFrameHeader_advanced() : * decode Frame Header, or require larger `srcSize`. * note : only works for formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless * @return : 0, `zfhPtr` is correctly filled, * >0, `srcSize` is too small, value is wanted `srcSize` amount, ** or an error code, which can be tested using ZSTD_isError() */ size_t ZSTD_getFrameHeader_advanced(ZSTD_FrameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format) { const BYTE* ip = (const BYTE*)src; size_t const minInputSize = ZSTD_startingInputLength(format); DEBUGLOG(5, "ZSTD_getFrameHeader_advanced: minInputSize = %zu, srcSize = %zu", minInputSize, srcSize); if (srcSize > 0) { /* note : technically could be considered an assert(), since it's an invalid entry */ RETURN_ERROR_IF(src==NULL, GENERIC, "invalid parameter : src==NULL, but srcSize>0"); } if (srcSize < minInputSize) { if (srcSize > 0 && format != ZSTD_f_zstd1_magicless) { /* when receiving less than @minInputSize bytes, * control these bytes at least correspond to a supported magic number * in order to error out early if they don't. **/ size_t const toCopy = MIN(4, srcSize); unsigned char hbuf[4]; MEM_writeLE32(hbuf, ZSTD_MAGICNUMBER); assert(src != NULL); ZSTD_memcpy(hbuf, src, toCopy); if ( MEM_readLE32(hbuf) != ZSTD_MAGICNUMBER ) { /* not a zstd frame : let's check if it's a skippable frame */ MEM_writeLE32(hbuf, ZSTD_MAGIC_SKIPPABLE_START); ZSTD_memcpy(hbuf, src, toCopy); if ((MEM_readLE32(hbuf) & ZSTD_MAGIC_SKIPPABLE_MASK) != ZSTD_MAGIC_SKIPPABLE_START) { RETURN_ERROR(prefix_unknown, "first bytes don't correspond to any supported magic number"); } } } return minInputSize; } ZSTD_memset(zfhPtr, 0, sizeof(*zfhPtr)); /* not strictly necessary, but static analyzers may not understand that zfhPtr will be read only if return value is zero, since they are 2 different signals */ if ( (format != ZSTD_f_zstd1_magicless) && (MEM_readLE32(src) != ZSTD_MAGICNUMBER) ) { if ((MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */ if (srcSize < ZSTD_SKIPPABLEHEADERSIZE) return ZSTD_SKIPPABLEHEADERSIZE; /* magic number + frame length */ ZSTD_memset(zfhPtr, 0, sizeof(*zfhPtr)); zfhPtr->frameType = ZSTD_skippableFrame; zfhPtr->dictID = MEM_readLE32(src) - ZSTD_MAGIC_SKIPPABLE_START; zfhPtr->headerSize = ZSTD_SKIPPABLEHEADERSIZE; zfhPtr->frameContentSize = MEM_readLE32((const char *)src + ZSTD_FRAMEIDSIZE); return 0; } RETURN_ERROR(prefix_unknown, ""); } /* ensure there is enough `srcSize` to fully read/decode frame header */ { size_t const fhsize = ZSTD_frameHeaderSize_internal(src, srcSize, format); if (srcSize < fhsize) return fhsize; zfhPtr->headerSize = (U32)fhsize; } { BYTE const fhdByte = ip[minInputSize-1]; size_t pos = minInputSize; U32 const dictIDSizeCode = fhdByte&3; U32 const checksumFlag = (fhdByte>>2)&1; U32 const singleSegment = (fhdByte>>5)&1; U32 const fcsID = fhdByte>>6; U64 windowSize = 0; U32 dictID = 0; U64 frameContentSize = ZSTD_CONTENTSIZE_UNKNOWN; RETURN_ERROR_IF((fhdByte & 0x08) != 0, frameParameter_unsupported, "reserved bits, must be zero"); if (!singleSegment) { BYTE const wlByte = ip[pos++]; U32 const windowLog = (wlByte >> 3) + ZSTD_WINDOWLOG_ABSOLUTEMIN; RETURN_ERROR_IF(windowLog > ZSTD_WINDOWLOG_MAX, frameParameter_windowTooLarge, ""); windowSize = (1ULL << windowLog); windowSize += (windowSize >> 3) * (wlByte&7); } switch(dictIDSizeCode) { default: assert(0); /* impossible */ ZSTD_FALLTHROUGH; case 0 : break; case 1 : dictID = ip[pos]; pos++; break; case 2 : dictID = MEM_readLE16(ip+pos); pos+=2; break; case 3 : dictID = MEM_readLE32(ip+pos); pos+=4; break; } switch(fcsID) { default: assert(0); /* impossible */ ZSTD_FALLTHROUGH; case 0 : if (singleSegment) frameContentSize = ip[pos]; break; case 1 : frameContentSize = MEM_readLE16(ip+pos)+256; break; case 2 : frameContentSize = MEM_readLE32(ip+pos); break; case 3 : frameContentSize = MEM_readLE64(ip+pos); break; } if (singleSegment) windowSize = frameContentSize; zfhPtr->frameType = ZSTD_frame; zfhPtr->frameContentSize = frameContentSize; zfhPtr->windowSize = windowSize; zfhPtr->blockSizeMax = (unsigned) MIN(windowSize, ZSTD_BLOCKSIZE_MAX); zfhPtr->dictID = dictID; zfhPtr->checksumFlag = checksumFlag; } return 0; } /* ZSTD_getFrameHeader() : * decode Frame Header, or require larger `srcSize`. * note : this function does not consume input, it only reads it. * @return : 0, `zfhPtr` is correctly filled, * >0, `srcSize` is too small, value is wanted `srcSize` amount, * or an error code, which can be tested using ZSTD_isError() */ size_t ZSTD_getFrameHeader(ZSTD_FrameHeader* zfhPtr, const void* src, size_t srcSize) { return ZSTD_getFrameHeader_advanced(zfhPtr, src, srcSize, ZSTD_f_zstd1); } /* ZSTD_getFrameContentSize() : * compatible with legacy mode * @return : decompressed size of the single frame pointed to be `src` if known, otherwise * - ZSTD_CONTENTSIZE_UNKNOWN if the size cannot be determined * - ZSTD_CONTENTSIZE_ERROR if an error occurred (e.g. invalid magic number, srcSize too small) */ unsigned long long ZSTD_getFrameContentSize(const void *src, size_t srcSize) { { ZSTD_FrameHeader zfh; if (ZSTD_getFrameHeader(&zfh, src, srcSize) != 0) return ZSTD_CONTENTSIZE_ERROR; if (zfh.frameType == ZSTD_skippableFrame) { return 0; } else { return zfh.frameContentSize; } } } static size_t readSkippableFrameSize(void const* src, size_t srcSize) { size_t const skippableHeaderSize = ZSTD_SKIPPABLEHEADERSIZE; U32 sizeU32; RETURN_ERROR_IF(srcSize < ZSTD_SKIPPABLEHEADERSIZE, srcSize_wrong, ""); sizeU32 = MEM_readLE32((BYTE const*)src + ZSTD_FRAMEIDSIZE); RETURN_ERROR_IF((U32)(sizeU32 + ZSTD_SKIPPABLEHEADERSIZE) < sizeU32, frameParameter_unsupported, ""); { size_t const skippableSize = skippableHeaderSize + sizeU32; RETURN_ERROR_IF(skippableSize > srcSize, srcSize_wrong, ""); return skippableSize; } } /*! ZSTD_readSkippableFrame() : * Retrieves content of a skippable frame, and writes it to dst buffer. * * The parameter magicVariant will receive the magicVariant that was supplied when the frame was written, * i.e. magicNumber - ZSTD_MAGIC_SKIPPABLE_START. This can be NULL if the caller is not interested * in the magicVariant. * * Returns an error if destination buffer is not large enough, or if this is not a valid skippable frame. * * @return : number of bytes written or a ZSTD error. */ size_t ZSTD_readSkippableFrame(void* dst, size_t dstCapacity, unsigned* magicVariant, /* optional, can be NULL */ const void* src, size_t srcSize) { RETURN_ERROR_IF(srcSize < ZSTD_SKIPPABLEHEADERSIZE, srcSize_wrong, ""); { U32 const magicNumber = MEM_readLE32(src); size_t skippableFrameSize = readSkippableFrameSize(src, srcSize); size_t skippableContentSize = skippableFrameSize - ZSTD_SKIPPABLEHEADERSIZE; /* check input validity */ RETURN_ERROR_IF(!ZSTD_isSkippableFrame(src, srcSize), frameParameter_unsupported, ""); RETURN_ERROR_IF(skippableFrameSize < ZSTD_SKIPPABLEHEADERSIZE || skippableFrameSize > srcSize, srcSize_wrong, ""); RETURN_ERROR_IF(skippableContentSize > dstCapacity, dstSize_tooSmall, ""); /* deliver payload */ if (skippableContentSize > 0 && dst != NULL) ZSTD_memcpy(dst, (const BYTE *)src + ZSTD_SKIPPABLEHEADERSIZE, skippableContentSize); if (magicVariant != NULL) *magicVariant = magicNumber - ZSTD_MAGIC_SKIPPABLE_START; return skippableContentSize; } } /* ZSTD_findDecompressedSize() : * `srcSize` must be the exact length of some number of ZSTD compressed and/or * skippable frames * note: compatible with legacy mode * @return : decompressed size of the frames contained */ unsigned long long ZSTD_findDecompressedSize(const void* src, size_t srcSize) { unsigned long long totalDstSize = 0; while (srcSize >= ZSTD_startingInputLength(ZSTD_f_zstd1)) { U32 const magicNumber = MEM_readLE32(src); if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { size_t const skippableSize = readSkippableFrameSize(src, srcSize); if (ZSTD_isError(skippableSize)) return ZSTD_CONTENTSIZE_ERROR; assert(skippableSize <= srcSize); src = (const BYTE *)src + skippableSize; srcSize -= skippableSize; continue; } { unsigned long long const fcs = ZSTD_getFrameContentSize(src, srcSize); if (fcs >= ZSTD_CONTENTSIZE_ERROR) return fcs; if (totalDstSize + fcs < totalDstSize) return ZSTD_CONTENTSIZE_ERROR; /* check for overflow */ totalDstSize += fcs; } /* skip to next frame */ { size_t const frameSrcSize = ZSTD_findFrameCompressedSize(src, srcSize); if (ZSTD_isError(frameSrcSize)) return ZSTD_CONTENTSIZE_ERROR; assert(frameSrcSize <= srcSize); src = (const BYTE *)src + frameSrcSize; srcSize -= frameSrcSize; } } /* while (srcSize >= ZSTD_frameHeaderSize_prefix) */ if (srcSize) return ZSTD_CONTENTSIZE_ERROR; return totalDstSize; } /* ZSTD_getDecompressedSize() : * compatible with legacy mode * @return : decompressed size if known, 0 otherwise note : 0 can mean any of the following : - frame content is empty - decompressed size field is not present in frame header - frame header unknown / not supported - frame header not complete (`srcSize` too small) */ unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize) { unsigned long long const ret = ZSTD_getFrameContentSize(src, srcSize); ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_ERROR < ZSTD_CONTENTSIZE_UNKNOWN); return (ret >= ZSTD_CONTENTSIZE_ERROR) ? 0 : ret; } /* ZSTD_decodeFrameHeader() : * `headerSize` must be the size provided by ZSTD_frameHeaderSize(). * If multiple DDict references are enabled, also will choose the correct DDict to use. * @return : 0 if success, or an error code, which can be tested using ZSTD_isError() */ static size_t ZSTD_decodeFrameHeader(ZSTD_DCtx* dctx, const void* src, size_t headerSize) { size_t const result = ZSTD_getFrameHeader_advanced(&(dctx->fParams), src, headerSize, dctx->format); if (ZSTD_isError(result)) return result; /* invalid header */ RETURN_ERROR_IF(result>0, srcSize_wrong, "headerSize too small"); /* Reference DDict requested by frame if dctx references multiple ddicts */ if (dctx->refMultipleDDicts == ZSTD_rmd_refMultipleDDicts && dctx->ddictSet) { ZSTD_DCtx_selectFrameDDict(dctx); } #ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION /* Skip the dictID check in fuzzing mode, because it makes the search * harder. */ RETURN_ERROR_IF(dctx->fParams.dictID && (dctx->dictID != dctx->fParams.dictID), dictionary_wrong, ""); #endif dctx->validateChecksum = (dctx->fParams.checksumFlag && !dctx->forceIgnoreChecksum) ? 1 : 0; if (dctx->validateChecksum) xxh64_reset(&dctx->xxhState, 0); dctx->processedCSize += headerSize; return 0; } static ZSTD_frameSizeInfo ZSTD_errorFrameSizeInfo(size_t ret) { ZSTD_frameSizeInfo frameSizeInfo; frameSizeInfo.compressedSize = ret; frameSizeInfo.decompressedBound = ZSTD_CONTENTSIZE_ERROR; return frameSizeInfo; } static ZSTD_frameSizeInfo ZSTD_findFrameSizeInfo(const void* src, size_t srcSize, ZSTD_format_e format) { ZSTD_frameSizeInfo frameSizeInfo; ZSTD_memset(&frameSizeInfo, 0, sizeof(ZSTD_frameSizeInfo)); if (format == ZSTD_f_zstd1 && (srcSize >= ZSTD_SKIPPABLEHEADERSIZE) && (MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { frameSizeInfo.compressedSize = readSkippableFrameSize(src, srcSize); assert(ZSTD_isError(frameSizeInfo.compressedSize) || frameSizeInfo.compressedSize <= srcSize); return frameSizeInfo; } else { const BYTE* ip = (const BYTE*)src; const BYTE* const ipstart = ip; size_t remainingSize = srcSize; size_t nbBlocks = 0; ZSTD_FrameHeader zfh; /* Extract Frame Header */ { size_t const ret = ZSTD_getFrameHeader_advanced(&zfh, src, srcSize, format); if (ZSTD_isError(ret)) return ZSTD_errorFrameSizeInfo(ret); if (ret > 0) return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong)); } ip += zfh.headerSize; remainingSize -= zfh.headerSize; /* Iterate over each block */ while (1) { blockProperties_t blockProperties; size_t const cBlockSize = ZSTD_getcBlockSize(ip, remainingSize, &blockProperties); if (ZSTD_isError(cBlockSize)) return ZSTD_errorFrameSizeInfo(cBlockSize); if (ZSTD_blockHeaderSize + cBlockSize > remainingSize) return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong)); ip += ZSTD_blockHeaderSize + cBlockSize; remainingSize -= ZSTD_blockHeaderSize + cBlockSize; nbBlocks++; if (blockProperties.lastBlock) break; } /* Final frame content checksum */ if (zfh.checksumFlag) { if (remainingSize < 4) return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong)); ip += 4; } frameSizeInfo.nbBlocks = nbBlocks; frameSizeInfo.compressedSize = (size_t)(ip - ipstart); frameSizeInfo.decompressedBound = (zfh.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN) ? zfh.frameContentSize : (unsigned long long)nbBlocks * zfh.blockSizeMax; return frameSizeInfo; } } static size_t ZSTD_findFrameCompressedSize_advanced(const void *src, size_t srcSize, ZSTD_format_e format) { ZSTD_frameSizeInfo const frameSizeInfo = ZSTD_findFrameSizeInfo(src, srcSize, format); return frameSizeInfo.compressedSize; } /* ZSTD_findFrameCompressedSize() : * See docs in zstd.h * Note: compatible with legacy mode */ size_t ZSTD_findFrameCompressedSize(const void *src, size_t srcSize) { return ZSTD_findFrameCompressedSize_advanced(src, srcSize, ZSTD_f_zstd1); } /* ZSTD_decompressBound() : * compatible with legacy mode * `src` must point to the start of a ZSTD frame or a skippable frame * `srcSize` must be at least as large as the frame contained * @return : the maximum decompressed size of the compressed source */ unsigned long long ZSTD_decompressBound(const void* src, size_t srcSize) { unsigned long long bound = 0; /* Iterate over each frame */ while (srcSize > 0) { ZSTD_frameSizeInfo const frameSizeInfo = ZSTD_findFrameSizeInfo(src, srcSize, ZSTD_f_zstd1); size_t const compressedSize = frameSizeInfo.compressedSize; unsigned long long const decompressedBound = frameSizeInfo.decompressedBound; if (ZSTD_isError(compressedSize) || decompressedBound == ZSTD_CONTENTSIZE_ERROR) return ZSTD_CONTENTSIZE_ERROR; assert(srcSize >= compressedSize); src = (const BYTE*)src + compressedSize; srcSize -= compressedSize; bound += decompressedBound; } return bound; } size_t ZSTD_decompressionMargin(void const* src, size_t srcSize) { size_t margin = 0; unsigned maxBlockSize = 0; /* Iterate over each frame */ while (srcSize > 0) { ZSTD_frameSizeInfo const frameSizeInfo = ZSTD_findFrameSizeInfo(src, srcSize, ZSTD_f_zstd1); size_t const compressedSize = frameSizeInfo.compressedSize; unsigned long long const decompressedBound = frameSizeInfo.decompressedBound; ZSTD_FrameHeader zfh; FORWARD_IF_ERROR(ZSTD_getFrameHeader(&zfh, src, srcSize), ""); if (ZSTD_isError(compressedSize) || decompressedBound == ZSTD_CONTENTSIZE_ERROR) return ERROR(corruption_detected); if (zfh.frameType == ZSTD_frame) { /* Add the frame header to our margin */ margin += zfh.headerSize; /* Add the checksum to our margin */ margin += zfh.checksumFlag ? 4 : 0; /* Add 3 bytes per block */ margin += 3 * frameSizeInfo.nbBlocks; /* Compute the max block size */ maxBlockSize = MAX(maxBlockSize, zfh.blockSizeMax); } else { assert(zfh.frameType == ZSTD_skippableFrame); /* Add the entire skippable frame size to our margin. */ margin += compressedSize; } assert(srcSize >= compressedSize); src = (const BYTE*)src + compressedSize; srcSize -= compressedSize; } /* Add the max block size back to the margin. */ margin += maxBlockSize; return margin; } /*-************************************************************* * Frame decoding ***************************************************************/ /* ZSTD_insertBlock() : * insert `src` block into `dctx` history. Useful to track uncompressed blocks. */ size_t ZSTD_insertBlock(ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize) { DEBUGLOG(5, "ZSTD_insertBlock: %u bytes", (unsigned)blockSize); ZSTD_checkContinuity(dctx, blockStart, blockSize); dctx->previousDstEnd = (const char*)blockStart + blockSize; return blockSize; } static size_t ZSTD_copyRawBlock(void* dst, size_t dstCapacity, const void* src, size_t srcSize) { DEBUGLOG(5, "ZSTD_copyRawBlock"); RETURN_ERROR_IF(srcSize > dstCapacity, dstSize_tooSmall, ""); if (dst == NULL) { if (srcSize == 0) return 0; RETURN_ERROR(dstBuffer_null, ""); } ZSTD_memmove(dst, src, srcSize); return srcSize; } static size_t ZSTD_setRleBlock(void* dst, size_t dstCapacity, BYTE b, size_t regenSize) { RETURN_ERROR_IF(regenSize > dstCapacity, dstSize_tooSmall, ""); if (dst == NULL) { if (regenSize == 0) return 0; RETURN_ERROR(dstBuffer_null, ""); } ZSTD_memset(dst, b, regenSize); return regenSize; } static void ZSTD_DCtx_trace_end(ZSTD_DCtx const* dctx, U64 uncompressedSize, U64 compressedSize, int streaming) { (void)dctx; (void)uncompressedSize; (void)compressedSize; (void)streaming; } /*! ZSTD_decompressFrame() : * @dctx must be properly initialized * will update *srcPtr and *srcSizePtr, * to make *srcPtr progress by one frame. */ static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void** srcPtr, size_t *srcSizePtr) { const BYTE* const istart = (const BYTE*)(*srcPtr); const BYTE* ip = istart; BYTE* const ostart = (BYTE*)dst; BYTE* const oend = dstCapacity != 0 ? ostart + dstCapacity : ostart; BYTE* op = ostart; size_t remainingSrcSize = *srcSizePtr; DEBUGLOG(4, "ZSTD_decompressFrame (srcSize:%i)", (int)*srcSizePtr); /* check */ RETURN_ERROR_IF( remainingSrcSize < ZSTD_FRAMEHEADERSIZE_MIN(dctx->format)+ZSTD_blockHeaderSize, srcSize_wrong, ""); /* Frame Header */ { size_t const frameHeaderSize = ZSTD_frameHeaderSize_internal( ip, ZSTD_FRAMEHEADERSIZE_PREFIX(dctx->format), dctx->format); if (ZSTD_isError(frameHeaderSize)) return frameHeaderSize; RETURN_ERROR_IF(remainingSrcSize < frameHeaderSize+ZSTD_blockHeaderSize, srcSize_wrong, ""); FORWARD_IF_ERROR( ZSTD_decodeFrameHeader(dctx, ip, frameHeaderSize) , ""); ip += frameHeaderSize; remainingSrcSize -= frameHeaderSize; } /* Shrink the blockSizeMax if enabled */ if (dctx->maxBlockSizeParam != 0) dctx->fParams.blockSizeMax = MIN(dctx->fParams.blockSizeMax, (unsigned)dctx->maxBlockSizeParam); /* Loop on each block */ while (1) { BYTE* oBlockEnd = oend; size_t decodedSize; blockProperties_t blockProperties; size_t const cBlockSize = ZSTD_getcBlockSize(ip, remainingSrcSize, &blockProperties); if (ZSTD_isError(cBlockSize)) return cBlockSize; ip += ZSTD_blockHeaderSize; remainingSrcSize -= ZSTD_blockHeaderSize; RETURN_ERROR_IF(cBlockSize > remainingSrcSize, srcSize_wrong, ""); if (ip >= op && ip < oBlockEnd) { /* We are decompressing in-place. Limit the output pointer so that we * don't overwrite the block that we are currently reading. This will * fail decompression if the input & output pointers aren't spaced * far enough apart. * * This is important to set, even when the pointers are far enough * apart, because ZSTD_decompressBlock_internal() can decide to store * literals in the output buffer, after the block it is decompressing. * Since we don't want anything to overwrite our input, we have to tell * ZSTD_decompressBlock_internal to never write past ip. * * See ZSTD_allocateLiteralsBuffer() for reference. */ oBlockEnd = op + (ip - op); } switch(blockProperties.blockType) { case bt_compressed: assert(dctx->isFrameDecompression == 1); decodedSize = ZSTD_decompressBlock_internal(dctx, op, (size_t)(oBlockEnd-op), ip, cBlockSize, not_streaming); break; case bt_raw : /* Use oend instead of oBlockEnd because this function is safe to overlap. It uses memmove. */ decodedSize = ZSTD_copyRawBlock(op, (size_t)(oend-op), ip, cBlockSize); break; case bt_rle : decodedSize = ZSTD_setRleBlock(op, (size_t)(oBlockEnd-op), *ip, blockProperties.origSize); break; case bt_reserved : default: RETURN_ERROR(corruption_detected, "invalid block type"); } FORWARD_IF_ERROR(decodedSize, "Block decompression failure"); DEBUGLOG(5, "Decompressed block of dSize = %u", (unsigned)decodedSize); if (dctx->validateChecksum) { xxh64_update(&dctx->xxhState, op, decodedSize); } if (decodedSize) /* support dst = NULL,0 */ { op += decodedSize; } assert(ip != NULL); ip += cBlockSize; remainingSrcSize -= cBlockSize; if (blockProperties.lastBlock) break; } if (dctx->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN) { RETURN_ERROR_IF((U64)(op-ostart) != dctx->fParams.frameContentSize, corruption_detected, ""); } if (dctx->fParams.checksumFlag) { /* Frame content checksum verification */ RETURN_ERROR_IF(remainingSrcSize<4, checksum_wrong, ""); if (!dctx->forceIgnoreChecksum) { U32 const checkCalc = (U32)xxh64_digest(&dctx->xxhState); U32 checkRead; checkRead = MEM_readLE32(ip); RETURN_ERROR_IF(checkRead != checkCalc, checksum_wrong, ""); } ip += 4; remainingSrcSize -= 4; } ZSTD_DCtx_trace_end(dctx, (U64)(op-ostart), (U64)(ip-istart), /* streaming */ 0); /* Allow caller to get size read */ DEBUGLOG(4, "ZSTD_decompressFrame: decompressed frame of size %i, consuming %i bytes of input", (int)(op-ostart), (int)(ip - (const BYTE*)*srcPtr)); *srcPtr = ip; *srcSizePtr = remainingSrcSize; return (size_t)(op-ostart); } static ZSTD_ALLOW_POINTER_OVERFLOW_ATTR size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, const void* dict, size_t dictSize, const ZSTD_DDict* ddict) { void* const dststart = dst; int moreThan1Frame = 0; DEBUGLOG(5, "ZSTD_decompressMultiFrame"); assert(dict==NULL || ddict==NULL); /* either dict or ddict set, not both */ if (ddict) { dict = ZSTD_DDict_dictContent(ddict); dictSize = ZSTD_DDict_dictSize(ddict); } while (srcSize >= ZSTD_startingInputLength(dctx->format)) { if (dctx->format == ZSTD_f_zstd1 && srcSize >= 4) { U32 const magicNumber = MEM_readLE32(src); DEBUGLOG(5, "reading magic number %08X", (unsigned)magicNumber); if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame detected : skip it */ size_t const skippableSize = readSkippableFrameSize(src, srcSize); FORWARD_IF_ERROR(skippableSize, "invalid skippable frame"); assert(skippableSize <= srcSize); src = (const BYTE *)src + skippableSize; srcSize -= skippableSize; continue; /* check next frame */ } } if (ddict) { /* we were called from ZSTD_decompress_usingDDict */ FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDDict(dctx, ddict), ""); } else { /* this will initialize correctly with no dict if dict == NULL, so * use this in all cases but ddict */ FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDict(dctx, dict, dictSize), ""); } ZSTD_checkContinuity(dctx, dst, dstCapacity); { const size_t res = ZSTD_decompressFrame(dctx, dst, dstCapacity, &src, &srcSize); RETURN_ERROR_IF( (ZSTD_getErrorCode(res) == ZSTD_error_prefix_unknown) && (moreThan1Frame==1), srcSize_wrong, "At least one frame successfully completed, " "but following bytes are garbage: " "it's more likely to be a srcSize error, " "specifying more input bytes than size of frame(s). " "Note: one could be unlucky, it might be a corruption error instead, " "happening right at the place where we expect zstd magic bytes. " "But this is _much_ less likely than a srcSize field error."); if (ZSTD_isError(res)) return res; assert(res <= dstCapacity); if (res != 0) dst = (BYTE*)dst + res; dstCapacity -= res; } moreThan1Frame = 1; } /* while (srcSize >= ZSTD_frameHeaderSize_prefix) */ RETURN_ERROR_IF(srcSize, srcSize_wrong, "input not entirely consumed"); return (size_t)((BYTE*)dst - (BYTE*)dststart); } size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, const void* dict, size_t dictSize) { return ZSTD_decompressMultiFrame(dctx, dst, dstCapacity, src, srcSize, dict, dictSize, NULL); } static ZSTD_DDict const* ZSTD_getDDict(ZSTD_DCtx* dctx) { switch (dctx->dictUses) { default: assert(0 /* Impossible */); ZSTD_FALLTHROUGH; case ZSTD_dont_use: ZSTD_clearDict(dctx); return NULL; case ZSTD_use_indefinitely: return dctx->ddict; case ZSTD_use_once: dctx->dictUses = ZSTD_dont_use; return dctx->ddict; } } size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize) { return ZSTD_decompress_usingDDict(dctx, dst, dstCapacity, src, srcSize, ZSTD_getDDict(dctx)); } size_t ZSTD_decompress(void* dst, size_t dstCapacity, const void* src, size_t srcSize) { #if defined(ZSTD_HEAPMODE) && (ZSTD_HEAPMODE>=1) size_t regenSize; ZSTD_DCtx* const dctx = ZSTD_createDCtx_internal(ZSTD_defaultCMem); RETURN_ERROR_IF(dctx==NULL, memory_allocation, "NULL pointer!"); regenSize = ZSTD_decompressDCtx(dctx, dst, dstCapacity, src, srcSize); ZSTD_freeDCtx(dctx); return regenSize; #else /* stack mode */ ZSTD_DCtx dctx; ZSTD_initDCtx_internal(&dctx); return ZSTD_decompressDCtx(&dctx, dst, dstCapacity, src, srcSize); #endif } /*-************************************** * Advanced Streaming Decompression API * Bufferless and synchronous ****************************************/ size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx) { return dctx->expected; } /* * Similar to ZSTD_nextSrcSizeToDecompress(), but when a block input can be streamed, we * allow taking a partial block as the input. Currently only raw uncompressed blocks can * be streamed. * * For blocks that can be streamed, this allows us to reduce the latency until we produce * output, and avoid copying the input. * * @param inputSize - The total amount of input that the caller currently has. */ static size_t ZSTD_nextSrcSizeToDecompressWithInputSize(ZSTD_DCtx* dctx, size_t inputSize) { if (!(dctx->stage == ZSTDds_decompressBlock || dctx->stage == ZSTDds_decompressLastBlock)) return dctx->expected; if (dctx->bType != bt_raw) return dctx->expected; return BOUNDED(1, inputSize, dctx->expected); } ZSTD_nextInputType_e ZSTD_nextInputType(ZSTD_DCtx* dctx) { switch(dctx->stage) { default: /* should not happen */ assert(0); ZSTD_FALLTHROUGH; case ZSTDds_getFrameHeaderSize: ZSTD_FALLTHROUGH; case ZSTDds_decodeFrameHeader: return ZSTDnit_frameHeader; case ZSTDds_decodeBlockHeader: return ZSTDnit_blockHeader; case ZSTDds_decompressBlock: return ZSTDnit_block; case ZSTDds_decompressLastBlock: return ZSTDnit_lastBlock; case ZSTDds_checkChecksum: return ZSTDnit_checksum; case ZSTDds_decodeSkippableHeader: ZSTD_FALLTHROUGH; case ZSTDds_skipFrame: return ZSTDnit_skippableFrame; } } static int ZSTD_isSkipFrame(ZSTD_DCtx* dctx) { return dctx->stage == ZSTDds_skipFrame; } /* ZSTD_decompressContinue() : * srcSize : must be the exact nb of bytes expected (see ZSTD_nextSrcSizeToDecompress()) * @return : nb of bytes generated into `dst` (necessarily <= `dstCapacity) * or an error code, which can be tested using ZSTD_isError() */ size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize) { DEBUGLOG(5, "ZSTD_decompressContinue (srcSize:%u)", (unsigned)srcSize); /* Sanity check */ RETURN_ERROR_IF(srcSize != ZSTD_nextSrcSizeToDecompressWithInputSize(dctx, srcSize), srcSize_wrong, "not allowed"); ZSTD_checkContinuity(dctx, dst, dstCapacity); dctx->processedCSize += srcSize; switch (dctx->stage) { case ZSTDds_getFrameHeaderSize : assert(src != NULL); if (dctx->format == ZSTD_f_zstd1) { /* allows header */ assert(srcSize >= ZSTD_FRAMEIDSIZE); /* to read skippable magic number */ if ((MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */ ZSTD_memcpy(dctx->headerBuffer, src, srcSize); dctx->expected = ZSTD_SKIPPABLEHEADERSIZE - srcSize; /* remaining to load to get full skippable frame header */ dctx->stage = ZSTDds_decodeSkippableHeader; return 0; } } dctx->headerSize = ZSTD_frameHeaderSize_internal(src, srcSize, dctx->format); if (ZSTD_isError(dctx->headerSize)) return dctx->headerSize; ZSTD_memcpy(dctx->headerBuffer, src, srcSize); dctx->expected = dctx->headerSize - srcSize; dctx->stage = ZSTDds_decodeFrameHeader; return 0; case ZSTDds_decodeFrameHeader: assert(src != NULL); ZSTD_memcpy(dctx->headerBuffer + (dctx->headerSize - srcSize), src, srcSize); FORWARD_IF_ERROR(ZSTD_decodeFrameHeader(dctx, dctx->headerBuffer, dctx->headerSize), ""); dctx->expected = ZSTD_blockHeaderSize; dctx->stage = ZSTDds_decodeBlockHeader; return 0; case ZSTDds_decodeBlockHeader: { blockProperties_t bp; size_t const cBlockSize = ZSTD_getcBlockSize(src, ZSTD_blockHeaderSize, &bp); if (ZSTD_isError(cBlockSize)) return cBlockSize; RETURN_ERROR_IF(cBlockSize > dctx->fParams.blockSizeMax, corruption_detected, "Block Size Exceeds Maximum"); dctx->expected = cBlockSize; dctx->bType = bp.blockType; dctx->rleSize = bp.origSize; if (cBlockSize) { dctx->stage = bp.lastBlock ? ZSTDds_decompressLastBlock : ZSTDds_decompressBlock; return 0; } /* empty block */ if (bp.lastBlock) { if (dctx->fParams.checksumFlag) { dctx->expected = 4; dctx->stage = ZSTDds_checkChecksum; } else { dctx->expected = 0; /* end of frame */ dctx->stage = ZSTDds_getFrameHeaderSize; } } else { dctx->expected = ZSTD_blockHeaderSize; /* jump to next header */ dctx->stage = ZSTDds_decodeBlockHeader; } return 0; } case ZSTDds_decompressLastBlock: case ZSTDds_decompressBlock: DEBUGLOG(5, "ZSTD_decompressContinue: case ZSTDds_decompressBlock"); { size_t rSize; switch(dctx->bType) { case bt_compressed: DEBUGLOG(5, "ZSTD_decompressContinue: case bt_compressed"); assert(dctx->isFrameDecompression == 1); rSize = ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize, is_streaming); dctx->expected = 0; /* Streaming not supported */ break; case bt_raw : assert(srcSize <= dctx->expected); rSize = ZSTD_copyRawBlock(dst, dstCapacity, src, srcSize); FORWARD_IF_ERROR(rSize, "ZSTD_copyRawBlock failed"); assert(rSize == srcSize); dctx->expected -= rSize; break; case bt_rle : rSize = ZSTD_setRleBlock(dst, dstCapacity, *(const BYTE*)src, dctx->rleSize); dctx->expected = 0; /* Streaming not supported */ break; case bt_reserved : /* should never happen */ default: RETURN_ERROR(corruption_detected, "invalid block type"); } FORWARD_IF_ERROR(rSize, ""); RETURN_ERROR_IF(rSize > dctx->fParams.blockSizeMax, corruption_detected, "Decompressed Block Size Exceeds Maximum"); DEBUGLOG(5, "ZSTD_decompressContinue: decoded size from block : %u", (unsigned)rSize); dctx->decodedSize += rSize; if (dctx->validateChecksum) xxh64_update(&dctx->xxhState, dst, rSize); dctx->previousDstEnd = (char*)dst + rSize; /* Stay on the same stage until we are finished streaming the block. */ if (dctx->expected > 0) { return rSize; } if (dctx->stage == ZSTDds_decompressLastBlock) { /* end of frame */ DEBUGLOG(4, "ZSTD_decompressContinue: decoded size from frame : %u", (unsigned)dctx->decodedSize); RETURN_ERROR_IF( dctx->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN && dctx->decodedSize != dctx->fParams.frameContentSize, corruption_detected, ""); if (dctx->fParams.checksumFlag) { /* another round for frame checksum */ dctx->expected = 4; dctx->stage = ZSTDds_checkChecksum; } else { ZSTD_DCtx_trace_end(dctx, dctx->decodedSize, dctx->processedCSize, /* streaming */ 1); dctx->expected = 0; /* ends here */ dctx->stage = ZSTDds_getFrameHeaderSize; } } else { dctx->stage = ZSTDds_decodeBlockHeader; dctx->expected = ZSTD_blockHeaderSize; } return rSize; } case ZSTDds_checkChecksum: assert(srcSize == 4); /* guaranteed by dctx->expected */ { if (dctx->validateChecksum) { U32 const h32 = (U32)xxh64_digest(&dctx->xxhState); U32 const check32 = MEM_readLE32(src); DEBUGLOG(4, "ZSTD_decompressContinue: checksum : calculated %08X :: %08X read", (unsigned)h32, (unsigned)check32); RETURN_ERROR_IF(check32 != h32, checksum_wrong, ""); } ZSTD_DCtx_trace_end(dctx, dctx->decodedSize, dctx->processedCSize, /* streaming */ 1); dctx->expected = 0; dctx->stage = ZSTDds_getFrameHeaderSize; return 0; } case ZSTDds_decodeSkippableHeader: assert(src != NULL); assert(srcSize <= ZSTD_SKIPPABLEHEADERSIZE); assert(dctx->format != ZSTD_f_zstd1_magicless); ZSTD_memcpy(dctx->headerBuffer + (ZSTD_SKIPPABLEHEADERSIZE - srcSize), src, srcSize); /* complete skippable header */ dctx->expected = MEM_readLE32(dctx->headerBuffer + ZSTD_FRAMEIDSIZE); /* note : dctx->expected can grow seriously large, beyond local buffer size */ dctx->stage = ZSTDds_skipFrame; return 0; case ZSTDds_skipFrame: dctx->expected = 0; dctx->stage = ZSTDds_getFrameHeaderSize; return 0; default: assert(0); /* impossible */ RETURN_ERROR(GENERIC, "impossible to reach"); /* some compilers require default to do something */ } } static size_t ZSTD_refDictContent(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) { dctx->dictEnd = dctx->previousDstEnd; dctx->virtualStart = (const char*)dict - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->prefixStart)); dctx->prefixStart = dict; dctx->previousDstEnd = (const char*)dict + dictSize; #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION dctx->dictContentBeginForFuzzing = dctx->prefixStart; dctx->dictContentEndForFuzzing = dctx->previousDstEnd; #endif return 0; } /*! ZSTD_loadDEntropy() : * dict : must point at beginning of a valid zstd dictionary. * @return : size of entropy tables read */ size_t ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy, const void* const dict, size_t const dictSize) { const BYTE* dictPtr = (const BYTE*)dict; const BYTE* const dictEnd = dictPtr + dictSize; RETURN_ERROR_IF(dictSize <= 8, dictionary_corrupted, "dict is too small"); assert(MEM_readLE32(dict) == ZSTD_MAGIC_DICTIONARY); /* dict must be valid */ dictPtr += 8; /* skip header = magic + dictID */ ZSTD_STATIC_ASSERT(offsetof(ZSTD_entropyDTables_t, OFTable) == offsetof(ZSTD_entropyDTables_t, LLTable) + sizeof(entropy->LLTable)); ZSTD_STATIC_ASSERT(offsetof(ZSTD_entropyDTables_t, MLTable) == offsetof(ZSTD_entropyDTables_t, OFTable) + sizeof(entropy->OFTable)); ZSTD_STATIC_ASSERT(sizeof(entropy->LLTable) + sizeof(entropy->OFTable) + sizeof(entropy->MLTable) >= HUF_DECOMPRESS_WORKSPACE_SIZE); { void* const workspace = &entropy->LLTable; /* use fse tables as temporary workspace; implies fse tables are grouped together */ size_t const workspaceSize = sizeof(entropy->LLTable) + sizeof(entropy->OFTable) + sizeof(entropy->MLTable); #ifdef HUF_FORCE_DECOMPRESS_X1 /* in minimal huffman, we always use X1 variants */ size_t const hSize = HUF_readDTableX1_wksp(entropy->hufTable, dictPtr, dictEnd - dictPtr, workspace, workspaceSize, /* flags */ 0); #else size_t const hSize = HUF_readDTableX2_wksp(entropy->hufTable, dictPtr, (size_t)(dictEnd - dictPtr), workspace, workspaceSize, /* flags */ 0); #endif RETURN_ERROR_IF(HUF_isError(hSize), dictionary_corrupted, ""); dictPtr += hSize; } { short offcodeNCount[MaxOff+1]; unsigned offcodeMaxValue = MaxOff, offcodeLog; size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, (size_t)(dictEnd-dictPtr)); RETURN_ERROR_IF(FSE_isError(offcodeHeaderSize), dictionary_corrupted, ""); RETURN_ERROR_IF(offcodeMaxValue > MaxOff, dictionary_corrupted, ""); RETURN_ERROR_IF(offcodeLog > OffFSELog, dictionary_corrupted, ""); ZSTD_buildFSETable( entropy->OFTable, offcodeNCount, offcodeMaxValue, OF_base, OF_bits, offcodeLog, entropy->workspace, sizeof(entropy->workspace), /* bmi2 */0); dictPtr += offcodeHeaderSize; } { short matchlengthNCount[MaxML+1]; unsigned matchlengthMaxValue = MaxML, matchlengthLog; size_t const matchlengthHeaderSize = FSE_readNCount(matchlengthNCount, &matchlengthMaxValue, &matchlengthLog, dictPtr, (size_t)(dictEnd-dictPtr)); RETURN_ERROR_IF(FSE_isError(matchlengthHeaderSize), dictionary_corrupted, ""); RETURN_ERROR_IF(matchlengthMaxValue > MaxML, dictionary_corrupted, ""); RETURN_ERROR_IF(matchlengthLog > MLFSELog, dictionary_corrupted, ""); ZSTD_buildFSETable( entropy->MLTable, matchlengthNCount, matchlengthMaxValue, ML_base, ML_bits, matchlengthLog, entropy->workspace, sizeof(entropy->workspace), /* bmi2 */ 0); dictPtr += matchlengthHeaderSize; } { short litlengthNCount[MaxLL+1]; unsigned litlengthMaxValue = MaxLL, litlengthLog; size_t const litlengthHeaderSize = FSE_readNCount(litlengthNCount, &litlengthMaxValue, &litlengthLog, dictPtr, (size_t)(dictEnd-dictPtr)); RETURN_ERROR_IF(FSE_isError(litlengthHeaderSize), dictionary_corrupted, ""); RETURN_ERROR_IF(litlengthMaxValue > MaxLL, dictionary_corrupted, ""); RETURN_ERROR_IF(litlengthLog > LLFSELog, dictionary_corrupted, ""); ZSTD_buildFSETable( entropy->LLTable, litlengthNCount, litlengthMaxValue, LL_base, LL_bits, litlengthLog, entropy->workspace, sizeof(entropy->workspace), /* bmi2 */ 0); dictPtr += litlengthHeaderSize; } RETURN_ERROR_IF(dictPtr+12 > dictEnd, dictionary_corrupted, ""); { int i; size_t const dictContentSize = (size_t)(dictEnd - (dictPtr+12)); for (i=0; i<3; i++) { U32 const rep = MEM_readLE32(dictPtr); dictPtr += 4; RETURN_ERROR_IF(rep==0 || rep > dictContentSize, dictionary_corrupted, ""); entropy->rep[i] = rep; } } return (size_t)(dictPtr - (const BYTE*)dict); } static size_t ZSTD_decompress_insertDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) { if (dictSize < 8) return ZSTD_refDictContent(dctx, dict, dictSize); { U32 const magic = MEM_readLE32(dict); if (magic != ZSTD_MAGIC_DICTIONARY) { return ZSTD_refDictContent(dctx, dict, dictSize); /* pure content mode */ } } dctx->dictID = MEM_readLE32((const char*)dict + ZSTD_FRAMEIDSIZE); /* load entropy tables */ { size_t const eSize = ZSTD_loadDEntropy(&dctx->entropy, dict, dictSize); RETURN_ERROR_IF(ZSTD_isError(eSize), dictionary_corrupted, ""); dict = (const char*)dict + eSize; dictSize -= eSize; } dctx->litEntropy = dctx->fseEntropy = 1; /* reference dictionary content */ return ZSTD_refDictContent(dctx, dict, dictSize); } size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx) { assert(dctx != NULL); dctx->expected = ZSTD_startingInputLength(dctx->format); /* dctx->format must be properly set */ dctx->stage = ZSTDds_getFrameHeaderSize; dctx->processedCSize = 0; dctx->decodedSize = 0; dctx->previousDstEnd = NULL; dctx->prefixStart = NULL; dctx->virtualStart = NULL; dctx->dictEnd = NULL; dctx->entropy.hufTable[0] = (HUF_DTable)((ZSTD_HUFFDTABLE_CAPACITY_LOG)*0x1000001); /* cover both little and big endian */ dctx->litEntropy = dctx->fseEntropy = 0; dctx->dictID = 0; dctx->bType = bt_reserved; dctx->isFrameDecompression = 1; ZSTD_STATIC_ASSERT(sizeof(dctx->entropy.rep) == sizeof(repStartValue)); ZSTD_memcpy(dctx->entropy.rep, repStartValue, sizeof(repStartValue)); /* initial repcodes */ dctx->LLTptr = dctx->entropy.LLTable; dctx->MLTptr = dctx->entropy.MLTable; dctx->OFTptr = dctx->entropy.OFTable; dctx->HUFptr = dctx->entropy.hufTable; return 0; } size_t ZSTD_decompressBegin_usingDict(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) { FORWARD_IF_ERROR( ZSTD_decompressBegin(dctx) , ""); if (dict && dictSize) RETURN_ERROR_IF( ZSTD_isError(ZSTD_decompress_insertDictionary(dctx, dict, dictSize)), dictionary_corrupted, ""); return 0; } /* ====== ZSTD_DDict ====== */ size_t ZSTD_decompressBegin_usingDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict) { DEBUGLOG(4, "ZSTD_decompressBegin_usingDDict"); assert(dctx != NULL); if (ddict) { const char* const dictStart = (const char*)ZSTD_DDict_dictContent(ddict); size_t const dictSize = ZSTD_DDict_dictSize(ddict); const void* const dictEnd = dictStart + dictSize; dctx->ddictIsCold = (dctx->dictEnd != dictEnd); DEBUGLOG(4, "DDict is %s", dctx->ddictIsCold ? "~cold~" : "hot!"); } FORWARD_IF_ERROR( ZSTD_decompressBegin(dctx) , ""); if (ddict) { /* NULL ddict is equivalent to no dictionary */ ZSTD_copyDDictParameters(dctx, ddict); } return 0; } /*! ZSTD_getDictID_fromDict() : * Provides the dictID stored within dictionary. * if @return == 0, the dictionary is not conformant with Zstandard specification. * It can still be loaded, but as a content-only dictionary. */ unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize) { if (dictSize < 8) return 0; if (MEM_readLE32(dict) != ZSTD_MAGIC_DICTIONARY) return 0; return MEM_readLE32((const char*)dict + ZSTD_FRAMEIDSIZE); } /*! ZSTD_getDictID_fromFrame() : * Provides the dictID required to decompress frame stored within `src`. * If @return == 0, the dictID could not be decoded. * This could for one of the following reasons : * - The frame does not require a dictionary (most common case). * - The frame was built with dictID intentionally removed. * Needed dictionary is a hidden piece of information. * Note : this use case also happens when using a non-conformant dictionary. * - `srcSize` is too small, and as a result, frame header could not be decoded. * Note : possible if `srcSize < ZSTD_FRAMEHEADERSIZE_MAX`. * - This is not a Zstandard frame. * When identifying the exact failure cause, it's possible to use * ZSTD_getFrameHeader(), which will provide a more precise error code. */ unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize) { ZSTD_FrameHeader zfp = { 0, 0, 0, ZSTD_frame, 0, 0, 0, 0, 0 }; size_t const hError = ZSTD_getFrameHeader(&zfp, src, srcSize); if (ZSTD_isError(hError)) return 0; return zfp.dictID; } /*! ZSTD_decompress_usingDDict() : * Decompression using a pre-digested Dictionary * Use dictionary without significant overhead. */ size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, const ZSTD_DDict* ddict) { /* pass content and size in case legacy frames are encountered */ return ZSTD_decompressMultiFrame(dctx, dst, dstCapacity, src, srcSize, NULL, 0, ddict); } /*===================================== * Streaming decompression *====================================*/ ZSTD_DStream* ZSTD_createDStream(void) { DEBUGLOG(3, "ZSTD_createDStream"); return ZSTD_createDCtx_internal(ZSTD_defaultCMem); } ZSTD_DStream* ZSTD_initStaticDStream(void *workspace, size_t workspaceSize) { return ZSTD_initStaticDCtx(workspace, workspaceSize); } ZSTD_DStream* ZSTD_createDStream_advanced(ZSTD_customMem customMem) { return ZSTD_createDCtx_internal(customMem); } size_t ZSTD_freeDStream(ZSTD_DStream* zds) { return ZSTD_freeDCtx(zds); } /* *** Initialization *** */ size_t ZSTD_DStreamInSize(void) { return ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize; } size_t ZSTD_DStreamOutSize(void) { return ZSTD_BLOCKSIZE_MAX; } size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType) { RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); ZSTD_clearDict(dctx); if (dict && dictSize != 0) { dctx->ddictLocal = ZSTD_createDDict_advanced(dict, dictSize, dictLoadMethod, dictContentType, dctx->customMem); RETURN_ERROR_IF(dctx->ddictLocal == NULL, memory_allocation, "NULL pointer!"); dctx->ddict = dctx->ddictLocal; dctx->dictUses = ZSTD_use_indefinitely; } return 0; } size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) { return ZSTD_DCtx_loadDictionary_advanced(dctx, dict, dictSize, ZSTD_dlm_byRef, ZSTD_dct_auto); } size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) { return ZSTD_DCtx_loadDictionary_advanced(dctx, dict, dictSize, ZSTD_dlm_byCopy, ZSTD_dct_auto); } size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType) { FORWARD_IF_ERROR(ZSTD_DCtx_loadDictionary_advanced(dctx, prefix, prefixSize, ZSTD_dlm_byRef, dictContentType), ""); dctx->dictUses = ZSTD_use_once; return 0; } size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize) { return ZSTD_DCtx_refPrefix_advanced(dctx, prefix, prefixSize, ZSTD_dct_rawContent); } /* ZSTD_initDStream_usingDict() : * return : expected size, aka ZSTD_startingInputLength(). * this function cannot fail */ size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize) { DEBUGLOG(4, "ZSTD_initDStream_usingDict"); FORWARD_IF_ERROR( ZSTD_DCtx_reset(zds, ZSTD_reset_session_only) , ""); FORWARD_IF_ERROR( ZSTD_DCtx_loadDictionary(zds, dict, dictSize) , ""); return ZSTD_startingInputLength(zds->format); } /* note : this variant can't fail */ size_t ZSTD_initDStream(ZSTD_DStream* zds) { DEBUGLOG(4, "ZSTD_initDStream"); FORWARD_IF_ERROR(ZSTD_DCtx_reset(zds, ZSTD_reset_session_only), ""); FORWARD_IF_ERROR(ZSTD_DCtx_refDDict(zds, NULL), ""); return ZSTD_startingInputLength(zds->format); } /* ZSTD_initDStream_usingDDict() : * ddict will just be referenced, and must outlive decompression session * this function cannot fail */ size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* dctx, const ZSTD_DDict* ddict) { DEBUGLOG(4, "ZSTD_initDStream_usingDDict"); FORWARD_IF_ERROR( ZSTD_DCtx_reset(dctx, ZSTD_reset_session_only) , ""); FORWARD_IF_ERROR( ZSTD_DCtx_refDDict(dctx, ddict) , ""); return ZSTD_startingInputLength(dctx->format); } /* ZSTD_resetDStream() : * return : expected size, aka ZSTD_startingInputLength(). * this function cannot fail */ size_t ZSTD_resetDStream(ZSTD_DStream* dctx) { DEBUGLOG(4, "ZSTD_resetDStream"); FORWARD_IF_ERROR(ZSTD_DCtx_reset(dctx, ZSTD_reset_session_only), ""); return ZSTD_startingInputLength(dctx->format); } size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict) { RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); ZSTD_clearDict(dctx); if (ddict) { dctx->ddict = ddict; dctx->dictUses = ZSTD_use_indefinitely; if (dctx->refMultipleDDicts == ZSTD_rmd_refMultipleDDicts) { if (dctx->ddictSet == NULL) { dctx->ddictSet = ZSTD_createDDictHashSet(dctx->customMem); if (!dctx->ddictSet) { RETURN_ERROR(memory_allocation, "Failed to allocate memory for hash set!"); } } assert(!dctx->staticSize); /* Impossible: ddictSet cannot have been allocated if static dctx */ FORWARD_IF_ERROR(ZSTD_DDictHashSet_addDDict(dctx->ddictSet, ddict, dctx->customMem), ""); } } return 0; } /* ZSTD_DCtx_setMaxWindowSize() : * note : no direct equivalence in ZSTD_DCtx_setParameter, * since this version sets windowSize, and the other sets windowLog */ size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize) { ZSTD_bounds const bounds = ZSTD_dParam_getBounds(ZSTD_d_windowLogMax); size_t const min = (size_t)1 << bounds.lowerBound; size_t const max = (size_t)1 << bounds.upperBound; RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); RETURN_ERROR_IF(maxWindowSize < min, parameter_outOfBound, ""); RETURN_ERROR_IF(maxWindowSize > max, parameter_outOfBound, ""); dctx->maxWindowSize = maxWindowSize; return 0; } size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format) { return ZSTD_DCtx_setParameter(dctx, ZSTD_d_format, (int)format); } ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam) { ZSTD_bounds bounds = { 0, 0, 0 }; switch(dParam) { case ZSTD_d_windowLogMax: bounds.lowerBound = ZSTD_WINDOWLOG_ABSOLUTEMIN; bounds.upperBound = ZSTD_WINDOWLOG_MAX; return bounds; case ZSTD_d_format: bounds.lowerBound = (int)ZSTD_f_zstd1; bounds.upperBound = (int)ZSTD_f_zstd1_magicless; ZSTD_STATIC_ASSERT(ZSTD_f_zstd1 < ZSTD_f_zstd1_magicless); return bounds; case ZSTD_d_stableOutBuffer: bounds.lowerBound = (int)ZSTD_bm_buffered; bounds.upperBound = (int)ZSTD_bm_stable; return bounds; case ZSTD_d_forceIgnoreChecksum: bounds.lowerBound = (int)ZSTD_d_validateChecksum; bounds.upperBound = (int)ZSTD_d_ignoreChecksum; return bounds; case ZSTD_d_refMultipleDDicts: bounds.lowerBound = (int)ZSTD_rmd_refSingleDDict; bounds.upperBound = (int)ZSTD_rmd_refMultipleDDicts; return bounds; case ZSTD_d_disableHuffmanAssembly: bounds.lowerBound = 0; bounds.upperBound = 1; return bounds; case ZSTD_d_maxBlockSize: bounds.lowerBound = ZSTD_BLOCKSIZE_MAX_MIN; bounds.upperBound = ZSTD_BLOCKSIZE_MAX; return bounds; default:; } bounds.error = ERROR(parameter_unsupported); return bounds; } /* ZSTD_dParam_withinBounds: * @return 1 if value is within dParam bounds, * 0 otherwise */ static int ZSTD_dParam_withinBounds(ZSTD_dParameter dParam, int value) { ZSTD_bounds const bounds = ZSTD_dParam_getBounds(dParam); if (ZSTD_isError(bounds.error)) return 0; if (value < bounds.lowerBound) return 0; if (value > bounds.upperBound) return 0; return 1; } #define CHECK_DBOUNDS(p,v) { \ RETURN_ERROR_IF(!ZSTD_dParam_withinBounds(p, v), parameter_outOfBound, ""); \ } size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value) { switch (param) { case ZSTD_d_windowLogMax: *value = (int)ZSTD_highbit32((U32)dctx->maxWindowSize); return 0; case ZSTD_d_format: *value = (int)dctx->format; return 0; case ZSTD_d_stableOutBuffer: *value = (int)dctx->outBufferMode; return 0; case ZSTD_d_forceIgnoreChecksum: *value = (int)dctx->forceIgnoreChecksum; return 0; case ZSTD_d_refMultipleDDicts: *value = (int)dctx->refMultipleDDicts; return 0; case ZSTD_d_disableHuffmanAssembly: *value = (int)dctx->disableHufAsm; return 0; case ZSTD_d_maxBlockSize: *value = dctx->maxBlockSizeParam; return 0; default:; } RETURN_ERROR(parameter_unsupported, ""); } size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter dParam, int value) { RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); switch(dParam) { case ZSTD_d_windowLogMax: if (value == 0) value = ZSTD_WINDOWLOG_LIMIT_DEFAULT; CHECK_DBOUNDS(ZSTD_d_windowLogMax, value); dctx->maxWindowSize = ((size_t)1) << value; return 0; case ZSTD_d_format: CHECK_DBOUNDS(ZSTD_d_format, value); dctx->format = (ZSTD_format_e)value; return 0; case ZSTD_d_stableOutBuffer: CHECK_DBOUNDS(ZSTD_d_stableOutBuffer, value); dctx->outBufferMode = (ZSTD_bufferMode_e)value; return 0; case ZSTD_d_forceIgnoreChecksum: CHECK_DBOUNDS(ZSTD_d_forceIgnoreChecksum, value); dctx->forceIgnoreChecksum = (ZSTD_forceIgnoreChecksum_e)value; return 0; case ZSTD_d_refMultipleDDicts: CHECK_DBOUNDS(ZSTD_d_refMultipleDDicts, value); if (dctx->staticSize != 0) { RETURN_ERROR(parameter_unsupported, "Static dctx does not support multiple DDicts!"); } dctx->refMultipleDDicts = (ZSTD_refMultipleDDicts_e)value; return 0; case ZSTD_d_disableHuffmanAssembly: CHECK_DBOUNDS(ZSTD_d_disableHuffmanAssembly, value); dctx->disableHufAsm = value != 0; return 0; case ZSTD_d_maxBlockSize: if (value != 0) CHECK_DBOUNDS(ZSTD_d_maxBlockSize, value); dctx->maxBlockSizeParam = value; return 0; default:; } RETURN_ERROR(parameter_unsupported, ""); } size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset) { if ( (reset == ZSTD_reset_session_only) || (reset == ZSTD_reset_session_and_parameters) ) { dctx->streamStage = zdss_init; dctx->noForwardProgress = 0; dctx->isFrameDecompression = 1; } if ( (reset == ZSTD_reset_parameters) || (reset == ZSTD_reset_session_and_parameters) ) { RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); ZSTD_clearDict(dctx); ZSTD_DCtx_resetParameters(dctx); } return 0; } size_t ZSTD_sizeof_DStream(const ZSTD_DStream* dctx) { return ZSTD_sizeof_DCtx(dctx); } static size_t ZSTD_decodingBufferSize_internal(unsigned long long windowSize, unsigned long long frameContentSize, size_t blockSizeMax) { size_t const blockSize = MIN((size_t)MIN(windowSize, ZSTD_BLOCKSIZE_MAX), blockSizeMax); /* We need blockSize + WILDCOPY_OVERLENGTH worth of buffer so that if a block * ends at windowSize + WILDCOPY_OVERLENGTH + 1 bytes, we can start writing * the block at the beginning of the output buffer, and maintain a full window. * * We need another blockSize worth of buffer so that we can store split * literals at the end of the block without overwriting the extDict window. */ unsigned long long const neededRBSize = windowSize + (blockSize * 2) + (WILDCOPY_OVERLENGTH * 2); unsigned long long const neededSize = MIN(frameContentSize, neededRBSize); size_t const minRBSize = (size_t) neededSize; RETURN_ERROR_IF((unsigned long long)minRBSize != neededSize, frameParameter_windowTooLarge, ""); return minRBSize; } size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize) { return ZSTD_decodingBufferSize_internal(windowSize, frameContentSize, ZSTD_BLOCKSIZE_MAX); } size_t ZSTD_estimateDStreamSize(size_t windowSize) { size_t const blockSize = MIN(windowSize, ZSTD_BLOCKSIZE_MAX); size_t const inBuffSize = blockSize; /* no block can be larger */ size_t const outBuffSize = ZSTD_decodingBufferSize_min(windowSize, ZSTD_CONTENTSIZE_UNKNOWN); return ZSTD_estimateDCtxSize() + inBuffSize + outBuffSize; } size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize) { U32 const windowSizeMax = 1U << ZSTD_WINDOWLOG_MAX; /* note : should be user-selectable, but requires an additional parameter (or a dctx) */ ZSTD_FrameHeader zfh; size_t const err = ZSTD_getFrameHeader(&zfh, src, srcSize); if (ZSTD_isError(err)) return err; RETURN_ERROR_IF(err>0, srcSize_wrong, ""); RETURN_ERROR_IF(zfh.windowSize > windowSizeMax, frameParameter_windowTooLarge, ""); return ZSTD_estimateDStreamSize((size_t)zfh.windowSize); } /* ***** Decompression ***** */ static int ZSTD_DCtx_isOverflow(ZSTD_DStream* zds, size_t const neededInBuffSize, size_t const neededOutBuffSize) { return (zds->inBuffSize + zds->outBuffSize) >= (neededInBuffSize + neededOutBuffSize) * ZSTD_WORKSPACETOOLARGE_FACTOR; } static void ZSTD_DCtx_updateOversizedDuration(ZSTD_DStream* zds, size_t const neededInBuffSize, size_t const neededOutBuffSize) { if (ZSTD_DCtx_isOverflow(zds, neededInBuffSize, neededOutBuffSize)) zds->oversizedDuration++; else zds->oversizedDuration = 0; } static int ZSTD_DCtx_isOversizedTooLong(ZSTD_DStream* zds) { return zds->oversizedDuration >= ZSTD_WORKSPACETOOLARGE_MAXDURATION; } /* Checks that the output buffer hasn't changed if ZSTD_obm_stable is used. */ static size_t ZSTD_checkOutBuffer(ZSTD_DStream const* zds, ZSTD_outBuffer const* output) { ZSTD_outBuffer const expect = zds->expectedOutBuffer; /* No requirement when ZSTD_obm_stable is not enabled. */ if (zds->outBufferMode != ZSTD_bm_stable) return 0; /* Any buffer is allowed in zdss_init, this must be the same for every other call until * the context is reset. */ if (zds->streamStage == zdss_init) return 0; /* The buffer must match our expectation exactly. */ if (expect.dst == output->dst && expect.pos == output->pos && expect.size == output->size) return 0; RETURN_ERROR(dstBuffer_wrong, "ZSTD_d_stableOutBuffer enabled but output differs!"); } /* Calls ZSTD_decompressContinue() with the right parameters for ZSTD_decompressStream() * and updates the stage and the output buffer state. This call is extracted so it can be * used both when reading directly from the ZSTD_inBuffer, and in buffered input mode. * NOTE: You must break after calling this function since the streamStage is modified. */ static size_t ZSTD_decompressContinueStream( ZSTD_DStream* zds, char** op, char* oend, void const* src, size_t srcSize) { int const isSkipFrame = ZSTD_isSkipFrame(zds); if (zds->outBufferMode == ZSTD_bm_buffered) { size_t const dstSize = isSkipFrame ? 0 : zds->outBuffSize - zds->outStart; size_t const decodedSize = ZSTD_decompressContinue(zds, zds->outBuff + zds->outStart, dstSize, src, srcSize); FORWARD_IF_ERROR(decodedSize, ""); if (!decodedSize && !isSkipFrame) { zds->streamStage = zdss_read; } else { zds->outEnd = zds->outStart + decodedSize; zds->streamStage = zdss_flush; } } else { /* Write directly into the output buffer */ size_t const dstSize = isSkipFrame ? 0 : (size_t)(oend - *op); size_t const decodedSize = ZSTD_decompressContinue(zds, *op, dstSize, src, srcSize); FORWARD_IF_ERROR(decodedSize, ""); *op += decodedSize; /* Flushing is not needed. */ zds->streamStage = zdss_read; assert(*op <= oend); assert(zds->outBufferMode == ZSTD_bm_stable); } return 0; } size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input) { const char* const src = (const char*)input->src; const char* const istart = input->pos != 0 ? src + input->pos : src; const char* const iend = input->size != 0 ? src + input->size : src; const char* ip = istart; char* const dst = (char*)output->dst; char* const ostart = output->pos != 0 ? dst + output->pos : dst; char* const oend = output->size != 0 ? dst + output->size : dst; char* op = ostart; U32 someMoreWork = 1; DEBUGLOG(5, "ZSTD_decompressStream"); assert(zds != NULL); RETURN_ERROR_IF( input->pos > input->size, srcSize_wrong, "forbidden. in: pos: %u vs size: %u", (U32)input->pos, (U32)input->size); RETURN_ERROR_IF( output->pos > output->size, dstSize_tooSmall, "forbidden. out: pos: %u vs size: %u", (U32)output->pos, (U32)output->size); DEBUGLOG(5, "input size : %u", (U32)(input->size - input->pos)); FORWARD_IF_ERROR(ZSTD_checkOutBuffer(zds, output), ""); while (someMoreWork) { switch(zds->streamStage) { case zdss_init : DEBUGLOG(5, "stage zdss_init => transparent reset "); zds->streamStage = zdss_loadHeader; zds->lhSize = zds->inPos = zds->outStart = zds->outEnd = 0; zds->hostageByte = 0; zds->expectedOutBuffer = *output; ZSTD_FALLTHROUGH; case zdss_loadHeader : DEBUGLOG(5, "stage zdss_loadHeader (srcSize : %u)", (U32)(iend - ip)); { size_t const hSize = ZSTD_getFrameHeader_advanced(&zds->fParams, zds->headerBuffer, zds->lhSize, zds->format); if (zds->refMultipleDDicts && zds->ddictSet) { ZSTD_DCtx_selectFrameDDict(zds); } if (ZSTD_isError(hSize)) { return hSize; /* error */ } if (hSize != 0) { /* need more input */ size_t const toLoad = hSize - zds->lhSize; /* if hSize!=0, hSize > zds->lhSize */ size_t const remainingInput = (size_t)(iend-ip); assert(iend >= ip); if (toLoad > remainingInput) { /* not enough input to load full header */ if (remainingInput > 0) { ZSTD_memcpy(zds->headerBuffer + zds->lhSize, ip, remainingInput); zds->lhSize += remainingInput; } input->pos = input->size; /* check first few bytes */ FORWARD_IF_ERROR( ZSTD_getFrameHeader_advanced(&zds->fParams, zds->headerBuffer, zds->lhSize, zds->format), "First few bytes detected incorrect" ); /* return hint input size */ return (MAX((size_t)ZSTD_FRAMEHEADERSIZE_MIN(zds->format), hSize) - zds->lhSize) + ZSTD_blockHeaderSize; /* remaining header bytes + next block header */ } assert(ip != NULL); ZSTD_memcpy(zds->headerBuffer + zds->lhSize, ip, toLoad); zds->lhSize = hSize; ip += toLoad; break; } } /* check for single-pass mode opportunity */ if (zds->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN && zds->fParams.frameType != ZSTD_skippableFrame && (U64)(size_t)(oend-op) >= zds->fParams.frameContentSize) { size_t const cSize = ZSTD_findFrameCompressedSize_advanced(istart, (size_t)(iend-istart), zds->format); if (cSize <= (size_t)(iend-istart)) { /* shortcut : using single-pass mode */ size_t const decompressedSize = ZSTD_decompress_usingDDict(zds, op, (size_t)(oend-op), istart, cSize, ZSTD_getDDict(zds)); if (ZSTD_isError(decompressedSize)) return decompressedSize; DEBUGLOG(4, "shortcut to single-pass ZSTD_decompress_usingDDict()"); assert(istart != NULL); ip = istart + cSize; op = op ? op + decompressedSize : op; /* can occur if frameContentSize = 0 (empty frame) */ zds->expected = 0; zds->streamStage = zdss_init; someMoreWork = 0; break; } } /* Check output buffer is large enough for ZSTD_odm_stable. */ if (zds->outBufferMode == ZSTD_bm_stable && zds->fParams.frameType != ZSTD_skippableFrame && zds->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN && (U64)(size_t)(oend-op) < zds->fParams.frameContentSize) { RETURN_ERROR(dstSize_tooSmall, "ZSTD_obm_stable passed but ZSTD_outBuffer is too small"); } /* Consume header (see ZSTDds_decodeFrameHeader) */ DEBUGLOG(4, "Consume header"); FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDDict(zds, ZSTD_getDDict(zds)), ""); if (zds->format == ZSTD_f_zstd1 && (MEM_readLE32(zds->headerBuffer) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */ zds->expected = MEM_readLE32(zds->headerBuffer + ZSTD_FRAMEIDSIZE); zds->stage = ZSTDds_skipFrame; } else { FORWARD_IF_ERROR(ZSTD_decodeFrameHeader(zds, zds->headerBuffer, zds->lhSize), ""); zds->expected = ZSTD_blockHeaderSize; zds->stage = ZSTDds_decodeBlockHeader; } /* control buffer memory usage */ DEBUGLOG(4, "Control max memory usage (%u KB <= max %u KB)", (U32)(zds->fParams.windowSize >>10), (U32)(zds->maxWindowSize >> 10) ); zds->fParams.windowSize = MAX(zds->fParams.windowSize, 1U << ZSTD_WINDOWLOG_ABSOLUTEMIN); RETURN_ERROR_IF(zds->fParams.windowSize > zds->maxWindowSize, frameParameter_windowTooLarge, ""); if (zds->maxBlockSizeParam != 0) zds->fParams.blockSizeMax = MIN(zds->fParams.blockSizeMax, (unsigned)zds->maxBlockSizeParam); /* Adapt buffer sizes to frame header instructions */ { size_t const neededInBuffSize = MAX(zds->fParams.blockSizeMax, 4 /* frame checksum */); size_t const neededOutBuffSize = zds->outBufferMode == ZSTD_bm_buffered ? ZSTD_decodingBufferSize_internal(zds->fParams.windowSize, zds->fParams.frameContentSize, zds->fParams.blockSizeMax) : 0; ZSTD_DCtx_updateOversizedDuration(zds, neededInBuffSize, neededOutBuffSize); { int const tooSmall = (zds->inBuffSize < neededInBuffSize) || (zds->outBuffSize < neededOutBuffSize); int const tooLarge = ZSTD_DCtx_isOversizedTooLong(zds); if (tooSmall || tooLarge) { size_t const bufferSize = neededInBuffSize + neededOutBuffSize; DEBUGLOG(4, "inBuff : from %u to %u", (U32)zds->inBuffSize, (U32)neededInBuffSize); DEBUGLOG(4, "outBuff : from %u to %u", (U32)zds->outBuffSize, (U32)neededOutBuffSize); if (zds->staticSize) { /* static DCtx */ DEBUGLOG(4, "staticSize : %u", (U32)zds->staticSize); assert(zds->staticSize >= sizeof(ZSTD_DCtx)); /* controlled at init */ RETURN_ERROR_IF( bufferSize > zds->staticSize - sizeof(ZSTD_DCtx), memory_allocation, ""); } else { ZSTD_customFree(zds->inBuff, zds->customMem); zds->inBuffSize = 0; zds->outBuffSize = 0; zds->inBuff = (char*)ZSTD_customMalloc(bufferSize, zds->customMem); RETURN_ERROR_IF(zds->inBuff == NULL, memory_allocation, ""); } zds->inBuffSize = neededInBuffSize; zds->outBuff = zds->inBuff + zds->inBuffSize; zds->outBuffSize = neededOutBuffSize; } } } zds->streamStage = zdss_read; ZSTD_FALLTHROUGH; case zdss_read: DEBUGLOG(5, "stage zdss_read"); { size_t const neededInSize = ZSTD_nextSrcSizeToDecompressWithInputSize(zds, (size_t)(iend - ip)); DEBUGLOG(5, "neededInSize = %u", (U32)neededInSize); if (neededInSize==0) { /* end of frame */ zds->streamStage = zdss_init; someMoreWork = 0; break; } if ((size_t)(iend-ip) >= neededInSize) { /* decode directly from src */ FORWARD_IF_ERROR(ZSTD_decompressContinueStream(zds, &op, oend, ip, neededInSize), ""); assert(ip != NULL); ip += neededInSize; /* Function modifies the stage so we must break */ break; } } if (ip==iend) { someMoreWork = 0; break; } /* no more input */ zds->streamStage = zdss_load; ZSTD_FALLTHROUGH; case zdss_load: { size_t const neededInSize = ZSTD_nextSrcSizeToDecompress(zds); size_t const toLoad = neededInSize - zds->inPos; int const isSkipFrame = ZSTD_isSkipFrame(zds); size_t loadedSize; /* At this point we shouldn't be decompressing a block that we can stream. */ assert(neededInSize == ZSTD_nextSrcSizeToDecompressWithInputSize(zds, (size_t)(iend - ip))); if (isSkipFrame) { loadedSize = MIN(toLoad, (size_t)(iend-ip)); } else { RETURN_ERROR_IF(toLoad > zds->inBuffSize - zds->inPos, corruption_detected, "should never happen"); loadedSize = ZSTD_limitCopy(zds->inBuff + zds->inPos, toLoad, ip, (size_t)(iend-ip)); } if (loadedSize != 0) { /* ip may be NULL */ ip += loadedSize; zds->inPos += loadedSize; } if (loadedSize < toLoad) { someMoreWork = 0; break; } /* not enough input, wait for more */ /* decode loaded input */ zds->inPos = 0; /* input is consumed */ FORWARD_IF_ERROR(ZSTD_decompressContinueStream(zds, &op, oend, zds->inBuff, neededInSize), ""); /* Function modifies the stage so we must break */ break; } case zdss_flush: { size_t const toFlushSize = zds->outEnd - zds->outStart; size_t const flushedSize = ZSTD_limitCopy(op, (size_t)(oend-op), zds->outBuff + zds->outStart, toFlushSize); op = op ? op + flushedSize : op; zds->outStart += flushedSize; if (flushedSize == toFlushSize) { /* flush completed */ zds->streamStage = zdss_read; if ( (zds->outBuffSize < zds->fParams.frameContentSize) && (zds->outStart + zds->fParams.blockSizeMax > zds->outBuffSize) ) { DEBUGLOG(5, "restart filling outBuff from beginning (left:%i, needed:%u)", (int)(zds->outBuffSize - zds->outStart), (U32)zds->fParams.blockSizeMax); zds->outStart = zds->outEnd = 0; } break; } } /* cannot complete flush */ someMoreWork = 0; break; default: assert(0); /* impossible */ RETURN_ERROR(GENERIC, "impossible to reach"); /* some compilers require default to do something */ } } /* result */ input->pos = (size_t)(ip - (const char*)(input->src)); output->pos = (size_t)(op - (char*)(output->dst)); /* Update the expected output buffer for ZSTD_obm_stable. */ zds->expectedOutBuffer = *output; if ((ip==istart) && (op==ostart)) { /* no forward progress */ zds->noForwardProgress ++; if (zds->noForwardProgress >= ZSTD_NO_FORWARD_PROGRESS_MAX) { RETURN_ERROR_IF(op==oend, noForwardProgress_destFull, ""); RETURN_ERROR_IF(ip==iend, noForwardProgress_inputEmpty, ""); assert(0); } } else { zds->noForwardProgress = 0; } { size_t nextSrcSizeHint = ZSTD_nextSrcSizeToDecompress(zds); if (!nextSrcSizeHint) { /* frame fully decoded */ if (zds->outEnd == zds->outStart) { /* output fully flushed */ if (zds->hostageByte) { if (input->pos >= input->size) { /* can't release hostage (not present) */ zds->streamStage = zdss_read; return 1; } input->pos++; /* release hostage */ } /* zds->hostageByte */ return 0; } /* zds->outEnd == zds->outStart */ if (!zds->hostageByte) { /* output not fully flushed; keep last byte as hostage; will be released when all output is flushed */ input->pos--; /* note : pos > 0, otherwise, impossible to finish reading last block */ zds->hostageByte=1; } return 1; } /* nextSrcSizeHint==0 */ nextSrcSizeHint += ZSTD_blockHeaderSize * (ZSTD_nextInputType(zds) == ZSTDnit_block); /* preload header of next block */ assert(zds->inPos <= nextSrcSizeHint); nextSrcSizeHint -= zds->inPos; /* part already loaded*/ return nextSrcSizeHint; } } size_t ZSTD_decompressStream_simpleArgs ( ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, size_t* dstPos, const void* src, size_t srcSize, size_t* srcPos) { ZSTD_outBuffer output; ZSTD_inBuffer input; output.dst = dst; output.size = dstCapacity; output.pos = *dstPos; input.src = src; input.size = srcSize; input.pos = *srcPos; { size_t const cErr = ZSTD_decompressStream(dctx, &output, &input); *dstPos = output.pos; *srcPos = input.pos; return cErr; } } |
| 10 10 10 5 1 7 4 8 1 2 4 1 2 1 2 3 1 1 4 1 5 5 1 1 3 17 3 5 5 5 8 10 20 4 1 2 2 1 1 18 12 24 7 35 35 35 35 35 35 34 2 34 35 35 106 1 105 103 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 | // SPDX-License-Identifier: GPL-2.0-or-later /* * Linux I2C core SMBus and SMBus emulation code * * This file contains the SMBus functions which are always included in the I2C * core because they can be emulated via I2C. SMBus specific extensions * (e.g. smbalert) are handled in a separate i2c-smbus module. * * All SMBus-related things are written by Frodo Looijaard <frodol@dds.nl> * SMBus 2.0 support by Mark Studebaker <mdsxyz123@yahoo.com> and * Jean Delvare <jdelvare@suse.de> */ #include <linux/device.h> #include <linux/err.h> #include <linux/i2c.h> #include <linux/i2c-smbus.h> #include <linux/property.h> #include <linux/slab.h> #include <linux/string_choices.h> #include "i2c-core.h" #define CREATE_TRACE_POINTS #include <trace/events/smbus.h> /* The SMBus parts */ #define POLY (0x1070U << 3) static u8 crc8(u16 data) { int i; for (i = 0; i < 8; i++) { if (data & 0x8000) data = data ^ POLY; data = data << 1; } return (u8)(data >> 8); } /** * i2c_smbus_pec - Incremental CRC8 over the given input data array * @crc: previous return crc8 value * @p: pointer to data buffer. * @count: number of bytes in data buffer. * * Incremental CRC8 over count bytes in the array pointed to by p */ u8 i2c_smbus_pec(u8 crc, u8 *p, size_t count) { int i; for (i = 0; i < count; i++) crc = crc8((crc ^ p[i]) << 8); return crc; } EXPORT_SYMBOL(i2c_smbus_pec); /* Assume a 7-bit address, which is reasonable for SMBus */ static u8 i2c_smbus_msg_pec(u8 pec, struct i2c_msg *msg) { /* The address will be sent first */ u8 addr = i2c_8bit_addr_from_msg(msg); pec = i2c_smbus_pec(pec, &addr, 1); /* The data buffer follows */ return i2c_smbus_pec(pec, msg->buf, msg->len); } /* Used for write only transactions */ static inline void i2c_smbus_add_pec(struct i2c_msg *msg) { msg->buf[msg->len] = i2c_smbus_msg_pec(0, msg); msg->len++; } /* Return <0 on CRC error If there was a write before this read (most cases) we need to take the partial CRC from the write part into account. Note that this function does modify the message (we need to decrease the message length to hide the CRC byte from the caller). */ static int i2c_smbus_check_pec(u8 cpec, struct i2c_msg *msg) { u8 rpec = msg->buf[--msg->len]; cpec = i2c_smbus_msg_pec(cpec, msg); if (rpec != cpec) { pr_debug("Bad PEC 0x%02x vs. 0x%02x\n", rpec, cpec); return -EBADMSG; } return 0; } /** * i2c_smbus_read_byte - SMBus "receive byte" protocol * @client: Handle to slave device * * This executes the SMBus "receive byte" protocol, returning negative errno * else the byte received from the device. */ s32 i2c_smbus_read_byte(const struct i2c_client *client) { union i2c_smbus_data data; int status; status = i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_READ, 0, I2C_SMBUS_BYTE, &data); return (status < 0) ? status : data.byte; } EXPORT_SYMBOL(i2c_smbus_read_byte); /** * i2c_smbus_write_byte - SMBus "send byte" protocol * @client: Handle to slave device * @value: Byte to be sent * * This executes the SMBus "send byte" protocol, returning negative errno * else zero on success. */ s32 i2c_smbus_write_byte(const struct i2c_client *client, u8 value) { return i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_WRITE, value, I2C_SMBUS_BYTE, NULL); } EXPORT_SYMBOL(i2c_smbus_write_byte); /** * i2c_smbus_read_byte_data - SMBus "read byte" protocol * @client: Handle to slave device * @command: Byte interpreted by slave * * This executes the SMBus "read byte" protocol, returning negative errno * else a data byte received from the device. */ s32 i2c_smbus_read_byte_data(const struct i2c_client *client, u8 command) { union i2c_smbus_data data; int status; status = i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_READ, command, I2C_SMBUS_BYTE_DATA, &data); return (status < 0) ? status : data.byte; } EXPORT_SYMBOL(i2c_smbus_read_byte_data); /** * i2c_smbus_write_byte_data - SMBus "write byte" protocol * @client: Handle to slave device * @command: Byte interpreted by slave * @value: Byte being written * * This executes the SMBus "write byte" protocol, returning negative errno * else zero on success. */ s32 i2c_smbus_write_byte_data(const struct i2c_client *client, u8 command, u8 value) { union i2c_smbus_data data; data.byte = value; return i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_WRITE, command, I2C_SMBUS_BYTE_DATA, &data); } EXPORT_SYMBOL(i2c_smbus_write_byte_data); /** * i2c_smbus_read_word_data - SMBus "read word" protocol * @client: Handle to slave device * @command: Byte interpreted by slave * * This executes the SMBus "read word" protocol, returning negative errno * else a 16-bit unsigned "word" received from the device. */ s32 i2c_smbus_read_word_data(const struct i2c_client *client, u8 command) { union i2c_smbus_data data; int status; status = i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_READ, command, I2C_SMBUS_WORD_DATA, &data); return (status < 0) ? status : data.word; } EXPORT_SYMBOL(i2c_smbus_read_word_data); /** * i2c_smbus_write_word_data - SMBus "write word" protocol * @client: Handle to slave device * @command: Byte interpreted by slave * @value: 16-bit "word" being written * * This executes the SMBus "write word" protocol, returning negative errno * else zero on success. */ s32 i2c_smbus_write_word_data(const struct i2c_client *client, u8 command, u16 value) { union i2c_smbus_data data; data.word = value; return i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_WRITE, command, I2C_SMBUS_WORD_DATA, &data); } EXPORT_SYMBOL(i2c_smbus_write_word_data); /** * i2c_smbus_read_block_data - SMBus "block read" protocol * @client: Handle to slave device * @command: Byte interpreted by slave * @values: Byte array into which data will be read; big enough to hold * the data returned by the slave. SMBus allows at most 32 bytes. * * This executes the SMBus "block read" protocol, returning negative errno * else the number of data bytes in the slave's response. * * Note that using this function requires that the client's adapter support * the I2C_FUNC_SMBUS_READ_BLOCK_DATA functionality. Not all adapter drivers * support this; its emulation through I2C messaging relies on a specific * mechanism (I2C_M_RECV_LEN) which may not be implemented. */ s32 i2c_smbus_read_block_data(const struct i2c_client *client, u8 command, u8 *values) { union i2c_smbus_data data; int status; status = i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_READ, command, I2C_SMBUS_BLOCK_DATA, &data); if (status) return status; memcpy(values, &data.block[1], data.block[0]); return data.block[0]; } EXPORT_SYMBOL(i2c_smbus_read_block_data); /** * i2c_smbus_write_block_data - SMBus "block write" protocol * @client: Handle to slave device * @command: Byte interpreted by slave * @length: Size of data block; SMBus allows at most 32 bytes * @values: Byte array which will be written. * * This executes the SMBus "block write" protocol, returning negative errno * else zero on success. */ s32 i2c_smbus_write_block_data(const struct i2c_client *client, u8 command, u8 length, const u8 *values) { union i2c_smbus_data data; if (length > I2C_SMBUS_BLOCK_MAX) length = I2C_SMBUS_BLOCK_MAX; data.block[0] = length; memcpy(&data.block[1], values, length); return i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_WRITE, command, I2C_SMBUS_BLOCK_DATA, &data); } EXPORT_SYMBOL(i2c_smbus_write_block_data); /* Returns the number of read bytes */ s32 i2c_smbus_read_i2c_block_data(const struct i2c_client *client, u8 command, u8 length, u8 *values) { union i2c_smbus_data data; int status; if (length > I2C_SMBUS_BLOCK_MAX) length = I2C_SMBUS_BLOCK_MAX; data.block[0] = length; status = i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_READ, command, I2C_SMBUS_I2C_BLOCK_DATA, &data); if (status < 0) return status; memcpy(values, &data.block[1], data.block[0]); return data.block[0]; } EXPORT_SYMBOL(i2c_smbus_read_i2c_block_data); s32 i2c_smbus_write_i2c_block_data(const struct i2c_client *client, u8 command, u8 length, const u8 *values) { union i2c_smbus_data data; if (length > I2C_SMBUS_BLOCK_MAX) length = I2C_SMBUS_BLOCK_MAX; data.block[0] = length; memcpy(data.block + 1, values, length); return i2c_smbus_xfer(client->adapter, client->addr, client->flags, I2C_SMBUS_WRITE, command, I2C_SMBUS_I2C_BLOCK_DATA, &data); } EXPORT_SYMBOL(i2c_smbus_write_i2c_block_data); static void i2c_smbus_try_get_dmabuf(struct i2c_msg *msg, u8 init_val) { bool is_read = msg->flags & I2C_M_RD; unsigned char *dma_buf; dma_buf = kzalloc(I2C_SMBUS_BLOCK_MAX + (is_read ? 2 : 3), GFP_KERNEL); if (!dma_buf) return; msg->buf = dma_buf; msg->flags |= I2C_M_DMA_SAFE; if (init_val) msg->buf[0] = init_val; } /* * Simulate a SMBus command using the I2C protocol. * No checking of parameters is done! */ static s32 i2c_smbus_xfer_emulated(struct i2c_adapter *adapter, u16 addr, unsigned short flags, char read_write, u8 command, int size, union i2c_smbus_data *data) { /* * So we need to generate a series of msgs. In the case of writing, we * need to use only one message; when reading, we need two. We * initialize most things with sane defaults, to keep the code below * somewhat simpler. */ unsigned char msgbuf0[I2C_SMBUS_BLOCK_MAX+3]; unsigned char msgbuf1[I2C_SMBUS_BLOCK_MAX+2]; int nmsgs = read_write == I2C_SMBUS_READ ? 2 : 1; u8 partial_pec = 0; int status; struct i2c_msg msg[2] = { { .addr = addr, .flags = flags, .len = 1, .buf = msgbuf0, }, { .addr = addr, .flags = flags | I2C_M_RD, .len = 0, .buf = msgbuf1, }, }; bool wants_pec = ((flags & I2C_CLIENT_PEC) && size != I2C_SMBUS_QUICK && size != I2C_SMBUS_I2C_BLOCK_DATA); msgbuf0[0] = command; switch (size) { case I2C_SMBUS_QUICK: msg[0].len = 0; /* Special case: The read/write field is used as data */ msg[0].flags = flags | (read_write == I2C_SMBUS_READ ? I2C_M_RD : 0); nmsgs = 1; break; case I2C_SMBUS_BYTE: if (read_write == I2C_SMBUS_READ) { /* Special case: only a read! */ msg[0].flags = I2C_M_RD | flags; nmsgs = 1; } break; case I2C_SMBUS_BYTE_DATA: if (read_write == I2C_SMBUS_READ) msg[1].len = 1; else { msg[0].len = 2; msgbuf0[1] = data->byte; } break; case I2C_SMBUS_WORD_DATA: if (read_write == I2C_SMBUS_READ) msg[1].len = 2; else { msg[0].len = 3; msgbuf0[1] = data->word & 0xff; msgbuf0[2] = data->word >> 8; } break; case I2C_SMBUS_PROC_CALL: nmsgs = 2; /* Special case */ read_write = I2C_SMBUS_READ; msg[0].len = 3; msg[1].len = 2; msgbuf0[1] = data->word & 0xff; msgbuf0[2] = data->word >> 8; break; case I2C_SMBUS_BLOCK_DATA: if (read_write == I2C_SMBUS_READ) { msg[1].flags |= I2C_M_RECV_LEN; msg[1].len = 1; /* block length will be added by the underlying bus driver */ i2c_smbus_try_get_dmabuf(&msg[1], 0); } else { msg[0].len = data->block[0] + 2; if (msg[0].len > I2C_SMBUS_BLOCK_MAX + 2) { dev_err(&adapter->dev, "Invalid block write size %d\n", data->block[0]); return -EINVAL; } i2c_smbus_try_get_dmabuf(&msg[0], command); memcpy(msg[0].buf + 1, data->block, msg[0].len - 1); } break; case I2C_SMBUS_BLOCK_PROC_CALL: nmsgs = 2; /* Another special case */ read_write = I2C_SMBUS_READ; if (data->block[0] > I2C_SMBUS_BLOCK_MAX) { dev_err(&adapter->dev, "Invalid block write size %d\n", data->block[0]); return -EINVAL; } msg[0].len = data->block[0] + 2; i2c_smbus_try_get_dmabuf(&msg[0], command); memcpy(msg[0].buf + 1, data->block, msg[0].len - 1); msg[1].flags |= I2C_M_RECV_LEN; msg[1].len = 1; /* block length will be added by the underlying bus driver */ i2c_smbus_try_get_dmabuf(&msg[1], 0); break; case I2C_SMBUS_I2C_BLOCK_DATA: if (data->block[0] > I2C_SMBUS_BLOCK_MAX) { dev_err(&adapter->dev, "Invalid block %s size %d\n", str_read_write(read_write == I2C_SMBUS_READ), data->block[0]); return -EINVAL; } if (read_write == I2C_SMBUS_READ) { msg[1].len = data->block[0]; i2c_smbus_try_get_dmabuf(&msg[1], 0); } else { msg[0].len = data->block[0] + 1; i2c_smbus_try_get_dmabuf(&msg[0], command); memcpy(msg[0].buf + 1, data->block + 1, data->block[0]); } break; default: dev_err(&adapter->dev, "Unsupported transaction %d\n", size); return -EOPNOTSUPP; } if (wants_pec) { /* Compute PEC if first message is a write */ if (!(msg[0].flags & I2C_M_RD)) { if (nmsgs == 1) /* Write only */ i2c_smbus_add_pec(&msg[0]); else /* Write followed by read */ partial_pec = i2c_smbus_msg_pec(0, &msg[0]); } /* Ask for PEC if last message is a read */ if (msg[nmsgs - 1].flags & I2C_M_RD) msg[nmsgs - 1].len++; } status = __i2c_transfer(adapter, msg, nmsgs); if (status < 0) goto cleanup; if (status != nmsgs) { status = -EIO; goto cleanup; } status = 0; /* Check PEC if last message is a read */ if (wants_pec && (msg[nmsgs - 1].flags & I2C_M_RD)) { status = i2c_smbus_check_pec(partial_pec, &msg[nmsgs - 1]); if (status < 0) goto cleanup; } if (read_write == I2C_SMBUS_READ) switch (size) { case I2C_SMBUS_BYTE: data->byte = msgbuf0[0]; break; case I2C_SMBUS_BYTE_DATA: data->byte = msgbuf1[0]; break; case I2C_SMBUS_WORD_DATA: case I2C_SMBUS_PROC_CALL: data->word = msgbuf1[0] | (msgbuf1[1] << 8); break; case I2C_SMBUS_I2C_BLOCK_DATA: memcpy(data->block + 1, msg[1].buf, data->block[0]); break; case I2C_SMBUS_BLOCK_DATA: case I2C_SMBUS_BLOCK_PROC_CALL: if (msg[1].buf[0] > I2C_SMBUS_BLOCK_MAX) { dev_err(&adapter->dev, "Invalid block size returned: %d\n", msg[1].buf[0]); status = -EPROTO; goto cleanup; } memcpy(data->block, msg[1].buf, msg[1].buf[0] + 1); break; } cleanup: if (msg[0].flags & I2C_M_DMA_SAFE) kfree(msg[0].buf); if (msg[1].flags & I2C_M_DMA_SAFE) kfree(msg[1].buf); return status; } /** * i2c_smbus_xfer - execute SMBus protocol operations * @adapter: Handle to I2C bus * @addr: Address of SMBus slave on that bus * @flags: I2C_CLIENT_* flags (usually zero or I2C_CLIENT_PEC) * @read_write: I2C_SMBUS_READ or I2C_SMBUS_WRITE * @command: Byte interpreted by slave, for protocols which use such bytes * @protocol: SMBus protocol operation to execute, such as I2C_SMBUS_PROC_CALL * @data: Data to be read or written * * This executes an SMBus protocol operation, and returns a negative * errno code else zero on success. */ s32 i2c_smbus_xfer(struct i2c_adapter *adapter, u16 addr, unsigned short flags, char read_write, u8 command, int protocol, union i2c_smbus_data *data) { s32 res; res = __i2c_lock_bus_helper(adapter); if (res) return res; res = __i2c_smbus_xfer(adapter, addr, flags, read_write, command, protocol, data); i2c_unlock_bus(adapter, I2C_LOCK_SEGMENT); return res; } EXPORT_SYMBOL(i2c_smbus_xfer); s32 __i2c_smbus_xfer(struct i2c_adapter *adapter, u16 addr, unsigned short flags, char read_write, u8 command, int protocol, union i2c_smbus_data *data) { int (*xfer_func)(struct i2c_adapter *adap, u16 addr, unsigned short flags, char read_write, u8 command, int size, union i2c_smbus_data *data); unsigned long orig_jiffies; int try; s32 res; res = __i2c_check_suspended(adapter); if (res) return res; /* If enabled, the following two tracepoints are conditional on * read_write and protocol. */ trace_smbus_write(adapter, addr, flags, read_write, command, protocol, data); trace_smbus_read(adapter, addr, flags, read_write, command, protocol); flags &= I2C_M_TEN | I2C_CLIENT_PEC | I2C_CLIENT_SCCB; xfer_func = adapter->algo->smbus_xfer; if (i2c_in_atomic_xfer_mode()) { if (adapter->algo->smbus_xfer_atomic) xfer_func = adapter->algo->smbus_xfer_atomic; else if (adapter->algo->master_xfer_atomic) xfer_func = NULL; /* fallback to I2C emulation */ } if (xfer_func) { /* Retry automatically on arbitration loss */ orig_jiffies = jiffies; for (res = 0, try = 0; try <= adapter->retries; try++) { res = xfer_func(adapter, addr, flags, read_write, command, protocol, data); if (res != -EAGAIN) break; if (time_after(jiffies, orig_jiffies + adapter->timeout)) break; } if (res != -EOPNOTSUPP || !adapter->algo->master_xfer) goto trace; /* * Fall back to i2c_smbus_xfer_emulated if the adapter doesn't * implement native support for the SMBus operation. */ } res = i2c_smbus_xfer_emulated(adapter, addr, flags, read_write, command, protocol, data); trace: /* If enabled, the reply tracepoint is conditional on read_write. */ trace_smbus_reply(adapter, addr, flags, read_write, command, protocol, data, res); trace_smbus_result(adapter, addr, flags, read_write, command, protocol, res); return res; } EXPORT_SYMBOL(__i2c_smbus_xfer); /** * i2c_smbus_read_i2c_block_data_or_emulated - read block or emulate * @client: Handle to slave device * @command: Byte interpreted by slave * @length: Size of data block; SMBus allows at most I2C_SMBUS_BLOCK_MAX bytes * @values: Byte array into which data will be read; big enough to hold * the data returned by the slave. SMBus allows at most * I2C_SMBUS_BLOCK_MAX bytes. * * This executes the SMBus "block read" protocol if supported by the adapter. * If block read is not supported, it emulates it using either word or byte * read protocols depending on availability. * * The addresses of the I2C slave device that are accessed with this function * must be mapped to a linear region, so that a block read will have the same * effect as a byte read. Before using this function you must double-check * if the I2C slave does support exchanging a block transfer with a byte * transfer. */ s32 i2c_smbus_read_i2c_block_data_or_emulated(const struct i2c_client *client, u8 command, u8 length, u8 *values) { u8 i = 0; int status; if (length > I2C_SMBUS_BLOCK_MAX) length = I2C_SMBUS_BLOCK_MAX; if (i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_READ_I2C_BLOCK)) return i2c_smbus_read_i2c_block_data(client, command, length, values); if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_READ_BYTE_DATA)) return -EOPNOTSUPP; if (i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_READ_WORD_DATA)) { while ((i + 2) <= length) { status = i2c_smbus_read_word_data(client, command + i); if (status < 0) return status; values[i] = status & 0xff; values[i + 1] = status >> 8; i += 2; } } while (i < length) { status = i2c_smbus_read_byte_data(client, command + i); if (status < 0) return status; values[i] = status; i++; } return i; } EXPORT_SYMBOL(i2c_smbus_read_i2c_block_data_or_emulated); /** * i2c_new_smbus_alert_device - get ara client for SMBus alert support * @adapter: the target adapter * @setup: setup data for the SMBus alert handler * Context: can sleep * * Setup handling of the SMBus alert protocol on a given I2C bus segment. * * Handling can be done either through our IRQ handler, or by the * adapter (from its handler, periodic polling, or whatever). * * This returns the ara client, which should be saved for later use with * i2c_handle_smbus_alert() and ultimately i2c_unregister_device(); or an * ERRPTR to indicate an error. */ struct i2c_client *i2c_new_smbus_alert_device(struct i2c_adapter *adapter, struct i2c_smbus_alert_setup *setup) { struct i2c_board_info ara_board_info = { I2C_BOARD_INFO("smbus_alert", 0x0c), .platform_data = setup, }; return i2c_new_client_device(adapter, &ara_board_info); } EXPORT_SYMBOL_GPL(i2c_new_smbus_alert_device); #if IS_ENABLED(CONFIG_I2C_SMBUS) int i2c_setup_smbus_alert(struct i2c_adapter *adapter) { struct device *parent = adapter->dev.parent; int irq; /* Adapter instantiated without parent, skip the SMBus alert setup */ if (!parent) return 0; /* Report serious errors */ irq = device_property_match_string(parent, "interrupt-names", "smbus_alert"); if (irq < 0 && irq != -EINVAL && irq != -ENODATA) return irq; /* Skip setup when no irq was found */ if (irq < 0 && !device_property_present(parent, "smbalert-gpios")) return 0; return PTR_ERR_OR_ZERO(i2c_new_smbus_alert_device(adapter, NULL)); } #endif |
| 1 1 1 45 43 3 34 4 1 5 5 5 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 | // SPDX-License-Identifier: GPL-2.0-only /* * Copyright (c) 2012 Smith Micro Software, Inc. * Copyright (c) 2012 Bjørn Mork <bjorn@mork.no> * * This driver is based on and reuse most of cdc_ncm, which is * Copyright (C) ST-Ericsson 2010-2012 */ #include <linux/module.h> #include <linux/netdevice.h> #include <linux/ethtool.h> #include <linux/if_vlan.h> #include <linux/ip.h> #include <linux/mii.h> #include <linux/usb.h> #include <linux/usb/cdc.h> #include <linux/usb/usbnet.h> #include <linux/usb/cdc-wdm.h> #include <linux/usb/cdc_ncm.h> #include <net/ipv6.h> #include <net/addrconf.h> #include <net/ipv6_stubs.h> #include <net/ndisc.h> /* alternative VLAN for IP session 0 if not untagged */ #define MBIM_IPS0_VID 4094 /* driver specific data - must match cdc_ncm usage */ struct cdc_mbim_state { struct cdc_ncm_ctx *ctx; atomic_t pmcount; struct usb_driver *subdriver; unsigned long _unused; unsigned long flags; }; /* flags for the cdc_mbim_state.flags field */ enum cdc_mbim_flags { FLAG_IPS0_VLAN = 1 << 0, /* IP session 0 is tagged */ }; /* using a counter to merge subdriver requests with our own into a combined state */ static int cdc_mbim_manage_power(struct usbnet *dev, int on) { struct cdc_mbim_state *info = (void *)&dev->data; int rv = 0; dev_dbg(&dev->intf->dev, "%s() pmcount=%d, on=%d\n", __func__, atomic_read(&info->pmcount), on); if ((on && atomic_add_return(1, &info->pmcount) == 1) || (!on && atomic_dec_and_test(&info->pmcount))) { /* need autopm_get/put here to ensure the usbcore sees the new value */ rv = usb_autopm_get_interface(dev->intf); dev->intf->needs_remote_wakeup = on; if (!rv) usb_autopm_put_interface(dev->intf); } return 0; } static int cdc_mbim_wdm_manage_power(struct usb_interface *intf, int status) { struct usbnet *dev = usb_get_intfdata(intf); /* can be called while disconnecting */ if (!dev) return 0; return cdc_mbim_manage_power(dev, status); } static int cdc_mbim_rx_add_vid(struct net_device *netdev, __be16 proto, u16 vid) { struct usbnet *dev = netdev_priv(netdev); struct cdc_mbim_state *info = (void *)&dev->data; /* creation of this VLAN is a request to tag IP session 0 */ if (vid == MBIM_IPS0_VID) info->flags |= FLAG_IPS0_VLAN; else if (vid >= 512) /* we don't map these to MBIM session */ return -EINVAL; return 0; } static int cdc_mbim_rx_kill_vid(struct net_device *netdev, __be16 proto, u16 vid) { struct usbnet *dev = netdev_priv(netdev); struct cdc_mbim_state *info = (void *)&dev->data; /* this is a request for an untagged IP session 0 */ if (vid == MBIM_IPS0_VID) info->flags &= ~FLAG_IPS0_VLAN; return 0; } static const struct net_device_ops cdc_mbim_netdev_ops = { .ndo_open = usbnet_open, .ndo_stop = usbnet_stop, .ndo_start_xmit = usbnet_start_xmit, .ndo_tx_timeout = usbnet_tx_timeout, .ndo_get_stats64 = dev_get_tstats64, .ndo_change_mtu = cdc_ncm_change_mtu, .ndo_set_mac_address = eth_mac_addr, .ndo_validate_addr = eth_validate_addr, .ndo_vlan_rx_add_vid = cdc_mbim_rx_add_vid, .ndo_vlan_rx_kill_vid = cdc_mbim_rx_kill_vid, }; /* Change the control interface altsetting and update the .driver_info * pointer if the matching entry after changing class codes points to * a different struct */ static int cdc_mbim_set_ctrlalt(struct usbnet *dev, struct usb_interface *intf, u8 alt) { struct usb_driver *driver = to_usb_driver(intf->dev.driver); const struct usb_device_id *id; struct driver_info *info; int ret; ret = usb_set_interface(dev->udev, intf->cur_altsetting->desc.bInterfaceNumber, alt); if (ret) return ret; id = usb_match_id(intf, driver->id_table); if (!id) return -ENODEV; info = (struct driver_info *)id->driver_info; if (info != dev->driver_info) { dev_dbg(&intf->dev, "driver_info updated to '%s'\n", info->description); dev->driver_info = info; } return 0; } static int cdc_mbim_bind(struct usbnet *dev, struct usb_interface *intf) { struct cdc_ncm_ctx *ctx; struct usb_driver *subdriver = ERR_PTR(-ENODEV); int ret = -ENODEV; u8 data_altsetting = 1; struct cdc_mbim_state *info = (void *)&dev->data; /* should we change control altsetting on a NCM/MBIM function? */ if (cdc_ncm_select_altsetting(intf) == CDC_NCM_COMM_ALTSETTING_MBIM) { data_altsetting = CDC_NCM_DATA_ALTSETTING_MBIM; ret = cdc_mbim_set_ctrlalt(dev, intf, CDC_NCM_COMM_ALTSETTING_MBIM); if (ret) goto err; ret = -ENODEV; } /* we will hit this for NCM/MBIM functions if prefer_mbim is false */ if (!cdc_ncm_comm_intf_is_mbim(intf->cur_altsetting)) goto err; ret = cdc_ncm_bind_common(dev, intf, data_altsetting, dev->driver_info->data); if (ret) goto err; ctx = info->ctx; /* The MBIM descriptor and the status endpoint are required */ if (ctx->mbim_desc && dev->status) subdriver = usb_cdc_wdm_register(ctx->control, &dev->status->desc, le16_to_cpu(ctx->mbim_desc->wMaxControlMessage), WWAN_PORT_MBIM, cdc_mbim_wdm_manage_power); if (IS_ERR(subdriver)) { ret = PTR_ERR(subdriver); cdc_ncm_unbind(dev, intf); goto err; } /* can't let usbnet use the interrupt endpoint */ dev->status = NULL; info->subdriver = subdriver; /* MBIM cannot do ARP */ dev->net->flags |= IFF_NOARP; /* no need to put the VLAN tci in the packet headers */ dev->net->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_FILTER; /* monitor VLAN additions and removals */ dev->net->netdev_ops = &cdc_mbim_netdev_ops; err: return ret; } static void cdc_mbim_unbind(struct usbnet *dev, struct usb_interface *intf) { struct cdc_mbim_state *info = (void *)&dev->data; struct cdc_ncm_ctx *ctx = info->ctx; /* disconnect subdriver from control interface */ if (info->subdriver && info->subdriver->disconnect) info->subdriver->disconnect(ctx->control); info->subdriver = NULL; /* let NCM unbind clean up both control and data interface */ cdc_ncm_unbind(dev, intf); } /* verify that the ethernet protocol is IPv4 or IPv6 */ static bool is_ip_proto(__be16 proto) { switch (proto) { case htons(ETH_P_IP): case htons(ETH_P_IPV6): return true; } return false; } static struct sk_buff *cdc_mbim_tx_fixup(struct usbnet *dev, struct sk_buff *skb, gfp_t flags) { struct sk_buff *skb_out; struct cdc_mbim_state *info = (void *)&dev->data; struct cdc_ncm_ctx *ctx = info->ctx; __le32 sign = cpu_to_le32(USB_CDC_MBIM_NDP16_IPS_SIGN); u16 tci = 0; bool is_ip; u8 *c; if (!ctx) goto error; if (skb) { if (skb->len <= ETH_HLEN) goto error; /* Some applications using e.g. packet sockets will * bypass the VLAN acceleration and create tagged * ethernet frames directly. We primarily look for * the accelerated out-of-band tag, but fall back if * required */ skb_reset_mac_header(skb); if (vlan_get_tag(skb, &tci) < 0 && skb->len > VLAN_ETH_HLEN && __vlan_get_tag(skb, &tci) == 0) { is_ip = is_ip_proto(vlan_eth_hdr(skb)->h_vlan_encapsulated_proto); skb_pull(skb, VLAN_ETH_HLEN); } else { is_ip = is_ip_proto(eth_hdr(skb)->h_proto); skb_pull(skb, ETH_HLEN); } /* Is IP session <0> tagged too? */ if (info->flags & FLAG_IPS0_VLAN) { /* drop all untagged packets */ if (!tci) goto error; /* map MBIM_IPS0_VID to IPS<0> */ if (tci == MBIM_IPS0_VID) tci = 0; } /* mapping VLANs to MBIM sessions: * no tag => IPS session <0> if !FLAG_IPS0_VLAN * 1 - 255 => IPS session <vlanid> * 256 - 511 => DSS session <vlanid - 256> * 512 - 4093 => unsupported, drop * 4094 => IPS session <0> if FLAG_IPS0_VLAN */ switch (tci & 0x0f00) { case 0x0000: /* VLAN ID 0 - 255 */ if (!is_ip) goto error; c = (u8 *)&sign; c[3] = tci; break; case 0x0100: /* VLAN ID 256 - 511 */ if (is_ip) goto error; sign = cpu_to_le32(USB_CDC_MBIM_NDP16_DSS_SIGN); c = (u8 *)&sign; c[3] = tci; break; default: netif_err(dev, tx_err, dev->net, "unsupported tci=0x%04x\n", tci); goto error; } } spin_lock_bh(&ctx->mtx); skb_out = cdc_ncm_fill_tx_frame(dev, skb, sign); spin_unlock_bh(&ctx->mtx); return skb_out; error: if (skb) dev_kfree_skb_any(skb); return NULL; } /* Some devices are known to send Neighbor Solicitation messages and * require Neighbor Advertisement replies. The IPv6 core will not * respond since IFF_NOARP is set, so we must handle them ourselves. */ static void do_neigh_solicit(struct usbnet *dev, u8 *buf, u16 tci) { struct ipv6hdr *iph = (void *)buf; struct nd_msg *msg = (void *)(iph + 1); struct net_device *netdev; struct inet6_dev *in6_dev; bool is_router; /* we'll only respond to requests from unicast addresses to * our solicited node addresses. */ if (!ipv6_addr_is_solict_mult(&iph->daddr) || !(ipv6_addr_type(&iph->saddr) & IPV6_ADDR_UNICAST)) return; /* need to send the NA on the VLAN dev, if any */ rcu_read_lock(); if (tci) { netdev = __vlan_find_dev_deep_rcu(dev->net, htons(ETH_P_8021Q), tci); if (!netdev) { rcu_read_unlock(); return; } } else { netdev = dev->net; } dev_hold(netdev); rcu_read_unlock(); in6_dev = in6_dev_get(netdev); if (!in6_dev) goto out; is_router = !!READ_ONCE(in6_dev->cnf.forwarding); in6_dev_put(in6_dev); /* ipv6_stub != NULL if in6_dev_get returned an inet6_dev */ ipv6_stub->ndisc_send_na(netdev, &iph->saddr, &msg->target, is_router /* router */, true /* solicited */, false /* override */, true /* inc_opt */); out: dev_put(netdev); } static bool is_neigh_solicit(u8 *buf, size_t len) { struct ipv6hdr *iph = (void *)buf; struct nd_msg *msg = (void *)(iph + 1); return (len >= sizeof(struct ipv6hdr) + sizeof(struct nd_msg) && iph->nexthdr == IPPROTO_ICMPV6 && msg->icmph.icmp6_code == 0 && msg->icmph.icmp6_type == NDISC_NEIGHBOUR_SOLICITATION); } static struct sk_buff *cdc_mbim_process_dgram(struct usbnet *dev, u8 *buf, size_t len, u16 tci) { __be16 proto = htons(ETH_P_802_3); struct sk_buff *skb = NULL; if (tci < 256 || tci == MBIM_IPS0_VID) { /* IPS session? */ if (len < sizeof(struct iphdr)) goto err; switch (*buf & 0xf0) { case 0x40: proto = htons(ETH_P_IP); break; case 0x60: if (is_neigh_solicit(buf, len)) do_neigh_solicit(dev, buf, tci); proto = htons(ETH_P_IPV6); break; default: goto err; } } skb = netdev_alloc_skb_ip_align(dev->net, len + ETH_HLEN); if (!skb) goto err; /* add an ethernet header */ skb_put(skb, ETH_HLEN); skb_reset_mac_header(skb); eth_hdr(skb)->h_proto = proto; eth_zero_addr(eth_hdr(skb)->h_source); memcpy(eth_hdr(skb)->h_dest, dev->net->dev_addr, ETH_ALEN); /* add datagram */ skb_put_data(skb, buf, len); /* map MBIM session to VLAN */ if (tci) __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), tci); err: return skb; } static int cdc_mbim_rx_fixup(struct usbnet *dev, struct sk_buff *skb_in) { struct sk_buff *skb; struct cdc_mbim_state *info = (void *)&dev->data; struct cdc_ncm_ctx *ctx = info->ctx; int len; int nframes; int x; int offset; struct usb_cdc_ncm_ndp16 *ndp16; struct usb_cdc_ncm_dpe16 *dpe16; int ndpoffset; int loopcount = 50; /* arbitrary max preventing infinite loop */ u32 payload = 0; u8 *c; u16 tci; ndpoffset = cdc_ncm_rx_verify_nth16(ctx, skb_in); if (ndpoffset < 0) goto error; next_ndp: nframes = cdc_ncm_rx_verify_ndp16(skb_in, ndpoffset); if (nframes < 0) goto error; ndp16 = (struct usb_cdc_ncm_ndp16 *)(skb_in->data + ndpoffset); switch (ndp16->dwSignature & cpu_to_le32(0x00ffffff)) { case cpu_to_le32(USB_CDC_MBIM_NDP16_IPS_SIGN): c = (u8 *)&ndp16->dwSignature; tci = c[3]; /* tag IPS<0> packets too if MBIM_IPS0_VID exists */ if (!tci && info->flags & FLAG_IPS0_VLAN) tci = MBIM_IPS0_VID; break; case cpu_to_le32(USB_CDC_MBIM_NDP16_DSS_SIGN): c = (u8 *)&ndp16->dwSignature; tci = c[3] + 256; break; default: netif_dbg(dev, rx_err, dev->net, "unsupported NDP signature <0x%08x>\n", le32_to_cpu(ndp16->dwSignature)); goto err_ndp; } dpe16 = ndp16->dpe16; for (x = 0; x < nframes; x++, dpe16++) { offset = le16_to_cpu(dpe16->wDatagramIndex); len = le16_to_cpu(dpe16->wDatagramLength); /* * CDC NCM ch. 3.7 * All entries after first NULL entry are to be ignored */ if ((offset == 0) || (len == 0)) { if (!x) goto err_ndp; /* empty NTB */ break; } /* sanity checking */ if (((offset + len) > skb_in->len) || (len > ctx->rx_max)) { netif_dbg(dev, rx_err, dev->net, "invalid frame detected (ignored) offset[%u]=%u, length=%u, skb=%p\n", x, offset, len, skb_in); if (!x) goto err_ndp; break; } else { skb = cdc_mbim_process_dgram(dev, skb_in->data + offset, len, tci); if (!skb) goto error; usbnet_skb_return(dev, skb); payload += len; /* count payload bytes in this NTB */ } } err_ndp: /* are there more NDPs to process? */ ndpoffset = le16_to_cpu(ndp16->wNextNdpIndex); if (ndpoffset && loopcount--) goto next_ndp; /* update stats */ ctx->rx_overhead += skb_in->len - payload; ctx->rx_ntbs++; return 1; error: return 0; } static int cdc_mbim_suspend(struct usb_interface *intf, pm_message_t message) { int ret = -ENODEV; struct usbnet *dev = usb_get_intfdata(intf); struct cdc_mbim_state *info = (void *)&dev->data; struct cdc_ncm_ctx *ctx = info->ctx; if (!ctx) goto error; /* * Both usbnet_suspend() and subdriver->suspend() MUST return 0 * in system sleep context, otherwise, the resume callback has * to recover device from previous suspend failure. */ ret = usbnet_suspend(intf, message); if (ret < 0) goto error; if (intf == ctx->control && info->subdriver && info->subdriver->suspend) ret = info->subdriver->suspend(intf, message); if (ret < 0) usbnet_resume(intf); error: return ret; } static int cdc_mbim_resume(struct usb_interface *intf) { int ret = 0; struct usbnet *dev = usb_get_intfdata(intf); struct cdc_mbim_state *info = (void *)&dev->data; struct cdc_ncm_ctx *ctx = info->ctx; bool callsub = (intf == ctx->control && info->subdriver && info->subdriver->resume); if (callsub) ret = info->subdriver->resume(intf); if (ret < 0) goto err; ret = usbnet_resume(intf); if (ret < 0 && callsub) info->subdriver->suspend(intf, PMSG_SUSPEND); err: return ret; } static const struct driver_info cdc_mbim_info = { .description = "CDC MBIM", .flags = FLAG_NO_SETINT | FLAG_MULTI_PACKET | FLAG_WWAN, .bind = cdc_mbim_bind, .unbind = cdc_mbim_unbind, .manage_power = cdc_mbim_manage_power, .rx_fixup = cdc_mbim_rx_fixup, .tx_fixup = cdc_mbim_tx_fixup, }; /* MBIM and NCM devices should not need a ZLP after NTBs with * dwNtbOutMaxSize length. Nevertheless, a number of devices from * different vendor IDs will fail unless we send ZLPs, forcing us * to make this the default. * * This default may cause a performance penalty for spec conforming * devices wanting to take advantage of optimizations possible without * ZLPs. A whitelist is added in an attempt to avoid this for devices * known to conform to the MBIM specification. * * All known devices supporting NCM compatibility mode are also * conforming to the NCM and MBIM specifications. For this reason, the * NCM subclass entry is also in the ZLP whitelist. */ static const struct driver_info cdc_mbim_info_zlp = { .description = "CDC MBIM", .flags = FLAG_NO_SETINT | FLAG_MULTI_PACKET | FLAG_WWAN | FLAG_SEND_ZLP, .bind = cdc_mbim_bind, .unbind = cdc_mbim_unbind, .manage_power = cdc_mbim_manage_power, .rx_fixup = cdc_mbim_rx_fixup, .tx_fixup = cdc_mbim_tx_fixup, }; /* The spefication explicitly allows NDPs to be placed anywhere in the * frame, but some devices fail unless the NDP is placed after the IP * packets. Using the CDC_NCM_FLAG_NDP_TO_END flags to force this * behaviour. * * Note: The current implementation of this feature restricts each NTB * to a single NDP, implying that multiplexed sessions cannot share an * NTB. This might affect performance for multiplexed sessions. */ static const struct driver_info cdc_mbim_info_ndp_to_end = { .description = "CDC MBIM", .flags = FLAG_NO_SETINT | FLAG_MULTI_PACKET | FLAG_WWAN, .bind = cdc_mbim_bind, .unbind = cdc_mbim_unbind, .manage_power = cdc_mbim_manage_power, .rx_fixup = cdc_mbim_rx_fixup, .tx_fixup = cdc_mbim_tx_fixup, .data = CDC_NCM_FLAG_NDP_TO_END, }; /* Some modems (e.g. Telit LE922A6) do not work properly with altsetting * toggle done in cdc_ncm_bind_common. CDC_MBIM_FLAG_AVOID_ALTSETTING_TOGGLE * flag is used to avoid this procedure. */ static const struct driver_info cdc_mbim_info_avoid_altsetting_toggle = { .description = "CDC MBIM", .flags = FLAG_NO_SETINT | FLAG_MULTI_PACKET | FLAG_WWAN | FLAG_SEND_ZLP, .bind = cdc_mbim_bind, .unbind = cdc_mbim_unbind, .manage_power = cdc_mbim_manage_power, .rx_fixup = cdc_mbim_rx_fixup, .tx_fixup = cdc_mbim_tx_fixup, .data = CDC_MBIM_FLAG_AVOID_ALTSETTING_TOGGLE, }; static const struct usb_device_id mbim_devs[] = { /* This duplicate NCM entry is intentional. MBIM devices can * be disguised as NCM by default, and this is necessary to * allow us to bind the correct driver_info to such devices. * * bind() will sort out this for us, selecting the correct * entry and reject the other */ { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_NCM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info, }, /* ZLP conformance whitelist: All Ericsson MBIM devices */ { USB_VENDOR_AND_INTERFACE_INFO(0x0bdb, USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info, }, /* Some Huawei devices, ME906s-158 (12d1:15c1) and E3372 * (12d1:157d), are known to fail unless the NDP is placed * after the IP packets. Applying the quirk to all Huawei * devices is broader than necessary, but harmless. */ { USB_VENDOR_AND_INTERFACE_INFO(0x12d1, USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info_ndp_to_end, }, /* The HP lt4132 (03f0:a31d) is a rebranded Huawei ME906s-158, * therefore it too requires the above "NDP to end" quirk. */ { USB_DEVICE_AND_INTERFACE_INFO(0x03f0, 0xa31d, USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info_ndp_to_end, }, /* Telit LE922A6 in MBIM composition */ { USB_DEVICE_AND_INTERFACE_INFO(0x1bc7, 0x1041, USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info_avoid_altsetting_toggle, }, /* Telit LN920 */ { USB_DEVICE_AND_INTERFACE_INFO(0x1bc7, 0x1061, USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info_avoid_altsetting_toggle, }, /* Telit FN990A */ { USB_DEVICE_AND_INTERFACE_INFO(0x1bc7, 0x1071, USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info_avoid_altsetting_toggle, }, /* Telit FE990A */ { USB_DEVICE_AND_INTERFACE_INFO(0x1bc7, 0x1081, USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info_avoid_altsetting_toggle, }, /* default entry */ { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_MBIM, USB_CDC_PROTO_NONE), .driver_info = (unsigned long)&cdc_mbim_info_zlp, }, { }, }; MODULE_DEVICE_TABLE(usb, mbim_devs); static struct usb_driver cdc_mbim_driver = { .name = "cdc_mbim", .id_table = mbim_devs, .probe = usbnet_probe, .disconnect = usbnet_disconnect, .suspend = cdc_mbim_suspend, .resume = cdc_mbim_resume, .reset_resume = cdc_mbim_resume, .supports_autosuspend = 1, .disable_hub_initiated_lpm = 1, }; module_usb_driver(cdc_mbim_driver); MODULE_AUTHOR("Greg Suarez <gsuarez@smithmicro.com>"); MODULE_AUTHOR("Bjørn Mork <bjorn@mork.no>"); MODULE_DESCRIPTION("USB CDC MBIM host driver"); MODULE_LICENSE("GPL"); |
| 4 1 1 1 1 1 1 1 4 4 1 4 1 4 4 1 4 3 118 119 39 39 101 102 29 29 468 66 11 59 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 | // SPDX-License-Identifier: GPL-2.0-or-later /* * Cryptographic API for algorithms (i.e., low-level API). * * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au> */ #include <crypto/algapi.h> #include <linux/err.h> #include <linux/errno.h> #include <linux/fips.h> #include <linux/init.h> #include <linux/kernel.h> #include <linux/list.h> #include <linux/module.h> #include <linux/rtnetlink.h> #include <linux/slab.h> #include <linux/string.h> #include <linux/workqueue.h> #include "internal.h" static LIST_HEAD(crypto_template_list); static inline void crypto_check_module_sig(struct module *mod) { if (fips_enabled && mod && !module_sig_ok(mod)) panic("Module %s signature verification failed in FIPS mode\n", module_name(mod)); } static int crypto_check_alg(struct crypto_alg *alg) { crypto_check_module_sig(alg->cra_module); if (!alg->cra_name[0] || !alg->cra_driver_name[0]) return -EINVAL; if (alg->cra_alignmask & (alg->cra_alignmask + 1)) return -EINVAL; /* General maximums for all algs. */ if (alg->cra_alignmask > MAX_ALGAPI_ALIGNMASK) return -EINVAL; if (alg->cra_blocksize > MAX_ALGAPI_BLOCKSIZE) return -EINVAL; /* Lower maximums for specific alg types. */ if (!alg->cra_type && (alg->cra_flags & CRYPTO_ALG_TYPE_MASK) == CRYPTO_ALG_TYPE_CIPHER) { if (alg->cra_alignmask > MAX_CIPHER_ALIGNMASK) return -EINVAL; if (alg->cra_blocksize > MAX_CIPHER_BLOCKSIZE) return -EINVAL; } if (alg->cra_priority < 0) return -EINVAL; refcount_set(&alg->cra_refcnt, 1); return 0; } static void crypto_free_instance(struct crypto_instance *inst) { inst->alg.cra_type->free(inst); } static void crypto_destroy_instance_workfn(struct work_struct *w) { struct crypto_template *tmpl = container_of(w, struct crypto_template, free_work); struct crypto_instance *inst; struct hlist_node *n; HLIST_HEAD(list); down_write(&crypto_alg_sem); hlist_for_each_entry_safe(inst, n, &tmpl->dead, list) { if (refcount_read(&inst->alg.cra_refcnt) != -1) continue; hlist_del(&inst->list); hlist_add_head(&inst->list, &list); } up_write(&crypto_alg_sem); hlist_for_each_entry_safe(inst, n, &list, list) crypto_free_instance(inst); } static void crypto_destroy_instance(struct crypto_alg *alg) { struct crypto_instance *inst = container_of(alg, struct crypto_instance, alg); struct crypto_template *tmpl = inst->tmpl; refcount_set(&alg->cra_refcnt, -1); schedule_work(&tmpl->free_work); } /* * This function adds a spawn to the list secondary_spawns which * will be used at the end of crypto_remove_spawns to unregister * instances, unless the spawn happens to be one that is depended * on by the new algorithm (nalg in crypto_remove_spawns). * * This function is also responsible for resurrecting any algorithms * in the dependency chain of nalg by unsetting n->dead. */ static struct list_head *crypto_more_spawns(struct crypto_alg *alg, struct list_head *stack, struct list_head *top, struct list_head *secondary_spawns) { struct crypto_spawn *spawn, *n; spawn = list_first_entry_or_null(stack, struct crypto_spawn, list); if (!spawn) return NULL; n = list_prev_entry(spawn, list); list_move(&spawn->list, secondary_spawns); if (list_is_last(&n->list, stack)) return top; n = list_next_entry(n, list); if (!spawn->dead) n->dead = false; return &n->inst->alg.cra_users; } static void crypto_remove_instance(struct crypto_instance *inst, struct list_head *list) { struct crypto_template *tmpl = inst->tmpl; if (crypto_is_dead(&inst->alg)) return; inst->alg.cra_flags |= CRYPTO_ALG_DEAD; if (!tmpl) return; list_del_init(&inst->alg.cra_list); hlist_del(&inst->list); hlist_add_head(&inst->list, &tmpl->dead); BUG_ON(!list_empty(&inst->alg.cra_users)); crypto_alg_put(&inst->alg); } /* * Given an algorithm alg, remove all algorithms that depend on it * through spawns. If nalg is not null, then exempt any algorithms * that is depended on by nalg. This is useful when nalg itself * depends on alg. */ void crypto_remove_spawns(struct crypto_alg *alg, struct list_head *list, struct crypto_alg *nalg) { u32 new_type = (nalg ?: alg)->cra_flags; struct crypto_spawn *spawn, *n; LIST_HEAD(secondary_spawns); struct list_head *spawns; LIST_HEAD(stack); LIST_HEAD(top); spawns = &alg->cra_users; list_for_each_entry_safe(spawn, n, spawns, list) { if ((spawn->alg->cra_flags ^ new_type) & spawn->mask) continue; list_move(&spawn->list, &top); } /* * Perform a depth-first walk starting from alg through * the cra_users tree. The list stack records the path * from alg to the current spawn. */ spawns = ⊤ do { while (!list_empty(spawns)) { struct crypto_instance *inst; spawn = list_first_entry(spawns, struct crypto_spawn, list); inst = spawn->inst; list_move(&spawn->list, &stack); spawn->dead = !spawn->registered || &inst->alg != nalg; if (!spawn->registered) break; BUG_ON(&inst->alg == alg); if (&inst->alg == nalg) break; spawns = &inst->alg.cra_users; /* * Even if spawn->registered is true, the * instance itself may still be unregistered. * This is because it may have failed during * registration. Therefore we still need to * make the following test. * * We may encounter an unregistered instance here, since * an instance's spawns are set up prior to the instance * being registered. An unregistered instance will have * NULL ->cra_users.next, since ->cra_users isn't * properly initialized until registration. But an * unregistered instance cannot have any users, so treat * it the same as ->cra_users being empty. */ if (spawns->next == NULL) break; } } while ((spawns = crypto_more_spawns(alg, &stack, &top, &secondary_spawns))); /* * Remove all instances that are marked as dead. Also * complete the resurrection of the others by moving them * back to the cra_users list. */ list_for_each_entry_safe(spawn, n, &secondary_spawns, list) { if (!spawn->dead) list_move(&spawn->list, &spawn->alg->cra_users); else if (spawn->registered) crypto_remove_instance(spawn->inst, list); } } EXPORT_SYMBOL_GPL(crypto_remove_spawns); static void crypto_alg_finish_registration(struct crypto_alg *alg, struct list_head *algs_to_put) __must_hold(&crypto_alg_sem) { struct crypto_alg *q; list_for_each_entry(q, &crypto_alg_list, cra_list) { if (q == alg) continue; if (crypto_is_moribund(q)) continue; if (crypto_is_larval(q)) continue; if (strcmp(alg->cra_name, q->cra_name)) continue; if (strcmp(alg->cra_driver_name, q->cra_driver_name) && q->cra_priority > alg->cra_priority) continue; crypto_remove_spawns(q, algs_to_put, alg); } crypto_notify(CRYPTO_MSG_ALG_LOADED, alg); } static struct crypto_larval *crypto_alloc_test_larval(struct crypto_alg *alg) { struct crypto_larval *larval; if (!IS_ENABLED(CONFIG_CRYPTO_SELFTESTS) || (alg->cra_flags & CRYPTO_ALG_INTERNAL)) return NULL; /* No self-test needed */ larval = crypto_larval_alloc(alg->cra_name, alg->cra_flags | CRYPTO_ALG_TESTED, 0); if (IS_ERR(larval)) return larval; larval->adult = crypto_mod_get(alg); if (!larval->adult) { kfree(larval); return ERR_PTR(-ENOENT); } refcount_set(&larval->alg.cra_refcnt, 1); memcpy(larval->alg.cra_driver_name, alg->cra_driver_name, CRYPTO_MAX_ALG_NAME); larval->alg.cra_priority = alg->cra_priority; return larval; } static struct crypto_larval * __crypto_register_alg(struct crypto_alg *alg, struct list_head *algs_to_put) __must_hold(&crypto_alg_sem) { struct crypto_alg *q; struct crypto_larval *larval; int ret = -EAGAIN; if (crypto_is_dead(alg)) goto err; INIT_LIST_HEAD(&alg->cra_users); ret = -EEXIST; list_for_each_entry(q, &crypto_alg_list, cra_list) { if (q == alg) goto err; if (crypto_is_moribund(q)) continue; if (crypto_is_larval(q)) { if (!strcmp(alg->cra_driver_name, q->cra_driver_name)) goto err; continue; } if (!strcmp(q->cra_driver_name, alg->cra_name) || !strcmp(q->cra_driver_name, alg->cra_driver_name) || !strcmp(q->cra_name, alg->cra_driver_name)) goto err; } larval = crypto_alloc_test_larval(alg); if (IS_ERR(larval)) goto out; list_add(&alg->cra_list, &crypto_alg_list); if (larval) { /* No cheating! */ alg->cra_flags &= ~CRYPTO_ALG_TESTED; list_add(&larval->alg.cra_list, &crypto_alg_list); } else { alg->cra_flags |= CRYPTO_ALG_TESTED; crypto_alg_finish_registration(alg, algs_to_put); } out: return larval; err: larval = ERR_PTR(ret); goto out; } void crypto_alg_tested(const char *name, int err) { struct crypto_larval *test; struct crypto_alg *alg; struct crypto_alg *q; LIST_HEAD(list); down_write(&crypto_alg_sem); list_for_each_entry(q, &crypto_alg_list, cra_list) { if (crypto_is_moribund(q) || !crypto_is_larval(q)) continue; test = (struct crypto_larval *)q; if (!strcmp(q->cra_driver_name, name)) goto found; } pr_err("alg: Unexpected test result for %s: %d\n", name, err); up_write(&crypto_alg_sem); return; found: q->cra_flags |= CRYPTO_ALG_DEAD; alg = test->adult; if (crypto_is_dead(alg)) goto complete; if (err == -ECANCELED) alg->cra_flags |= CRYPTO_ALG_FIPS_INTERNAL; else if (err) goto complete; else alg->cra_flags &= ~CRYPTO_ALG_FIPS_INTERNAL; alg->cra_flags |= CRYPTO_ALG_TESTED; crypto_alg_finish_registration(alg, &list); complete: list_del_init(&test->alg.cra_list); complete_all(&test->completion); up_write(&crypto_alg_sem); crypto_alg_put(&test->alg); crypto_remove_final(&list); } EXPORT_SYMBOL_GPL(crypto_alg_tested); void crypto_remove_final(struct list_head *list) { struct crypto_alg *alg; struct crypto_alg *n; list_for_each_entry_safe(alg, n, list, cra_list) { list_del_init(&alg->cra_list); crypto_alg_put(alg); } } EXPORT_SYMBOL_GPL(crypto_remove_final); static void crypto_free_alg(struct crypto_alg *alg) { unsigned int algsize = alg->cra_type->algsize; u8 *p = (u8 *)alg - algsize; crypto_destroy_alg(alg); kfree(p); } int crypto_register_alg(struct crypto_alg *alg) { struct crypto_larval *larval; bool test_started = false; LIST_HEAD(algs_to_put); int err; alg->cra_flags &= ~CRYPTO_ALG_DEAD; err = crypto_check_alg(alg); if (err) return err; if (alg->cra_flags & CRYPTO_ALG_DUP_FIRST && !WARN_ON_ONCE(alg->cra_destroy)) { unsigned int algsize = alg->cra_type->algsize; u8 *p = (u8 *)alg - algsize; p = kmemdup(p, algsize + sizeof(*alg), GFP_KERNEL); if (!p) return -ENOMEM; alg = (void *)(p + algsize); alg->cra_destroy = crypto_free_alg; } down_write(&crypto_alg_sem); larval = __crypto_register_alg(alg, &algs_to_put); if (!IS_ERR_OR_NULL(larval)) { test_started = crypto_boot_test_finished(); larval->test_started = test_started; } up_write(&crypto_alg_sem); if (IS_ERR(larval)) { crypto_alg_put(alg); return PTR_ERR(larval); } if (test_started) crypto_schedule_test(larval); else crypto_remove_final(&algs_to_put); return 0; } EXPORT_SYMBOL_GPL(crypto_register_alg); static int crypto_remove_alg(struct crypto_alg *alg, struct list_head *list) { if (unlikely(list_empty(&alg->cra_list))) return -ENOENT; alg->cra_flags |= CRYPTO_ALG_DEAD; list_del_init(&alg->cra_list); crypto_remove_spawns(alg, list, NULL); return 0; } void crypto_unregister_alg(struct crypto_alg *alg) { int ret; LIST_HEAD(list); down_write(&crypto_alg_sem); ret = crypto_remove_alg(alg, &list); up_write(&crypto_alg_sem); if (WARN(ret, "Algorithm %s is not registered", alg->cra_driver_name)) return; WARN_ON(!alg->cra_destroy && refcount_read(&alg->cra_refcnt) != 1); list_add(&alg->cra_list, &list); crypto_remove_final(&list); } EXPORT_SYMBOL_GPL(crypto_unregister_alg); int crypto_register_algs(struct crypto_alg *algs, int count) { int i, ret; for (i = 0; i < count; i++) { ret = crypto_register_alg(&algs[i]); if (ret) { crypto_unregister_algs(algs, i); return ret; } } return 0; } EXPORT_SYMBOL_GPL(crypto_register_algs); void crypto_unregister_algs(struct crypto_alg *algs, int count) { int i; for (i = count - 1; i >= 0; --i) crypto_unregister_alg(&algs[i]); } EXPORT_SYMBOL_GPL(crypto_unregister_algs); int crypto_register_template(struct crypto_template *tmpl) { struct crypto_template *q; int err = -EEXIST; INIT_WORK(&tmpl->free_work, crypto_destroy_instance_workfn); down_write(&crypto_alg_sem); crypto_check_module_sig(tmpl->module); list_for_each_entry(q, &crypto_template_list, list) { if (q == tmpl) goto out; } list_add(&tmpl->list, &crypto_template_list); err = 0; out: up_write(&crypto_alg_sem); return err; } EXPORT_SYMBOL_GPL(crypto_register_template); int crypto_register_templates(struct crypto_template *tmpls, int count) { int i, err; for (i = 0; i < count; i++) { err = crypto_register_template(&tmpls[i]); if (err) goto out; } return 0; out: for (--i; i >= 0; --i) crypto_unregister_template(&tmpls[i]); return err; } EXPORT_SYMBOL_GPL(crypto_register_templates); void crypto_unregister_template(struct crypto_template *tmpl) { struct crypto_instance *inst; struct hlist_node *n; struct hlist_head *list; LIST_HEAD(users); down_write(&crypto_alg_sem); BUG_ON(list_empty(&tmpl->list)); list_del_init(&tmpl->list); list = &tmpl->instances; hlist_for_each_entry(inst, list, list) { int err = crypto_remove_alg(&inst->alg, &users); BUG_ON(err); } up_write(&crypto_alg_sem); hlist_for_each_entry_safe(inst, n, list, list) { BUG_ON(refcount_read(&inst->alg.cra_refcnt) != 1); crypto_free_instance(inst); } crypto_remove_final(&users); flush_work(&tmpl->free_work); } EXPORT_SYMBOL_GPL(crypto_unregister_template); void crypto_unregister_templates(struct crypto_template *tmpls, int count) { int i; for (i = count - 1; i >= 0; --i) crypto_unregister_template(&tmpls[i]); } EXPORT_SYMBOL_GPL(crypto_unregister_templates); static struct crypto_template *__crypto_lookup_template(const char *name) { struct crypto_template *q, *tmpl = NULL; down_read(&crypto_alg_sem); list_for_each_entry(q, &crypto_template_list, list) { if (strcmp(q->name, name)) continue; if (unlikely(!crypto_tmpl_get(q))) continue; tmpl = q; break; } up_read(&crypto_alg_sem); return tmpl; } struct crypto_template *crypto_lookup_template(const char *name) { return try_then_request_module(__crypto_lookup_template(name), "crypto-%s", name); } EXPORT_SYMBOL_GPL(crypto_lookup_template); int crypto_register_instance(struct crypto_template *tmpl, struct crypto_instance *inst) { struct crypto_larval *larval; struct crypto_spawn *spawn; u32 fips_internal = 0; LIST_HEAD(algs_to_put); int err; err = crypto_check_alg(&inst->alg); if (err) return err; inst->alg.cra_module = tmpl->module; inst->alg.cra_flags |= CRYPTO_ALG_INSTANCE; inst->alg.cra_destroy = crypto_destroy_instance; down_write(&crypto_alg_sem); larval = ERR_PTR(-EAGAIN); for (spawn = inst->spawns; spawn;) { struct crypto_spawn *next; if (spawn->dead) goto unlock; next = spawn->next; spawn->inst = inst; spawn->registered = true; fips_internal |= spawn->alg->cra_flags; crypto_mod_put(spawn->alg); spawn = next; } inst->alg.cra_flags |= (fips_internal & CRYPTO_ALG_FIPS_INTERNAL); larval = __crypto_register_alg(&inst->alg, &algs_to_put); if (IS_ERR(larval)) goto unlock; else if (larval) larval->test_started = true; hlist_add_head(&inst->list, &tmpl->instances); inst->tmpl = tmpl; unlock: up_write(&crypto_alg_sem); if (IS_ERR(larval)) return PTR_ERR(larval); if (larval) crypto_schedule_test(larval); else crypto_remove_final(&algs_to_put); return 0; } EXPORT_SYMBOL_GPL(crypto_register_instance); void crypto_unregister_instance(struct crypto_instance *inst) { LIST_HEAD(list); down_write(&crypto_alg_sem); crypto_remove_spawns(&inst->alg, &list, NULL); crypto_remove_instance(inst, &list); up_write(&crypto_alg_sem); crypto_remove_final(&list); } EXPORT_SYMBOL_GPL(crypto_unregister_instance); int crypto_grab_spawn(struct crypto_spawn *spawn, struct crypto_instance *inst, const char *name, u32 type, u32 mask) { struct crypto_alg *alg; int err = -EAGAIN; if (WARN_ON_ONCE(inst == NULL)) return -EINVAL; /* Allow the result of crypto_attr_alg_name() to be passed directly */ if (IS_ERR(name)) return PTR_ERR(name); alg = crypto_find_alg(name, spawn->frontend, type | CRYPTO_ALG_FIPS_INTERNAL, mask); if (IS_ERR(alg)) return PTR_ERR(alg); down_write(&crypto_alg_sem); if (!crypto_is_moribund(alg)) { list_add(&spawn->list, &alg->cra_users); spawn->alg = alg; spawn->mask = mask; spawn->next = inst->spawns; inst->spawns = spawn; inst->alg.cra_flags |= (alg->cra_flags & CRYPTO_ALG_INHERITED_FLAGS); err = 0; } up_write(&crypto_alg_sem); if (err) crypto_mod_put(alg); return err; } EXPORT_SYMBOL_GPL(crypto_grab_spawn); void crypto_drop_spawn(struct crypto_spawn *spawn) { if (!spawn->alg) /* not yet initialized? */ return; down_write(&crypto_alg_sem); if (!spawn->dead) list_del(&spawn->list); up_write(&crypto_alg_sem); if (!spawn->registered) crypto_mod_put(spawn->alg); } EXPORT_SYMBOL_GPL(crypto_drop_spawn); static struct crypto_alg *crypto_spawn_alg(struct crypto_spawn *spawn) { struct crypto_alg *alg = ERR_PTR(-EAGAIN); struct crypto_alg *target; bool shoot = false; down_read(&crypto_alg_sem); if (!spawn->dead) { alg = spawn->alg; if (!crypto_mod_get(alg)) { target = crypto_alg_get(alg); shoot = true; alg = ERR_PTR(-EAGAIN); } } up_read(&crypto_alg_sem); if (shoot) { crypto_shoot_alg(target); crypto_alg_put(target); } return alg; } struct crypto_tfm *crypto_spawn_tfm(struct crypto_spawn *spawn, u32 type, u32 mask) { struct crypto_alg *alg; struct crypto_tfm *tfm; alg = crypto_spawn_alg(spawn); if (IS_ERR(alg)) return ERR_CAST(alg); tfm = ERR_PTR(-EINVAL); if (unlikely((alg->cra_flags ^ type) & mask)) goto out_put_alg; tfm = __crypto_alloc_tfm(alg, type, mask); if (IS_ERR(tfm)) goto out_put_alg; return tfm; out_put_alg: crypto_mod_put(alg); return tfm; } EXPORT_SYMBOL_GPL(crypto_spawn_tfm); void *crypto_spawn_tfm2(struct crypto_spawn *spawn) { struct crypto_alg *alg; struct crypto_tfm *tfm; alg = crypto_spawn_alg(spawn); if (IS_ERR(alg)) return ERR_CAST(alg); tfm = crypto_create_tfm(alg, spawn->frontend); if (IS_ERR(tfm)) goto out_put_alg; return tfm; out_put_alg: crypto_mod_put(alg); return tfm; } EXPORT_SYMBOL_GPL(crypto_spawn_tfm2); int crypto_register_notifier(struct notifier_block *nb) { return blocking_notifier_chain_register(&crypto_chain, nb); } EXPORT_SYMBOL_GPL(crypto_register_notifier); int crypto_unregister_notifier(struct notifier_block *nb) { return blocking_notifier_chain_unregister(&crypto_chain, nb); } EXPORT_SYMBOL_GPL(crypto_unregister_notifier); struct crypto_attr_type *crypto_get_attr_type(struct rtattr **tb) { struct rtattr *rta = tb[0]; struct crypto_attr_type *algt; if (!rta) return ERR_PTR(-ENOENT); if (RTA_PAYLOAD(rta) < sizeof(*algt)) return ERR_PTR(-EINVAL); if (rta->rta_type != CRYPTOA_TYPE) return ERR_PTR(-EINVAL); algt = RTA_DATA(rta); return algt; } EXPORT_SYMBOL_GPL(crypto_get_attr_type); /** * crypto_check_attr_type() - check algorithm type and compute inherited mask * @tb: the template parameters * @type: the algorithm type the template would be instantiated as * @mask_ret: (output) the mask that should be passed to crypto_grab_*() * to restrict the flags of any inner algorithms * * Validate that the algorithm type the user requested is compatible with the * one the template would actually be instantiated as. E.g., if the user is * doing crypto_alloc_shash("cbc(aes)", ...), this would return an error because * the "cbc" template creates an "skcipher" algorithm, not an "shash" algorithm. * * Also compute the mask to use to restrict the flags of any inner algorithms. * * Return: 0 on success; -errno on failure */ int crypto_check_attr_type(struct rtattr **tb, u32 type, u32 *mask_ret) { struct crypto_attr_type *algt; algt = crypto_get_attr_type(tb); if (IS_ERR(algt)) return PTR_ERR(algt); if ((algt->type ^ type) & algt->mask) return -EINVAL; *mask_ret = crypto_algt_inherited_mask(algt); return 0; } EXPORT_SYMBOL_GPL(crypto_check_attr_type); const char *crypto_attr_alg_name(struct rtattr *rta) { struct crypto_attr_alg *alga; if (!rta) return ERR_PTR(-ENOENT); if (RTA_PAYLOAD(rta) < sizeof(*alga)) return ERR_PTR(-EINVAL); if (rta->rta_type != CRYPTOA_ALG) return ERR_PTR(-EINVAL); alga = RTA_DATA(rta); alga->name[CRYPTO_MAX_ALG_NAME - 1] = 0; return alga->name; } EXPORT_SYMBOL_GPL(crypto_attr_alg_name); int __crypto_inst_setname(struct crypto_instance *inst, const char *name, const char *driver, struct crypto_alg *alg) { if (snprintf(inst->alg.cra_name, CRYPTO_MAX_ALG_NAME, "%s(%s)", name, alg->cra_name) >= CRYPTO_MAX_ALG_NAME) return -ENAMETOOLONG; if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME, "%s(%s)", driver, alg->cra_driver_name) >= CRYPTO_MAX_ALG_NAME) return -ENAMETOOLONG; return 0; } EXPORT_SYMBOL_GPL(__crypto_inst_setname); void crypto_init_queue(struct crypto_queue *queue, unsigned int max_qlen) { INIT_LIST_HEAD(&queue->list); queue->backlog = &queue->list; queue->qlen = 0; queue->max_qlen = max_qlen; } EXPORT_SYMBOL_GPL(crypto_init_queue); int crypto_enqueue_request(struct crypto_queue *queue, struct crypto_async_request *request) { int err = -EINPROGRESS; if (unlikely(queue->qlen >= queue->max_qlen)) { if (!(request->flags & CRYPTO_TFM_REQ_MAY_BACKLOG)) { err = -ENOSPC; goto out; } err = -EBUSY; if (queue->backlog == &queue->list) queue->backlog = &request->list; } queue->qlen++; list_add_tail(&request->list, &queue->list); out: return err; } EXPORT_SYMBOL_GPL(crypto_enqueue_request); void crypto_enqueue_request_head(struct crypto_queue *queue, struct crypto_async_request *request) { if (unlikely(queue->qlen >= queue->max_qlen)) queue->backlog = queue->backlog->prev; queue->qlen++; list_add(&request->list, &queue->list); } EXPORT_SYMBOL_GPL(crypto_enqueue_request_head); struct crypto_async_request *crypto_dequeue_request(struct crypto_queue *queue) { struct list_head *request; if (unlikely(!queue->qlen)) return NULL; queue->qlen--; if (queue->backlog != &queue->list) queue->backlog = queue->backlog->next; request = queue->list.next; list_del_init(request); return list_entry(request, struct crypto_async_request, list); } EXPORT_SYMBOL_GPL(crypto_dequeue_request); static inline void crypto_inc_byte(u8 *a, unsigned int size) { u8 *b = (a + size); u8 c; for (; size; size--) { c = *--b + 1; *b = c; if (c) break; } } void crypto_inc(u8 *a, unsigned int size) { __be32 *b = (__be32 *)(a + size); u32 c; if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) || IS_ALIGNED((unsigned long)b, __alignof__(*b))) for (; size >= 4; size -= 4) { c = be32_to_cpu(*--b) + 1; *b = cpu_to_be32(c); if (likely(c)) return; } crypto_inc_byte(a, size); } EXPORT_SYMBOL_GPL(crypto_inc); unsigned int crypto_alg_extsize(struct crypto_alg *alg) { return alg->cra_ctxsize + (alg->cra_alignmask & ~(crypto_tfm_ctx_alignment() - 1)); } EXPORT_SYMBOL_GPL(crypto_alg_extsize); int crypto_type_has_alg(const char *name, const struct crypto_type *frontend, u32 type, u32 mask) { int ret = 0; struct crypto_alg *alg = crypto_find_alg(name, frontend, type, mask); if (!IS_ERR(alg)) { crypto_mod_put(alg); ret = 1; } return ret; } EXPORT_SYMBOL_GPL(crypto_type_has_alg); static void __init crypto_start_tests(void) { if (!IS_BUILTIN(CONFIG_CRYPTO_ALGAPI)) return; if (!IS_ENABLED(CONFIG_CRYPTO_SELFTESTS)) return; set_crypto_boot_test_finished(); for (;;) { struct crypto_larval *larval = NULL; struct crypto_alg *q; down_write(&crypto_alg_sem); list_for_each_entry(q, &crypto_alg_list, cra_list) { struct crypto_larval *l; if (!crypto_is_larval(q)) continue; l = (void *)q; if (!crypto_is_test_larval(l)) continue; if (l->test_started) continue; l->test_started = true; larval = l; break; } up_write(&crypto_alg_sem); if (!larval) break; crypto_schedule_test(larval); } } static int __init crypto_algapi_init(void) { crypto_init_proc(); crypto_start_tests(); return 0; } static void __exit crypto_algapi_exit(void) { crypto_exit_proc(); } /* * We run this at late_initcall so that all the built-in algorithms * have had a chance to register themselves first. */ late_initcall(crypto_algapi_init); module_exit(crypto_algapi_exit); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("Cryptographic algorithms API"); MODULE_SOFTDEP("pre: cryptomgr"); |
| 8 8 8 8 8 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 | /* SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause */ /* ****************************************************************** * FSE : Finite State Entropy codec * Public Prototypes declaration * Copyright (c) Meta Platforms, Inc. and affiliates. * * You can contact the author at : * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy * * This source code is licensed under both the BSD-style license (found in the * LICENSE file in the root directory of this source tree) and the GPLv2 (found * in the COPYING file in the root directory of this source tree). * You may select, at your option, one of the above-listed licenses. ****************************************************************** */ #ifndef FSE_H #define FSE_H /*-***************************************** * Dependencies ******************************************/ #include "zstd_deps.h" /* size_t, ptrdiff_t */ /*-***************************************** * FSE_PUBLIC_API : control library symbols visibility ******************************************/ #if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4) # define FSE_PUBLIC_API __attribute__ ((visibility ("default"))) #elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */ # define FSE_PUBLIC_API __declspec(dllexport) #elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1) # define FSE_PUBLIC_API __declspec(dllimport) /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/ #else # define FSE_PUBLIC_API #endif /*------ Version ------*/ #define FSE_VERSION_MAJOR 0 #define FSE_VERSION_MINOR 9 #define FSE_VERSION_RELEASE 0 #define FSE_LIB_VERSION FSE_VERSION_MAJOR.FSE_VERSION_MINOR.FSE_VERSION_RELEASE #define FSE_QUOTE(str) #str #define FSE_EXPAND_AND_QUOTE(str) FSE_QUOTE(str) #define FSE_VERSION_STRING FSE_EXPAND_AND_QUOTE(FSE_LIB_VERSION) #define FSE_VERSION_NUMBER (FSE_VERSION_MAJOR *100*100 + FSE_VERSION_MINOR *100 + FSE_VERSION_RELEASE) FSE_PUBLIC_API unsigned FSE_versionNumber(void); /*< library version number; to be used when checking dll version */ /*-***************************************** * Tool functions ******************************************/ FSE_PUBLIC_API size_t FSE_compressBound(size_t size); /* maximum compressed size */ /* Error Management */ FSE_PUBLIC_API unsigned FSE_isError(size_t code); /* tells if a return value is an error code */ FSE_PUBLIC_API const char* FSE_getErrorName(size_t code); /* provides error code string (useful for debugging) */ /*-***************************************** * FSE detailed API ******************************************/ /*! FSE_compress() does the following: 1. count symbol occurrence from source[] into table count[] (see hist.h) 2. normalize counters so that sum(count[]) == Power_of_2 (2^tableLog) 3. save normalized counters to memory buffer using writeNCount() 4. build encoding table 'CTable' from normalized counters 5. encode the data stream using encoding table 'CTable' FSE_decompress() does the following: 1. read normalized counters with readNCount() 2. build decoding table 'DTable' from normalized counters 3. decode the data stream using decoding table 'DTable' The following API allows targeting specific sub-functions for advanced tasks. For example, it's possible to compress several blocks using the same 'CTable', or to save and provide normalized distribution using external method. */ /* *** COMPRESSION *** */ /*! FSE_optimalTableLog(): dynamically downsize 'tableLog' when conditions are met. It saves CPU time, by using smaller tables, while preserving or even improving compression ratio. @return : recommended tableLog (necessarily <= 'maxTableLog') */ FSE_PUBLIC_API unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue); /*! FSE_normalizeCount(): normalize counts so that sum(count[]) == Power_of_2 (2^tableLog) 'normalizedCounter' is a table of short, of minimum size (maxSymbolValue+1). useLowProbCount is a boolean parameter which trades off compressed size for faster header decoding. When it is set to 1, the compressed data will be slightly smaller. And when it is set to 0, FSE_readNCount() and FSE_buildDTable() will be faster. If you are compressing a small amount of data (< 2 KB) then useLowProbCount=0 is a good default, since header deserialization makes a big speed difference. Otherwise, useLowProbCount=1 is a good default, since the speed difference is small. @return : tableLog, or an errorCode, which can be tested using FSE_isError() */ FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog, const unsigned* count, size_t srcSize, unsigned maxSymbolValue, unsigned useLowProbCount); /*! FSE_NCountWriteBound(): Provides the maximum possible size of an FSE normalized table, given 'maxSymbolValue' and 'tableLog'. Typically useful for allocation purpose. */ FSE_PUBLIC_API size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog); /*! FSE_writeNCount(): Compactly save 'normalizedCounter' into 'buffer'. @return : size of the compressed table, or an errorCode, which can be tested using FSE_isError(). */ FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog); /*! Constructor and Destructor of FSE_CTable. Note that FSE_CTable size depends on 'tableLog' and 'maxSymbolValue' */ typedef unsigned FSE_CTable; /* don't allocate that. It's only meant to be more restrictive than void* */ /*! FSE_buildCTable(): Builds `ct`, which must be already allocated, using FSE_createCTable(). @return : 0, or an errorCode, which can be tested using FSE_isError() */ FSE_PUBLIC_API size_t FSE_buildCTable(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog); /*! FSE_compress_usingCTable(): Compress `src` using `ct` into `dst` which must be already allocated. @return : size of compressed data (<= `dstCapacity`), or 0 if compressed data could not fit into `dst`, or an errorCode, which can be tested using FSE_isError() */ FSE_PUBLIC_API size_t FSE_compress_usingCTable (void* dst, size_t dstCapacity, const void* src, size_t srcSize, const FSE_CTable* ct); /*! Tutorial : ---------- The first step is to count all symbols. FSE_count() does this job very fast. Result will be saved into 'count', a table of unsigned int, which must be already allocated, and have 'maxSymbolValuePtr[0]+1' cells. 'src' is a table of bytes of size 'srcSize'. All values within 'src' MUST be <= maxSymbolValuePtr[0] maxSymbolValuePtr[0] will be updated, with its real value (necessarily <= original value) FSE_count() will return the number of occurrence of the most frequent symbol. This can be used to know if there is a single symbol within 'src', and to quickly evaluate its compressibility. If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()). The next step is to normalize the frequencies. FSE_normalizeCount() will ensure that sum of frequencies is == 2 ^'tableLog'. It also guarantees a minimum of 1 to any Symbol with frequency >= 1. You can use 'tableLog'==0 to mean "use default tableLog value". If you are unsure of which tableLog value to use, you can ask FSE_optimalTableLog(), which will provide the optimal valid tableLog given sourceSize, maxSymbolValue, and a user-defined maximum (0 means "default"). The result of FSE_normalizeCount() will be saved into a table, called 'normalizedCounter', which is a table of signed short. 'normalizedCounter' must be already allocated, and have at least 'maxSymbolValue+1' cells. The return value is tableLog if everything proceeded as expected. It is 0 if there is a single symbol within distribution. If there is an error (ex: invalid tableLog value), the function will return an ErrorCode (which can be tested using FSE_isError()). 'normalizedCounter' can be saved in a compact manner to a memory area using FSE_writeNCount(). 'buffer' must be already allocated. For guaranteed success, buffer size must be at least FSE_headerBound(). The result of the function is the number of bytes written into 'buffer'. If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError(); ex : buffer size too small). 'normalizedCounter' can then be used to create the compression table 'CTable'. The space required by 'CTable' must be already allocated, using FSE_createCTable(). You can then use FSE_buildCTable() to fill 'CTable'. If there is an error, both functions will return an ErrorCode (which can be tested using FSE_isError()). 'CTable' can then be used to compress 'src', with FSE_compress_usingCTable(). Similar to FSE_count(), the convention is that 'src' is assumed to be a table of char of size 'srcSize' The function returns the size of compressed data (without header), necessarily <= `dstCapacity`. If it returns '0', compressed data could not fit into 'dst'. If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()). */ /* *** DECOMPRESSION *** */ /*! FSE_readNCount(): Read compactly saved 'normalizedCounter' from 'rBuffer'. @return : size read from 'rBuffer', or an errorCode, which can be tested using FSE_isError(). maxSymbolValuePtr[0] and tableLogPtr[0] will also be updated with their respective values */ FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter, unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, const void* rBuffer, size_t rBuffSize); /*! FSE_readNCount_bmi2(): * Same as FSE_readNCount() but pass bmi2=1 when your CPU supports BMI2 and 0 otherwise. */ FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter, unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, const void* rBuffer, size_t rBuffSize, int bmi2); typedef unsigned FSE_DTable; /* don't allocate that. It's just a way to be more restrictive than void* */ /*! Tutorial : ---------- (Note : these functions only decompress FSE-compressed blocks. If block is uncompressed, use memcpy() instead If block is a single repeated byte, use memset() instead ) The first step is to obtain the normalized frequencies of symbols. This can be performed by FSE_readNCount() if it was saved using FSE_writeNCount(). 'normalizedCounter' must be already allocated, and have at least 'maxSymbolValuePtr[0]+1' cells of signed short. In practice, that means it's necessary to know 'maxSymbolValue' beforehand, or size the table to handle worst case situations (typically 256). FSE_readNCount() will provide 'tableLog' and 'maxSymbolValue'. The result of FSE_readNCount() is the number of bytes read from 'rBuffer'. Note that 'rBufferSize' must be at least 4 bytes, even if useful information is less than that. If there is an error, the function will return an error code, which can be tested using FSE_isError(). The next step is to build the decompression tables 'FSE_DTable' from 'normalizedCounter'. This is performed by the function FSE_buildDTable(). The space required by 'FSE_DTable' must be already allocated using FSE_createDTable(). If there is an error, the function will return an error code, which can be tested using FSE_isError(). `FSE_DTable` can then be used to decompress `cSrc`, with FSE_decompress_usingDTable(). `cSrcSize` must be strictly correct, otherwise decompression will fail. FSE_decompress_usingDTable() result will tell how many bytes were regenerated (<=`dstCapacity`). If there is an error, the function will return an error code, which can be tested using FSE_isError(). (ex: dst buffer too small) */ #endif /* FSE_H */ #if !defined(FSE_H_FSE_STATIC_LINKING_ONLY) #define FSE_H_FSE_STATIC_LINKING_ONLY #include "bitstream.h" /* ***************************************** * Static allocation *******************************************/ /* FSE buffer bounds */ #define FSE_NCOUNTBOUND 512 #define FSE_BLOCKBOUND(size) ((size) + ((size)>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */) #define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size)) /* Macro version, useful for static allocation */ /* It is possible to statically allocate FSE CTable/DTable as a table of FSE_CTable/FSE_DTable using below macros */ #define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) (1 + (1<<((maxTableLog)-1)) + (((maxSymbolValue)+1)*2)) #define FSE_DTABLE_SIZE_U32(maxTableLog) (1 + (1<<(maxTableLog))) /* or use the size to malloc() space directly. Pay attention to alignment restrictions though */ #define FSE_CTABLE_SIZE(maxTableLog, maxSymbolValue) (FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(FSE_CTable)) #define FSE_DTABLE_SIZE(maxTableLog) (FSE_DTABLE_SIZE_U32(maxTableLog) * sizeof(FSE_DTable)) /* ***************************************** * FSE advanced API ***************************************** */ unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus); /*< same as FSE_optimalTableLog(), which used `minus==2` */ size_t FSE_buildCTable_rle (FSE_CTable* ct, unsigned char symbolValue); /*< build a fake FSE_CTable, designed to compress always the same symbolValue */ /* FSE_buildCTable_wksp() : * Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`). * `wkspSize` must be >= `FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)` of `unsigned`. * See FSE_buildCTable_wksp() for breakdown of workspace usage. */ #define FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog) (((maxSymbolValue + 2) + (1ull << (tableLog)))/2 + sizeof(U64)/sizeof(U32) /* additional 8 bytes for potential table overwrite */) #define FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) (sizeof(unsigned) * FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)) size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); #define FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) (sizeof(short) * (maxSymbolValue + 1) + (1ULL << maxTableLog) + 8) #define FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ((FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) + sizeof(unsigned) - 1) / sizeof(unsigned)) FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); /*< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */ #define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + 1 + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1) #define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned)) size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2); /*< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)`. * Set bmi2 to 1 if your CPU supports BMI2 or 0 if it doesn't */ typedef enum { FSE_repeat_none, /*< Cannot use the previous table */ FSE_repeat_check, /*< Can use the previous table but it must be checked */ FSE_repeat_valid /*< Can use the previous table and it is assumed to be valid */ } FSE_repeat; /* ***************************************** * FSE symbol compression API *******************************************/ /*! This API consists of small unitary functions, which highly benefit from being inlined. Hence their body are included in next section. */ typedef struct { ptrdiff_t value; const void* stateTable; const void* symbolTT; unsigned stateLog; } FSE_CState_t; static void FSE_initCState(FSE_CState_t* CStatePtr, const FSE_CTable* ct); static void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* CStatePtr, unsigned symbol); static void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* CStatePtr); /*< These functions are inner components of FSE_compress_usingCTable(). They allow the creation of custom streams, mixing multiple tables and bit sources. A key property to keep in mind is that encoding and decoding are done **in reverse direction**. So the first symbol you will encode is the last you will decode, like a LIFO stack. You will need a few variables to track your CStream. They are : FSE_CTable ct; // Provided by FSE_buildCTable() BIT_CStream_t bitStream; // bitStream tracking structure FSE_CState_t state; // State tracking structure (can have several) The first thing to do is to init bitStream and state. size_t errorCode = BIT_initCStream(&bitStream, dstBuffer, maxDstSize); FSE_initCState(&state, ct); Note that BIT_initCStream() can produce an error code, so its result should be tested, using FSE_isError(); You can then encode your input data, byte after byte. FSE_encodeSymbol() outputs a maximum of 'tableLog' bits at a time. Remember decoding will be done in reverse direction. FSE_encodeByte(&bitStream, &state, symbol); At any time, you can also add any bit sequence. Note : maximum allowed nbBits is 25, for compatibility with 32-bits decoders BIT_addBits(&bitStream, bitField, nbBits); The above methods don't commit data to memory, they just store it into local register, for speed. Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t). Writing data to memory is a manual operation, performed by the flushBits function. BIT_flushBits(&bitStream); Your last FSE encoding operation shall be to flush your last state value(s). FSE_flushState(&bitStream, &state); Finally, you must close the bitStream. The function returns the size of CStream in bytes. If data couldn't fit into dstBuffer, it will return a 0 ( == not compressible) If there is an error, it returns an errorCode (which can be tested using FSE_isError()). size_t size = BIT_closeCStream(&bitStream); */ /* ***************************************** * FSE symbol decompression API *******************************************/ typedef struct { size_t state; const void* table; /* precise table may vary, depending on U16 */ } FSE_DState_t; static void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt); static unsigned char FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD); static unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr); /*< Let's now decompose FSE_decompress_usingDTable() into its unitary components. You will decode FSE-encoded symbols from the bitStream, and also any other bitFields you put in, **in reverse order**. You will need a few variables to track your bitStream. They are : BIT_DStream_t DStream; // Stream context FSE_DState_t DState; // State context. Multiple ones are possible FSE_DTable* DTablePtr; // Decoding table, provided by FSE_buildDTable() The first thing to do is to init the bitStream. errorCode = BIT_initDStream(&DStream, srcBuffer, srcSize); You should then retrieve your initial state(s) (in reverse flushing order if you have several ones) : errorCode = FSE_initDState(&DState, &DStream, DTablePtr); You can then decode your data, symbol after symbol. For information the maximum number of bits read by FSE_decodeSymbol() is 'tableLog'. Keep in mind that symbols are decoded in reverse order, like a LIFO stack (last in, first out). unsigned char symbol = FSE_decodeSymbol(&DState, &DStream); You can retrieve any bitfield you eventually stored into the bitStream (in reverse order) Note : maximum allowed nbBits is 25, for 32-bits compatibility size_t bitField = BIT_readBits(&DStream, nbBits); All above operations only read from local register (which size depends on size_t). Refueling the register from memory is manually performed by the reload method. endSignal = FSE_reloadDStream(&DStream); BIT_reloadDStream() result tells if there is still some more data to read from DStream. BIT_DStream_unfinished : there is still some data left into the DStream. BIT_DStream_endOfBuffer : Dstream reached end of buffer. Its container may no longer be completely filled. BIT_DStream_completed : Dstream reached its exact end, corresponding in general to decompression completed. BIT_DStream_tooFar : Dstream went too far. Decompression result is corrupted. When reaching end of buffer (BIT_DStream_endOfBuffer), progress slowly, notably if you decode multiple symbols per loop, to properly detect the exact end of stream. After each decoded symbol, check if DStream is fully consumed using this simple test : BIT_reloadDStream(&DStream) >= BIT_DStream_completed When it's done, verify decompression is fully completed, by checking both DStream and the relevant states. Checking if DStream has reached its end is performed by : BIT_endOfDStream(&DStream); Check also the states. There might be some symbols left there, if some high probability ones (>50%) are possible. FSE_endOfDState(&DState); */ /* ***************************************** * FSE unsafe API *******************************************/ static unsigned char FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD); /* faster, but works only if nbBits is always >= 1 (otherwise, result will be corrupted) */ /* ***************************************** * Implementation of inlined functions *******************************************/ typedef struct { int deltaFindState; U32 deltaNbBits; } FSE_symbolCompressionTransform; /* total 8 bytes */ MEM_STATIC void FSE_initCState(FSE_CState_t* statePtr, const FSE_CTable* ct) { const void* ptr = ct; const U16* u16ptr = (const U16*) ptr; const U32 tableLog = MEM_read16(ptr); statePtr->value = (ptrdiff_t)1<<tableLog; statePtr->stateTable = u16ptr+2; statePtr->symbolTT = ct + 1 + (tableLog ? (1<<(tableLog-1)) : 1); statePtr->stateLog = tableLog; } /*! FSE_initCState2() : * Same as FSE_initCState(), but the first symbol to include (which will be the last to be read) * uses the smallest state value possible, saving the cost of this symbol */ MEM_STATIC void FSE_initCState2(FSE_CState_t* statePtr, const FSE_CTable* ct, U32 symbol) { FSE_initCState(statePtr, ct); { const FSE_symbolCompressionTransform symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol]; const U16* stateTable = (const U16*)(statePtr->stateTable); U32 nbBitsOut = (U32)((symbolTT.deltaNbBits + (1<<15)) >> 16); statePtr->value = (nbBitsOut << 16) - symbolTT.deltaNbBits; statePtr->value = stateTable[(statePtr->value >> nbBitsOut) + symbolTT.deltaFindState]; } } MEM_STATIC void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* statePtr, unsigned symbol) { FSE_symbolCompressionTransform const symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol]; const U16* const stateTable = (const U16*)(statePtr->stateTable); U32 const nbBitsOut = (U32)((statePtr->value + symbolTT.deltaNbBits) >> 16); BIT_addBits(bitC, (BitContainerType)statePtr->value, nbBitsOut); statePtr->value = stateTable[ (statePtr->value >> nbBitsOut) + symbolTT.deltaFindState]; } MEM_STATIC void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* statePtr) { BIT_addBits(bitC, (BitContainerType)statePtr->value, statePtr->stateLog); BIT_flushBits(bitC); } /* FSE_getMaxNbBits() : * Approximate maximum cost of a symbol, in bits. * Fractional get rounded up (i.e. a symbol with a normalized frequency of 3 gives the same result as a frequency of 2) * note 1 : assume symbolValue is valid (<= maxSymbolValue) * note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */ MEM_STATIC U32 FSE_getMaxNbBits(const void* symbolTTPtr, U32 symbolValue) { const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr; return (symbolTT[symbolValue].deltaNbBits + ((1<<16)-1)) >> 16; } /* FSE_bitCost() : * Approximate symbol cost, as fractional value, using fixed-point format (accuracyLog fractional bits) * note 1 : assume symbolValue is valid (<= maxSymbolValue) * note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */ MEM_STATIC U32 FSE_bitCost(const void* symbolTTPtr, U32 tableLog, U32 symbolValue, U32 accuracyLog) { const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr; U32 const minNbBits = symbolTT[symbolValue].deltaNbBits >> 16; U32 const threshold = (minNbBits+1) << 16; assert(tableLog < 16); assert(accuracyLog < 31-tableLog); /* ensure enough room for renormalization double shift */ { U32 const tableSize = 1 << tableLog; U32 const deltaFromThreshold = threshold - (symbolTT[symbolValue].deltaNbBits + tableSize); U32 const normalizedDeltaFromThreshold = (deltaFromThreshold << accuracyLog) >> tableLog; /* linear interpolation (very approximate) */ U32 const bitMultiplier = 1 << accuracyLog; assert(symbolTT[symbolValue].deltaNbBits + tableSize <= threshold); assert(normalizedDeltaFromThreshold <= bitMultiplier); return (minNbBits+1)*bitMultiplier - normalizedDeltaFromThreshold; } } /* ====== Decompression ====== */ typedef struct { U16 tableLog; U16 fastMode; } FSE_DTableHeader; /* sizeof U32 */ typedef struct { unsigned short newState; unsigned char symbol; unsigned char nbBits; } FSE_decode_t; /* size == U32 */ MEM_STATIC void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt) { const void* ptr = dt; const FSE_DTableHeader* const DTableH = (const FSE_DTableHeader*)ptr; DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog); BIT_reloadDStream(bitD); DStatePtr->table = dt + 1; } MEM_STATIC BYTE FSE_peekSymbol(const FSE_DState_t* DStatePtr) { FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state]; return DInfo.symbol; } MEM_STATIC void FSE_updateState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD) { FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state]; U32 const nbBits = DInfo.nbBits; size_t const lowBits = BIT_readBits(bitD, nbBits); DStatePtr->state = DInfo.newState + lowBits; } MEM_STATIC BYTE FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD) { FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state]; U32 const nbBits = DInfo.nbBits; BYTE const symbol = DInfo.symbol; size_t const lowBits = BIT_readBits(bitD, nbBits); DStatePtr->state = DInfo.newState + lowBits; return symbol; } /*! FSE_decodeSymbolFast() : unsafe, only works if no symbol has a probability > 50% */ MEM_STATIC BYTE FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD) { FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state]; U32 const nbBits = DInfo.nbBits; BYTE const symbol = DInfo.symbol; size_t const lowBits = BIT_readBitsFast(bitD, nbBits); DStatePtr->state = DInfo.newState + lowBits; return symbol; } MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr) { return DStatePtr->state == 0; } #ifndef FSE_COMMONDEFS_ONLY /* ************************************************************** * Tuning parameters ****************************************************************/ /*!MEMORY_USAGE : * Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.) * Increasing memory usage improves compression ratio * Reduced memory usage can improve speed, due to cache effect * Recommended max value is 14, for 16KB, which nicely fits into Intel x86 L1 cache */ #ifndef FSE_MAX_MEMORY_USAGE # define FSE_MAX_MEMORY_USAGE 14 #endif #ifndef FSE_DEFAULT_MEMORY_USAGE # define FSE_DEFAULT_MEMORY_USAGE 13 #endif #if (FSE_DEFAULT_MEMORY_USAGE > FSE_MAX_MEMORY_USAGE) # error "FSE_DEFAULT_MEMORY_USAGE must be <= FSE_MAX_MEMORY_USAGE" #endif /*!FSE_MAX_SYMBOL_VALUE : * Maximum symbol value authorized. * Required for proper stack allocation */ #ifndef FSE_MAX_SYMBOL_VALUE # define FSE_MAX_SYMBOL_VALUE 255 #endif /* ************************************************************** * template functions type & suffix ****************************************************************/ #define FSE_FUNCTION_TYPE BYTE #define FSE_FUNCTION_EXTENSION #define FSE_DECODE_TYPE FSE_decode_t #endif /* !FSE_COMMONDEFS_ONLY */ /* *************************************************************** * Constants *****************************************************************/ #define FSE_MAX_TABLELOG (FSE_MAX_MEMORY_USAGE-2) #define FSE_MAX_TABLESIZE (1U<<FSE_MAX_TABLELOG) #define FSE_MAXTABLESIZE_MASK (FSE_MAX_TABLESIZE-1) #define FSE_DEFAULT_TABLELOG (FSE_DEFAULT_MEMORY_USAGE-2) #define FSE_MIN_TABLELOG 5 #define FSE_TABLELOG_ABSOLUTE_MAX 15 #if FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX # error "FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX is not supported" #endif #define FSE_TABLESTEP(tableSize) (((tableSize)>>1) + ((tableSize)>>3) + 3) #endif /* FSE_STATIC_LINKING_ONLY */ |
| 5 5 155 155 31 31 5 30 5 28 3 31 4 30 31 31 3 30 4 3 35 35 4 35 35 35 35 35 31 6 15 10 31 4 31 31 12 30 34 34 34 3 31 31 31 170 170 170 170 8 8 8 170 112 123 84 170 33 33 179 170 170 154 34 155 32 155 2 2 1 2 155 156 37 37 148 148 33 33 33 208 54 54 28 6 32 153 153 153 96 100 67 134 134 134 134 1 1 1 1 134 8 2 3 6 2 242 121 196 76 168 164 12 148 148 148 148 148 148 148 148 148 125 178 141 125 148 148 148 148 148 41 148 148 26 26 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 | // SPDX-License-Identifier: GPL-2.0 /* * Copyright (c) 2000-2003 Silicon Graphics, Inc. * All Rights Reserved. */ #include "xfs_platform.h" #include "xfs_fs.h" #include "xfs_format.h" #include "xfs_log_format.h" #include "xfs_shared.h" #include "xfs_trans_resv.h" #include "xfs_bit.h" #include "xfs_mount.h" #include "xfs_defer.h" #include "xfs_inode.h" #include "xfs_bmap.h" #include "xfs_quota.h" #include "xfs_trans.h" #include "xfs_buf_item.h" #include "xfs_trans_space.h" #include "xfs_trans_priv.h" #include "xfs_qm.h" #include "xfs_trace.h" #include "xfs_log.h" #include "xfs_bmap_btree.h" #include "xfs_error.h" #include "xfs_health.h" /* * Lock order: * * ip->i_lock * qi->qi_tree_lock * dquot->q_qlock * dquot->q_flush (xfs_dqflock() and friends) * qi->qi_lru_lock * * If two dquots need to be locked the order is user before group/project, * otherwise by the lowest id first, see xfs_dqlock2. */ struct kmem_cache *xfs_dqtrx_cache; static struct kmem_cache *xfs_dquot_cache; static struct lock_class_key xfs_dquot_group_class; static struct lock_class_key xfs_dquot_project_class; /* Record observations of quota corruption with the health tracking system. */ static void xfs_dquot_mark_sick( struct xfs_dquot *dqp) { struct xfs_mount *mp = dqp->q_mount; switch (dqp->q_type) { case XFS_DQTYPE_USER: xfs_fs_mark_sick(mp, XFS_SICK_FS_UQUOTA); break; case XFS_DQTYPE_GROUP: xfs_fs_mark_sick(mp, XFS_SICK_FS_GQUOTA); break; case XFS_DQTYPE_PROJ: xfs_fs_mark_sick(mp, XFS_SICK_FS_PQUOTA); break; default: ASSERT(0); break; } } /* * Detach the dquot buffer if it's still attached, because we can get called * through dqpurge after a log shutdown. Caller must hold the dqflock or have * otherwise isolated the dquot. */ void xfs_dquot_detach_buf( struct xfs_dquot *dqp) { struct xfs_dq_logitem *qlip = &dqp->q_logitem; struct xfs_buf *bp = NULL; spin_lock(&qlip->qli_lock); if (qlip->qli_item.li_buf) { bp = qlip->qli_item.li_buf; qlip->qli_item.li_buf = NULL; } spin_unlock(&qlip->qli_lock); if (bp) { xfs_buf_lock(bp); list_del_init(&qlip->qli_item.li_bio_list); xfs_buf_relse(bp); } } /* * This is called to free all the memory associated with a dquot */ void xfs_qm_dqdestroy( struct xfs_dquot *dqp) { ASSERT(list_empty(&dqp->q_lru)); ASSERT(dqp->q_logitem.qli_item.li_buf == NULL); kvfree(dqp->q_logitem.qli_item.li_lv_shadow); mutex_destroy(&dqp->q_qlock); XFS_STATS_DEC(dqp->q_mount, xs_qm_dquot); kmem_cache_free(xfs_dquot_cache, dqp); } /* * If default limits are in force, push them into the dquot now. * We overwrite the dquot limits only if they are zero and this * is not the root dquot. */ void xfs_qm_adjust_dqlimits( struct xfs_dquot *dq) { struct xfs_mount *mp = dq->q_mount; struct xfs_quotainfo *q = mp->m_quotainfo; struct xfs_def_quota *defq; int prealloc = 0; ASSERT(dq->q_id); defq = xfs_get_defquota(q, xfs_dquot_type(dq)); if (!dq->q_blk.softlimit) { dq->q_blk.softlimit = defq->blk.soft; prealloc = 1; } if (!dq->q_blk.hardlimit) { dq->q_blk.hardlimit = defq->blk.hard; prealloc = 1; } if (!dq->q_ino.softlimit) dq->q_ino.softlimit = defq->ino.soft; if (!dq->q_ino.hardlimit) dq->q_ino.hardlimit = defq->ino.hard; if (!dq->q_rtb.softlimit) dq->q_rtb.softlimit = defq->rtb.soft; if (!dq->q_rtb.hardlimit) dq->q_rtb.hardlimit = defq->rtb.hard; if (prealloc) xfs_dquot_set_prealloc_limits(dq); } /* Set the expiration time of a quota's grace period. */ time64_t xfs_dquot_set_timeout( struct xfs_mount *mp, time64_t timeout) { struct xfs_quotainfo *qi = mp->m_quotainfo; return clamp_t(time64_t, timeout, qi->qi_expiry_min, qi->qi_expiry_max); } /* Set the length of the default grace period. */ time64_t xfs_dquot_set_grace_period( time64_t grace) { return clamp_t(time64_t, grace, XFS_DQ_GRACE_MIN, XFS_DQ_GRACE_MAX); } /* * Determine if this quota counter is over either limit and set the quota * timers as appropriate. */ static inline void xfs_qm_adjust_res_timer( struct xfs_mount *mp, struct xfs_dquot_res *res, struct xfs_quota_limits *qlim) { ASSERT(res->hardlimit == 0 || res->softlimit <= res->hardlimit); if ((res->softlimit && res->count > res->softlimit) || (res->hardlimit && res->count > res->hardlimit)) { if (res->timer == 0) res->timer = xfs_dquot_set_timeout(mp, ktime_get_real_seconds() + qlim->time); } else { res->timer = 0; } } /* * Check the limits and timers of a dquot and start or reset timers * if necessary. * This gets called even when quota enforcement is OFF, which makes our * life a little less complicated. (We just don't reject any quota * reservations in that case, when enforcement is off). * We also return 0 as the values of the timers in Q_GETQUOTA calls, when * enforcement's off. * In contrast, warnings are a little different in that they don't * 'automatically' get started when limits get exceeded. They do * get reset to zero, however, when we find the count to be under * the soft limit (they are only ever set non-zero via userspace). */ void xfs_qm_adjust_dqtimers( struct xfs_dquot *dq) { struct xfs_mount *mp = dq->q_mount; struct xfs_quotainfo *qi = mp->m_quotainfo; struct xfs_def_quota *defq; ASSERT(dq->q_id); defq = xfs_get_defquota(qi, xfs_dquot_type(dq)); xfs_qm_adjust_res_timer(dq->q_mount, &dq->q_blk, &defq->blk); xfs_qm_adjust_res_timer(dq->q_mount, &dq->q_ino, &defq->ino); xfs_qm_adjust_res_timer(dq->q_mount, &dq->q_rtb, &defq->rtb); } /* * initialize a buffer full of dquots and log the whole thing */ void xfs_qm_init_dquot_blk( struct xfs_trans *tp, xfs_dqid_t id, xfs_dqtype_t type, struct xfs_buf *bp) { struct xfs_mount *mp = tp->t_mountp; struct xfs_quotainfo *q = mp->m_quotainfo; struct xfs_dqblk *d; xfs_dqid_t curid; unsigned int qflag; unsigned int blftype; int i; ASSERT(tp); ASSERT(xfs_buf_islocked(bp)); switch (type) { case XFS_DQTYPE_USER: qflag = XFS_UQUOTA_CHKD; blftype = XFS_BLF_UDQUOT_BUF; break; case XFS_DQTYPE_PROJ: qflag = XFS_PQUOTA_CHKD; blftype = XFS_BLF_PDQUOT_BUF; break; case XFS_DQTYPE_GROUP: qflag = XFS_GQUOTA_CHKD; blftype = XFS_BLF_GDQUOT_BUF; break; default: ASSERT(0); return; } d = bp->b_addr; /* * ID of the first dquot in the block - id's are zero based. */ curid = id - (id % q->qi_dqperchunk); memset(d, 0, BBTOB(q->qi_dqchunklen)); for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) { d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); d->dd_diskdq.d_version = XFS_DQUOT_VERSION; d->dd_diskdq.d_id = cpu_to_be32(curid); d->dd_diskdq.d_type = type; if (curid > 0 && xfs_has_bigtime(mp)) d->dd_diskdq.d_type |= XFS_DQTYPE_BIGTIME; if (xfs_has_crc(mp)) { uuid_copy(&d->dd_uuid, &mp->m_sb.sb_meta_uuid); xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk), XFS_DQUOT_CRC_OFF); } } xfs_trans_dquot_buf(tp, bp, blftype); /* * quotacheck uses delayed writes to update all the dquots on disk in an * efficient manner instead of logging the individual dquot changes as * they are made. However if we log the buffer allocated here and crash * after quotacheck while the logged initialisation is still in the * active region of the log, log recovery can replay the dquot buffer * initialisation over the top of the checked dquots and corrupt quota * accounting. * * To avoid this problem, quotacheck cannot log the initialised buffer. * We must still dirty the buffer and write it back before the * allocation transaction clears the log. Therefore, mark the buffer as * ordered instead of logging it directly. This is safe for quotacheck * because it detects and repairs allocated but initialized dquot blocks * in the quota inodes. */ if (!(mp->m_qflags & qflag)) xfs_trans_ordered_buf(tp, bp); else xfs_trans_log_buf(tp, bp, 0, BBTOB(q->qi_dqchunklen) - 1); } static void xfs_dquot_set_prealloc( struct xfs_dquot_pre *pre, const struct xfs_dquot_res *res) { xfs_qcnt_t space; pre->q_prealloc_hi_wmark = res->hardlimit; pre->q_prealloc_lo_wmark = res->softlimit; space = div_u64(pre->q_prealloc_hi_wmark, 100); if (!pre->q_prealloc_lo_wmark) pre->q_prealloc_lo_wmark = space * 95; pre->q_low_space[XFS_QLOWSP_1_PCNT] = space; pre->q_low_space[XFS_QLOWSP_3_PCNT] = space * 3; pre->q_low_space[XFS_QLOWSP_5_PCNT] = space * 5; } /* * Initialize the dynamic speculative preallocation thresholds. The lo/hi * watermarks correspond to the soft and hard limits by default. If a soft limit * is not specified, we use 95% of the hard limit. */ void xfs_dquot_set_prealloc_limits(struct xfs_dquot *dqp) { xfs_dquot_set_prealloc(&dqp->q_blk_prealloc, &dqp->q_blk); xfs_dquot_set_prealloc(&dqp->q_rtb_prealloc, &dqp->q_rtb); } /* * Ensure that the given in-core dquot has a buffer on disk backing it, and * return the buffer locked and held. This is called when the bmapi finds a * hole. */ STATIC int xfs_dquot_disk_alloc( struct xfs_dquot *dqp, struct xfs_buf **bpp) { struct xfs_bmbt_irec map; struct xfs_trans *tp; struct xfs_mount *mp = dqp->q_mount; struct xfs_buf *bp; xfs_dqtype_t qtype = xfs_dquot_type(dqp); struct xfs_inode *quotip = xfs_quota_inode(mp, qtype); int nmaps = 1; int error; trace_xfs_dqalloc(dqp); error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_dqalloc, XFS_QM_DQALLOC_SPACE_RES(mp), 0, 0, &tp); if (error) return error; xfs_ilock(quotip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, quotip, 0); if (!xfs_this_quota_on(dqp->q_mount, qtype)) { /* * Return if this type of quotas is turned off while we didn't * have an inode lock */ error = -ESRCH; goto err_cancel; } error = xfs_iext_count_extend(tp, quotip, XFS_DATA_FORK, XFS_IEXT_ADD_NOSPLIT_CNT); if (error) goto err_cancel; /* Create the block mapping. */ error = xfs_bmapi_write(tp, quotip, dqp->q_fileoffset, XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, 0, &map, &nmaps); if (error) goto err_cancel; ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB); ASSERT((map.br_startblock != DELAYSTARTBLOCK) && (map.br_startblock != HOLESTARTBLOCK)); /* * Keep track of the blkno to save a lookup later */ dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); /* now we can just get the buffer (there's nothing to read yet) */ error = xfs_trans_get_buf(tp, mp->m_ddev_targp, dqp->q_blkno, mp->m_quotainfo->qi_dqchunklen, 0, &bp); if (error) goto err_cancel; bp->b_ops = &xfs_dquot_buf_ops; /* * Make a chunk of dquots out of this buffer and log * the entire thing. */ xfs_qm_init_dquot_blk(tp, dqp->q_id, qtype, bp); xfs_buf_set_ref(bp, XFS_DQUOT_REF); /* * Hold the buffer and join it to the dfops so that we'll still own * the buffer when we return to the caller. The buffer disposal on * error must be paid attention to very carefully, as it has been * broken since commit efa092f3d4c6 "[XFS] Fixes a bug in the quota * code when allocating a new dquot record" in 2005, and the later * conversion to xfs_defer_ops in commit 310a75a3c6c747 failed to keep * the buffer locked across the _defer_finish call. We can now do * this correctly with xfs_defer_bjoin. * * Above, we allocated a disk block for the dquot information and used * get_buf to initialize the dquot. If the _defer_finish fails, the old * transaction is gone but the new buffer is not joined or held to any * transaction, so we must _buf_relse it. * * If everything succeeds, the caller of this function is returned a * buffer that is locked and held to the transaction. The caller * is responsible for unlocking any buffer passed back, either * manually or by committing the transaction. On error, the buffer is * released and not passed back. * * Keep the quota inode ILOCKed until after the transaction commit to * maintain the atomicity of bmap/rmap updates. */ xfs_trans_bhold(tp, bp); error = xfs_trans_commit(tp); xfs_iunlock(quotip, XFS_ILOCK_EXCL); if (error) { xfs_buf_relse(bp); return error; } *bpp = bp; return 0; err_cancel: xfs_trans_cancel(tp); xfs_iunlock(quotip, XFS_ILOCK_EXCL); return error; } /* * Read in the in-core dquot's on-disk metadata and return the buffer. * Returns ENOENT to signal a hole. */ STATIC int xfs_dquot_disk_read( struct xfs_mount *mp, struct xfs_dquot *dqp, struct xfs_buf **bpp) { struct xfs_bmbt_irec map; struct xfs_buf *bp; xfs_dqtype_t qtype = xfs_dquot_type(dqp); struct xfs_inode *quotip = xfs_quota_inode(mp, qtype); uint lock_mode; int nmaps = 1; int error; lock_mode = xfs_ilock_data_map_shared(quotip); if (!xfs_this_quota_on(mp, qtype)) { /* * Return if this type of quotas is turned off while we * didn't have the quota inode lock. */ xfs_iunlock(quotip, lock_mode); return -ESRCH; } /* * Find the block map; no allocations yet */ error = xfs_bmapi_read(quotip, dqp->q_fileoffset, XFS_DQUOT_CLUSTER_SIZE_FSB, &map, &nmaps, 0); xfs_iunlock(quotip, lock_mode); if (error) return error; ASSERT(nmaps == 1); ASSERT(map.br_blockcount >= 1); ASSERT(map.br_startblock != DELAYSTARTBLOCK); if (map.br_startblock == HOLESTARTBLOCK) return -ENOENT; trace_xfs_dqtobp_read(dqp); /* * store the blkno etc so that we don't have to do the * mapping all the time */ dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno, mp->m_quotainfo->qi_dqchunklen, 0, &bp, &xfs_dquot_buf_ops); if (xfs_metadata_is_sick(error)) xfs_dquot_mark_sick(dqp); if (error) { ASSERT(bp == NULL); return error; } ASSERT(xfs_buf_islocked(bp)); xfs_buf_set_ref(bp, XFS_DQUOT_REF); *bpp = bp; return 0; } /* Allocate and initialize everything we need for an incore dquot. */ STATIC struct xfs_dquot * xfs_dquot_alloc( struct xfs_mount *mp, xfs_dqid_t id, xfs_dqtype_t type) { struct xfs_dquot *dqp; dqp = kmem_cache_zalloc(xfs_dquot_cache, GFP_KERNEL | __GFP_NOFAIL); dqp->q_type = type; dqp->q_id = id; dqp->q_mount = mp; INIT_LIST_HEAD(&dqp->q_lru); mutex_init(&dqp->q_qlock); init_waitqueue_head(&dqp->q_pinwait); dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk; /* * Offset of dquot in the (fixed sized) dquot chunk. */ dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) * sizeof(struct xfs_dqblk); /* * Because we want to use a counting completion, complete * the flush completion once to allow a single access to * the flush completion without blocking. */ init_completion(&dqp->q_flush); complete(&dqp->q_flush); /* * Make sure group quotas have a different lock class than user * quotas. */ switch (type) { case XFS_DQTYPE_USER: /* uses the default lock class */ break; case XFS_DQTYPE_GROUP: lockdep_set_class(&dqp->q_qlock, &xfs_dquot_group_class); break; case XFS_DQTYPE_PROJ: lockdep_set_class(&dqp->q_qlock, &xfs_dquot_project_class); break; default: ASSERT(0); break; } xfs_qm_dquot_logitem_init(dqp); XFS_STATS_INC(mp, xs_qm_dquot); return dqp; } /* Check the ondisk dquot's id and type match what the incore dquot expects. */ static bool xfs_dquot_check_type( struct xfs_dquot *dqp, struct xfs_disk_dquot *ddqp) { uint8_t ddqp_type; uint8_t dqp_type; ddqp_type = ddqp->d_type & XFS_DQTYPE_REC_MASK; dqp_type = xfs_dquot_type(dqp); if (be32_to_cpu(ddqp->d_id) != dqp->q_id) return false; /* * V5 filesystems always expect an exact type match. V4 filesystems * expect an exact match for user dquots and for non-root group and * project dquots. */ if (xfs_has_crc(dqp->q_mount) || dqp_type == XFS_DQTYPE_USER || dqp->q_id != 0) return ddqp_type == dqp_type; /* * V4 filesystems support either group or project quotas, but not both * at the same time. The non-user quota file can be switched between * group and project quota uses depending on the mount options, which * means that we can encounter the other type when we try to load quota * defaults. Quotacheck will soon reset the entire quota file * (including the root dquot) anyway, but don't log scary corruption * reports to dmesg. */ return ddqp_type == XFS_DQTYPE_GROUP || ddqp_type == XFS_DQTYPE_PROJ; } /* Copy the in-core quota fields in from the on-disk buffer. */ STATIC int xfs_dquot_from_disk( struct xfs_dquot *dqp, struct xfs_buf *bp) { struct xfs_dqblk *dqb = xfs_buf_offset(bp, dqp->q_bufoffset); struct xfs_disk_dquot *ddqp = &dqb->dd_diskdq; /* * Ensure that we got the type and ID we were looking for. * Everything else was checked by the dquot buffer verifier. */ if (!xfs_dquot_check_type(dqp, ddqp)) { xfs_alert_tag(bp->b_mount, XFS_PTAG_VERIFIER_ERROR, "Metadata corruption detected at %pS, quota %u", __this_address, dqp->q_id); xfs_alert(bp->b_mount, "Unmount and run xfs_repair"); xfs_dquot_mark_sick(dqp); return -EFSCORRUPTED; } /* copy everything from disk dquot to the incore dquot */ dqp->q_type = ddqp->d_type; dqp->q_blk.hardlimit = be64_to_cpu(ddqp->d_blk_hardlimit); dqp->q_blk.softlimit = be64_to_cpu(ddqp->d_blk_softlimit); dqp->q_ino.hardlimit = be64_to_cpu(ddqp->d_ino_hardlimit); dqp->q_ino.softlimit = be64_to_cpu(ddqp->d_ino_softlimit); dqp->q_rtb.hardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit); dqp->q_rtb.softlimit = be64_to_cpu(ddqp->d_rtb_softlimit); dqp->q_blk.count = be64_to_cpu(ddqp->d_bcount); dqp->q_ino.count = be64_to_cpu(ddqp->d_icount); dqp->q_rtb.count = be64_to_cpu(ddqp->d_rtbcount); dqp->q_blk.timer = xfs_dquot_from_disk_ts(ddqp, ddqp->d_btimer); dqp->q_ino.timer = xfs_dquot_from_disk_ts(ddqp, ddqp->d_itimer); dqp->q_rtb.timer = xfs_dquot_from_disk_ts(ddqp, ddqp->d_rtbtimer); /* * Reservation counters are defined as reservation plus current usage * to avoid having to add every time. */ dqp->q_blk.reserved = dqp->q_blk.count; dqp->q_ino.reserved = dqp->q_ino.count; dqp->q_rtb.reserved = dqp->q_rtb.count; /* initialize the dquot speculative prealloc thresholds */ xfs_dquot_set_prealloc_limits(dqp); return 0; } /* Copy the in-core quota fields into the on-disk buffer. */ void xfs_dquot_to_disk( struct xfs_disk_dquot *ddqp, struct xfs_dquot *dqp) { ddqp->d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); ddqp->d_version = XFS_DQUOT_VERSION; ddqp->d_type = dqp->q_type; ddqp->d_id = cpu_to_be32(dqp->q_id); ddqp->d_pad0 = 0; ddqp->d_pad = 0; ddqp->d_blk_hardlimit = cpu_to_be64(dqp->q_blk.hardlimit); ddqp->d_blk_softlimit = cpu_to_be64(dqp->q_blk.softlimit); ddqp->d_ino_hardlimit = cpu_to_be64(dqp->q_ino.hardlimit); ddqp->d_ino_softlimit = cpu_to_be64(dqp->q_ino.softlimit); ddqp->d_rtb_hardlimit = cpu_to_be64(dqp->q_rtb.hardlimit); ddqp->d_rtb_softlimit = cpu_to_be64(dqp->q_rtb.softlimit); ddqp->d_bcount = cpu_to_be64(dqp->q_blk.count); ddqp->d_icount = cpu_to_be64(dqp->q_ino.count); ddqp->d_rtbcount = cpu_to_be64(dqp->q_rtb.count); ddqp->d_bwarns = 0; ddqp->d_iwarns = 0; ddqp->d_rtbwarns = 0; ddqp->d_btimer = xfs_dquot_to_disk_ts(dqp, dqp->q_blk.timer); ddqp->d_itimer = xfs_dquot_to_disk_ts(dqp, dqp->q_ino.timer); ddqp->d_rtbtimer = xfs_dquot_to_disk_ts(dqp, dqp->q_rtb.timer); } /* * Read in the ondisk dquot using dqtobp() then copy it to an incore version, * and release the buffer immediately. If @can_alloc is true, fill any * holes in the on-disk metadata. */ static int xfs_qm_dqread( struct xfs_mount *mp, xfs_dqid_t id, xfs_dqtype_t type, bool can_alloc, struct xfs_dquot **dqpp) { struct xfs_dquot *dqp; struct xfs_buf *bp; int error; dqp = xfs_dquot_alloc(mp, id, type); trace_xfs_dqread(dqp); /* Try to read the buffer, allocating if necessary. */ error = xfs_dquot_disk_read(mp, dqp, &bp); if (error == -ENOENT && can_alloc) error = xfs_dquot_disk_alloc(dqp, &bp); if (error) goto err; /* * At this point we should have a clean locked buffer. Copy the data * to the incore dquot and release the buffer since the incore dquot * has its own locking protocol so we needn't tie up the buffer any * further. */ ASSERT(xfs_buf_islocked(bp)); error = xfs_dquot_from_disk(dqp, bp); xfs_buf_relse(bp); if (error) goto err; *dqpp = dqp; return error; err: trace_xfs_dqread_fail(dqp); xfs_qm_dqdestroy(dqp); *dqpp = NULL; return error; } /* * Advance to the next id in the current chunk, or if at the * end of the chunk, skip ahead to first id in next allocated chunk * using the SEEK_DATA interface. */ static int xfs_dq_get_next_id( struct xfs_mount *mp, xfs_dqtype_t type, xfs_dqid_t *id) { struct xfs_inode *quotip = xfs_quota_inode(mp, type); xfs_dqid_t next_id = *id + 1; /* simple advance */ uint lock_flags; struct xfs_bmbt_irec got; struct xfs_iext_cursor cur; xfs_fsblock_t start; int error = 0; /* If we'd wrap past the max ID, stop */ if (next_id < *id) return -ENOENT; /* If new ID is within the current chunk, advancing it sufficed */ if (next_id % mp->m_quotainfo->qi_dqperchunk) { *id = next_id; return 0; } /* Nope, next_id is now past the current chunk, so find the next one */ start = (xfs_fsblock_t)next_id / mp->m_quotainfo->qi_dqperchunk; lock_flags = xfs_ilock_data_map_shared(quotip); error = xfs_iread_extents(NULL, quotip, XFS_DATA_FORK); if (error) return error; if (xfs_iext_lookup_extent(quotip, "ip->i_df, start, &cur, &got)) { /* contiguous chunk, bump startoff for the id calculation */ if (got.br_startoff < start) got.br_startoff = start; *id = got.br_startoff * mp->m_quotainfo->qi_dqperchunk; } else { error = -ENOENT; } xfs_iunlock(quotip, lock_flags); return error; } /* * Look up the dquot in the in-core cache. If found, the dquot is returned * locked and ready to go. */ static struct xfs_dquot * xfs_qm_dqget_cache_lookup( struct xfs_mount *mp, xfs_dqid_t id, xfs_dqtype_t type) { struct xfs_quotainfo *qi = mp->m_quotainfo; struct radix_tree_root *tree = xfs_dquot_tree(qi, type); struct xfs_dquot *dqp; restart: mutex_lock(&qi->qi_tree_lock); dqp = radix_tree_lookup(tree, id); if (!dqp) { mutex_unlock(&qi->qi_tree_lock); XFS_STATS_INC(mp, xs_qm_dqcachemisses); return NULL; } if (!lockref_get_not_dead(&dqp->q_lockref)) { mutex_unlock(&qi->qi_tree_lock); trace_xfs_dqget_freeing(dqp); delay(1); goto restart; } mutex_unlock(&qi->qi_tree_lock); trace_xfs_dqget_hit(dqp); XFS_STATS_INC(mp, xs_qm_dqcachehits); return dqp; } /* * Try to insert a new dquot into the in-core cache. If an error occurs the * caller should throw away the dquot and start over. Otherwise, the dquot * is returned (and held by the cache) as if there had been a cache hit. * * The insert needs to be done under memalloc_nofs context because the radix * tree can do memory allocation during insert. The qi->qi_tree_lock is taken in * memory reclaim when freeing unused dquots, so we cannot have the radix tree * node allocation recursing into filesystem reclaim whilst we hold the * qi_tree_lock. */ static int xfs_qm_dqget_cache_insert( struct xfs_mount *mp, xfs_dqid_t id, xfs_dqtype_t type, struct xfs_dquot *dqp) { struct xfs_quotainfo *qi = mp->m_quotainfo; struct radix_tree_root *tree = xfs_dquot_tree(qi, type); unsigned int nofs_flags; int error; nofs_flags = memalloc_nofs_save(); mutex_lock(&qi->qi_tree_lock); error = radix_tree_insert(tree, id, dqp); if (unlikely(error)) { trace_xfs_dqget_dup(dqp); goto out_unlock; } lockref_init(&dqp->q_lockref); qi->qi_dquots++; out_unlock: mutex_unlock(&qi->qi_tree_lock); memalloc_nofs_restore(nofs_flags); return error; } /* Check our input parameters. */ static int xfs_qm_dqget_checks( struct xfs_mount *mp, xfs_dqtype_t type) { switch (type) { case XFS_DQTYPE_USER: if (!XFS_IS_UQUOTA_ON(mp)) return -ESRCH; return 0; case XFS_DQTYPE_GROUP: if (!XFS_IS_GQUOTA_ON(mp)) return -ESRCH; return 0; case XFS_DQTYPE_PROJ: if (!XFS_IS_PQUOTA_ON(mp)) return -ESRCH; return 0; default: WARN_ON_ONCE(0); return -EINVAL; } } /* * Given the file system, id, and type (UDQUOT/GDQUOT/PDQUOT), return a * dquot, doing an allocation (if requested) as needed. */ int xfs_qm_dqget( struct xfs_mount *mp, xfs_dqid_t id, xfs_dqtype_t type, bool can_alloc, struct xfs_dquot **O_dqpp) { struct xfs_dquot *dqp; int error; error = xfs_qm_dqget_checks(mp, type); if (error) return error; restart: dqp = xfs_qm_dqget_cache_lookup(mp, id, type); if (dqp) goto found; error = xfs_qm_dqread(mp, id, type, can_alloc, &dqp); if (error) return error; error = xfs_qm_dqget_cache_insert(mp, id, type, dqp); if (error) { xfs_qm_dqdestroy(dqp); if (error == -EEXIST) { /* * Duplicate found. Just throw away the new dquot and * start over. */ XFS_STATS_INC(mp, xs_qm_dquot_dups); goto restart; } return error; } trace_xfs_dqget_miss(dqp); found: *O_dqpp = dqp; return 0; } /* * Given a dquot id and type, read and initialize a dquot from the on-disk * metadata. This function is only for use during quota initialization so * it ignores the dquot cache assuming that the dquot shrinker isn't set up. * The caller is responsible for _qm_dqdestroy'ing the returned dquot. */ int xfs_qm_dqget_uncached( struct xfs_mount *mp, xfs_dqid_t id, xfs_dqtype_t type, struct xfs_dquot **dqpp) { int error; error = xfs_qm_dqget_checks(mp, type); if (error) return error; return xfs_qm_dqread(mp, id, type, 0, dqpp); } /* Return the quota id for a given inode and type. */ xfs_dqid_t xfs_qm_id_for_quotatype( struct xfs_inode *ip, xfs_dqtype_t type) { switch (type) { case XFS_DQTYPE_USER: return i_uid_read(VFS_I(ip)); case XFS_DQTYPE_GROUP: return i_gid_read(VFS_I(ip)); case XFS_DQTYPE_PROJ: return ip->i_projid; } ASSERT(0); return 0; } /* * Return the dquot for a given inode and type. If @can_alloc is true, then * allocate blocks if needed. The inode's ILOCK must be held and it must not * have already had an inode attached. */ int xfs_qm_dqget_inode( struct xfs_inode *ip, xfs_dqtype_t type, bool can_alloc, struct xfs_dquot **dqpp) { struct xfs_mount *mp = ip->i_mount; struct xfs_dquot *dqp; xfs_dqid_t id; int error; ASSERT(!*dqpp); xfs_assert_ilocked(ip, XFS_ILOCK_EXCL); error = xfs_qm_dqget_checks(mp, type); if (error) return error; xfs_assert_ilocked(ip, XFS_ILOCK_EXCL); ASSERT(xfs_inode_dquot(ip, type) == NULL); ASSERT(!xfs_is_metadir_inode(ip)); id = xfs_qm_id_for_quotatype(ip, type); restart: dqp = xfs_qm_dqget_cache_lookup(mp, id, type); if (dqp) goto found; /* * Dquot cache miss. We don't want to keep the inode lock across * a (potential) disk read. Also we don't want to deal with the lock * ordering between quotainode and this inode. OTOH, dropping the inode * lock here means dealing with a chown that can happen before * we re-acquire the lock. */ xfs_iunlock(ip, XFS_ILOCK_EXCL); error = xfs_qm_dqread(mp, id, type, can_alloc, &dqp); xfs_ilock(ip, XFS_ILOCK_EXCL); if (error) return error; /* * A dquot could be attached to this inode by now, since we had * dropped the ilock. */ if (xfs_this_quota_on(mp, type)) { struct xfs_dquot *dqp1; dqp1 = xfs_inode_dquot(ip, type); if (dqp1) { xfs_qm_dqdestroy(dqp); dqp = dqp1; goto dqret; } } else { /* inode stays locked on return */ xfs_qm_dqdestroy(dqp); return -ESRCH; } error = xfs_qm_dqget_cache_insert(mp, id, type, dqp); if (error) { xfs_qm_dqdestroy(dqp); if (error == -EEXIST) { /* * Duplicate found. Just throw away the new dquot and * start over. */ XFS_STATS_INC(mp, xs_qm_dquot_dups); goto restart; } return error; } dqret: xfs_assert_ilocked(ip, XFS_ILOCK_EXCL); trace_xfs_dqget_miss(dqp); found: trace_xfs_dqattach_get(dqp); *dqpp = dqp; return 0; } /* * Starting at @id and progressing upwards, look for an initialized incore * dquot, lock it, and return it. */ int xfs_qm_dqget_next( struct xfs_mount *mp, xfs_dqid_t id, xfs_dqtype_t type, struct xfs_dquot **dqpp) { struct xfs_dquot *dqp; int error = 0; *dqpp = NULL; for (; !error; error = xfs_dq_get_next_id(mp, type, &id)) { error = xfs_qm_dqget(mp, id, type, false, &dqp); if (error == -ENOENT) continue; else if (error != 0) break; mutex_lock(&dqp->q_qlock); if (!XFS_IS_DQUOT_UNINITIALIZED(dqp)) { *dqpp = dqp; return 0; } mutex_unlock(&dqp->q_qlock); xfs_qm_dqrele(dqp); } return error; } /* * Release a reference to the dquot. */ void xfs_qm_dqrele( struct xfs_dquot *dqp) { if (!dqp) return; trace_xfs_dqrele(dqp); if (lockref_put_or_lock(&dqp->q_lockref)) return; if (!--dqp->q_lockref.count) { struct xfs_quotainfo *qi = dqp->q_mount->m_quotainfo; trace_xfs_dqrele_free(dqp); if (list_lru_add_obj(&qi->qi_lru, &dqp->q_lru)) XFS_STATS_INC(dqp->q_mount, xs_qm_dquot_unused); } spin_unlock(&dqp->q_lockref.lock); } /* * This is the dquot flushing I/O completion routine. It is called * from interrupt level when the buffer containing the dquot is * flushed to disk. It is responsible for removing the dquot logitem * from the AIL if it has not been re-logged, and unlocking the dquot's * flush lock. This behavior is very similar to that of inodes.. */ static void xfs_qm_dqflush_done( struct xfs_log_item *lip) { struct xfs_dq_logitem *qlip = container_of(lip, struct xfs_dq_logitem, qli_item); struct xfs_dquot *dqp = qlip->qli_dquot; struct xfs_ail *ailp = lip->li_ailp; struct xfs_buf *bp = NULL; xfs_lsn_t tail_lsn; /* * We only want to pull the item from the AIL if its * location in the log has not changed since we started the flush. * Thus, we only bother if the dquot's lsn has * not changed. First we check the lsn outside the lock * since it's cheaper, and then we recheck while * holding the lock before removing the dquot from the AIL. */ if (test_bit(XFS_LI_IN_AIL, &lip->li_flags) && (lip->li_lsn == qlip->qli_flush_lsn || test_bit(XFS_LI_FAILED, &lip->li_flags))) { spin_lock(&ailp->ail_lock); clear_bit(XFS_LI_FAILED, &lip->li_flags); if (lip->li_lsn == qlip->qli_flush_lsn) { /* xfs_ail_update_finish() drops the AIL lock */ tail_lsn = xfs_ail_delete_one(ailp, lip); xfs_ail_update_finish(ailp, tail_lsn); } else { spin_unlock(&ailp->ail_lock); } } /* * If this dquot hasn't been dirtied since initiating the last dqflush, * release the buffer reference. We already unlinked this dquot item * from the buffer. */ spin_lock(&qlip->qli_lock); if (!qlip->qli_dirty) { bp = lip->li_buf; lip->li_buf = NULL; } spin_unlock(&qlip->qli_lock); if (bp) xfs_buf_rele(bp); /* * Release the dq's flush lock since we're done with it. */ xfs_dqfunlock(dqp); } void xfs_buf_dquot_iodone( struct xfs_buf *bp) { struct xfs_log_item *lip, *n; list_for_each_entry_safe(lip, n, &bp->b_li_list, li_bio_list) { list_del_init(&lip->li_bio_list); xfs_qm_dqflush_done(lip); } } /* Check incore dquot for errors before we flush. */ static xfs_failaddr_t xfs_qm_dqflush_check( struct xfs_dquot *dqp) { xfs_dqtype_t type = xfs_dquot_type(dqp); if (type != XFS_DQTYPE_USER && type != XFS_DQTYPE_GROUP && type != XFS_DQTYPE_PROJ) return __this_address; if (dqp->q_id == 0) return NULL; if (dqp->q_blk.softlimit && dqp->q_blk.count > dqp->q_blk.softlimit && !dqp->q_blk.timer) return __this_address; if (dqp->q_ino.softlimit && dqp->q_ino.count > dqp->q_ino.softlimit && !dqp->q_ino.timer) return __this_address; if (dqp->q_rtb.softlimit && dqp->q_rtb.count > dqp->q_rtb.softlimit && !dqp->q_rtb.timer) return __this_address; /* bigtime flag should never be set on root dquots */ if (dqp->q_type & XFS_DQTYPE_BIGTIME) { if (!xfs_has_bigtime(dqp->q_mount)) return __this_address; if (dqp->q_id == 0) return __this_address; } return NULL; } /* * Get the buffer containing the on-disk dquot. * * Requires dquot flush lock, will clear the dirty flag, delete the quota log * item from the AIL, and shut down the system if something goes wrong. */ static int xfs_dquot_read_buf( struct xfs_trans *tp, struct xfs_dquot *dqp, struct xfs_buf **bpp) { struct xfs_mount *mp = dqp->q_mount; struct xfs_buf *bp = NULL; int error; error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, dqp->q_blkno, mp->m_quotainfo->qi_dqchunklen, 0, &bp, &xfs_dquot_buf_ops); if (xfs_metadata_is_sick(error)) xfs_dquot_mark_sick(dqp); if (error) goto out_abort; *bpp = bp; return 0; out_abort: dqp->q_flags &= ~XFS_DQFLAG_DIRTY; xfs_trans_ail_delete(&dqp->q_logitem.qli_item, 0); xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); return error; } /* * Attach a dquot buffer to this dquot to avoid allocating a buffer during a * dqflush, since dqflush can be called from reclaim context. Caller must hold * the dqlock. */ int xfs_dquot_attach_buf( struct xfs_trans *tp, struct xfs_dquot *dqp) { struct xfs_dq_logitem *qlip = &dqp->q_logitem; struct xfs_log_item *lip = &qlip->qli_item; int error; spin_lock(&qlip->qli_lock); if (!lip->li_buf) { struct xfs_buf *bp = NULL; spin_unlock(&qlip->qli_lock); error = xfs_dquot_read_buf(tp, dqp, &bp); if (error) return error; /* * Hold the dquot buffer so that we retain our ref to it after * detaching it from the transaction, then give that ref to the * dquot log item so that the AIL does not have to read the * dquot buffer to push this item. */ xfs_buf_hold(bp); xfs_trans_brelse(tp, bp); spin_lock(&qlip->qli_lock); lip->li_buf = bp; } qlip->qli_dirty = true; spin_unlock(&qlip->qli_lock); return 0; } /* * Get a new reference the dquot buffer attached to this dquot for a dqflush * operation. * * Returns 0 and a NULL bp if none was attached to the dquot; 0 and a locked * bp; or -EAGAIN if the buffer could not be locked. */ int xfs_dquot_use_attached_buf( struct xfs_dquot *dqp, struct xfs_buf **bpp) { struct xfs_buf *bp = dqp->q_logitem.qli_item.li_buf; /* * A NULL buffer can happen if the dquot dirty flag was set but the * filesystem shut down before transaction commit happened. In that * case we're not going to flush anyway. */ if (!bp) { ASSERT(xfs_is_shutdown(dqp->q_mount)); *bpp = NULL; return 0; } if (!xfs_buf_trylock(bp)) return -EAGAIN; xfs_buf_hold(bp); *bpp = bp; return 0; } /* * Write a modified dquot to disk. * The dquot must be locked and the flush lock too taken by caller. * The flush lock will not be unlocked until the dquot reaches the disk, * but the dquot is free to be unlocked and modified by the caller * in the interim. Dquot is still locked on return. This behavior is * identical to that of inodes. */ int xfs_qm_dqflush( struct xfs_dquot *dqp, struct xfs_buf *bp) { struct xfs_mount *mp = dqp->q_mount; struct xfs_dq_logitem *qlip = &dqp->q_logitem; struct xfs_log_item *lip = &qlip->qli_item; struct xfs_dqblk *dqblk; xfs_failaddr_t fa; int error; ASSERT(XFS_DQ_IS_LOCKED(dqp)); ASSERT(!completion_done(&dqp->q_flush)); ASSERT(atomic_read(&dqp->q_pincount) == 0); trace_xfs_dqflush(dqp); fa = xfs_qm_dqflush_check(dqp); if (fa) { xfs_alert(mp, "corrupt dquot ID 0x%x in memory at %pS", dqp->q_id, fa); xfs_dquot_mark_sick(dqp); error = -EFSCORRUPTED; goto out_abort; } /* Flush the incore dquot to the ondisk buffer. */ dqblk = xfs_buf_offset(bp, dqp->q_bufoffset); xfs_dquot_to_disk(&dqblk->dd_diskdq, dqp); /* * Clear the dirty field and remember the flush lsn for later use. */ dqp->q_flags &= ~XFS_DQFLAG_DIRTY; /* * We hold the dquot lock, so nobody can dirty it while we're * scheduling the write out. Clear the dirty-since-flush flag. */ spin_lock(&qlip->qli_lock); qlip->qli_dirty = false; spin_unlock(&qlip->qli_lock); xfs_trans_ail_copy_lsn(mp->m_ail, &qlip->qli_flush_lsn, &lip->li_lsn); /* * copy the lsn into the on-disk dquot now while we have the in memory * dquot here. This can't be done later in the write verifier as we * can't get access to the log item at that point in time. * * We also calculate the CRC here so that the on-disk dquot in the * buffer always has a valid CRC. This ensures there is no possibility * of a dquot without an up-to-date CRC getting to disk. */ if (xfs_has_crc(mp)) { dqblk->dd_lsn = cpu_to_be64(lip->li_lsn); xfs_update_cksum((char *)dqblk, sizeof(struct xfs_dqblk), XFS_DQUOT_CRC_OFF); } /* * Attach the dquot to the buffer so that we can remove this dquot from * the AIL and release the flush lock once the dquot is synced to disk. */ bp->b_iodone = xfs_buf_dquot_iodone; list_add_tail(&lip->li_bio_list, &bp->b_li_list); /* * If the buffer is pinned then push on the log so we won't * get stuck waiting in the write for too long. */ if (xfs_buf_ispinned(bp)) { trace_xfs_dqflush_force(dqp); xfs_log_force(mp, 0); } trace_xfs_dqflush_done(dqp); return 0; out_abort: dqp->q_flags &= ~XFS_DQFLAG_DIRTY; xfs_trans_ail_delete(lip, 0); xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); xfs_dqfunlock(dqp); return error; } /* * Lock two xfs_dquot structures. * * To avoid deadlocks we always lock the quota structure with * the lowerd id first. */ void xfs_dqlock2( struct xfs_dquot *d1, struct xfs_dquot *d2) { if (d1 && d2) { ASSERT(d1 != d2); if (d1->q_id > d2->q_id) { mutex_lock(&d2->q_qlock); mutex_lock_nested(&d1->q_qlock, XFS_QLOCK_NESTED); } else { mutex_lock(&d1->q_qlock); mutex_lock_nested(&d2->q_qlock, XFS_QLOCK_NESTED); } } else if (d1) { mutex_lock(&d1->q_qlock); } else if (d2) { mutex_lock(&d2->q_qlock); } } static int xfs_dqtrx_cmp( const void *a, const void *b) { const struct xfs_dqtrx *qa = a; const struct xfs_dqtrx *qb = b; if (qa->qt_dquot->q_id > qb->qt_dquot->q_id) return 1; if (qa->qt_dquot->q_id < qb->qt_dquot->q_id) return -1; return 0; } void xfs_dqlockn( struct xfs_dqtrx *q) { unsigned int i; BUILD_BUG_ON(XFS_QM_TRANS_MAXDQS > MAX_LOCKDEP_SUBCLASSES); /* Sort in order of dquot id, do not allow duplicates */ for (i = 0; i < XFS_QM_TRANS_MAXDQS && q[i].qt_dquot != NULL; i++) { unsigned int j; for (j = 0; j < i; j++) ASSERT(q[i].qt_dquot != q[j].qt_dquot); } if (i == 0) return; sort(q, i, sizeof(struct xfs_dqtrx), xfs_dqtrx_cmp, NULL); mutex_lock(&q[0].qt_dquot->q_qlock); for (i = 1; i < XFS_QM_TRANS_MAXDQS && q[i].qt_dquot != NULL; i++) mutex_lock_nested(&q[i].qt_dquot->q_qlock, XFS_QLOCK_NESTED + i - 1); } int __init xfs_qm_init(void) { xfs_dquot_cache = kmem_cache_create("xfs_dquot", sizeof(struct xfs_dquot), 0, 0, NULL); if (!xfs_dquot_cache) goto out; xfs_dqtrx_cache = kmem_cache_create("xfs_dqtrx", sizeof(struct xfs_dquot_acct), 0, 0, NULL); if (!xfs_dqtrx_cache) goto out_free_dquot_cache; return 0; out_free_dquot_cache: kmem_cache_destroy(xfs_dquot_cache); out: return -ENOMEM; } void xfs_qm_exit(void) { kmem_cache_destroy(xfs_dqtrx_cache); kmem_cache_destroy(xfs_dquot_cache); } |
| 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 17 17 2 15 23 22 23 23 8 16 15 16 16 7 7 1 6 6 14 14 14 5 16 14 2 2 13 15 1 14 14 5 5 16 1 14 14 14 14 14 5 5 5 2 2 2 2 1 1 1 1 1 1 2 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 | // SPDX-License-Identifier: GPL-2.0-or-later /* * Core registration and callback routines for MTD * drivers and users. * * Copyright © 1999-2010 David Woodhouse <dwmw2@infradead.org> * Copyright © 2006 Red Hat UK Limited */ #include <linux/module.h> #include <linux/kernel.h> #include <linux/ptrace.h> #include <linux/seq_file.h> #include <linux/string.h> #include <linux/timer.h> #include <linux/major.h> #include <linux/fs.h> #include <linux/err.h> #include <linux/ioctl.h> #include <linux/init.h> #include <linux/of.h> #include <linux/proc_fs.h> #include <linux/idr.h> #include <linux/backing-dev.h> #include <linux/gfp.h> #include <linux/random.h> #include <linux/slab.h> #include <linux/reboot.h> #include <linux/leds.h> #include <linux/debugfs.h> #include <linux/nvmem-provider.h> #include <linux/root_dev.h> #include <linux/error-injection.h> #include <linux/mtd/mtd.h> #include <linux/mtd/partitions.h> #include "mtdcore.h" struct backing_dev_info *mtd_bdi; #ifdef CONFIG_PM_SLEEP static int mtd_cls_suspend(struct device *dev) { struct mtd_info *mtd = dev_get_drvdata(dev); return mtd ? mtd_suspend(mtd) : 0; } static int mtd_cls_resume(struct device *dev) { struct mtd_info *mtd = dev_get_drvdata(dev); if (mtd) mtd_resume(mtd); return 0; } static SIMPLE_DEV_PM_OPS(mtd_cls_pm_ops, mtd_cls_suspend, mtd_cls_resume); #define MTD_CLS_PM_OPS (&mtd_cls_pm_ops) #else #define MTD_CLS_PM_OPS NULL #endif static struct class mtd_class = { .name = "mtd", .pm = MTD_CLS_PM_OPS, }; static DEFINE_IDR(mtd_idr); /* These are exported solely for the purpose of mtd_blkdevs.c. You should not use them for _anything_ else */ DEFINE_MUTEX(mtd_table_mutex); EXPORT_SYMBOL_GPL(mtd_table_mutex); struct mtd_info *__mtd_next_device(int i) { return idr_get_next(&mtd_idr, &i); } EXPORT_SYMBOL_GPL(__mtd_next_device); static LIST_HEAD(mtd_notifiers); #define MTD_DEVT(index) MKDEV(MTD_CHAR_MAJOR, (index)*2) /* REVISIT once MTD uses the driver model better, whoever allocates * the mtd_info will probably want to use the release() hook... */ static void mtd_release(struct device *dev) { struct mtd_info *mtd = dev_get_drvdata(dev); dev_t index = MTD_DEVT(mtd->index); idr_remove(&mtd_idr, mtd->index); of_node_put(mtd_get_of_node(mtd)); if (mtd_is_partition(mtd)) release_mtd_partition(mtd); /* remove /dev/mtdXro node */ device_destroy(&mtd_class, index + 1); } static void mtd_device_release(struct kref *kref) { struct mtd_info *mtd = container_of(kref, struct mtd_info, refcnt); bool is_partition = mtd_is_partition(mtd); debugfs_remove_recursive(mtd->dbg.dfs_dir); /* Try to remove the NVMEM provider */ nvmem_unregister(mtd->nvmem); device_unregister(&mtd->dev); /* * Clear dev so mtd can be safely re-registered later if desired. * Should not be done for partition, * as it was already destroyed in device_unregister(). */ if (!is_partition) memset(&mtd->dev, 0, sizeof(mtd->dev)); module_put(THIS_MODULE); } #define MTD_DEVICE_ATTR_RO(name) \ static DEVICE_ATTR(name, 0444, mtd_##name##_show, NULL) #define MTD_DEVICE_ATTR_RW(name) \ static DEVICE_ATTR(name, 0644, mtd_##name##_show, mtd_##name##_store) static ssize_t mtd_type_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); char *type; switch (mtd->type) { case MTD_ABSENT: type = "absent"; break; case MTD_RAM: type = "ram"; break; case MTD_ROM: type = "rom"; break; case MTD_NORFLASH: type = "nor"; break; case MTD_NANDFLASH: type = "nand"; break; case MTD_DATAFLASH: type = "dataflash"; break; case MTD_UBIVOLUME: type = "ubi"; break; case MTD_MLCNANDFLASH: type = "mlc-nand"; break; default: type = "unknown"; } return sysfs_emit(buf, "%s\n", type); } MTD_DEVICE_ATTR_RO(type); static ssize_t mtd_flags_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "0x%lx\n", (unsigned long)mtd->flags); } MTD_DEVICE_ATTR_RO(flags); static ssize_t mtd_size_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%llu\n", (unsigned long long)mtd->size); } MTD_DEVICE_ATTR_RO(size); static ssize_t mtd_erasesize_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%lu\n", (unsigned long)mtd->erasesize); } MTD_DEVICE_ATTR_RO(erasesize); static ssize_t mtd_writesize_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%lu\n", (unsigned long)mtd->writesize); } MTD_DEVICE_ATTR_RO(writesize); static ssize_t mtd_subpagesize_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); unsigned int subpagesize = mtd->writesize >> mtd->subpage_sft; return sysfs_emit(buf, "%u\n", subpagesize); } MTD_DEVICE_ATTR_RO(subpagesize); static ssize_t mtd_oobsize_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%lu\n", (unsigned long)mtd->oobsize); } MTD_DEVICE_ATTR_RO(oobsize); static ssize_t mtd_oobavail_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%u\n", mtd->oobavail); } MTD_DEVICE_ATTR_RO(oobavail); static ssize_t mtd_numeraseregions_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%u\n", mtd->numeraseregions); } MTD_DEVICE_ATTR_RO(numeraseregions); static ssize_t mtd_name_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%s\n", mtd->name); } MTD_DEVICE_ATTR_RO(name); static ssize_t mtd_ecc_strength_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%u\n", mtd->ecc_strength); } MTD_DEVICE_ATTR_RO(ecc_strength); static ssize_t mtd_bitflip_threshold_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%u\n", mtd->bitflip_threshold); } static ssize_t mtd_bitflip_threshold_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct mtd_info *mtd = dev_get_drvdata(dev); unsigned int bitflip_threshold; int retval; retval = kstrtouint(buf, 0, &bitflip_threshold); if (retval) return retval; mtd->bitflip_threshold = bitflip_threshold; return count; } MTD_DEVICE_ATTR_RW(bitflip_threshold); static ssize_t mtd_ecc_step_size_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); return sysfs_emit(buf, "%u\n", mtd->ecc_step_size); } MTD_DEVICE_ATTR_RO(ecc_step_size); static ssize_t mtd_corrected_bits_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); struct mtd_ecc_stats *ecc_stats = &mtd->ecc_stats; return sysfs_emit(buf, "%u\n", ecc_stats->corrected); } MTD_DEVICE_ATTR_RO(corrected_bits); /* ecc stats corrected */ static ssize_t mtd_ecc_failures_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); struct mtd_ecc_stats *ecc_stats = &mtd->ecc_stats; return sysfs_emit(buf, "%u\n", ecc_stats->failed); } MTD_DEVICE_ATTR_RO(ecc_failures); /* ecc stats errors */ static ssize_t mtd_bad_blocks_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); struct mtd_ecc_stats *ecc_stats = &mtd->ecc_stats; return sysfs_emit(buf, "%u\n", ecc_stats->badblocks); } MTD_DEVICE_ATTR_RO(bad_blocks); static ssize_t mtd_bbt_blocks_show(struct device *dev, struct device_attribute *attr, char *buf) { struct mtd_info *mtd = dev_get_drvdata(dev); struct mtd_ecc_stats *ecc_stats = &mtd->ecc_stats; return sysfs_emit(buf, "%u\n", ecc_stats->bbtblocks); } MTD_DEVICE_ATTR_RO(bbt_blocks); static struct attribute *mtd_attrs[] = { &dev_attr_type.attr, &dev_attr_flags.attr, &dev_attr_size.attr, &dev_attr_erasesize.attr, &dev_attr_writesize.attr, &dev_attr_subpagesize.attr, &dev_attr_oobsize.attr, &dev_attr_oobavail.attr, &dev_attr_numeraseregions.attr, &dev_attr_name.attr, &dev_attr_ecc_strength.attr, &dev_attr_ecc_step_size.attr, &dev_attr_corrected_bits.attr, &dev_attr_ecc_failures.attr, &dev_attr_bad_blocks.attr, &dev_attr_bbt_blocks.attr, &dev_attr_bitflip_threshold.attr, NULL, }; ATTRIBUTE_GROUPS(mtd); static const struct device_type mtd_devtype = { .name = "mtd", .groups = mtd_groups, .release = mtd_release, }; static bool mtd_expert_analysis_mode; #ifdef CONFIG_DEBUG_FS bool mtd_check_expert_analysis_mode(void) { const char *mtd_expert_analysis_warning = "Bad block checks have been entirely disabled.\n" "This is only reserved for post-mortem forensics and debug purposes.\n" "Never enable this mode if you do not know what you are doing!\n"; return WARN_ONCE(mtd_expert_analysis_mode, mtd_expert_analysis_warning); } EXPORT_SYMBOL_GPL(mtd_check_expert_analysis_mode); #endif static struct dentry *dfs_dir_mtd; static int mtd_ooblayout_show(struct seq_file *s, void *p, int (*iter)(struct mtd_info *, int section, struct mtd_oob_region *region)) { struct mtd_info *mtd = s->private; int section; for (section = 0;; section++) { struct mtd_oob_region region; int err; err = iter(mtd, section, ®ion); if (err) { if (err == -ERANGE) break; return err; } seq_printf(s, "%-3d %4u %4u\n", section, region.offset, region.length); } return 0; } static int mtd_ooblayout_ecc_show(struct seq_file *s, void *p) { return mtd_ooblayout_show(s, p, mtd_ooblayout_ecc); } DEFINE_SHOW_ATTRIBUTE(mtd_ooblayout_ecc); static int mtd_ooblayout_free_show(struct seq_file *s, void *p) { return mtd_ooblayout_show(s, p, mtd_ooblayout_free); } DEFINE_SHOW_ATTRIBUTE(mtd_ooblayout_free); static void mtd_debugfs_populate(struct mtd_info *mtd) { struct device *dev = &mtd->dev; struct mtd_oob_region region; if (IS_ERR_OR_NULL(dfs_dir_mtd)) return; mtd->dbg.dfs_dir = debugfs_create_dir(dev_name(dev), dfs_dir_mtd); if (IS_ERR_OR_NULL(mtd->dbg.dfs_dir)) return; /* Create ooblayout files only if at least one region is present. */ if (mtd_ooblayout_ecc(mtd, 0, ®ion) == 0) debugfs_create_file("ooblayout_ecc", 0444, mtd->dbg.dfs_dir, mtd, &mtd_ooblayout_ecc_fops); if (mtd_ooblayout_free(mtd, 0, ®ion) == 0) debugfs_create_file("ooblayout_free", 0444, mtd->dbg.dfs_dir, mtd, &mtd_ooblayout_free_fops); } #ifndef CONFIG_MMU unsigned mtd_mmap_capabilities(struct mtd_info *mtd) { switch (mtd->type) { case MTD_RAM: return NOMMU_MAP_COPY | NOMMU_MAP_DIRECT | NOMMU_MAP_EXEC | NOMMU_MAP_READ | NOMMU_MAP_WRITE; case MTD_ROM: return NOMMU_MAP_COPY | NOMMU_MAP_DIRECT | NOMMU_MAP_EXEC | NOMMU_MAP_READ; default: return NOMMU_MAP_COPY; } } EXPORT_SYMBOL_GPL(mtd_mmap_capabilities); #endif static int mtd_reboot_notifier(struct notifier_block *n, unsigned long state, void *cmd) { struct mtd_info *mtd; mtd = container_of(n, struct mtd_info, reboot_notifier); mtd->_reboot(mtd); return NOTIFY_DONE; } /** * mtd_wunit_to_pairing_info - get pairing information of a wunit * @mtd: pointer to new MTD device info structure * @wunit: write unit we are interested in * @info: returned pairing information * * Retrieve pairing information associated to the wunit. * This is mainly useful when dealing with MLC/TLC NANDs where pages can be * paired together, and where programming a page may influence the page it is * paired with. * The notion of page is replaced by the term wunit (write-unit) to stay * consistent with the ->writesize field. * * The @wunit argument can be extracted from an absolute offset using * mtd_offset_to_wunit(). @info is filled with the pairing information attached * to @wunit. * * From the pairing info the MTD user can find all the wunits paired with * @wunit using the following loop: * * for (i = 0; i < mtd_pairing_groups(mtd); i++) { * info.pair = i; * mtd_pairing_info_to_wunit(mtd, &info); * ... * } */ int mtd_wunit_to_pairing_info(struct mtd_info *mtd, int wunit, struct mtd_pairing_info *info) { struct mtd_info *master = mtd_get_master(mtd); int npairs = mtd_wunit_per_eb(master) / mtd_pairing_groups(master); if (wunit < 0 || wunit >= npairs) return -EINVAL; if (master->pairing && master->pairing->get_info) return master->pairing->get_info(master, wunit, info); info->group = 0; info->pair = wunit; return 0; } EXPORT_SYMBOL_GPL(mtd_wunit_to_pairing_info); /** * mtd_pairing_info_to_wunit - get wunit from pairing information * @mtd: pointer to new MTD device info structure * @info: pairing information struct * * Returns a positive number representing the wunit associated to the info * struct, or a negative error code. * * This is the reverse of mtd_wunit_to_pairing_info(), and can help one to * iterate over all wunits of a given pair (see mtd_wunit_to_pairing_info() * doc). * * It can also be used to only program the first page of each pair (i.e. * page attached to group 0), which allows one to use an MLC NAND in * software-emulated SLC mode: * * info.group = 0; * npairs = mtd_wunit_per_eb(mtd) / mtd_pairing_groups(mtd); * for (info.pair = 0; info.pair < npairs; info.pair++) { * wunit = mtd_pairing_info_to_wunit(mtd, &info); * mtd_write(mtd, mtd_wunit_to_offset(mtd, blkoffs, wunit), * mtd->writesize, &retlen, buf + (i * mtd->writesize)); * } */ int mtd_pairing_info_to_wunit(struct mtd_info *mtd, const struct mtd_pairing_info *info) { struct mtd_info *master = mtd_get_master(mtd); int ngroups = mtd_pairing_groups(master); int npairs = mtd_wunit_per_eb(master) / ngroups; if (!info || info->pair < 0 || info->pair >= npairs || info->group < 0 || info->group >= ngroups) return -EINVAL; if (master->pairing && master->pairing->get_wunit) return mtd->pairing->get_wunit(master, info); return info->pair; } EXPORT_SYMBOL_GPL(mtd_pairing_info_to_wunit); /** * mtd_pairing_groups - get the number of pairing groups * @mtd: pointer to new MTD device info structure * * Returns the number of pairing groups. * * This number is usually equal to the number of bits exposed by a single * cell, and can be used in conjunction with mtd_pairing_info_to_wunit() * to iterate over all pages of a given pair. */ int mtd_pairing_groups(struct mtd_info *mtd) { struct mtd_info *master = mtd_get_master(mtd); if (!master->pairing || !master->pairing->ngroups) return 1; return master->pairing->ngroups; } EXPORT_SYMBOL_GPL(mtd_pairing_groups); static int mtd_nvmem_reg_read(void *priv, unsigned int offset, void *val, size_t bytes) { struct mtd_info *mtd = priv; size_t retlen; int err; err = mtd_read(mtd, offset, bytes, &retlen, val); if (err && err != -EUCLEAN) return err; return retlen == bytes ? 0 : -EIO; } static int mtd_nvmem_add(struct mtd_info *mtd) { struct device_node *node = mtd_get_of_node(mtd); struct nvmem_config config = {}; config.id = NVMEM_DEVID_NONE; config.dev = &mtd->dev; config.name = dev_name(&mtd->dev); config.owner = THIS_MODULE; config.add_legacy_fixed_of_cells = of_device_is_compatible(node, "nvmem-cells"); config.reg_read = mtd_nvmem_reg_read; config.size = mtd->size; config.word_size = 1; config.stride = 1; config.read_only = true; config.root_only = true; config.ignore_wp = true; config.priv = mtd; mtd->nvmem = nvmem_register(&config); if (IS_ERR(mtd->nvmem)) { /* Just ignore if there is no NVMEM support in the kernel */ if (PTR_ERR(mtd->nvmem) == -EOPNOTSUPP) mtd->nvmem = NULL; else return dev_err_probe(&mtd->dev, PTR_ERR(mtd->nvmem), "Failed to register NVMEM device\n"); } return 0; } static void mtd_check_of_node(struct mtd_info *mtd) { struct device_node *partitions, *parent_dn, *mtd_dn = NULL; const char *pname, *prefix = "partition-"; int plen, mtd_name_len, offset, prefix_len; /* Check if MTD already has a device node */ if (mtd_get_of_node(mtd)) return; if (!mtd_is_partition(mtd)) return; parent_dn = of_node_get(mtd_get_of_node(mtd->parent)); if (!parent_dn) return; if (mtd_is_partition(mtd->parent)) partitions = of_node_get(parent_dn); else partitions = of_get_child_by_name(parent_dn, "partitions"); if (!partitions) goto exit_parent; prefix_len = strlen(prefix); mtd_name_len = strlen(mtd->name); /* Search if a partition is defined with the same name */ for_each_child_of_node(partitions, mtd_dn) { /* Skip partition with no/wrong prefix */ if (!of_node_name_prefix(mtd_dn, prefix)) continue; /* Label have priority. Check that first */ if (!of_property_read_string(mtd_dn, "label", &pname)) { offset = 0; } else { pname = mtd_dn->name; offset = prefix_len; } plen = strlen(pname) - offset; if (plen == mtd_name_len && !strncmp(mtd->name, pname + offset, plen)) { mtd_set_of_node(mtd, mtd_dn); of_node_put(mtd_dn); break; } } of_node_put(partitions); exit_parent: of_node_put(parent_dn); } /** * add_mtd_device - register an MTD device * @mtd: pointer to new MTD device info structure * * Add a device to the list of MTD devices present in the system, and * notify each currently active MTD 'user' of its arrival. Returns * zero on success or non-zero on failure. */ int add_mtd_device(struct mtd_info *mtd) { struct device_node *np = mtd_get_of_node(mtd); struct mtd_info *master = mtd_get_master(mtd); struct mtd_notifier *not; int i, error, ofidx; /* * May occur, for instance, on buggy drivers which call * mtd_device_parse_register() multiple times on the same master MTD, * especially with CONFIG_MTD_PARTITIONED_MASTER=y. */ if (WARN_ONCE(mtd->dev.type, "MTD already registered\n")) return -EEXIST; BUG_ON(mtd->writesize == 0); /* * MTD drivers should implement ->_{write,read}() or * ->_{write,read}_oob(), but not both. */ if (WARN_ON((mtd->_write && mtd->_write_oob) || (mtd->_read && mtd->_read_oob))) return -EINVAL; if (WARN_ON((!mtd->erasesize || !master->_erase) && !(mtd->flags & MTD_NO_ERASE))) return -EINVAL; /* * MTD_SLC_ON_MLC_EMULATION can only be set on partitions, when the * master is an MLC NAND and has a proper pairing scheme defined. * We also reject masters that implement ->_writev() for now, because * NAND controller drivers don't implement this hook, and adding the * SLC -> MLC address/length conversion to this path is useless if we * don't have a user. */ if (mtd->flags & MTD_SLC_ON_MLC_EMULATION && (!mtd_is_partition(mtd) || master->type != MTD_MLCNANDFLASH || !master->pairing || master->_writev)) return -EINVAL; mutex_lock(&mtd_table_mutex); ofidx = -1; if (np) ofidx = of_alias_get_id(np, "mtd"); if (ofidx >= 0) i = idr_alloc(&mtd_idr, mtd, ofidx, ofidx + 1, GFP_KERNEL); else i = idr_alloc(&mtd_idr, mtd, 0, 0, GFP_KERNEL); if (i < 0) { error = i; goto fail_locked; } mtd->index = i; kref_init(&mtd->refcnt); /* default value if not set by driver */ if (mtd->bitflip_threshold == 0) mtd->bitflip_threshold = mtd->ecc_strength; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) { int ngroups = mtd_pairing_groups(master); mtd->erasesize /= ngroups; mtd->size = (u64)mtd_div_by_eb(mtd->size, master) * mtd->erasesize; } if (is_power_of_2(mtd->erasesize)) mtd->erasesize_shift = ffs(mtd->erasesize) - 1; else mtd->erasesize_shift = 0; if (is_power_of_2(mtd->writesize)) mtd->writesize_shift = ffs(mtd->writesize) - 1; else mtd->writesize_shift = 0; mtd->erasesize_mask = (1 << mtd->erasesize_shift) - 1; mtd->writesize_mask = (1 << mtd->writesize_shift) - 1; /* Some chips always power up locked. Unlock them now */ if ((mtd->flags & MTD_WRITEABLE) && (mtd->flags & MTD_POWERUP_LOCK)) { error = mtd_unlock(mtd, 0, mtd->size); if (error && error != -EOPNOTSUPP) printk(KERN_WARNING "%s: unlock failed, writes may not work\n", mtd->name); /* Ignore unlock failures? */ error = 0; } /* Caller should have set dev.parent to match the * physical device, if appropriate. */ mtd->dev.type = &mtd_devtype; mtd->dev.class = &mtd_class; mtd->dev.devt = MTD_DEVT(i); error = dev_set_name(&mtd->dev, "mtd%d", i); if (error) goto fail_devname; dev_set_drvdata(&mtd->dev, mtd); mtd_check_of_node(mtd); of_node_get(mtd_get_of_node(mtd)); error = device_register(&mtd->dev); if (error) { put_device(&mtd->dev); goto fail_added; } /* Add the nvmem provider */ error = mtd_nvmem_add(mtd); if (error) goto fail_nvmem_add; mtd_debugfs_populate(mtd); device_create(&mtd_class, mtd->dev.parent, MTD_DEVT(i) + 1, NULL, "mtd%dro", i); pr_debug("mtd: Giving out device %d to %s\n", i, mtd->name); /* No need to get a refcount on the module containing the notifier, since we hold the mtd_table_mutex */ list_for_each_entry(not, &mtd_notifiers, list) not->add(mtd); mutex_unlock(&mtd_table_mutex); if (of_property_read_bool(mtd_get_of_node(mtd), "linux,rootfs")) { if (IS_BUILTIN(CONFIG_MTD)) { pr_info("mtd: setting mtd%d (%s) as root device\n", mtd->index, mtd->name); ROOT_DEV = MKDEV(MTD_BLOCK_MAJOR, mtd->index); } else { pr_warn("mtd: can't set mtd%d (%s) as root device - mtd must be builtin\n", mtd->index, mtd->name); } } /* We _know_ we aren't being removed, because our caller is still holding us here. So none of this try_ nonsense, and no bitching about it either. :) */ __module_get(THIS_MODULE); return 0; fail_nvmem_add: device_unregister(&mtd->dev); fail_added: of_node_put(mtd_get_of_node(mtd)); fail_devname: idr_remove(&mtd_idr, i); fail_locked: mutex_unlock(&mtd_table_mutex); return error; } /** * del_mtd_device - unregister an MTD device * @mtd: pointer to MTD device info structure * * Remove a device from the list of MTD devices present in the system, * and notify each currently active MTD 'user' of its departure. * Returns zero on success or 1 on failure, which currently will happen * if the requested device does not appear to be present in the list. */ int del_mtd_device(struct mtd_info *mtd) { int ret; struct mtd_notifier *not; mutex_lock(&mtd_table_mutex); if (idr_find(&mtd_idr, mtd->index) != mtd) { ret = -ENODEV; goto out_error; } /* No need to get a refcount on the module containing the notifier, since we hold the mtd_table_mutex */ list_for_each_entry(not, &mtd_notifiers, list) not->remove(mtd); kref_put(&mtd->refcnt, mtd_device_release); ret = 0; out_error: mutex_unlock(&mtd_table_mutex); return ret; } /* * Set a few defaults based on the parent devices, if not provided by the * driver */ static void mtd_set_dev_defaults(struct mtd_info *mtd) { if (mtd->dev.parent) { if (!mtd->owner && mtd->dev.parent->driver) mtd->owner = mtd->dev.parent->driver->owner; if (!mtd->name) mtd->name = dev_name(mtd->dev.parent); } else { pr_debug("mtd device won't show a device symlink in sysfs\n"); } INIT_LIST_HEAD(&mtd->partitions); mutex_init(&mtd->master.partitions_lock); mutex_init(&mtd->master.chrdev_lock); } static ssize_t mtd_otp_size(struct mtd_info *mtd, bool is_user) { struct otp_info *info; ssize_t size = 0; unsigned int i; size_t retlen; int ret; info = kmalloc(PAGE_SIZE, GFP_KERNEL); if (!info) return -ENOMEM; if (is_user) ret = mtd_get_user_prot_info(mtd, PAGE_SIZE, &retlen, info); else ret = mtd_get_fact_prot_info(mtd, PAGE_SIZE, &retlen, info); if (ret) goto err; for (i = 0; i < retlen / sizeof(*info); i++) size += info[i].length; kfree(info); return size; err: kfree(info); /* ENODATA means there is no OTP region. */ return ret == -ENODATA ? 0 : ret; } static struct nvmem_device *mtd_otp_nvmem_register(struct mtd_info *mtd, const char *compatible, int size, nvmem_reg_read_t reg_read) { struct nvmem_device *nvmem = NULL; struct nvmem_config config = {}; struct device_node *np; /* DT binding is optional */ np = of_get_compatible_child(mtd->dev.of_node, compatible); /* OTP nvmem will be registered on the physical device */ config.dev = mtd->dev.parent; config.name = compatible; config.id = NVMEM_DEVID_AUTO; config.owner = THIS_MODULE; config.add_legacy_fixed_of_cells = !mtd_type_is_nand(mtd); config.type = NVMEM_TYPE_OTP; config.root_only = true; config.ignore_wp = true; config.reg_read = reg_read; config.size = size; config.of_node = np; config.priv = mtd; nvmem = nvmem_register(&config); /* Just ignore if there is no NVMEM support in the kernel */ if (IS_ERR(nvmem) && PTR_ERR(nvmem) == -EOPNOTSUPP) nvmem = NULL; of_node_put(np); return nvmem; } static int mtd_nvmem_user_otp_reg_read(void *priv, unsigned int offset, void *val, size_t bytes) { struct mtd_info *mtd = priv; size_t retlen; int ret; ret = mtd_read_user_prot_reg(mtd, offset, bytes, &retlen, val); if (ret) return ret; return retlen == bytes ? 0 : -EIO; } static int mtd_nvmem_fact_otp_reg_read(void *priv, unsigned int offset, void *val, size_t bytes) { struct mtd_info *mtd = priv; size_t retlen; int ret; ret = mtd_read_fact_prot_reg(mtd, offset, bytes, &retlen, val); if (ret) return ret; return retlen == bytes ? 0 : -EIO; } static int mtd_otp_nvmem_add(struct mtd_info *mtd) { struct device *dev = mtd->dev.parent; struct nvmem_device *nvmem; ssize_t size; int err; if (mtd->_get_user_prot_info && mtd->_read_user_prot_reg) { size = mtd_otp_size(mtd, true); if (size < 0) { err = size; goto err; } if (size > 0) { nvmem = mtd_otp_nvmem_register(mtd, "user-otp", size, mtd_nvmem_user_otp_reg_read); if (IS_ERR(nvmem)) { err = PTR_ERR(nvmem); goto err; } mtd->otp_user_nvmem = nvmem; } } if (mtd->_get_fact_prot_info && mtd->_read_fact_prot_reg) { size = mtd_otp_size(mtd, false); if (size < 0) { err = size; goto err; } if (size > 0) { /* * The factory OTP contains thing such as a unique serial * number and is small, so let's read it out and put it * into the entropy pool. */ void *otp; otp = kmalloc(size, GFP_KERNEL); if (!otp) { err = -ENOMEM; goto err; } err = mtd_nvmem_fact_otp_reg_read(mtd, 0, otp, size); if (err < 0) { kfree(otp); goto err; } add_device_randomness(otp, err); kfree(otp); nvmem = mtd_otp_nvmem_register(mtd, "factory-otp", size, mtd_nvmem_fact_otp_reg_read); if (IS_ERR(nvmem)) { err = PTR_ERR(nvmem); goto err; } mtd->otp_factory_nvmem = nvmem; } } return 0; err: nvmem_unregister(mtd->otp_user_nvmem); /* Don't report error if OTP is not supported. */ if (err == -EOPNOTSUPP) return 0; return dev_err_probe(dev, err, "Failed to register OTP NVMEM device\n"); } /** * mtd_device_parse_register - parse partitions and register an MTD device. * * @mtd: the MTD device to register * @types: the list of MTD partition probes to try, see * 'parse_mtd_partitions()' for more information * @parser_data: MTD partition parser-specific data * @parts: fallback partition information to register, if parsing fails; * only valid if %nr_parts > %0 * @nr_parts: the number of partitions in parts, if zero then the full * MTD device is registered if no partition info is found * * This function aggregates MTD partitions parsing (done by * 'parse_mtd_partitions()') and MTD device and partitions registering. It * basically follows the most common pattern found in many MTD drivers: * * * If the MTD_PARTITIONED_MASTER option is set, then the device as a whole is * registered first. * * Then It tries to probe partitions on MTD device @mtd using parsers * specified in @types (if @types is %NULL, then the default list of parsers * is used, see 'parse_mtd_partitions()' for more information). If none are * found this functions tries to fallback to information specified in * @parts/@nr_parts. * * If no partitions were found this function just registers the MTD device * @mtd and exits. * * Returns zero in case of success and a negative error code in case of failure. */ int mtd_device_parse_register(struct mtd_info *mtd, const char * const *types, struct mtd_part_parser_data *parser_data, const struct mtd_partition *parts, int nr_parts) { int ret, err; mtd_set_dev_defaults(mtd); ret = mtd_otp_nvmem_add(mtd); if (ret) goto out; if (IS_ENABLED(CONFIG_MTD_PARTITIONED_MASTER)) { ret = add_mtd_device(mtd); if (ret) goto out; } /* Prefer parsed partitions over driver-provided fallback */ ret = parse_mtd_partitions(mtd, types, parser_data); if (ret == -EPROBE_DEFER) goto out; if (ret > 0) ret = 0; else if (nr_parts) ret = add_mtd_partitions(mtd, parts, nr_parts); else if (!device_is_registered(&mtd->dev)) ret = add_mtd_device(mtd); else ret = 0; if (ret) goto out; /* * FIXME: some drivers unfortunately call this function more than once. * So we have to check if we've already assigned the reboot notifier. * * Generally, we can make multiple calls work for most cases, but it * does cause problems with parse_mtd_partitions() above (e.g., * cmdlineparts will register partitions more than once). */ WARN_ONCE(mtd->_reboot && mtd->reboot_notifier.notifier_call, "MTD already registered\n"); if (mtd->_reboot && !mtd->reboot_notifier.notifier_call) { mtd->reboot_notifier.notifier_call = mtd_reboot_notifier; register_reboot_notifier(&mtd->reboot_notifier); } out: if (ret) { nvmem_unregister(mtd->otp_user_nvmem); nvmem_unregister(mtd->otp_factory_nvmem); } if (ret && device_is_registered(&mtd->dev)) { err = del_mtd_device(mtd); if (err) pr_err("Error when deleting MTD device (%d)\n", err); } return ret; } EXPORT_SYMBOL_GPL(mtd_device_parse_register); /** * mtd_device_unregister - unregister an existing MTD device. * * @master: the MTD device to unregister. This will unregister both the master * and any partitions if registered. */ int mtd_device_unregister(struct mtd_info *master) { int err; if (master->_reboot) { unregister_reboot_notifier(&master->reboot_notifier); memset(&master->reboot_notifier, 0, sizeof(master->reboot_notifier)); } nvmem_unregister(master->otp_user_nvmem); nvmem_unregister(master->otp_factory_nvmem); err = del_mtd_partitions(master); if (err) return err; if (!device_is_registered(&master->dev)) return 0; return del_mtd_device(master); } EXPORT_SYMBOL_GPL(mtd_device_unregister); /** * register_mtd_user - register a 'user' of MTD devices. * @new: pointer to notifier info structure * * Registers a pair of callbacks function to be called upon addition * or removal of MTD devices. Causes the 'add' callback to be immediately * invoked for each MTD device currently present in the system. */ void register_mtd_user (struct mtd_notifier *new) { struct mtd_info *mtd; mutex_lock(&mtd_table_mutex); list_add(&new->list, &mtd_notifiers); __module_get(THIS_MODULE); mtd_for_each_device(mtd) new->add(mtd); mutex_unlock(&mtd_table_mutex); } EXPORT_SYMBOL_GPL(register_mtd_user); /** * unregister_mtd_user - unregister a 'user' of MTD devices. * @old: pointer to notifier info structure * * Removes a callback function pair from the list of 'users' to be * notified upon addition or removal of MTD devices. Causes the * 'remove' callback to be immediately invoked for each MTD device * currently present in the system. */ int unregister_mtd_user (struct mtd_notifier *old) { struct mtd_info *mtd; mutex_lock(&mtd_table_mutex); module_put(THIS_MODULE); mtd_for_each_device(mtd) old->remove(mtd); list_del(&old->list); mutex_unlock(&mtd_table_mutex); return 0; } EXPORT_SYMBOL_GPL(unregister_mtd_user); /** * get_mtd_device - obtain a validated handle for an MTD device * @mtd: last known address of the required MTD device * @num: internal device number of the required MTD device * * Given a number and NULL address, return the num'th entry in the device * table, if any. Given an address and num == -1, search the device table * for a device with that address and return if it's still present. Given * both, return the num'th driver only if its address matches. Return * error code if not. */ struct mtd_info *get_mtd_device(struct mtd_info *mtd, int num) { struct mtd_info *ret = NULL, *other; int err = -ENODEV; mutex_lock(&mtd_table_mutex); if (num == -1) { mtd_for_each_device(other) { if (other == mtd) { ret = mtd; break; } } } else if (num >= 0) { ret = idr_find(&mtd_idr, num); if (mtd && mtd != ret) ret = NULL; } if (!ret) { ret = ERR_PTR(err); goto out; } err = __get_mtd_device(ret); if (err) ret = ERR_PTR(err); out: mutex_unlock(&mtd_table_mutex); return ret; } EXPORT_SYMBOL_GPL(get_mtd_device); int __get_mtd_device(struct mtd_info *mtd) { struct mtd_info *master = mtd_get_master(mtd); int err; if (master->_get_device) { err = master->_get_device(mtd); if (err) return err; } if (!try_module_get(master->owner)) { if (master->_put_device) master->_put_device(master); return -ENODEV; } while (mtd) { if (mtd != master) kref_get(&mtd->refcnt); mtd = mtd->parent; } if (IS_ENABLED(CONFIG_MTD_PARTITIONED_MASTER)) kref_get(&master->refcnt); return 0; } EXPORT_SYMBOL_GPL(__get_mtd_device); /** * of_get_mtd_device_by_node - obtain an MTD device associated with a given node * * @np: device tree node */ struct mtd_info *of_get_mtd_device_by_node(struct device_node *np) { struct mtd_info *mtd = NULL; struct mtd_info *tmp; int err; mutex_lock(&mtd_table_mutex); err = -EPROBE_DEFER; mtd_for_each_device(tmp) { if (mtd_get_of_node(tmp) == np) { mtd = tmp; err = __get_mtd_device(mtd); break; } } mutex_unlock(&mtd_table_mutex); return err ? ERR_PTR(err) : mtd; } EXPORT_SYMBOL_GPL(of_get_mtd_device_by_node); /** * get_mtd_device_nm - obtain a validated handle for an MTD device by * device name * @name: MTD device name to open * * This function returns MTD device description structure in case of * success and an error code in case of failure. */ struct mtd_info *get_mtd_device_nm(const char *name) { int err = -ENODEV; struct mtd_info *mtd = NULL, *other; mutex_lock(&mtd_table_mutex); mtd_for_each_device(other) { if (!strcmp(name, other->name)) { mtd = other; break; } } if (!mtd) goto out_unlock; err = __get_mtd_device(mtd); if (err) goto out_unlock; mutex_unlock(&mtd_table_mutex); return mtd; out_unlock: mutex_unlock(&mtd_table_mutex); return ERR_PTR(err); } EXPORT_SYMBOL_GPL(get_mtd_device_nm); void put_mtd_device(struct mtd_info *mtd) { mutex_lock(&mtd_table_mutex); __put_mtd_device(mtd); mutex_unlock(&mtd_table_mutex); } EXPORT_SYMBOL_GPL(put_mtd_device); void __put_mtd_device(struct mtd_info *mtd) { struct mtd_info *master = mtd_get_master(mtd); while (mtd) { /* kref_put() can relese mtd, so keep a reference mtd->parent */ struct mtd_info *parent = mtd->parent; if (mtd != master) kref_put(&mtd->refcnt, mtd_device_release); mtd = parent; } if (IS_ENABLED(CONFIG_MTD_PARTITIONED_MASTER)) kref_put(&master->refcnt, mtd_device_release); module_put(master->owner); /* must be the last as master can be freed in the _put_device */ if (master->_put_device) master->_put_device(master); } EXPORT_SYMBOL_GPL(__put_mtd_device); /* * Erase is an synchronous operation. Device drivers are epected to return a * negative error code if the operation failed and update instr->fail_addr * to point the portion that was not properly erased. */ int mtd_erase(struct mtd_info *mtd, struct erase_info *instr) { struct mtd_info *master = mtd_get_master(mtd); u64 mst_ofs = mtd_get_master_ofs(mtd, 0); struct erase_info adjinstr; int ret; instr->fail_addr = MTD_FAIL_ADDR_UNKNOWN; adjinstr = *instr; if (!mtd->erasesize || !master->_erase) return -ENOTSUPP; if (instr->addr >= mtd->size || instr->len > mtd->size - instr->addr) return -EINVAL; if (!(mtd->flags & MTD_WRITEABLE)) return -EROFS; if (!instr->len) return 0; ledtrig_mtd_activity(); if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) { adjinstr.addr = (loff_t)mtd_div_by_eb(instr->addr, mtd) * master->erasesize; adjinstr.len = ((u64)mtd_div_by_eb(instr->addr + instr->len, mtd) * master->erasesize) - adjinstr.addr; } adjinstr.addr += mst_ofs; ret = master->_erase(master, &adjinstr); if (adjinstr.fail_addr != MTD_FAIL_ADDR_UNKNOWN) { instr->fail_addr = adjinstr.fail_addr - mst_ofs; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) { instr->fail_addr = mtd_div_by_eb(instr->fail_addr, master); instr->fail_addr *= mtd->erasesize; } } return ret; } EXPORT_SYMBOL_GPL(mtd_erase); ALLOW_ERROR_INJECTION(mtd_erase, ERRNO); /* * This stuff for eXecute-In-Place. phys is optional and may be set to NULL. */ int mtd_point(struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, void **virt, resource_size_t *phys) { struct mtd_info *master = mtd_get_master(mtd); *retlen = 0; *virt = NULL; if (phys) *phys = 0; if (!master->_point) return -EOPNOTSUPP; if (from < 0 || from >= mtd->size || len > mtd->size - from) return -EINVAL; if (!len) return 0; from = mtd_get_master_ofs(mtd, from); return master->_point(master, from, len, retlen, virt, phys); } EXPORT_SYMBOL_GPL(mtd_point); /* We probably shouldn't allow XIP if the unpoint isn't a NULL */ int mtd_unpoint(struct mtd_info *mtd, loff_t from, size_t len) { struct mtd_info *master = mtd_get_master(mtd); if (!master->_unpoint) return -EOPNOTSUPP; if (from < 0 || from >= mtd->size || len > mtd->size - from) return -EINVAL; if (!len) return 0; return master->_unpoint(master, mtd_get_master_ofs(mtd, from), len); } EXPORT_SYMBOL_GPL(mtd_unpoint); /* * Allow NOMMU mmap() to directly map the device (if not NULL) * - return the address to which the offset maps * - return -ENOSYS to indicate refusal to do the mapping */ unsigned long mtd_get_unmapped_area(struct mtd_info *mtd, unsigned long len, unsigned long offset, unsigned long flags) { size_t retlen; void *virt; int ret; ret = mtd_point(mtd, offset, len, &retlen, &virt, NULL); if (ret) return ret; if (retlen != len) { mtd_unpoint(mtd, offset, retlen); return -ENOSYS; } return (unsigned long)virt; } EXPORT_SYMBOL_GPL(mtd_get_unmapped_area); static void mtd_update_ecc_stats(struct mtd_info *mtd, struct mtd_info *master, const struct mtd_ecc_stats *old_stats) { struct mtd_ecc_stats diff; if (master == mtd) return; diff = master->ecc_stats; diff.failed -= old_stats->failed; diff.corrected -= old_stats->corrected; while (mtd->parent) { mtd->ecc_stats.failed += diff.failed; mtd->ecc_stats.corrected += diff.corrected; mtd = mtd->parent; } } int mtd_read(struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char *buf) { struct mtd_oob_ops ops = { .len = len, .datbuf = buf, }; int ret; ret = mtd_read_oob(mtd, from, &ops); *retlen = ops.retlen; WARN_ON_ONCE(*retlen != len && mtd_is_bitflip_or_eccerr(ret)); return ret; } EXPORT_SYMBOL_GPL(mtd_read); ALLOW_ERROR_INJECTION(mtd_read, ERRNO); int mtd_write(struct mtd_info *mtd, loff_t to, size_t len, size_t *retlen, const u_char *buf) { struct mtd_oob_ops ops = { .len = len, .datbuf = (u8 *)buf, }; int ret; ret = mtd_write_oob(mtd, to, &ops); *retlen = ops.retlen; return ret; } EXPORT_SYMBOL_GPL(mtd_write); ALLOW_ERROR_INJECTION(mtd_write, ERRNO); /* * In blackbox flight recorder like scenarios we want to make successful writes * in interrupt context. panic_write() is only intended to be called when its * known the kernel is about to panic and we need the write to succeed. Since * the kernel is not going to be running for much longer, this function can * break locks and delay to ensure the write succeeds (but not sleep). */ int mtd_panic_write(struct mtd_info *mtd, loff_t to, size_t len, size_t *retlen, const u_char *buf) { struct mtd_info *master = mtd_get_master(mtd); *retlen = 0; if (!master->_panic_write) return -EOPNOTSUPP; if (to < 0 || to >= mtd->size || len > mtd->size - to) return -EINVAL; if (!(mtd->flags & MTD_WRITEABLE)) return -EROFS; if (!len) return 0; if (!master->oops_panic_write) master->oops_panic_write = true; return master->_panic_write(master, mtd_get_master_ofs(mtd, to), len, retlen, buf); } EXPORT_SYMBOL_GPL(mtd_panic_write); static int mtd_check_oob_ops(struct mtd_info *mtd, loff_t offs, struct mtd_oob_ops *ops) { /* * Some users are setting ->datbuf or ->oobbuf to NULL, but are leaving * ->len or ->ooblen uninitialized. Force ->len and ->ooblen to 0 in * this case. */ if (!ops->datbuf) ops->len = 0; if (!ops->oobbuf) ops->ooblen = 0; if (offs < 0 || offs + ops->len > mtd->size) return -EINVAL; if (ops->ooblen) { size_t maxooblen; if (ops->ooboffs >= mtd_oobavail(mtd, ops)) return -EINVAL; maxooblen = ((size_t)(mtd_div_by_ws(mtd->size, mtd) - mtd_div_by_ws(offs, mtd)) * mtd_oobavail(mtd, ops)) - ops->ooboffs; if (ops->ooblen > maxooblen) return -EINVAL; } return 0; } static int mtd_read_oob_std(struct mtd_info *mtd, loff_t from, struct mtd_oob_ops *ops) { struct mtd_info *master = mtd_get_master(mtd); int ret; from = mtd_get_master_ofs(mtd, from); if (master->_read_oob) ret = master->_read_oob(master, from, ops); else ret = master->_read(master, from, ops->len, &ops->retlen, ops->datbuf); return ret; } static int mtd_write_oob_std(struct mtd_info *mtd, loff_t to, struct mtd_oob_ops *ops) { struct mtd_info *master = mtd_get_master(mtd); int ret; to = mtd_get_master_ofs(mtd, to); if (master->_write_oob) ret = master->_write_oob(master, to, ops); else ret = master->_write(master, to, ops->len, &ops->retlen, ops->datbuf); return ret; } static int mtd_io_emulated_slc(struct mtd_info *mtd, loff_t start, bool read, struct mtd_oob_ops *ops) { struct mtd_info *master = mtd_get_master(mtd); int ngroups = mtd_pairing_groups(master); int npairs = mtd_wunit_per_eb(master) / ngroups; struct mtd_oob_ops adjops = *ops; unsigned int wunit, oobavail; struct mtd_pairing_info info; int max_bitflips = 0; u32 ebofs, pageofs; loff_t base, pos; ebofs = mtd_mod_by_eb(start, mtd); base = (loff_t)mtd_div_by_eb(start, mtd) * master->erasesize; info.group = 0; info.pair = mtd_div_by_ws(ebofs, mtd); pageofs = mtd_mod_by_ws(ebofs, mtd); oobavail = mtd_oobavail(mtd, ops); while (ops->retlen < ops->len || ops->oobretlen < ops->ooblen) { int ret; if (info.pair >= npairs) { info.pair = 0; base += master->erasesize; } wunit = mtd_pairing_info_to_wunit(master, &info); pos = mtd_wunit_to_offset(mtd, base, wunit); adjops.len = ops->len - ops->retlen; if (adjops.len > mtd->writesize - pageofs) adjops.len = mtd->writesize - pageofs; adjops.ooblen = ops->ooblen - ops->oobretlen; if (adjops.ooblen > oobavail - adjops.ooboffs) adjops.ooblen = oobavail - adjops.ooboffs; if (read) { ret = mtd_read_oob_std(mtd, pos + pageofs, &adjops); if (ret > 0) max_bitflips = max(max_bitflips, ret); } else { ret = mtd_write_oob_std(mtd, pos + pageofs, &adjops); } if (ret < 0) return ret; max_bitflips = max(max_bitflips, ret); ops->retlen += adjops.retlen; ops->oobretlen += adjops.oobretlen; adjops.datbuf += adjops.retlen; adjops.oobbuf += adjops.oobretlen; adjops.ooboffs = 0; pageofs = 0; info.pair++; } return max_bitflips; } int mtd_read_oob(struct mtd_info *mtd, loff_t from, struct mtd_oob_ops *ops) { struct mtd_info *master = mtd_get_master(mtd); struct mtd_ecc_stats old_stats = master->ecc_stats; int ret_code; ops->retlen = ops->oobretlen = 0; ret_code = mtd_check_oob_ops(mtd, from, ops); if (ret_code) return ret_code; ledtrig_mtd_activity(); /* Check the validity of a potential fallback on mtd->_read */ if (!master->_read_oob && (!master->_read || ops->oobbuf)) return -EOPNOTSUPP; if (ops->stats) memset(ops->stats, 0, sizeof(*ops->stats)); if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) ret_code = mtd_io_emulated_slc(mtd, from, true, ops); else ret_code = mtd_read_oob_std(mtd, from, ops); mtd_update_ecc_stats(mtd, master, &old_stats); /* * In cases where ops->datbuf != NULL, mtd->_read_oob() has semantics * similar to mtd->_read(), returning a non-negative integer * representing max bitflips. In other cases, mtd->_read_oob() may * return -EUCLEAN. In all cases, perform similar logic to mtd_read(). */ if (unlikely(ret_code < 0)) return ret_code; if (mtd->ecc_strength == 0) return 0; /* device lacks ecc */ if (ops->stats) ops->stats->max_bitflips = ret_code; return ret_code >= mtd->bitflip_threshold ? -EUCLEAN : 0; } EXPORT_SYMBOL_GPL(mtd_read_oob); int mtd_write_oob(struct mtd_info *mtd, loff_t to, struct mtd_oob_ops *ops) { struct mtd_info *master = mtd_get_master(mtd); int ret; ops->retlen = ops->oobretlen = 0; if (!(mtd->flags & MTD_WRITEABLE)) return -EROFS; ret = mtd_check_oob_ops(mtd, to, ops); if (ret) return ret; ledtrig_mtd_activity(); /* Check the validity of a potential fallback on mtd->_write */ if (!master->_write_oob && (!master->_write || ops->oobbuf)) return -EOPNOTSUPP; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) return mtd_io_emulated_slc(mtd, to, false, ops); return mtd_write_oob_std(mtd, to, ops); } EXPORT_SYMBOL_GPL(mtd_write_oob); /** * mtd_ooblayout_ecc - Get the OOB region definition of a specific ECC section * @mtd: MTD device structure * @section: ECC section. Depending on the layout you may have all the ECC * bytes stored in a single contiguous section, or one section * per ECC chunk (and sometime several sections for a single ECC * ECC chunk) * @oobecc: OOB region struct filled with the appropriate ECC position * information * * This function returns ECC section information in the OOB area. If you want * to get all the ECC bytes information, then you should call * mtd_ooblayout_ecc(mtd, section++, oobecc) until it returns -ERANGE. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_ecc(struct mtd_info *mtd, int section, struct mtd_oob_region *oobecc) { struct mtd_info *master = mtd_get_master(mtd); memset(oobecc, 0, sizeof(*oobecc)); if (!master || section < 0) return -EINVAL; if (!master->ooblayout || !master->ooblayout->ecc) return -ENOTSUPP; return master->ooblayout->ecc(master, section, oobecc); } EXPORT_SYMBOL_GPL(mtd_ooblayout_ecc); /** * mtd_ooblayout_free - Get the OOB region definition of a specific free * section * @mtd: MTD device structure * @section: Free section you are interested in. Depending on the layout * you may have all the free bytes stored in a single contiguous * section, or one section per ECC chunk plus an extra section * for the remaining bytes (or other funky layout). * @oobfree: OOB region struct filled with the appropriate free position * information * * This function returns free bytes position in the OOB area. If you want * to get all the free bytes information, then you should call * mtd_ooblayout_free(mtd, section++, oobfree) until it returns -ERANGE. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_free(struct mtd_info *mtd, int section, struct mtd_oob_region *oobfree) { struct mtd_info *master = mtd_get_master(mtd); memset(oobfree, 0, sizeof(*oobfree)); if (!master || section < 0) return -EINVAL; if (!master->ooblayout || !master->ooblayout->free) return -ENOTSUPP; return master->ooblayout->free(master, section, oobfree); } EXPORT_SYMBOL_GPL(mtd_ooblayout_free); /** * mtd_ooblayout_find_region - Find the region attached to a specific byte * @mtd: mtd info structure * @byte: the byte we are searching for * @sectionp: pointer where the section id will be stored * @oobregion: used to retrieve the ECC position * @iter: iterator function. Should be either mtd_ooblayout_free or * mtd_ooblayout_ecc depending on the region type you're searching for * * This function returns the section id and oobregion information of a * specific byte. For example, say you want to know where the 4th ECC byte is * stored, you'll use: * * mtd_ooblayout_find_region(mtd, 3, §ion, &oobregion, mtd_ooblayout_ecc); * * Returns zero on success, a negative error code otherwise. */ static int mtd_ooblayout_find_region(struct mtd_info *mtd, int byte, int *sectionp, struct mtd_oob_region *oobregion, int (*iter)(struct mtd_info *, int section, struct mtd_oob_region *oobregion)) { int pos = 0, ret, section = 0; memset(oobregion, 0, sizeof(*oobregion)); while (1) { ret = iter(mtd, section, oobregion); if (ret) return ret; if (pos + oobregion->length > byte) break; pos += oobregion->length; section++; } /* * Adjust region info to make it start at the beginning at the * 'start' ECC byte. */ oobregion->offset += byte - pos; oobregion->length -= byte - pos; *sectionp = section; return 0; } /** * mtd_ooblayout_find_eccregion - Find the ECC region attached to a specific * ECC byte * @mtd: mtd info structure * @eccbyte: the byte we are searching for * @section: pointer where the section id will be stored * @oobregion: OOB region information * * Works like mtd_ooblayout_find_region() except it searches for a specific ECC * byte. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_find_eccregion(struct mtd_info *mtd, int eccbyte, int *section, struct mtd_oob_region *oobregion) { return mtd_ooblayout_find_region(mtd, eccbyte, section, oobregion, mtd_ooblayout_ecc); } EXPORT_SYMBOL_GPL(mtd_ooblayout_find_eccregion); /** * mtd_ooblayout_get_bytes - Extract OOB bytes from the oob buffer * @mtd: mtd info structure * @buf: destination buffer to store OOB bytes * @oobbuf: OOB buffer * @start: first byte to retrieve * @nbytes: number of bytes to retrieve * @iter: section iterator * * Extract bytes attached to a specific category (ECC or free) * from the OOB buffer and copy them into buf. * * Returns zero on success, a negative error code otherwise. */ static int mtd_ooblayout_get_bytes(struct mtd_info *mtd, u8 *buf, const u8 *oobbuf, int start, int nbytes, int (*iter)(struct mtd_info *, int section, struct mtd_oob_region *oobregion)) { struct mtd_oob_region oobregion; int section, ret; ret = mtd_ooblayout_find_region(mtd, start, §ion, &oobregion, iter); while (!ret) { int cnt; cnt = min_t(int, nbytes, oobregion.length); memcpy(buf, oobbuf + oobregion.offset, cnt); buf += cnt; nbytes -= cnt; if (!nbytes) break; ret = iter(mtd, ++section, &oobregion); } return ret; } /** * mtd_ooblayout_set_bytes - put OOB bytes into the oob buffer * @mtd: mtd info structure * @buf: source buffer to get OOB bytes from * @oobbuf: OOB buffer * @start: first OOB byte to set * @nbytes: number of OOB bytes to set * @iter: section iterator * * Fill the OOB buffer with data provided in buf. The category (ECC or free) * is selected by passing the appropriate iterator. * * Returns zero on success, a negative error code otherwise. */ static int mtd_ooblayout_set_bytes(struct mtd_info *mtd, const u8 *buf, u8 *oobbuf, int start, int nbytes, int (*iter)(struct mtd_info *, int section, struct mtd_oob_region *oobregion)) { struct mtd_oob_region oobregion; int section, ret; ret = mtd_ooblayout_find_region(mtd, start, §ion, &oobregion, iter); while (!ret) { int cnt; cnt = min_t(int, nbytes, oobregion.length); memcpy(oobbuf + oobregion.offset, buf, cnt); buf += cnt; nbytes -= cnt; if (!nbytes) break; ret = iter(mtd, ++section, &oobregion); } return ret; } /** * mtd_ooblayout_count_bytes - count the number of bytes in a OOB category * @mtd: mtd info structure * @iter: category iterator * * Count the number of bytes in a given category. * * Returns a positive value on success, a negative error code otherwise. */ static int mtd_ooblayout_count_bytes(struct mtd_info *mtd, int (*iter)(struct mtd_info *, int section, struct mtd_oob_region *oobregion)) { struct mtd_oob_region oobregion; int section = 0, ret, nbytes = 0; while (1) { ret = iter(mtd, section++, &oobregion); if (ret) { if (ret == -ERANGE) ret = nbytes; break; } nbytes += oobregion.length; } return ret; } /** * mtd_ooblayout_get_eccbytes - extract ECC bytes from the oob buffer * @mtd: mtd info structure * @eccbuf: destination buffer to store ECC bytes * @oobbuf: OOB buffer * @start: first ECC byte to retrieve * @nbytes: number of ECC bytes to retrieve * * Works like mtd_ooblayout_get_bytes(), except it acts on ECC bytes. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_get_eccbytes(struct mtd_info *mtd, u8 *eccbuf, const u8 *oobbuf, int start, int nbytes) { return mtd_ooblayout_get_bytes(mtd, eccbuf, oobbuf, start, nbytes, mtd_ooblayout_ecc); } EXPORT_SYMBOL_GPL(mtd_ooblayout_get_eccbytes); /** * mtd_ooblayout_set_eccbytes - set ECC bytes into the oob buffer * @mtd: mtd info structure * @eccbuf: source buffer to get ECC bytes from * @oobbuf: OOB buffer * @start: first ECC byte to set * @nbytes: number of ECC bytes to set * * Works like mtd_ooblayout_set_bytes(), except it acts on ECC bytes. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_set_eccbytes(struct mtd_info *mtd, const u8 *eccbuf, u8 *oobbuf, int start, int nbytes) { return mtd_ooblayout_set_bytes(mtd, eccbuf, oobbuf, start, nbytes, mtd_ooblayout_ecc); } EXPORT_SYMBOL_GPL(mtd_ooblayout_set_eccbytes); /** * mtd_ooblayout_get_databytes - extract data bytes from the oob buffer * @mtd: mtd info structure * @databuf: destination buffer to store ECC bytes * @oobbuf: OOB buffer * @start: first ECC byte to retrieve * @nbytes: number of ECC bytes to retrieve * * Works like mtd_ooblayout_get_bytes(), except it acts on free bytes. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_get_databytes(struct mtd_info *mtd, u8 *databuf, const u8 *oobbuf, int start, int nbytes) { return mtd_ooblayout_get_bytes(mtd, databuf, oobbuf, start, nbytes, mtd_ooblayout_free); } EXPORT_SYMBOL_GPL(mtd_ooblayout_get_databytes); /** * mtd_ooblayout_set_databytes - set data bytes into the oob buffer * @mtd: mtd info structure * @databuf: source buffer to get data bytes from * @oobbuf: OOB buffer * @start: first ECC byte to set * @nbytes: number of ECC bytes to set * * Works like mtd_ooblayout_set_bytes(), except it acts on free bytes. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_set_databytes(struct mtd_info *mtd, const u8 *databuf, u8 *oobbuf, int start, int nbytes) { return mtd_ooblayout_set_bytes(mtd, databuf, oobbuf, start, nbytes, mtd_ooblayout_free); } EXPORT_SYMBOL_GPL(mtd_ooblayout_set_databytes); /** * mtd_ooblayout_count_freebytes - count the number of free bytes in OOB * @mtd: mtd info structure * * Works like mtd_ooblayout_count_bytes(), except it count free bytes. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_count_freebytes(struct mtd_info *mtd) { return mtd_ooblayout_count_bytes(mtd, mtd_ooblayout_free); } EXPORT_SYMBOL_GPL(mtd_ooblayout_count_freebytes); /** * mtd_ooblayout_count_eccbytes - count the number of ECC bytes in OOB * @mtd: mtd info structure * * Works like mtd_ooblayout_count_bytes(), except it count ECC bytes. * * Returns zero on success, a negative error code otherwise. */ int mtd_ooblayout_count_eccbytes(struct mtd_info *mtd) { return mtd_ooblayout_count_bytes(mtd, mtd_ooblayout_ecc); } EXPORT_SYMBOL_GPL(mtd_ooblayout_count_eccbytes); /* * Method to access the protection register area, present in some flash * devices. The user data is one time programmable but the factory data is read * only. */ int mtd_get_fact_prot_info(struct mtd_info *mtd, size_t len, size_t *retlen, struct otp_info *buf) { struct mtd_info *master = mtd_get_master(mtd); if (!master->_get_fact_prot_info) return -EOPNOTSUPP; if (!len) return 0; return master->_get_fact_prot_info(master, len, retlen, buf); } EXPORT_SYMBOL_GPL(mtd_get_fact_prot_info); int mtd_read_fact_prot_reg(struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char *buf) { struct mtd_info *master = mtd_get_master(mtd); *retlen = 0; if (!master->_read_fact_prot_reg) return -EOPNOTSUPP; if (!len) return 0; return master->_read_fact_prot_reg(master, from, len, retlen, buf); } EXPORT_SYMBOL_GPL(mtd_read_fact_prot_reg); int mtd_get_user_prot_info(struct mtd_info *mtd, size_t len, size_t *retlen, struct otp_info *buf) { struct mtd_info *master = mtd_get_master(mtd); if (!master->_get_user_prot_info) return -EOPNOTSUPP; if (!len) return 0; return master->_get_user_prot_info(master, len, retlen, buf); } EXPORT_SYMBOL_GPL(mtd_get_user_prot_info); int mtd_read_user_prot_reg(struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char *buf) { struct mtd_info *master = mtd_get_master(mtd); *retlen = 0; if (!master->_read_user_prot_reg) return -EOPNOTSUPP; if (!len) return 0; return master->_read_user_prot_reg(master, from, len, retlen, buf); } EXPORT_SYMBOL_GPL(mtd_read_user_prot_reg); int mtd_write_user_prot_reg(struct mtd_info *mtd, loff_t to, size_t len, size_t *retlen, const u_char *buf) { struct mtd_info *master = mtd_get_master(mtd); int ret; *retlen = 0; if (!master->_write_user_prot_reg) return -EOPNOTSUPP; if (!len) return 0; ret = master->_write_user_prot_reg(master, to, len, retlen, buf); if (ret) return ret; /* * If no data could be written at all, we are out of memory and * must return -ENOSPC. */ return (*retlen) ? 0 : -ENOSPC; } EXPORT_SYMBOL_GPL(mtd_write_user_prot_reg); int mtd_lock_user_prot_reg(struct mtd_info *mtd, loff_t from, size_t len) { struct mtd_info *master = mtd_get_master(mtd); if (!master->_lock_user_prot_reg) return -EOPNOTSUPP; if (!len) return 0; return master->_lock_user_prot_reg(master, from, len); } EXPORT_SYMBOL_GPL(mtd_lock_user_prot_reg); int mtd_erase_user_prot_reg(struct mtd_info *mtd, loff_t from, size_t len) { struct mtd_info *master = mtd_get_master(mtd); if (!master->_erase_user_prot_reg) return -EOPNOTSUPP; if (!len) return 0; return master->_erase_user_prot_reg(master, from, len); } EXPORT_SYMBOL_GPL(mtd_erase_user_prot_reg); /* Chip-supported device locking */ int mtd_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct mtd_info *master = mtd_get_master(mtd); if (!master->_lock) return -EOPNOTSUPP; if (ofs < 0 || ofs >= mtd->size || len > mtd->size - ofs) return -EINVAL; if (!len) return 0; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) { ofs = (loff_t)mtd_div_by_eb(ofs, mtd) * master->erasesize; len = (u64)mtd_div_by_eb(len, mtd) * master->erasesize; } return master->_lock(master, mtd_get_master_ofs(mtd, ofs), len); } EXPORT_SYMBOL_GPL(mtd_lock); int mtd_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct mtd_info *master = mtd_get_master(mtd); if (!master->_unlock) return -EOPNOTSUPP; if (ofs < 0 || ofs >= mtd->size || len > mtd->size - ofs) return -EINVAL; if (!len) return 0; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) { ofs = (loff_t)mtd_div_by_eb(ofs, mtd) * master->erasesize; len = (u64)mtd_div_by_eb(len, mtd) * master->erasesize; } return master->_unlock(master, mtd_get_master_ofs(mtd, ofs), len); } EXPORT_SYMBOL_GPL(mtd_unlock); int mtd_is_locked(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct mtd_info *master = mtd_get_master(mtd); if (!master->_is_locked) return -EOPNOTSUPP; if (ofs < 0 || ofs >= mtd->size || len > mtd->size - ofs) return -EINVAL; if (!len) return 0; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) { ofs = (loff_t)mtd_div_by_eb(ofs, mtd) * master->erasesize; len = (u64)mtd_div_by_eb(len, mtd) * master->erasesize; } return master->_is_locked(master, mtd_get_master_ofs(mtd, ofs), len); } EXPORT_SYMBOL_GPL(mtd_is_locked); int mtd_block_isreserved(struct mtd_info *mtd, loff_t ofs) { struct mtd_info *master = mtd_get_master(mtd); if (ofs < 0 || ofs >= mtd->size) return -EINVAL; if (!master->_block_isreserved) return 0; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) ofs = (loff_t)mtd_div_by_eb(ofs, mtd) * master->erasesize; return master->_block_isreserved(master, mtd_get_master_ofs(mtd, ofs)); } EXPORT_SYMBOL_GPL(mtd_block_isreserved); int mtd_block_isbad(struct mtd_info *mtd, loff_t ofs) { struct mtd_info *master = mtd_get_master(mtd); if (ofs < 0 || ofs >= mtd->size) return -EINVAL; if (!master->_block_isbad) return 0; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) ofs = (loff_t)mtd_div_by_eb(ofs, mtd) * master->erasesize; return master->_block_isbad(master, mtd_get_master_ofs(mtd, ofs)); } EXPORT_SYMBOL_GPL(mtd_block_isbad); int mtd_block_markbad(struct mtd_info *mtd, loff_t ofs) { struct mtd_info *master = mtd_get_master(mtd); loff_t moffs; int ret; if (!master->_block_markbad) return -EOPNOTSUPP; if (ofs < 0 || ofs >= mtd->size) return -EINVAL; if (!(mtd->flags & MTD_WRITEABLE)) return -EROFS; if (mtd->flags & MTD_SLC_ON_MLC_EMULATION) ofs = (loff_t)mtd_div_by_eb(ofs, mtd) * master->erasesize; moffs = mtd_get_master_ofs(mtd, ofs); if (master->_block_isbad) { ret = master->_block_isbad(master, moffs); if (ret > 0) return 0; } ret = master->_block_markbad(master, moffs); if (ret) return ret; while (mtd->parent) { mtd->ecc_stats.badblocks++; mtd = mtd->parent; } return 0; } EXPORT_SYMBOL_GPL(mtd_block_markbad); ALLOW_ERROR_INJECTION(mtd_block_markbad, ERRNO); /* * default_mtd_writev - the default writev method * @mtd: mtd device description object pointer * @vecs: the vectors to write * @count: count of vectors in @vecs * @to: the MTD device offset to write to * @retlen: on exit contains the count of bytes written to the MTD device. * * This function returns zero in case of success and a negative error code in * case of failure. */ static int default_mtd_writev(struct mtd_info *mtd, const struct kvec *vecs, unsigned long count, loff_t to, size_t *retlen) { unsigned long i; size_t totlen = 0, thislen; int ret = 0; for (i = 0; i < count; i++) { if (!vecs[i].iov_len) continue; ret = mtd_write(mtd, to, vecs[i].iov_len, &thislen, vecs[i].iov_base); totlen += thislen; if (ret || thislen != vecs[i].iov_len) break; to += vecs[i].iov_len; } *retlen = totlen; return ret; } /* * mtd_writev - the vector-based MTD write method * @mtd: mtd device description object pointer * @vecs: the vectors to write * @count: count of vectors in @vecs * @to: the MTD device offset to write to * @retlen: on exit contains the count of bytes written to the MTD device. * * This function returns zero in case of success and a negative error code in * case of failure. */ int mtd_writev(struct mtd_info *mtd, const struct kvec *vecs, unsigned long count, loff_t to, size_t *retlen) { struct mtd_info *master = mtd_get_master(mtd); *retlen = 0; if (!(mtd->flags & MTD_WRITEABLE)) return -EROFS; if (!master->_writev) return default_mtd_writev(mtd, vecs, count, to, retlen); return master->_writev(master, vecs, count, mtd_get_master_ofs(mtd, to), retlen); } EXPORT_SYMBOL_GPL(mtd_writev); /** * mtd_kmalloc_up_to - allocate a contiguous buffer up to the specified size * @mtd: mtd device description object pointer * @size: a pointer to the ideal or maximum size of the allocation, points * to the actual allocation size on success. * * This routine attempts to allocate a contiguous kernel buffer up to * the specified size, backing off the size of the request exponentially * until the request succeeds or until the allocation size falls below * the system page size. This attempts to make sure it does not adversely * impact system performance, so when allocating more than one page, we * ask the memory allocator to avoid re-trying, swapping, writing back * or performing I/O. * * Note, this function also makes sure that the allocated buffer is aligned to * the MTD device's min. I/O unit, i.e. the "mtd->writesize" value. * * This is called, for example by mtd_{read,write} and jffs2_scan_medium, * to handle smaller (i.e. degraded) buffer allocations under low- or * fragmented-memory situations where such reduced allocations, from a * requested ideal, are allowed. * * Returns a pointer to the allocated buffer on success; otherwise, NULL. */ void *mtd_kmalloc_up_to(const struct mtd_info *mtd, size_t *size) { gfp_t flags = __GFP_NOWARN | __GFP_DIRECT_RECLAIM | __GFP_NORETRY; size_t min_alloc = max_t(size_t, mtd->writesize, PAGE_SIZE); void *kbuf; *size = min_t(size_t, *size, KMALLOC_MAX_SIZE); while (*size > min_alloc) { kbuf = kmalloc(*size, flags); if (kbuf) return kbuf; *size >>= 1; *size = ALIGN(*size, mtd->writesize); } /* * For the last resort allocation allow 'kmalloc()' to do all sorts of * things (write-back, dropping caches, etc) by using GFP_KERNEL. */ return kmalloc(*size, GFP_KERNEL); } EXPORT_SYMBOL_GPL(mtd_kmalloc_up_to); #ifdef CONFIG_PROC_FS /*====================================================================*/ /* Support for /proc/mtd */ static int mtd_proc_show(struct seq_file *m, void *v) { struct mtd_info *mtd; seq_puts(m, "dev: size erasesize name\n"); mutex_lock(&mtd_table_mutex); mtd_for_each_device(mtd) { seq_printf(m, "mtd%d: %8.8llx %8.8x \"%s\"\n", mtd->index, (unsigned long long)mtd->size, mtd->erasesize, mtd->name); } mutex_unlock(&mtd_table_mutex); return 0; } #endif /* CONFIG_PROC_FS */ /*====================================================================*/ /* Init code */ static struct backing_dev_info * __init mtd_bdi_init(const char *name) { struct backing_dev_info *bdi; int ret; bdi = bdi_alloc(NUMA_NO_NODE); if (!bdi) return ERR_PTR(-ENOMEM); bdi->ra_pages = 0; bdi->io_pages = 0; /* * We put '-0' suffix to the name to get the same name format as we * used to get. Since this is called only once, we get a unique name. */ ret = bdi_register(bdi, "%.28s-0", name); if (ret) bdi_put(bdi); return ret ? ERR_PTR(ret) : bdi; } static struct proc_dir_entry *proc_mtd; static int __init init_mtd(void) { int ret; ret = class_register(&mtd_class); if (ret) goto err_reg; mtd_bdi = mtd_bdi_init("mtd"); if (IS_ERR(mtd_bdi)) { ret = PTR_ERR(mtd_bdi); goto err_bdi; } proc_mtd = proc_create_single("mtd", 0, NULL, mtd_proc_show); ret = init_mtdchar(); if (ret) goto out_procfs; dfs_dir_mtd = debugfs_create_dir("mtd", NULL); debugfs_create_bool("expert_analysis_mode", 0600, dfs_dir_mtd, &mtd_expert_analysis_mode); return 0; out_procfs: if (proc_mtd) remove_proc_entry("mtd", NULL); bdi_unregister(mtd_bdi); bdi_put(mtd_bdi); err_bdi: class_unregister(&mtd_class); err_reg: pr_err("Error registering mtd class or bdi: %d\n", ret); return ret; } static void __exit cleanup_mtd(void) { debugfs_remove_recursive(dfs_dir_mtd); cleanup_mtdchar(); if (proc_mtd) remove_proc_entry("mtd", NULL); class_unregister(&mtd_class); bdi_unregister(mtd_bdi); bdi_put(mtd_bdi); idr_destroy(&mtd_idr); } module_init(init_mtd); module_exit(cleanup_mtd); MODULE_LICENSE("GPL"); MODULE_AUTHOR("David Woodhouse <dwmw2@infradead.org>"); MODULE_DESCRIPTION("Core MTD registration and access routines"); |
| 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 | // SPDX-License-Identifier: GPL-2.0-only /* * Copyright (C) 2010 IBM Corporation * Copyright (C) 2010 Politecnico di Torino, Italy * TORSEC group -- https://security.polito.it * * Authors: * Mimi Zohar <zohar@us.ibm.com> * Roberto Sassu <roberto.sassu@polito.it> * * See Documentation/security/keys/trusted-encrypted.rst */ #include <linux/uaccess.h> #include <linux/err.h> #include <keys/trusted-type.h> #include <keys/encrypted-type.h> #include "encrypted.h" /* * request_trusted_key - request the trusted key * * Trusted keys are sealed to PCRs and other metadata. Although userspace * manages both trusted/encrypted key-types, like the encrypted key type * data, trusted key type data is not visible decrypted from userspace. */ struct key *request_trusted_key(const char *trusted_desc, const u8 **master_key, size_t *master_keylen) { struct trusted_key_payload *tpayload; struct key *tkey; tkey = request_key(&key_type_trusted, trusted_desc, NULL); if (IS_ERR(tkey)) goto error; down_read(&tkey->sem); tpayload = tkey->payload.data[0]; *master_key = tpayload->key; *master_keylen = tpayload->key_len; error: return tkey; } |
| 682 682 673 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 | // SPDX-License-Identifier: GPL-2.0 #include <linux/export.h> #include <linux/spinlock.h> #include <linux/atomic.h> /* * This is an implementation of the notion of "decrement a * reference count, and return locked if it decremented to zero". * * NOTE NOTE NOTE! This is _not_ equivalent to * * if (atomic_dec_and_test(&atomic)) { * spin_lock(&lock); * return 1; * } * return 0; * * because the spin-lock and the decrement must be * "atomic". */ int atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock) { /* Subtract 1 from counter unless that drops it to 0 (ie. it was 1) */ if (atomic_add_unless(atomic, -1, 1)) return 0; /* Otherwise do it the slow way */ spin_lock(lock); if (atomic_dec_and_test(atomic)) return 1; spin_unlock(lock); return 0; } EXPORT_SYMBOL(atomic_dec_and_lock); int _atomic_dec_and_lock_irqsave(atomic_t *atomic, spinlock_t *lock, unsigned long *flags) { /* Subtract 1 from counter unless that drops it to 0 (ie. it was 1) */ if (atomic_add_unless(atomic, -1, 1)) return 0; /* Otherwise do it the slow way */ spin_lock_irqsave(lock, *flags); if (atomic_dec_and_test(atomic)) return 1; spin_unlock_irqrestore(lock, *flags); return 0; } EXPORT_SYMBOL(_atomic_dec_and_lock_irqsave); int atomic_dec_and_raw_lock(atomic_t *atomic, raw_spinlock_t *lock) { /* Subtract 1 from counter unless that drops it to 0 (ie. it was 1) */ if (atomic_add_unless(atomic, -1, 1)) return 0; /* Otherwise do it the slow way */ raw_spin_lock(lock); if (atomic_dec_and_test(atomic)) return 1; raw_spin_unlock(lock); return 0; } EXPORT_SYMBOL(atomic_dec_and_raw_lock); int _atomic_dec_and_raw_lock_irqsave(atomic_t *atomic, raw_spinlock_t *lock, unsigned long *flags) { /* Subtract 1 from counter unless that drops it to 0 (ie. it was 1) */ if (atomic_add_unless(atomic, -1, 1)) return 0; /* Otherwise do it the slow way */ raw_spin_lock_irqsave(lock, *flags); if (atomic_dec_and_test(atomic)) return 1; raw_spin_unlock_irqrestore(lock, *flags); return 0; } EXPORT_SYMBOL(_atomic_dec_and_raw_lock_irqsave); |
| 46 46 46 1424 1421 1423 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 | /* SPDX-License-Identifier: GPL-2.0-only */ /* * AppArmor security module * * This file contains AppArmor task related definitions and mediation * * Copyright 2017 Canonical Ltd. */ #ifndef __AA_TASK_H #define __AA_TASK_H static inline struct aa_task_ctx *task_ctx(struct task_struct *task) { return task->security + apparmor_blob_sizes.lbs_task; } /* * struct aa_task_ctx - information for current task label change * @nnp: snapshot of label at time of no_new_privs * @onexec: profile to transition to on next exec (MAY BE NULL) * @previous: profile the task may return to (MAY BE NULL) * @token: magic value the task must know for returning to @previous_profile */ struct aa_task_ctx { struct aa_label *nnp; struct aa_label *onexec; struct aa_label *previous; u64 token; }; int aa_replace_current_label(struct aa_label *label); void aa_set_current_onexec(struct aa_label *label, bool stack); int aa_set_current_hat(struct aa_label *label, u64 token); int aa_restore_previous_label(u64 cookie); struct aa_label *aa_get_task_label(struct task_struct *task); /** * aa_free_task_ctx - free a task_ctx * @ctx: task_ctx to free (MAYBE NULL) */ static inline void aa_free_task_ctx(struct aa_task_ctx *ctx) { if (ctx) { aa_put_label(ctx->nnp); aa_put_label(ctx->previous); aa_put_label(ctx->onexec); } } /** * aa_dup_task_ctx - duplicate a task context, incrementing reference counts * @new: a blank task context (NOT NULL) * @old: the task context to copy (NOT NULL) */ static inline void aa_dup_task_ctx(struct aa_task_ctx *new, const struct aa_task_ctx *old) { *new = *old; aa_get_label(new->nnp); aa_get_label(new->previous); aa_get_label(new->onexec); } /** * aa_clear_task_ctx_trans - clear transition tracking info from the ctx * @ctx: task context to clear (NOT NULL) */ static inline void aa_clear_task_ctx_trans(struct aa_task_ctx *ctx) { AA_BUG(!ctx); aa_put_label(ctx->previous); aa_put_label(ctx->onexec); ctx->previous = NULL; ctx->onexec = NULL; ctx->token = 0; } #define AA_PTRACE_TRACE MAY_WRITE #define AA_PTRACE_READ MAY_READ #define AA_MAY_BE_TRACED AA_MAY_APPEND #define AA_MAY_BE_READ AA_MAY_CREATE #define PTRACE_PERM_SHIFT 2 #define AA_PTRACE_PERM_MASK (AA_PTRACE_READ | AA_PTRACE_TRACE | \ AA_MAY_BE_READ | AA_MAY_BE_TRACED) #define AA_SIGNAL_PERM_MASK (MAY_READ | MAY_WRITE) #define AA_SFS_SIG_MASK "hup int quit ill trap abrt bus fpe kill usr1 " \ "segv usr2 pipe alrm term stkflt chld cont stop stp ttin ttou urg " \ "xcpu xfsz vtalrm prof winch io pwr sys emt lost" int aa_may_ptrace(const struct cred *tracer_cred, struct aa_label *tracer, const struct cred *tracee_cred, struct aa_label *tracee, u32 request); #define AA_USERNS_CREATE 8 int aa_profile_ns_perm(struct aa_profile *profile, struct apparmor_audit_data *ad, u32 request); #endif /* __AA_TASK_H */ |
| 1 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 | // SPDX-License-Identifier: GPL-2.0-or-later /* * * Broadcom Blutonium firmware driver * * Copyright (C) 2003 Maxim Krasnyansky <maxk@qualcomm.com> * Copyright (C) 2003 Marcel Holtmann <marcel@holtmann.org> */ #include <linux/module.h> #include <linux/kernel.h> #include <linux/init.h> #include <linux/slab.h> #include <linux/types.h> #include <linux/errno.h> #include <linux/device.h> #include <linux/firmware.h> #include <linux/usb.h> #include <net/bluetooth/bluetooth.h> #define VERSION "1.2" static const struct usb_device_id bcm203x_table[] = { /* Broadcom Blutonium (BCM2033) */ { USB_DEVICE(0x0a5c, 0x2033) }, { } /* Terminating entry */ }; MODULE_DEVICE_TABLE(usb, bcm203x_table); #define BCM203X_ERROR 0 #define BCM203X_RESET 1 #define BCM203X_LOAD_MINIDRV 2 #define BCM203X_SELECT_MEMORY 3 #define BCM203X_CHECK_MEMORY 4 #define BCM203X_LOAD_FIRMWARE 5 #define BCM203X_CHECK_FIRMWARE 6 #define BCM203X_IN_EP 0x81 #define BCM203X_OUT_EP 0x02 struct bcm203x_data { struct usb_device *udev; unsigned long state; struct work_struct work; atomic_t shutdown; struct urb *urb; unsigned char *buffer; unsigned char *fw_data; unsigned int fw_size; unsigned int fw_sent; }; static void bcm203x_complete(struct urb *urb) { struct bcm203x_data *data = urb->context; struct usb_device *udev = urb->dev; int len; BT_DBG("udev %p urb %p", udev, urb); if (urb->status) { BT_ERR("URB failed with status %d", urb->status); data->state = BCM203X_ERROR; return; } switch (data->state) { case BCM203X_LOAD_MINIDRV: memcpy(data->buffer, "#", 1); usb_fill_bulk_urb(urb, udev, usb_sndbulkpipe(udev, BCM203X_OUT_EP), data->buffer, 1, bcm203x_complete, data); data->state = BCM203X_SELECT_MEMORY; /* use workqueue to have a small delay */ schedule_work(&data->work); break; case BCM203X_SELECT_MEMORY: usb_fill_int_urb(urb, udev, usb_rcvintpipe(udev, BCM203X_IN_EP), data->buffer, 32, bcm203x_complete, data, 1); data->state = BCM203X_CHECK_MEMORY; if (usb_submit_urb(data->urb, GFP_ATOMIC) < 0) BT_ERR("Can't submit URB"); break; case BCM203X_CHECK_MEMORY: if (data->buffer[0] != '#') { BT_ERR("Memory select failed"); data->state = BCM203X_ERROR; break; } data->state = BCM203X_LOAD_FIRMWARE; fallthrough; case BCM203X_LOAD_FIRMWARE: if (data->fw_sent == data->fw_size) { usb_fill_int_urb(urb, udev, usb_rcvintpipe(udev, BCM203X_IN_EP), data->buffer, 32, bcm203x_complete, data, 1); data->state = BCM203X_CHECK_FIRMWARE; } else { len = min_t(uint, data->fw_size - data->fw_sent, 4096); usb_fill_bulk_urb(urb, udev, usb_sndbulkpipe(udev, BCM203X_OUT_EP), data->fw_data + data->fw_sent, len, bcm203x_complete, data); data->fw_sent += len; } if (usb_submit_urb(data->urb, GFP_ATOMIC) < 0) BT_ERR("Can't submit URB"); break; case BCM203X_CHECK_FIRMWARE: if (data->buffer[0] != '.') { BT_ERR("Firmware loading failed"); data->state = BCM203X_ERROR; break; } data->state = BCM203X_RESET; break; } } static void bcm203x_work(struct work_struct *work) { struct bcm203x_data *data = container_of(work, struct bcm203x_data, work); if (atomic_read(&data->shutdown)) return; if (usb_submit_urb(data->urb, GFP_KERNEL) < 0) BT_ERR("Can't submit URB"); } static int bcm203x_probe(struct usb_interface *intf, const struct usb_device_id *id) { const struct firmware *firmware; struct usb_device *udev = interface_to_usbdev(intf); struct bcm203x_data *data; int size; BT_DBG("intf %p id %p", intf, id); if (intf->cur_altsetting->desc.bInterfaceNumber != 0) return -ENODEV; data = devm_kzalloc(&intf->dev, sizeof(*data), GFP_KERNEL); if (!data) return -ENOMEM; data->udev = udev; data->state = BCM203X_LOAD_MINIDRV; data->urb = usb_alloc_urb(0, GFP_KERNEL); if (!data->urb) return -ENOMEM; if (request_firmware(&firmware, "BCM2033-MD.hex", &udev->dev) < 0) { BT_ERR("Mini driver request failed"); usb_free_urb(data->urb); return -EIO; } BT_DBG("minidrv data %p size %zu", firmware->data, firmware->size); size = max_t(uint, firmware->size, 4096); data->buffer = kmalloc(size, GFP_KERNEL); if (!data->buffer) { BT_ERR("Can't allocate memory for mini driver"); release_firmware(firmware); usb_free_urb(data->urb); return -ENOMEM; } memcpy(data->buffer, firmware->data, firmware->size); usb_fill_bulk_urb(data->urb, udev, usb_sndbulkpipe(udev, BCM203X_OUT_EP), data->buffer, firmware->size, bcm203x_complete, data); release_firmware(firmware); if (request_firmware(&firmware, "BCM2033-FW.bin", &udev->dev) < 0) { BT_ERR("Firmware request failed"); usb_free_urb(data->urb); kfree(data->buffer); return -EIO; } BT_DBG("firmware data %p size %zu", firmware->data, firmware->size); data->fw_data = kmemdup(firmware->data, firmware->size, GFP_KERNEL); if (!data->fw_data) { BT_ERR("Can't allocate memory for firmware image"); release_firmware(firmware); usb_free_urb(data->urb); kfree(data->buffer); return -ENOMEM; } data->fw_size = firmware->size; data->fw_sent = 0; release_firmware(firmware); INIT_WORK(&data->work, bcm203x_work); usb_set_intfdata(intf, data); /* use workqueue to have a small delay */ schedule_work(&data->work); return 0; } static void bcm203x_disconnect(struct usb_interface *intf) { struct bcm203x_data *data = usb_get_intfdata(intf); BT_DBG("intf %p", intf); atomic_inc(&data->shutdown); cancel_work_sync(&data->work); usb_kill_urb(data->urb); usb_set_intfdata(intf, NULL); usb_free_urb(data->urb); kfree(data->fw_data); kfree(data->buffer); } static struct usb_driver bcm203x_driver = { .name = "bcm203x", .probe = bcm203x_probe, .disconnect = bcm203x_disconnect, .id_table = bcm203x_table, .disable_hub_initiated_lpm = 1, }; module_usb_driver(bcm203x_driver); MODULE_AUTHOR("Marcel Holtmann <marcel@holtmann.org>"); MODULE_DESCRIPTION("Broadcom Blutonium firmware driver ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); MODULE_FIRMWARE("BCM2033-MD.hex"); MODULE_FIRMWARE("BCM2033-FW.bin"); |
| 14 14 2 162 161 161 161 161 161 161 4 2 2 2 2 7 1 6 57 64 64 63 63 52 10 9 9 61 14 61 15 46 45 1 23 33 2 2 8 25 1 14 1 22 22 63 3 9 3 89 90 28 17 84 89 16 16 16 16 10 10 10 7 5 14 12 4 11 11 11 2 9 11 20 4 16 16 4 4 4 1 3 3 1 1 1 1 8 7 11 12 9 3 11 9 2 7 7 7 2276 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 | // SPDX-License-Identifier: GPL-2.0 /* Copyright (C) B.A.T.M.A.N. contributors: * * Marek Lindner, Simon Wunderlich */ #include "mesh-interface.h" #include "main.h" #include <linux/atomic.h> #include <linux/byteorder/generic.h> #include <linux/cache.h> #include <linux/compiler.h> #include <linux/container_of.h> #include <linux/cpumask.h> #include <linux/errno.h> #include <linux/etherdevice.h> #include <linux/ethtool.h> #include <linux/gfp.h> #include <linux/if_ether.h> #include <linux/if_vlan.h> #include <linux/jiffies.h> #include <linux/kref.h> #include <linux/list.h> #include <linux/lockdep.h> #include <linux/netdevice.h> #include <linux/netlink.h> #include <linux/percpu.h> #include <linux/random.h> #include <linux/rculist.h> #include <linux/rcupdate.h> #include <linux/skbuff.h> #include <linux/slab.h> #include <linux/socket.h> #include <linux/spinlock.h> #include <linux/stddef.h> #include <linux/string.h> #include <linux/types.h> #include <net/netlink.h> #include <net/rtnetlink.h> #include <uapi/linux/batadv_packet.h> #include <uapi/linux/batman_adv.h> #include "bat_algo.h" #include "bridge_loop_avoidance.h" #include "distributed-arp-table.h" #include "gateway_client.h" #include "hard-interface.h" #include "multicast.h" #include "send.h" #include "translation-table.h" /** * batadv_skb_head_push() - Increase header size and move (push) head pointer * @skb: packet buffer which should be modified * @len: number of bytes to add * * Return: 0 on success or negative error number in case of failure */ int batadv_skb_head_push(struct sk_buff *skb, unsigned int len) { int result; /* TODO: We must check if we can release all references to non-payload * data using __skb_header_release in our skbs to allow skb_cow_header * to work optimally. This means that those skbs are not allowed to read * or write any data which is before the current position of skb->data * after that call and thus allow other skbs with the same data buffer * to write freely in that area. */ result = skb_cow_head(skb, len); if (result < 0) return result; skb_push(skb, len); return 0; } /** * batadv_sum_counter() - Sum the cpu-local counters for index 'idx' * @bat_priv: the bat priv with all the mesh interface information * @idx: index of counter to sum up * * Return: sum of all cpu-local counters */ static u64 batadv_sum_counter(struct batadv_priv *bat_priv, size_t idx) { u64 *counters, sum = 0; int cpu; for_each_possible_cpu(cpu) { counters = per_cpu_ptr(bat_priv->bat_counters, cpu); sum += counters[idx]; } return sum; } static struct net_device_stats *batadv_interface_stats(struct net_device *dev) { struct batadv_priv *bat_priv = netdev_priv(dev); struct net_device_stats *stats = &dev->stats; stats->tx_packets = batadv_sum_counter(bat_priv, BATADV_CNT_TX); stats->tx_bytes = batadv_sum_counter(bat_priv, BATADV_CNT_TX_BYTES); stats->tx_dropped = batadv_sum_counter(bat_priv, BATADV_CNT_TX_DROPPED); stats->rx_packets = batadv_sum_counter(bat_priv, BATADV_CNT_RX); stats->rx_bytes = batadv_sum_counter(bat_priv, BATADV_CNT_RX_BYTES); return stats; } static int batadv_interface_set_mac_addr(struct net_device *dev, void *p) { struct batadv_priv *bat_priv = netdev_priv(dev); struct batadv_meshif_vlan *vlan; struct sockaddr *addr = p; u8 old_addr[ETH_ALEN]; if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; ether_addr_copy(old_addr, dev->dev_addr); eth_hw_addr_set(dev, addr->sa_data); /* only modify transtable if it has been initialized before */ if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE) return 0; rcu_read_lock(); hlist_for_each_entry_rcu(vlan, &bat_priv->meshif_vlan_list, list) { batadv_tt_local_remove(bat_priv, old_addr, vlan->vid, "mac address changed", false); batadv_tt_local_add(dev, addr->sa_data, vlan->vid, BATADV_NULL_IFINDEX, BATADV_NO_MARK); } rcu_read_unlock(); return 0; } static int batadv_interface_change_mtu(struct net_device *dev, int new_mtu) { struct batadv_priv *bat_priv = netdev_priv(dev); /* check ranges */ if (new_mtu < ETH_MIN_MTU || new_mtu > batadv_hardif_min_mtu(dev)) return -EINVAL; WRITE_ONCE(dev->mtu, new_mtu); bat_priv->mtu_set_by_user = new_mtu; return 0; } /** * batadv_interface_set_rx_mode() - set the rx mode of a device * @dev: registered network device to modify * * We do not actually need to set any rx filters for the virtual batman * mesh interface. However a dummy handler enables a user to set static * multicast listeners for instance. */ static void batadv_interface_set_rx_mode(struct net_device *dev) { } static netdev_tx_t batadv_interface_tx(struct sk_buff *skb, struct net_device *mesh_iface) { struct ethhdr *ethhdr; struct batadv_priv *bat_priv = netdev_priv(mesh_iface); struct batadv_hard_iface *primary_if = NULL; struct batadv_bcast_packet *bcast_packet; static const u8 stp_addr[ETH_ALEN] = {0x01, 0x80, 0xC2, 0x00, 0x00, 0x00}; static const u8 ectp_addr[ETH_ALEN] = {0xCF, 0x00, 0x00, 0x00, 0x00, 0x00}; enum batadv_dhcp_recipient dhcp_rcp = BATADV_DHCP_NO; u8 *dst_hint = NULL, chaddr[ETH_ALEN]; struct vlan_ethhdr *vhdr; unsigned int header_len = 0; int data_len = skb->len, ret; unsigned long brd_delay = 0; bool do_bcast = false, client_added; unsigned short vid; u32 seqno; int gw_mode; enum batadv_forw_mode forw_mode = BATADV_FORW_BCAST; int mcast_is_routable = 0; int network_offset = ETH_HLEN; __be16 proto; if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE) goto dropped; /* reset control block to avoid left overs from previous users */ memset(skb->cb, 0, sizeof(struct batadv_skb_cb)); netif_trans_update(mesh_iface); vid = batadv_get_vid(skb, 0); skb_reset_mac_header(skb); ethhdr = eth_hdr(skb); proto = ethhdr->h_proto; switch (ntohs(proto)) { case ETH_P_8021Q: if (!pskb_may_pull(skb, sizeof(*vhdr))) goto dropped; vhdr = vlan_eth_hdr(skb); proto = vhdr->h_vlan_encapsulated_proto; /* drop batman-in-batman packets to prevent loops */ if (proto != htons(ETH_P_BATMAN)) { network_offset += VLAN_HLEN; break; } fallthrough; case ETH_P_BATMAN: goto dropped; } skb_set_network_header(skb, network_offset); if (batadv_bla_tx(bat_priv, skb, vid)) goto dropped; /* skb->data might have been reallocated by batadv_bla_tx() */ ethhdr = eth_hdr(skb); /* Register the client MAC in the transtable */ if (!is_multicast_ether_addr(ethhdr->h_source) && !batadv_bla_is_loopdetect_mac(ethhdr->h_source)) { client_added = batadv_tt_local_add(mesh_iface, ethhdr->h_source, vid, skb->skb_iif, skb->mark); if (!client_added) goto dropped; } /* Snoop address candidates from DHCPACKs for early DAT filling */ batadv_dat_snoop_outgoing_dhcp_ack(bat_priv, skb, proto, vid); /* don't accept stp packets. STP does not help in meshes. * better use the bridge loop avoidance ... * * The same goes for ECTP sent at least by some Cisco Switches, * it might confuse the mesh when used with bridge loop avoidance. */ if (batadv_compare_eth(ethhdr->h_dest, stp_addr)) goto dropped; if (batadv_compare_eth(ethhdr->h_dest, ectp_addr)) goto dropped; gw_mode = atomic_read(&bat_priv->gw.mode); if (is_multicast_ether_addr(ethhdr->h_dest)) { /* if gw mode is off, broadcast every packet */ if (gw_mode == BATADV_GW_MODE_OFF) { do_bcast = true; goto send; } dhcp_rcp = batadv_gw_dhcp_recipient_get(skb, &header_len, chaddr); /* skb->data may have been modified by * batadv_gw_dhcp_recipient_get() */ ethhdr = eth_hdr(skb); /* if gw_mode is on, broadcast any non-DHCP message. * All the DHCP packets are going to be sent as unicast */ if (dhcp_rcp == BATADV_DHCP_NO) { do_bcast = true; goto send; } if (dhcp_rcp == BATADV_DHCP_TO_CLIENT) dst_hint = chaddr; else if ((gw_mode == BATADV_GW_MODE_SERVER) && (dhcp_rcp == BATADV_DHCP_TO_SERVER)) /* gateways should not forward any DHCP message if * directed to a DHCP server */ goto dropped; send: if (do_bcast && !is_broadcast_ether_addr(ethhdr->h_dest)) { forw_mode = batadv_mcast_forw_mode(bat_priv, skb, vid, &mcast_is_routable); switch (forw_mode) { case BATADV_FORW_BCAST: break; case BATADV_FORW_UCASTS: case BATADV_FORW_MCAST: do_bcast = false; break; case BATADV_FORW_NONE: fallthrough; default: goto dropped; } } } batadv_skb_set_priority(skb, 0); /* ethernet packet should be broadcasted */ if (do_bcast) { primary_if = batadv_primary_if_get_selected(bat_priv); if (!primary_if) goto dropped; /* in case of ARP request, we do not immediately broadcasti the * packet, instead we first wait for DAT to try to retrieve the * correct ARP entry */ if (batadv_dat_snoop_outgoing_arp_request(bat_priv, skb)) brd_delay = msecs_to_jiffies(ARP_REQ_DELAY); if (batadv_skb_head_push(skb, sizeof(*bcast_packet)) < 0) goto dropped; bcast_packet = (struct batadv_bcast_packet *)skb->data; bcast_packet->version = BATADV_COMPAT_VERSION; bcast_packet->ttl = BATADV_TTL - 1; /* batman packet type: broadcast */ bcast_packet->packet_type = BATADV_BCAST; bcast_packet->reserved = 0; /* hw address of first interface is the orig mac because only * this mac is known throughout the mesh */ ether_addr_copy(bcast_packet->orig, primary_if->net_dev->dev_addr); /* set broadcast sequence number */ seqno = atomic_inc_return(&bat_priv->bcast_seqno); bcast_packet->seqno = htonl(seqno); batadv_send_bcast_packet(bat_priv, skb, brd_delay, true); /* unicast packet */ } else { /* DHCP packets going to a server will use the GW feature */ if (dhcp_rcp == BATADV_DHCP_TO_SERVER) { ret = batadv_gw_out_of_range(bat_priv, skb); if (ret) goto dropped; ret = batadv_send_skb_via_gw(bat_priv, skb, vid); } else if (forw_mode == BATADV_FORW_UCASTS) { ret = batadv_mcast_forw_send(bat_priv, skb, vid, mcast_is_routable); } else if (forw_mode == BATADV_FORW_MCAST) { ret = batadv_mcast_forw_mcsend(bat_priv, skb); } else { if (batadv_dat_snoop_outgoing_arp_request(bat_priv, skb)) goto dropped; batadv_dat_snoop_outgoing_arp_reply(bat_priv, skb); ret = batadv_send_skb_via_tt(bat_priv, skb, dst_hint, vid); } if (ret != NET_XMIT_SUCCESS) goto dropped_freed; } batadv_inc_counter(bat_priv, BATADV_CNT_TX); batadv_add_counter(bat_priv, BATADV_CNT_TX_BYTES, data_len); goto end; dropped: kfree_skb(skb); dropped_freed: batadv_inc_counter(bat_priv, BATADV_CNT_TX_DROPPED); end: batadv_hardif_put(primary_if); return NETDEV_TX_OK; } /** * batadv_interface_rx() - receive ethernet frame on local batman-adv interface * @mesh_iface: local interface which will receive the ethernet frame * @skb: ethernet frame for @mesh_iface * @hdr_size: size of already parsed batman-adv header * @orig_node: originator from which the batman-adv packet was sent * * Sends an ethernet frame to the receive path of the local @mesh_iface. * skb->data has still point to the batman-adv header with the size @hdr_size. * The caller has to have parsed this header already and made sure that at least * @hdr_size bytes are still available for pull in @skb. * * The packet may still get dropped. This can happen when the encapsulated * ethernet frame is invalid or contains again an batman-adv packet. Also * unicast packets will be dropped directly when it was sent between two * isolated clients. */ void batadv_interface_rx(struct net_device *mesh_iface, struct sk_buff *skb, int hdr_size, struct batadv_orig_node *orig_node) { struct batadv_bcast_packet *batadv_bcast_packet; struct batadv_priv *bat_priv = netdev_priv(mesh_iface); struct vlan_ethhdr *vhdr; struct ethhdr *ethhdr; unsigned short vid; int packet_type; batadv_bcast_packet = (struct batadv_bcast_packet *)skb->data; packet_type = batadv_bcast_packet->packet_type; skb_pull_rcsum(skb, hdr_size); skb_reset_mac_header(skb); /* clean the netfilter state now that the batman-adv header has been * removed */ nf_reset_ct(skb); if (unlikely(!pskb_may_pull(skb, ETH_HLEN))) goto dropped; vid = batadv_get_vid(skb, 0); ethhdr = eth_hdr(skb); switch (ntohs(ethhdr->h_proto)) { case ETH_P_8021Q: if (!pskb_may_pull(skb, VLAN_ETH_HLEN)) goto dropped; vhdr = skb_vlan_eth_hdr(skb); /* drop batman-in-batman packets to prevent loops */ if (vhdr->h_vlan_encapsulated_proto != htons(ETH_P_BATMAN)) break; fallthrough; case ETH_P_BATMAN: goto dropped; } /* skb->dev & skb->pkt_type are set here */ skb->protocol = eth_type_trans(skb, mesh_iface); skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN); batadv_inc_counter(bat_priv, BATADV_CNT_RX); batadv_add_counter(bat_priv, BATADV_CNT_RX_BYTES, skb->len + ETH_HLEN); /* Let the bridge loop avoidance check the packet. If will * not handle it, we can safely push it up. */ if (batadv_bla_rx(bat_priv, skb, vid, packet_type)) goto out; if (orig_node) batadv_tt_add_temporary_global_entry(bat_priv, orig_node, ethhdr->h_source, vid); if (is_multicast_ether_addr(ethhdr->h_dest)) { /* set the mark on broadcast packets if AP isolation is ON and * the packet is coming from an "isolated" client */ if (batadv_vlan_ap_isola_get(bat_priv, vid) && batadv_tt_global_is_isolated(bat_priv, ethhdr->h_source, vid)) { /* save bits in skb->mark not covered by the mask and * apply the mark on the rest */ skb->mark &= ~bat_priv->isolation_mark_mask; skb->mark |= bat_priv->isolation_mark; } } else if (batadv_is_ap_isolated(bat_priv, ethhdr->h_source, ethhdr->h_dest, vid)) { goto dropped; } netif_rx(skb); goto out; dropped: kfree_skb(skb); out: return; } /** * batadv_meshif_vlan_release() - release vlan from lists and queue for free * after rcu grace period * @ref: kref pointer of the vlan object */ void batadv_meshif_vlan_release(struct kref *ref) { struct batadv_meshif_vlan *vlan; vlan = container_of(ref, struct batadv_meshif_vlan, refcount); spin_lock_bh(&vlan->bat_priv->meshif_vlan_list_lock); hlist_del_rcu(&vlan->list); spin_unlock_bh(&vlan->bat_priv->meshif_vlan_list_lock); kfree_rcu(vlan, rcu); } /** * batadv_meshif_vlan_get() - get the vlan object for a specific vid * @bat_priv: the bat priv with all the mesh interface information * @vid: the identifier of the vlan object to retrieve * * Return: the private data of the vlan matching the vid passed as argument or * NULL otherwise. The refcounter of the returned object is incremented by 1. */ struct batadv_meshif_vlan *batadv_meshif_vlan_get(struct batadv_priv *bat_priv, unsigned short vid) { struct batadv_meshif_vlan *vlan_tmp, *vlan = NULL; rcu_read_lock(); hlist_for_each_entry_rcu(vlan_tmp, &bat_priv->meshif_vlan_list, list) { if (vlan_tmp->vid != vid) continue; if (!kref_get_unless_zero(&vlan_tmp->refcount)) continue; vlan = vlan_tmp; break; } rcu_read_unlock(); return vlan; } /** * batadv_meshif_create_vlan() - allocate the needed resources for a new vlan * @bat_priv: the bat priv with all the mesh interface information * @vid: the VLAN identifier * * Return: 0 on success, a negative error otherwise. */ int batadv_meshif_create_vlan(struct batadv_priv *bat_priv, unsigned short vid) { struct batadv_meshif_vlan *vlan; spin_lock_bh(&bat_priv->meshif_vlan_list_lock); vlan = batadv_meshif_vlan_get(bat_priv, vid); if (vlan) { batadv_meshif_vlan_put(vlan); spin_unlock_bh(&bat_priv->meshif_vlan_list_lock); return -EEXIST; } vlan = kzalloc(sizeof(*vlan), GFP_ATOMIC); if (!vlan) { spin_unlock_bh(&bat_priv->meshif_vlan_list_lock); return -ENOMEM; } vlan->bat_priv = bat_priv; vlan->vid = vid; kref_init(&vlan->refcount); atomic_set(&vlan->ap_isolation, 0); kref_get(&vlan->refcount); hlist_add_head_rcu(&vlan->list, &bat_priv->meshif_vlan_list); spin_unlock_bh(&bat_priv->meshif_vlan_list_lock); /* add a new TT local entry. This one will be marked with the NOPURGE * flag */ batadv_tt_local_add(bat_priv->mesh_iface, bat_priv->mesh_iface->dev_addr, vid, BATADV_NULL_IFINDEX, BATADV_NO_MARK); /* don't return reference to new meshif_vlan */ batadv_meshif_vlan_put(vlan); return 0; } /** * batadv_meshif_destroy_vlan() - remove and destroy a meshif_vlan object * @bat_priv: the bat priv with all the mesh interface information * @vlan: the object to remove */ static void batadv_meshif_destroy_vlan(struct batadv_priv *bat_priv, struct batadv_meshif_vlan *vlan) { /* explicitly remove the associated TT local entry because it is marked * with the NOPURGE flag */ batadv_tt_local_remove(bat_priv, bat_priv->mesh_iface->dev_addr, vlan->vid, "vlan interface destroyed", false); batadv_meshif_vlan_put(vlan); } /** * batadv_interface_add_vid() - ndo_add_vid API implementation * @dev: the netdev of the mesh interface * @proto: protocol of the vlan id * @vid: identifier of the new vlan * * Set up all the internal structures for handling the new vlan on top of the * mesh interface * * Return: 0 on success or a negative error code in case of failure. */ static int batadv_interface_add_vid(struct net_device *dev, __be16 proto, unsigned short vid) { struct batadv_priv *bat_priv = netdev_priv(dev); struct batadv_meshif_vlan *vlan; /* only 802.1Q vlans are supported. * batman-adv does not know how to handle other types */ if (proto != htons(ETH_P_8021Q)) return -EINVAL; /* VID 0 is only used to indicate "priority tag" frames which only * contain priority information and no VID. No management structures * should be created for this VID and it should be handled like an * untagged frame. */ if (vid == 0) return 0; vid |= BATADV_VLAN_HAS_TAG; /* if a new vlan is getting created and it already exists, it means that * it was not deleted yet. batadv_meshif_vlan_get() increases the * refcount in order to revive the object. * * if it does not exist then create it. */ vlan = batadv_meshif_vlan_get(bat_priv, vid); if (!vlan) return batadv_meshif_create_vlan(bat_priv, vid); /* add a new TT local entry. This one will be marked with the NOPURGE * flag. This must be added again, even if the vlan object already * exists, because the entry was deleted by kill_vid() */ batadv_tt_local_add(bat_priv->mesh_iface, bat_priv->mesh_iface->dev_addr, vid, BATADV_NULL_IFINDEX, BATADV_NO_MARK); return 0; } /** * batadv_interface_kill_vid() - ndo_kill_vid API implementation * @dev: the netdev of the mesh interface * @proto: protocol of the vlan id * @vid: identifier of the deleted vlan * * Destroy all the internal structures used to handle the vlan identified by vid * on top of the mesh interface * * Return: 0 on success, -EINVAL if the specified prototype is not ETH_P_8021Q * or -ENOENT if the specified vlan id wasn't registered. */ static int batadv_interface_kill_vid(struct net_device *dev, __be16 proto, unsigned short vid) { struct batadv_priv *bat_priv = netdev_priv(dev); struct batadv_meshif_vlan *vlan; /* only 802.1Q vlans are supported. batman-adv does not know how to * handle other types */ if (proto != htons(ETH_P_8021Q)) return -EINVAL; /* "priority tag" frames are handled like "untagged" frames * and no meshif_vlan needs to be destroyed */ if (vid == 0) return 0; vlan = batadv_meshif_vlan_get(bat_priv, vid | BATADV_VLAN_HAS_TAG); if (!vlan) return -ENOENT; batadv_meshif_destroy_vlan(bat_priv, vlan); /* finally free the vlan object */ batadv_meshif_vlan_put(vlan); return 0; } /* batman-adv network devices have devices nesting below it and are a special * "super class" of normal network devices; split their locks off into a * separate class since they always nest. */ static struct lock_class_key batadv_netdev_xmit_lock_key; static struct lock_class_key batadv_netdev_addr_lock_key; /** * batadv_set_lockdep_class_one() - Set lockdep class for a single tx queue * @dev: device which owns the tx queue * @txq: tx queue to modify * @_unused: always NULL */ static void batadv_set_lockdep_class_one(struct net_device *dev, struct netdev_queue *txq, void *_unused) { lockdep_set_class(&txq->_xmit_lock, &batadv_netdev_xmit_lock_key); } /** * batadv_set_lockdep_class() - Set txq and addr_list lockdep class * @dev: network device to modify */ static void batadv_set_lockdep_class(struct net_device *dev) { lockdep_set_class(&dev->addr_list_lock, &batadv_netdev_addr_lock_key); netdev_for_each_tx_queue(dev, batadv_set_lockdep_class_one, NULL); } /** * batadv_meshif_init_late() - late stage initialization of mesh interface * @dev: registered network device to modify * * Return: error code on failures */ static int batadv_meshif_init_late(struct net_device *dev) { struct batadv_priv *bat_priv; u32 random_seqno; int ret; size_t cnt_len = sizeof(u64) * BATADV_CNT_NUM; batadv_set_lockdep_class(dev); bat_priv = netdev_priv(dev); bat_priv->mesh_iface = dev; /* batadv_interface_stats() needs to be available as soon as * register_netdevice() has been called */ bat_priv->bat_counters = __alloc_percpu(cnt_len, __alignof__(u64)); if (!bat_priv->bat_counters) return -ENOMEM; atomic_set(&bat_priv->aggregated_ogms, 1); atomic_set(&bat_priv->bonding, 0); #ifdef CONFIG_BATMAN_ADV_BLA atomic_set(&bat_priv->bridge_loop_avoidance, 1); #endif #ifdef CONFIG_BATMAN_ADV_DAT atomic_set(&bat_priv->distributed_arp_table, 1); #endif #ifdef CONFIG_BATMAN_ADV_MCAST atomic_set(&bat_priv->multicast_mode, 1); atomic_set(&bat_priv->multicast_fanout, 16); atomic_set(&bat_priv->mcast.num_want_all_unsnoopables, 0); atomic_set(&bat_priv->mcast.num_want_all_ipv4, 0); atomic_set(&bat_priv->mcast.num_want_all_ipv6, 0); atomic_set(&bat_priv->mcast.num_no_mc_ptype_capa, 0); #endif atomic_set(&bat_priv->gw.mode, BATADV_GW_MODE_OFF); atomic_set(&bat_priv->gw.bandwidth_down, 100); atomic_set(&bat_priv->gw.bandwidth_up, 20); atomic_set(&bat_priv->orig_interval, 1000); atomic_set(&bat_priv->hop_penalty, 30); #ifdef CONFIG_BATMAN_ADV_DEBUG atomic_set(&bat_priv->log_level, 0); #endif atomic_set(&bat_priv->fragmentation, 1); atomic_set(&bat_priv->packet_size_max, BATADV_MAX_MTU); atomic_set(&bat_priv->bcast_queue_left, BATADV_BCAST_QUEUE_LEN); atomic_set(&bat_priv->batman_queue_left, BATADV_BATMAN_QUEUE_LEN); atomic_set(&bat_priv->mesh_state, BATADV_MESH_INACTIVE); atomic_set(&bat_priv->bcast_seqno, 1); atomic_set(&bat_priv->tt.vn, 0); atomic_set(&bat_priv->tt.ogm_append_cnt, 0); #ifdef CONFIG_BATMAN_ADV_BLA atomic_set(&bat_priv->bla.num_requests, 0); #endif atomic_set(&bat_priv->tp_num, 0); WRITE_ONCE(bat_priv->tt.local_changes, 0); bat_priv->tt.last_changeset = NULL; bat_priv->tt.last_changeset_len = 0; bat_priv->isolation_mark = 0; bat_priv->isolation_mark_mask = 0; /* randomize initial seqno to avoid collision */ get_random_bytes(&random_seqno, sizeof(random_seqno)); atomic_set(&bat_priv->frag_seqno, random_seqno); bat_priv->primary_if = NULL; if (!bat_priv->algo_ops) { ret = batadv_algo_select(bat_priv, batadv_routing_algo); if (ret < 0) goto free_bat_counters; } ret = batadv_mesh_init(dev); if (ret < 0) goto free_bat_counters; return 0; free_bat_counters: free_percpu(bat_priv->bat_counters); bat_priv->bat_counters = NULL; return ret; } /** * batadv_meshif_slave_add() - Add a slave interface to a batadv_mesh_interface * @dev: batadv_mesh_interface used as master interface * @slave_dev: net_device which should become the slave interface * @extack: extended ACK report struct * * Return: 0 if successful or error otherwise. */ static int batadv_meshif_slave_add(struct net_device *dev, struct net_device *slave_dev, struct netlink_ext_ack *extack) { struct batadv_hard_iface *hard_iface; int ret = -EINVAL; hard_iface = batadv_hardif_get_by_netdev(slave_dev); if (!hard_iface || hard_iface->mesh_iface) goto out; ret = batadv_hardif_enable_interface(hard_iface, dev); out: batadv_hardif_put(hard_iface); return ret; } /** * batadv_meshif_slave_del() - Delete a slave iface from a batadv_mesh_interface * @dev: batadv_mesh_interface used as master interface * @slave_dev: net_device which should be removed from the master interface * * Return: 0 if successful or error otherwise. */ static int batadv_meshif_slave_del(struct net_device *dev, struct net_device *slave_dev) { struct batadv_hard_iface *hard_iface; int ret = -EINVAL; hard_iface = batadv_hardif_get_by_netdev(slave_dev); if (!hard_iface || hard_iface->mesh_iface != dev) goto out; batadv_hardif_disable_interface(hard_iface); ret = 0; out: batadv_hardif_put(hard_iface); return ret; } static const struct net_device_ops batadv_netdev_ops = { .ndo_init = batadv_meshif_init_late, .ndo_get_stats = batadv_interface_stats, .ndo_vlan_rx_add_vid = batadv_interface_add_vid, .ndo_vlan_rx_kill_vid = batadv_interface_kill_vid, .ndo_set_mac_address = batadv_interface_set_mac_addr, .ndo_change_mtu = batadv_interface_change_mtu, .ndo_set_rx_mode = batadv_interface_set_rx_mode, .ndo_start_xmit = batadv_interface_tx, .ndo_validate_addr = eth_validate_addr, .ndo_add_slave = batadv_meshif_slave_add, .ndo_del_slave = batadv_meshif_slave_del, }; static void batadv_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { strscpy(info->driver, "B.A.T.M.A.N. advanced", sizeof(info->driver)); strscpy(info->version, BATADV_SOURCE_VERSION, sizeof(info->version)); strscpy(info->fw_version, "N/A", sizeof(info->fw_version)); strscpy(info->bus_info, "batman", sizeof(info->bus_info)); } /* Inspired by drivers/net/ethernet/dlink/sundance.c:1702 * Declare each description string in struct.name[] to get fixed sized buffer * and compile time checking for strings longer than ETH_GSTRING_LEN. */ static const struct { const char name[ETH_GSTRING_LEN]; } batadv_counters_strings[] = { { "tx" }, { "tx_bytes" }, { "tx_dropped" }, { "rx" }, { "rx_bytes" }, { "forward" }, { "forward_bytes" }, { "mgmt_tx" }, { "mgmt_tx_bytes" }, { "mgmt_rx" }, { "mgmt_rx_bytes" }, { "frag_tx" }, { "frag_tx_bytes" }, { "frag_rx" }, { "frag_rx_bytes" }, { "frag_fwd" }, { "frag_fwd_bytes" }, { "tt_request_tx" }, { "tt_request_rx" }, { "tt_response_tx" }, { "tt_response_rx" }, { "tt_roam_adv_tx" }, { "tt_roam_adv_rx" }, #ifdef CONFIG_BATMAN_ADV_MCAST { "mcast_tx" }, { "mcast_tx_bytes" }, { "mcast_tx_local" }, { "mcast_tx_local_bytes" }, { "mcast_rx" }, { "mcast_rx_bytes" }, { "mcast_rx_local" }, { "mcast_rx_local_bytes" }, { "mcast_fwd" }, { "mcast_fwd_bytes" }, #endif #ifdef CONFIG_BATMAN_ADV_DAT { "dat_get_tx" }, { "dat_get_rx" }, { "dat_put_tx" }, { "dat_put_rx" }, { "dat_cached_reply_tx" }, #endif }; static void batadv_get_strings(struct net_device *dev, u32 stringset, u8 *data) { if (stringset == ETH_SS_STATS) memcpy(data, batadv_counters_strings, sizeof(batadv_counters_strings)); } static void batadv_get_ethtool_stats(struct net_device *dev, struct ethtool_stats *stats, u64 *data) { struct batadv_priv *bat_priv = netdev_priv(dev); int i; for (i = 0; i < BATADV_CNT_NUM; i++) data[i] = batadv_sum_counter(bat_priv, i); } static int batadv_get_sset_count(struct net_device *dev, int stringset) { if (stringset == ETH_SS_STATS) return BATADV_CNT_NUM; return -EOPNOTSUPP; } static const struct ethtool_ops batadv_ethtool_ops = { .get_drvinfo = batadv_get_drvinfo, .get_link = ethtool_op_get_link, .get_strings = batadv_get_strings, .get_ethtool_stats = batadv_get_ethtool_stats, .get_sset_count = batadv_get_sset_count, }; /** * batadv_meshif_free() - Deconstructor of batadv_mesh_interface * @dev: Device to cleanup and remove */ static void batadv_meshif_free(struct net_device *dev) { batadv_mesh_free(dev); /* some scheduled RCU callbacks need the bat_priv struct to accomplish * their tasks. Wait for them all to be finished before freeing the * netdev and its private data (bat_priv) */ rcu_barrier(); } /** * batadv_meshif_init_early() - early stage initialization of mesh interface * @dev: registered network device to modify */ static void batadv_meshif_init_early(struct net_device *dev) { ether_setup(dev); dev->netdev_ops = &batadv_netdev_ops; dev->needs_free_netdev = true; dev->priv_destructor = batadv_meshif_free; dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER; dev->priv_flags |= IFF_NO_QUEUE; dev->lltx = true; dev->netns_immutable = true; /* can't call min_mtu, because the needed variables * have not been initialized yet */ dev->mtu = ETH_DATA_LEN; dev->max_mtu = BATADV_MAX_MTU; /* generate random address */ eth_hw_addr_random(dev); dev->ethtool_ops = &batadv_ethtool_ops; } /** * batadv_meshif_validate() - validate configuration of new batadv link * @tb: IFLA_INFO_DATA netlink attributes * @data: enum batadv_ifla_attrs attributes * @extack: extended ACK report struct * * Return: 0 if successful or error otherwise. */ static int batadv_meshif_validate(struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { struct batadv_algo_ops *algo_ops; if (!data) return 0; if (data[IFLA_BATADV_ALGO_NAME]) { algo_ops = batadv_algo_get(nla_data(data[IFLA_BATADV_ALGO_NAME])); if (!algo_ops) return -EINVAL; } return 0; } /** * batadv_meshif_newlink() - pre-initialize and register new batadv link * @dev: network device to register * @params: rtnl newlink parameters * @extack: extended ACK report struct * * Return: 0 if successful or error otherwise. */ static int batadv_meshif_newlink(struct net_device *dev, struct rtnl_newlink_params *params, struct netlink_ext_ack *extack) { struct batadv_priv *bat_priv = netdev_priv(dev); struct nlattr **data = params->data; const char *algo_name; int err; if (data && data[IFLA_BATADV_ALGO_NAME]) { algo_name = nla_data(data[IFLA_BATADV_ALGO_NAME]); err = batadv_algo_select(bat_priv, algo_name); if (err) return -EINVAL; } return register_netdevice(dev); } /** * batadv_meshif_destroy_netlink() - deletion of batadv_mesh_interface via * netlink * @mesh_iface: the to-be-removed batman-adv interface * @head: list pointer */ static void batadv_meshif_destroy_netlink(struct net_device *mesh_iface, struct list_head *head) { struct batadv_priv *bat_priv = netdev_priv(mesh_iface); struct batadv_hard_iface *hard_iface; struct batadv_meshif_vlan *vlan; while (!list_empty(&mesh_iface->adj_list.lower)) { hard_iface = netdev_adjacent_get_private(mesh_iface->adj_list.lower.next); batadv_hardif_disable_interface(hard_iface); } /* destroy the "untagged" VLAN */ vlan = batadv_meshif_vlan_get(bat_priv, BATADV_NO_FLAGS); if (vlan) { batadv_meshif_destroy_vlan(bat_priv, vlan); batadv_meshif_vlan_put(vlan); } unregister_netdevice_queue(mesh_iface, head); } /** * batadv_meshif_is_valid() - Check whether device is a batadv mesh interface * @net_dev: device which should be checked * * Return: true when net_dev is a batman-adv interface, false otherwise */ bool batadv_meshif_is_valid(const struct net_device *net_dev) { if (net_dev->netdev_ops->ndo_start_xmit == batadv_interface_tx) return true; return false; } static const struct nla_policy batadv_ifla_policy[IFLA_BATADV_MAX + 1] = { [IFLA_BATADV_ALGO_NAME] = { .type = NLA_NUL_STRING }, }; struct rtnl_link_ops batadv_link_ops __read_mostly = { .kind = "batadv", .priv_size = sizeof(struct batadv_priv), .setup = batadv_meshif_init_early, .maxtype = IFLA_BATADV_MAX, .policy = batadv_ifla_policy, .validate = batadv_meshif_validate, .newlink = batadv_meshif_newlink, .dellink = batadv_meshif_destroy_netlink, }; |
| 7140 443 6781 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 | /* SPDX-License-Identifier: GPL-2.0 */ #ifndef _ASM_PGTABLE_INVERT_H #define _ASM_PGTABLE_INVERT_H 1 #ifndef __ASSEMBLER__ /* * A clear pte value is special, and doesn't get inverted. * * Note that even users that only pass a pgprot_t (rather * than a full pte) won't trigger the special zero case, * because even PAGE_NONE has _PAGE_PROTNONE | _PAGE_ACCESSED * set. So the all zero case really is limited to just the * cleared page table entry case. */ static inline bool __pte_needs_invert(u64 val) { return val && !(val & _PAGE_PRESENT); } /* Get a mask to xor with the page table entry to get the correct pfn. */ static inline u64 protnone_mask(u64 val) { return __pte_needs_invert(val) ? ~0ull : 0; } static inline u64 flip_protnone_guard(u64 oldval, u64 val, u64 mask) { /* * When a PTE transitions from NONE to !NONE or vice-versa * invert the PFN part to stop speculation. * pte_pfn undoes this when needed. */ if (__pte_needs_invert(oldval) != __pte_needs_invert(val)) val = (val & ~mask) | (~val & mask); return val; } #endif /* __ASSEMBLER__ */ #endif |
| 11 156 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 | /* SPDX-License-Identifier: GPL-2.0-or-later */ /* * Copyright (c) 2014 Mahesh Bandewar <maheshb@google.com> */ #ifndef __IPVLAN_H #define __IPVLAN_H #include <linux/kernel.h> #include <linux/types.h> #include <linux/module.h> #include <linux/init.h> #include <linux/rculist.h> #include <linux/notifier.h> #include <linux/netdevice.h> #include <linux/etherdevice.h> #include <linux/if_arp.h> #include <linux/if_link.h> #include <linux/if_vlan.h> #include <linux/ip.h> #include <linux/inetdevice.h> #include <linux/netfilter.h> #include <net/ip.h> #include <net/ip6_route.h> #include <net/netns/generic.h> #include <net/rtnetlink.h> #include <net/route.h> #include <net/addrconf.h> #include <net/l3mdev.h> #define IPVLAN_DRV "ipvlan" #define IPV_DRV_VER "0.1" #define IPVLAN_HASH_SIZE (1 << BITS_PER_BYTE) #define IPVLAN_HASH_MASK (IPVLAN_HASH_SIZE - 1) #define IPVLAN_MAC_FILTER_BITS 8 #define IPVLAN_MAC_FILTER_SIZE (1 << IPVLAN_MAC_FILTER_BITS) #define IPVLAN_MAC_FILTER_MASK (IPVLAN_MAC_FILTER_SIZE - 1) #define IPVLAN_QBACKLOG_LIMIT 1000 typedef enum { IPVL_IPV6 = 0, IPVL_ICMPV6, IPVL_IPV4, IPVL_ARP, } ipvl_hdr_type; struct ipvl_pcpu_stats { u64_stats_t rx_pkts; u64_stats_t rx_bytes; u64_stats_t rx_mcast; u64_stats_t tx_pkts; u64_stats_t tx_bytes; struct u64_stats_sync syncp; u32 rx_errs; u32 tx_drps; }; struct ipvl_port; struct ipvl_dev { struct net_device *dev; struct list_head pnode; struct ipvl_port *port; struct net_device *phy_dev; struct list_head addrs; struct ipvl_pcpu_stats __percpu *pcpu_stats; DECLARE_BITMAP(mac_filters, IPVLAN_MAC_FILTER_SIZE); netdev_features_t sfeatures; u32 msg_enable; }; struct ipvl_addr { struct ipvl_dev *master; /* Back pointer to master */ union { struct in6_addr ip6; /* IPv6 address on logical interface */ struct in_addr ip4; /* IPv4 address on logical interface */ } ipu; #define ip6addr ipu.ip6 #define ip4addr ipu.ip4 struct hlist_node hlnode; /* Hash-table linkage */ struct list_head anode; /* logical-interface linkage */ ipvl_hdr_type atype; struct rcu_head rcu; }; struct ipvl_port { struct net_device *dev; possible_net_t pnet; struct hlist_head hlhead[IPVLAN_HASH_SIZE]; spinlock_t addrs_lock; /* guards hash-table and addrs */ struct list_head ipvlans; u16 mode; u16 flags; u16 dev_id_start; struct work_struct wq; struct sk_buff_head backlog; int count; struct ida ida; netdevice_tracker dev_tracker; }; struct ipvl_skb_cb { bool tx_pkt; }; #define IPVL_SKB_CB(_skb) ((struct ipvl_skb_cb *)&((_skb)->cb[0])) static inline struct ipvl_port *ipvlan_port_get_rcu(const struct net_device *d) { return rcu_dereference(d->rx_handler_data); } static inline struct ipvl_port *ipvlan_port_get_rcu_bh(const struct net_device *d) { return rcu_dereference_bh(d->rx_handler_data); } static inline struct ipvl_port *ipvlan_port_get_rtnl(const struct net_device *d) { return rtnl_dereference(d->rx_handler_data); } static inline bool ipvlan_is_private(const struct ipvl_port *port) { return !!(port->flags & IPVLAN_F_PRIVATE); } static inline void ipvlan_mark_private(struct ipvl_port *port) { port->flags |= IPVLAN_F_PRIVATE; } static inline void ipvlan_clear_private(struct ipvl_port *port) { port->flags &= ~IPVLAN_F_PRIVATE; } static inline bool ipvlan_is_vepa(const struct ipvl_port *port) { return !!(port->flags & IPVLAN_F_VEPA); } static inline void ipvlan_mark_vepa(struct ipvl_port *port) { port->flags |= IPVLAN_F_VEPA; } static inline void ipvlan_clear_vepa(struct ipvl_port *port) { port->flags &= ~IPVLAN_F_VEPA; } void ipvlan_init_secret(void); unsigned int ipvlan_mac_hash(const unsigned char *addr); rx_handler_result_t ipvlan_handle_frame(struct sk_buff **pskb); void ipvlan_process_multicast(struct work_struct *work); int ipvlan_queue_xmit(struct sk_buff *skb, struct net_device *dev); void ipvlan_ht_addr_add(struct ipvl_dev *ipvlan, struct ipvl_addr *addr); struct ipvl_addr *ipvlan_find_addr(const struct ipvl_dev *ipvlan, const void *iaddr, bool is_v6); bool ipvlan_addr_busy(struct ipvl_port *port, void *iaddr, bool is_v6); void ipvlan_ht_addr_del(struct ipvl_addr *addr); struct ipvl_addr *ipvlan_addr_lookup(struct ipvl_port *port, void *lyr3h, int addr_type, bool use_dest); void *ipvlan_get_L3_hdr(struct ipvl_port *port, struct sk_buff *skb, int *type); void ipvlan_count_rx(const struct ipvl_dev *ipvlan, unsigned int len, bool success, bool mcast); int ipvlan_link_new(struct net_device *dev, struct rtnl_newlink_params *params, struct netlink_ext_ack *extack); void ipvlan_link_delete(struct net_device *dev, struct list_head *head); void ipvlan_link_setup(struct net_device *dev); int ipvlan_link_register(struct rtnl_link_ops *ops); #ifdef CONFIG_IPVLAN_L3S int ipvlan_l3s_register(struct ipvl_port *port); void ipvlan_l3s_unregister(struct ipvl_port *port); void ipvlan_migrate_l3s_hook(struct net *oldnet, struct net *newnet); int ipvlan_l3s_init(void); void ipvlan_l3s_cleanup(void); #else static inline int ipvlan_l3s_register(struct ipvl_port *port) { return -ENOTSUPP; } static inline void ipvlan_l3s_unregister(struct ipvl_port *port) { } static inline void ipvlan_migrate_l3s_hook(struct net *oldnet, struct net *newnet) { } static inline int ipvlan_l3s_init(void) { return 0; } static inline void ipvlan_l3s_cleanup(void) { } #endif /* CONFIG_IPVLAN_L3S */ static inline bool netif_is_ipvlan_port(const struct net_device *dev) { return rcu_access_pointer(dev->rx_handler) == ipvlan_handle_frame; } #endif /* __IPVLAN_H */ |
| 1 1 1 2 2 2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 | // SPDX-License-Identifier: GPL-2.0+ /* * Driver for SCM Microsystems (a.k.a. Shuttle) USB-ATAPI cable * * Current development and maintenance by: * (c) 2000, 2001 Robert Baruch (autophile@starband.net) * (c) 2004, 2005 Daniel Drake <dsd@gentoo.org> * * Developed with the assistance of: * (c) 2002 Alan Stern <stern@rowland.org> * * Flash support based on earlier work by: * (c) 2002 Thomas Kreiling <usbdev@sm04.de> * * Many originally ATAPI devices were slightly modified to meet the USB * market by using some kind of translation from ATAPI to USB on the host, * and the peripheral would translate from USB back to ATAPI. * * SCM Microsystems (www.scmmicro.com) makes a device, sold to OEM's only, * which does the USB-to-ATAPI conversion. By obtaining the data sheet on * their device under nondisclosure agreement, I have been able to write * this driver for Linux. * * The chip used in the device can also be used for EPP and ISA translation * as well. This driver is only guaranteed to work with the ATAPI * translation. * * See the Kconfig help text for a list of devices known to be supported by * this driver. */ #include <linux/errno.h> #include <linux/module.h> #include <linux/slab.h> #include <linux/string_choices.h> #include <linux/cdrom.h> #include <scsi/scsi.h> #include <scsi/scsi_cmnd.h> #include "usb.h" #include "transport.h" #include "protocol.h" #include "debug.h" #include "scsiglue.h" #define DRV_NAME "ums-usbat" MODULE_DESCRIPTION("Driver for SCM Microsystems (a.k.a. Shuttle) USB-ATAPI cable"); MODULE_AUTHOR("Daniel Drake <dsd@gentoo.org>, Robert Baruch <autophile@starband.net>"); MODULE_LICENSE("GPL"); MODULE_IMPORT_NS("USB_STORAGE"); /* Supported device types */ #define USBAT_DEV_HP8200 0x01 #define USBAT_DEV_FLASH 0x02 #define USBAT_EPP_PORT 0x10 #define USBAT_EPP_REGISTER 0x30 #define USBAT_ATA 0x40 #define USBAT_ISA 0x50 /* Commands (need to be logically OR'd with an access type */ #define USBAT_CMD_READ_REG 0x00 #define USBAT_CMD_WRITE_REG 0x01 #define USBAT_CMD_READ_BLOCK 0x02 #define USBAT_CMD_WRITE_BLOCK 0x03 #define USBAT_CMD_COND_READ_BLOCK 0x04 #define USBAT_CMD_COND_WRITE_BLOCK 0x05 #define USBAT_CMD_WRITE_REGS 0x07 /* Commands (these don't need an access type) */ #define USBAT_CMD_EXEC_CMD 0x80 #define USBAT_CMD_SET_FEAT 0x81 #define USBAT_CMD_UIO 0x82 /* Methods of accessing UIO register */ #define USBAT_UIO_READ 1 #define USBAT_UIO_WRITE 0 /* Qualifier bits */ #define USBAT_QUAL_FCQ 0x20 /* full compare */ #define USBAT_QUAL_ALQ 0x10 /* auto load subcount */ /* USBAT Flash Media status types */ #define USBAT_FLASH_MEDIA_NONE 0 #define USBAT_FLASH_MEDIA_CF 1 /* USBAT Flash Media change types */ #define USBAT_FLASH_MEDIA_SAME 0 #define USBAT_FLASH_MEDIA_CHANGED 1 /* USBAT ATA registers */ #define USBAT_ATA_DATA 0x10 /* read/write data (R/W) */ #define USBAT_ATA_FEATURES 0x11 /* set features (W) */ #define USBAT_ATA_ERROR 0x11 /* error (R) */ #define USBAT_ATA_SECCNT 0x12 /* sector count (R/W) */ #define USBAT_ATA_SECNUM 0x13 /* sector number (R/W) */ #define USBAT_ATA_LBA_ME 0x14 /* cylinder low (R/W) */ #define USBAT_ATA_LBA_HI 0x15 /* cylinder high (R/W) */ #define USBAT_ATA_DEVICE 0x16 /* head/device selection (R/W) */ #define USBAT_ATA_STATUS 0x17 /* device status (R) */ #define USBAT_ATA_CMD 0x17 /* device command (W) */ #define USBAT_ATA_ALTSTATUS 0x0E /* status (no clear IRQ) (R) */ /* USBAT User I/O Data registers */ #define USBAT_UIO_EPAD 0x80 /* Enable Peripheral Control Signals */ #define USBAT_UIO_CDT 0x40 /* Card Detect (Read Only) */ /* CDT = ACKD & !UI1 & !UI0 */ #define USBAT_UIO_1 0x20 /* I/O 1 */ #define USBAT_UIO_0 0x10 /* I/O 0 */ #define USBAT_UIO_EPP_ATA 0x08 /* 1=EPP mode, 0=ATA mode */ #define USBAT_UIO_UI1 0x04 /* Input 1 */ #define USBAT_UIO_UI0 0x02 /* Input 0 */ #define USBAT_UIO_INTR_ACK 0x01 /* Interrupt (ATA/ISA)/Acknowledge (EPP) */ /* USBAT User I/O Enable registers */ #define USBAT_UIO_DRVRST 0x80 /* Reset Peripheral */ #define USBAT_UIO_ACKD 0x40 /* Enable Card Detect */ #define USBAT_UIO_OE1 0x20 /* I/O 1 set=output/clr=input */ /* If ACKD=1, set OE1 to 1 also. */ #define USBAT_UIO_OE0 0x10 /* I/O 0 set=output/clr=input */ #define USBAT_UIO_ADPRST 0x01 /* Reset SCM chip */ /* USBAT Features */ #define USBAT_FEAT_ETEN 0x80 /* External trigger enable */ #define USBAT_FEAT_U1 0x08 #define USBAT_FEAT_U0 0x04 #define USBAT_FEAT_ET1 0x02 #define USBAT_FEAT_ET2 0x01 struct usbat_info { int devicetype; /* Used for Flash readers only */ unsigned long sectors; /* total sector count */ unsigned long ssize; /* sector size in bytes */ unsigned char sense_key; unsigned long sense_asc; /* additional sense code */ unsigned long sense_ascq; /* additional sense code qualifier */ }; #define short_pack(LSB,MSB) ( ((u16)(LSB)) | ( ((u16)(MSB))<<8 ) ) #define LSB_of(s) ((s)&0xFF) #define MSB_of(s) ((s)>>8) static int transferred = 0; static int usbat_flash_transport(struct scsi_cmnd * srb, struct us_data *us); static int usbat_hp8200e_transport(struct scsi_cmnd *srb, struct us_data *us); static int init_usbat_cd(struct us_data *us); static int init_usbat_flash(struct us_data *us); /* * The table of devices */ #define UNUSUAL_DEV(id_vendor, id_product, bcdDeviceMin, bcdDeviceMax, \ vendorName, productName, useProtocol, useTransport, \ initFunction, flags) \ { USB_DEVICE_VER(id_vendor, id_product, bcdDeviceMin, bcdDeviceMax), \ .driver_info = (flags) } static const struct usb_device_id usbat_usb_ids[] = { # include "unusual_usbat.h" { } /* Terminating entry */ }; MODULE_DEVICE_TABLE(usb, usbat_usb_ids); #undef UNUSUAL_DEV /* * The flags table */ #define UNUSUAL_DEV(idVendor, idProduct, bcdDeviceMin, bcdDeviceMax, \ vendor_name, product_name, use_protocol, use_transport, \ init_function, Flags) \ { \ .vendorName = vendor_name, \ .productName = product_name, \ .useProtocol = use_protocol, \ .useTransport = use_transport, \ .initFunction = init_function, \ } static const struct us_unusual_dev usbat_unusual_dev_list[] = { # include "unusual_usbat.h" { } /* Terminating entry */ }; #undef UNUSUAL_DEV /* * Convenience function to produce an ATA read/write sectors command * Use cmd=0x20 for read, cmd=0x30 for write */ static void usbat_pack_ata_sector_cmd(unsigned char *buf, unsigned char thistime, u32 sector, unsigned char cmd) { buf[0] = 0; buf[1] = thistime; buf[2] = sector & 0xFF; buf[3] = (sector >> 8) & 0xFF; buf[4] = (sector >> 16) & 0xFF; buf[5] = 0xE0 | ((sector >> 24) & 0x0F); buf[6] = cmd; } /* * Convenience function to get the device type (flash or hp8200) */ static int usbat_get_device_type(struct us_data *us) { return ((struct usbat_info*)us->extra)->devicetype; } /* * Read a register from the device */ static int usbat_read(struct us_data *us, unsigned char access, unsigned char reg, unsigned char *content) { return usb_stor_ctrl_transfer(us, us->recv_ctrl_pipe, access | USBAT_CMD_READ_REG, 0xC0, (u16)reg, 0, content, 1); } /* * Write to a register on the device */ static int usbat_write(struct us_data *us, unsigned char access, unsigned char reg, unsigned char content) { return usb_stor_ctrl_transfer(us, us->send_ctrl_pipe, access | USBAT_CMD_WRITE_REG, 0x40, short_pack(reg, content), 0, NULL, 0); } /* * Convenience function to perform a bulk read */ static int usbat_bulk_read(struct us_data *us, void* buf, unsigned int len, int use_sg) { if (len == 0) return USB_STOR_XFER_GOOD; usb_stor_dbg(us, "len = %d\n", len); return usb_stor_bulk_transfer_sg(us, us->recv_bulk_pipe, buf, len, use_sg, NULL); } /* * Convenience function to perform a bulk write */ static int usbat_bulk_write(struct us_data *us, void* buf, unsigned int len, int use_sg) { if (len == 0) return USB_STOR_XFER_GOOD; usb_stor_dbg(us, "len = %d\n", len); return usb_stor_bulk_transfer_sg(us, us->send_bulk_pipe, buf, len, use_sg, NULL); } /* * Some USBAT-specific commands can only be executed over a command transport * This transport allows one (len=8) or two (len=16) vendor-specific commands * to be executed. */ static int usbat_execute_command(struct us_data *us, unsigned char *commands, unsigned int len) { return usb_stor_ctrl_transfer(us, us->send_ctrl_pipe, USBAT_CMD_EXEC_CMD, 0x40, 0, 0, commands, len); } /* * Read the status register */ static int usbat_get_status(struct us_data *us, unsigned char *status) { int rc; rc = usbat_read(us, USBAT_ATA, USBAT_ATA_STATUS, status); usb_stor_dbg(us, "0x%02X\n", *status); return rc; } /* * Check the device status */ static int usbat_check_status(struct us_data *us) { unsigned char *reply = us->iobuf; int rc; rc = usbat_get_status(us, reply); if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_FAILED; /* error/check condition (0x51 is ok) */ if (*reply & 0x01 && *reply != 0x51) return USB_STOR_TRANSPORT_FAILED; /* device fault */ if (*reply & 0x20) return USB_STOR_TRANSPORT_FAILED; return USB_STOR_TRANSPORT_GOOD; } /* * Stores critical information in internal registers in preparation for the execution * of a conditional usbat_read_blocks or usbat_write_blocks call. */ static int usbat_set_shuttle_features(struct us_data *us, unsigned char external_trigger, unsigned char epp_control, unsigned char mask_byte, unsigned char test_pattern, unsigned char subcountH, unsigned char subcountL) { unsigned char *command = us->iobuf; command[0] = 0x40; command[1] = USBAT_CMD_SET_FEAT; /* * The only bit relevant to ATA access is bit 6 * which defines 8 bit data access (set) or 16 bit (unset) */ command[2] = epp_control; /* * If FCQ is set in the qualifier (defined in R/W cmd), then bits U0, U1, * ET1 and ET2 define an external event to be checked for on event of a * _read_blocks or _write_blocks operation. The read/write will not take * place unless the defined trigger signal is active. */ command[3] = external_trigger; /* * The resultant byte of the mask operation (see mask_byte) is compared for * equivalence with this test pattern. If equal, the read/write will take * place. */ command[4] = test_pattern; /* * This value is logically ANDed with the status register field specified * in the read/write command. */ command[5] = mask_byte; /* * If ALQ is set in the qualifier, this field contains the address of the * registers where the byte count should be read for transferring the data. * If ALQ is not set, then this field contains the number of bytes to be * transferred. */ command[6] = subcountL; command[7] = subcountH; return usbat_execute_command(us, command, 8); } /* * Block, waiting for an ATA device to become not busy or to report * an error condition. */ static int usbat_wait_not_busy(struct us_data *us, int minutes) { int i; int result; unsigned char *status = us->iobuf; /* * Synchronizing cache on a CDR could take a heck of a long time, * but probably not more than 10 minutes or so. On the other hand, * doing a full blank on a CDRW at speed 1 will take about 75 * minutes! */ for (i=0; i<1200+minutes*60; i++) { result = usbat_get_status(us, status); if (result!=USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (*status & 0x01) { /* check condition */ result = usbat_read(us, USBAT_ATA, 0x10, status); return USB_STOR_TRANSPORT_FAILED; } if (*status & 0x20) /* device fault */ return USB_STOR_TRANSPORT_FAILED; if ((*status & 0x80)==0x00) { /* not busy */ usb_stor_dbg(us, "Waited not busy for %d steps\n", i); return USB_STOR_TRANSPORT_GOOD; } if (i<500) msleep(10); /* 5 seconds */ else if (i<700) msleep(50); /* 10 seconds */ else if (i<1200) msleep(100); /* 50 seconds */ else msleep(1000); /* X minutes */ } usb_stor_dbg(us, "Waited not busy for %d minutes, timing out\n", minutes); return USB_STOR_TRANSPORT_FAILED; } /* * Read block data from the data register */ static int usbat_read_block(struct us_data *us, void* buf, unsigned short len, int use_sg) { int result; unsigned char *command = us->iobuf; if (!len) return USB_STOR_TRANSPORT_GOOD; command[0] = 0xC0; command[1] = USBAT_ATA | USBAT_CMD_READ_BLOCK; command[2] = USBAT_ATA_DATA; command[3] = 0; command[4] = 0; command[5] = 0; command[6] = LSB_of(len); command[7] = MSB_of(len); result = usbat_execute_command(us, command, 8); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; result = usbat_bulk_read(us, buf, len, use_sg); return (result == USB_STOR_XFER_GOOD ? USB_STOR_TRANSPORT_GOOD : USB_STOR_TRANSPORT_ERROR); } /* * Write block data via the data register */ static int usbat_write_block(struct us_data *us, unsigned char access, void* buf, unsigned short len, int minutes, int use_sg) { int result; unsigned char *command = us->iobuf; if (!len) return USB_STOR_TRANSPORT_GOOD; command[0] = 0x40; command[1] = access | USBAT_CMD_WRITE_BLOCK; command[2] = USBAT_ATA_DATA; command[3] = 0; command[4] = 0; command[5] = 0; command[6] = LSB_of(len); command[7] = MSB_of(len); result = usbat_execute_command(us, command, 8); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; result = usbat_bulk_write(us, buf, len, use_sg); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; return usbat_wait_not_busy(us, minutes); } /* * Process read and write requests */ static int usbat_hp8200e_rw_block_test(struct us_data *us, unsigned char access, unsigned char *registers, unsigned char *data_out, unsigned short num_registers, unsigned char data_reg, unsigned char status_reg, unsigned char timeout, unsigned char qualifier, int direction, void *buf, unsigned short len, int use_sg, int minutes) { int result; unsigned int pipe = (direction == DMA_FROM_DEVICE) ? us->recv_bulk_pipe : us->send_bulk_pipe; unsigned char *command = us->iobuf; int i, j; int cmdlen; unsigned char *data = us->iobuf; unsigned char *status = us->iobuf; BUG_ON(num_registers > US_IOBUF_SIZE/2); for (i=0; i<20; i++) { /* * The first time we send the full command, which consists * of downloading the SCSI command followed by downloading * the data via a write-and-test. Any other time we only * send the command to download the data -- the SCSI command * is still 'active' in some sense in the device. * * We're only going to try sending the data 10 times. After * that, we just return a failure. */ if (i==0) { cmdlen = 16; /* * Write to multiple registers * Not really sure the 0x07, 0x17, 0xfc, 0xe7 is * necessary here, but that's what came out of the * trace every single time. */ command[0] = 0x40; command[1] = access | USBAT_CMD_WRITE_REGS; command[2] = 0x07; command[3] = 0x17; command[4] = 0xFC; command[5] = 0xE7; command[6] = LSB_of(num_registers*2); command[7] = MSB_of(num_registers*2); } else cmdlen = 8; /* Conditionally read or write blocks */ command[cmdlen-8] = (direction==DMA_TO_DEVICE ? 0x40 : 0xC0); command[cmdlen-7] = access | (direction==DMA_TO_DEVICE ? USBAT_CMD_COND_WRITE_BLOCK : USBAT_CMD_COND_READ_BLOCK); command[cmdlen-6] = data_reg; command[cmdlen-5] = status_reg; command[cmdlen-4] = timeout; command[cmdlen-3] = qualifier; command[cmdlen-2] = LSB_of(len); command[cmdlen-1] = MSB_of(len); result = usbat_execute_command(us, command, cmdlen); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (i==0) { for (j=0; j<num_registers; j++) { data[j<<1] = registers[j]; data[1+(j<<1)] = data_out[j]; } result = usbat_bulk_write(us, data, num_registers*2, 0); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; } result = usb_stor_bulk_transfer_sg(us, pipe, buf, len, use_sg, NULL); /* * If we get a stall on the bulk download, we'll retry * the bulk download -- but not the SCSI command because * in some sense the SCSI command is still 'active' and * waiting for the data. Don't ask me why this should be; * I'm only following what the Windoze driver did. * * Note that a stall for the test-and-read/write command means * that the test failed. In this case we're testing to make * sure that the device is error-free * (i.e. bit 0 -- CHK -- of status is 0). The most likely * hypothesis is that the USBAT chip somehow knows what * the device will accept, but doesn't give the device any * data until all data is received. Thus, the device would * still be waiting for the first byte of data if a stall * occurs, even if the stall implies that some data was * transferred. */ if (result == USB_STOR_XFER_SHORT || result == USB_STOR_XFER_STALLED) { /* * If we're reading and we stalled, then clear * the bulk output pipe only the first time. */ if (direction==DMA_FROM_DEVICE && i==0) { if (usb_stor_clear_halt(us, us->send_bulk_pipe) < 0) return USB_STOR_TRANSPORT_ERROR; } /* * Read status: is the device angry, or just busy? */ result = usbat_read(us, USBAT_ATA, direction==DMA_TO_DEVICE ? USBAT_ATA_STATUS : USBAT_ATA_ALTSTATUS, status); if (result!=USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (*status & 0x01) /* check condition */ return USB_STOR_TRANSPORT_FAILED; if (*status & 0x20) /* device fault */ return USB_STOR_TRANSPORT_FAILED; usb_stor_dbg(us, "Redoing %s\n", str_write_read(direction == DMA_TO_DEVICE)); } else if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; else return usbat_wait_not_busy(us, minutes); } usb_stor_dbg(us, "Bummer! %s bulk data 20 times failed\n", direction == DMA_TO_DEVICE ? "Writing" : "Reading"); return USB_STOR_TRANSPORT_FAILED; } /* * Write to multiple registers: * Allows us to write specific data to any registers. The data to be written * gets packed in this sequence: reg0, data0, reg1, data1, ..., regN, dataN * which gets sent through bulk out. * Not designed for large transfers of data! */ static int usbat_multiple_write(struct us_data *us, unsigned char *registers, unsigned char *data_out, unsigned short num_registers) { int i, result; unsigned char *data = us->iobuf; unsigned char *command = us->iobuf; BUG_ON(num_registers > US_IOBUF_SIZE/2); /* Write to multiple registers, ATA access */ command[0] = 0x40; command[1] = USBAT_ATA | USBAT_CMD_WRITE_REGS; /* No relevance */ command[2] = 0; command[3] = 0; command[4] = 0; command[5] = 0; /* Number of bytes to be transferred (incl. addresses and data) */ command[6] = LSB_of(num_registers*2); command[7] = MSB_of(num_registers*2); /* The setup command */ result = usbat_execute_command(us, command, 8); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; /* Create the reg/data, reg/data sequence */ for (i=0; i<num_registers; i++) { data[i<<1] = registers[i]; data[1+(i<<1)] = data_out[i]; } /* Send the data */ result = usbat_bulk_write(us, data, num_registers*2, 0); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_get_device_type(us) == USBAT_DEV_HP8200) return usbat_wait_not_busy(us, 0); else return USB_STOR_TRANSPORT_GOOD; } /* * Conditionally read blocks from device: * Allows us to read blocks from a specific data register, based upon the * condition that a status register can be successfully masked with a status * qualifier. If this condition is not initially met, the read will wait * up until a maximum amount of time has elapsed, as specified by timeout. * The read will start when the condition is met, otherwise the command aborts. * * The qualifier defined here is not the value that is masked, it defines * conditions for the write to take place. The actual masked qualifier (and * other related details) are defined beforehand with _set_shuttle_features(). */ static int usbat_read_blocks(struct us_data *us, void* buffer, int len, int use_sg) { int result; unsigned char *command = us->iobuf; command[0] = 0xC0; command[1] = USBAT_ATA | USBAT_CMD_COND_READ_BLOCK; command[2] = USBAT_ATA_DATA; command[3] = USBAT_ATA_STATUS; command[4] = 0xFD; /* Timeout (ms); */ command[5] = USBAT_QUAL_FCQ; command[6] = LSB_of(len); command[7] = MSB_of(len); /* Multiple block read setup command */ result = usbat_execute_command(us, command, 8); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_FAILED; /* Read the blocks we just asked for */ result = usbat_bulk_read(us, buffer, len, use_sg); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_FAILED; return USB_STOR_TRANSPORT_GOOD; } /* * Conditionally write blocks to device: * Allows us to write blocks to a specific data register, based upon the * condition that a status register can be successfully masked with a status * qualifier. If this condition is not initially met, the write will wait * up until a maximum amount of time has elapsed, as specified by timeout. * The read will start when the condition is met, otherwise the command aborts. * * The qualifier defined here is not the value that is masked, it defines * conditions for the write to take place. The actual masked qualifier (and * other related details) are defined beforehand with _set_shuttle_features(). */ static int usbat_write_blocks(struct us_data *us, void* buffer, int len, int use_sg) { int result; unsigned char *command = us->iobuf; command[0] = 0x40; command[1] = USBAT_ATA | USBAT_CMD_COND_WRITE_BLOCK; command[2] = USBAT_ATA_DATA; command[3] = USBAT_ATA_STATUS; command[4] = 0xFD; /* Timeout (ms) */ command[5] = USBAT_QUAL_FCQ; command[6] = LSB_of(len); command[7] = MSB_of(len); /* Multiple block write setup command */ result = usbat_execute_command(us, command, 8); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_FAILED; /* Write the data */ result = usbat_bulk_write(us, buffer, len, use_sg); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_FAILED; return USB_STOR_TRANSPORT_GOOD; } /* * Read the User IO register */ static int usbat_read_user_io(struct us_data *us, unsigned char *data_flags) { int result; result = usb_stor_ctrl_transfer(us, us->recv_ctrl_pipe, USBAT_CMD_UIO, 0xC0, 0, 0, data_flags, USBAT_UIO_READ); usb_stor_dbg(us, "UIO register reads %02X\n", *data_flags); return result; } /* * Write to the User IO register */ static int usbat_write_user_io(struct us_data *us, unsigned char enable_flags, unsigned char data_flags) { return usb_stor_ctrl_transfer(us, us->send_ctrl_pipe, USBAT_CMD_UIO, 0x40, short_pack(enable_flags, data_flags), 0, NULL, USBAT_UIO_WRITE); } /* * Reset the device * Often needed on media change. */ static int usbat_device_reset(struct us_data *us) { int rc; /* * Reset peripheral, enable peripheral control signals * (bring reset signal up) */ rc = usbat_write_user_io(us, USBAT_UIO_DRVRST | USBAT_UIO_OE1 | USBAT_UIO_OE0, USBAT_UIO_EPAD | USBAT_UIO_1); if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; /* * Enable peripheral control signals * (bring reset signal down) */ rc = usbat_write_user_io(us, USBAT_UIO_OE1 | USBAT_UIO_OE0, USBAT_UIO_EPAD | USBAT_UIO_1); if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; return USB_STOR_TRANSPORT_GOOD; } /* * Enable card detect */ static int usbat_device_enable_cdt(struct us_data *us) { int rc; /* Enable peripheral control signals and card detect */ rc = usbat_write_user_io(us, USBAT_UIO_ACKD | USBAT_UIO_OE1 | USBAT_UIO_OE0, USBAT_UIO_EPAD | USBAT_UIO_1); if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; return USB_STOR_TRANSPORT_GOOD; } /* * Determine if media is present. */ static int usbat_flash_check_media_present(struct us_data *us, unsigned char *uio) { if (*uio & USBAT_UIO_UI0) { usb_stor_dbg(us, "no media detected\n"); return USBAT_FLASH_MEDIA_NONE; } return USBAT_FLASH_MEDIA_CF; } /* * Determine if media has changed since last operation */ static int usbat_flash_check_media_changed(struct us_data *us, unsigned char *uio) { if (*uio & USBAT_UIO_0) { usb_stor_dbg(us, "media change detected\n"); return USBAT_FLASH_MEDIA_CHANGED; } return USBAT_FLASH_MEDIA_SAME; } /* * Check for media change / no media and handle the situation appropriately */ static int usbat_flash_check_media(struct us_data *us, struct usbat_info *info) { int rc; unsigned char *uio = us->iobuf; rc = usbat_read_user_io(us, uio); if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; /* Check for media existence */ rc = usbat_flash_check_media_present(us, uio); if (rc == USBAT_FLASH_MEDIA_NONE) { info->sense_key = 0x02; info->sense_asc = 0x3A; info->sense_ascq = 0x00; return USB_STOR_TRANSPORT_FAILED; } /* Check for media change */ rc = usbat_flash_check_media_changed(us, uio); if (rc == USBAT_FLASH_MEDIA_CHANGED) { /* Reset and re-enable card detect */ rc = usbat_device_reset(us); if (rc != USB_STOR_TRANSPORT_GOOD) return rc; rc = usbat_device_enable_cdt(us); if (rc != USB_STOR_TRANSPORT_GOOD) return rc; msleep(50); rc = usbat_read_user_io(us, uio); if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; info->sense_key = UNIT_ATTENTION; info->sense_asc = 0x28; info->sense_ascq = 0x00; return USB_STOR_TRANSPORT_FAILED; } return USB_STOR_TRANSPORT_GOOD; } /* * Determine whether we are controlling a flash-based reader/writer, * or a HP8200-based CD drive. * Sets transport functions as appropriate. */ static int usbat_identify_device(struct us_data *us, struct usbat_info *info) { int rc; unsigned char status; if (!us || !info) return USB_STOR_TRANSPORT_ERROR; rc = usbat_device_reset(us); if (rc != USB_STOR_TRANSPORT_GOOD) return rc; msleep(500); /* * In attempt to distinguish between HP CDRW's and Flash readers, we now * execute the IDENTIFY PACKET DEVICE command. On ATA devices (i.e. flash * readers), this command should fail with error. On ATAPI devices (i.e. * CDROM drives), it should succeed. */ rc = usbat_write(us, USBAT_ATA, USBAT_ATA_CMD, 0xA1); if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; rc = usbat_get_status(us, &status); if (rc != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; /* Check for error bit, or if the command 'fell through' */ if (status == 0xA1 || !(status & 0x01)) { /* Device is HP 8200 */ usb_stor_dbg(us, "Detected HP8200 CDRW\n"); info->devicetype = USBAT_DEV_HP8200; } else { /* Device is a CompactFlash reader/writer */ usb_stor_dbg(us, "Detected Flash reader/writer\n"); info->devicetype = USBAT_DEV_FLASH; } return USB_STOR_TRANSPORT_GOOD; } /* * Set the transport function based on the device type */ static int usbat_set_transport(struct us_data *us, struct usbat_info *info, int devicetype) { if (!info->devicetype) info->devicetype = devicetype; if (!info->devicetype) usbat_identify_device(us, info); switch (info->devicetype) { default: return USB_STOR_TRANSPORT_ERROR; case USBAT_DEV_HP8200: us->transport = usbat_hp8200e_transport; break; case USBAT_DEV_FLASH: us->transport = usbat_flash_transport; break; } return 0; } /* * Read the media capacity */ static int usbat_flash_get_sector_count(struct us_data *us, struct usbat_info *info) { unsigned char registers[3] = { USBAT_ATA_SECCNT, USBAT_ATA_DEVICE, USBAT_ATA_CMD, }; unsigned char command[3] = { 0x01, 0xA0, 0xEC }; unsigned char *reply; unsigned char status; int rc; if (!us || !info) return USB_STOR_TRANSPORT_ERROR; reply = kmalloc(512, GFP_NOIO); if (!reply) return USB_STOR_TRANSPORT_ERROR; /* ATA command : IDENTIFY DEVICE */ rc = usbat_multiple_write(us, registers, command, 3); if (rc != USB_STOR_XFER_GOOD) { usb_stor_dbg(us, "Gah! identify_device failed\n"); rc = USB_STOR_TRANSPORT_ERROR; goto leave; } /* Read device status */ if (usbat_get_status(us, &status) != USB_STOR_XFER_GOOD) { rc = USB_STOR_TRANSPORT_ERROR; goto leave; } msleep(100); /* Read the device identification data */ rc = usbat_read_block(us, reply, 512, 0); if (rc != USB_STOR_TRANSPORT_GOOD) goto leave; info->sectors = ((u32)(reply[117]) << 24) | ((u32)(reply[116]) << 16) | ((u32)(reply[115]) << 8) | ((u32)(reply[114]) ); rc = USB_STOR_TRANSPORT_GOOD; leave: kfree(reply); return rc; } /* * Read data from device */ static int usbat_flash_read_data(struct us_data *us, struct usbat_info *info, u32 sector, u32 sectors) { unsigned char registers[7] = { USBAT_ATA_FEATURES, USBAT_ATA_SECCNT, USBAT_ATA_SECNUM, USBAT_ATA_LBA_ME, USBAT_ATA_LBA_HI, USBAT_ATA_DEVICE, USBAT_ATA_STATUS, }; unsigned char command[7]; unsigned char *buffer; unsigned char thistime; unsigned int totallen, alloclen; int len, result; unsigned int sg_offset = 0; struct scatterlist *sg = NULL; result = usbat_flash_check_media(us, info); if (result != USB_STOR_TRANSPORT_GOOD) return result; /* * we're working in LBA mode. according to the ATA spec, * we can support up to 28-bit addressing. I don't know if Jumpshot * supports beyond 24-bit addressing. It's kind of hard to test * since it requires > 8GB CF card. */ if (sector > 0x0FFFFFFF) return USB_STOR_TRANSPORT_ERROR; totallen = sectors * info->ssize; /* * Since we don't read more than 64 KB at a time, we have to create * a bounce buffer and move the data a piece at a time between the * bounce buffer and the actual transfer buffer. */ alloclen = min(totallen, 65536u); buffer = kmalloc(alloclen, GFP_NOIO); if (buffer == NULL) return USB_STOR_TRANSPORT_ERROR; do { /* * loop, never allocate or transfer more than 64k at once * (min(128k, 255*info->ssize) is the real limit) */ len = min(totallen, alloclen); thistime = (len / info->ssize) & 0xff; /* ATA command 0x20 (READ SECTORS) */ usbat_pack_ata_sector_cmd(command, thistime, sector, 0x20); /* Write/execute ATA read command */ result = usbat_multiple_write(us, registers, command, 7); if (result != USB_STOR_TRANSPORT_GOOD) goto leave; /* Read the data we just requested */ result = usbat_read_blocks(us, buffer, len, 0); if (result != USB_STOR_TRANSPORT_GOOD) goto leave; usb_stor_dbg(us, "%d bytes\n", len); /* Store the data in the transfer buffer */ usb_stor_access_xfer_buf(buffer, len, us->srb, &sg, &sg_offset, TO_XFER_BUF); sector += thistime; totallen -= len; } while (totallen > 0); kfree(buffer); return USB_STOR_TRANSPORT_GOOD; leave: kfree(buffer); return USB_STOR_TRANSPORT_ERROR; } /* * Write data to device */ static int usbat_flash_write_data(struct us_data *us, struct usbat_info *info, u32 sector, u32 sectors) { unsigned char registers[7] = { USBAT_ATA_FEATURES, USBAT_ATA_SECCNT, USBAT_ATA_SECNUM, USBAT_ATA_LBA_ME, USBAT_ATA_LBA_HI, USBAT_ATA_DEVICE, USBAT_ATA_STATUS, }; unsigned char command[7]; unsigned char *buffer; unsigned char thistime; unsigned int totallen, alloclen; int len, result; unsigned int sg_offset = 0; struct scatterlist *sg = NULL; result = usbat_flash_check_media(us, info); if (result != USB_STOR_TRANSPORT_GOOD) return result; /* * we're working in LBA mode. according to the ATA spec, * we can support up to 28-bit addressing. I don't know if the device * supports beyond 24-bit addressing. It's kind of hard to test * since it requires > 8GB media. */ if (sector > 0x0FFFFFFF) return USB_STOR_TRANSPORT_ERROR; totallen = sectors * info->ssize; /* * Since we don't write more than 64 KB at a time, we have to create * a bounce buffer and move the data a piece at a time between the * bounce buffer and the actual transfer buffer. */ alloclen = min(totallen, 65536u); buffer = kmalloc(alloclen, GFP_NOIO); if (buffer == NULL) return USB_STOR_TRANSPORT_ERROR; do { /* * loop, never allocate or transfer more than 64k at once * (min(128k, 255*info->ssize) is the real limit) */ len = min(totallen, alloclen); thistime = (len / info->ssize) & 0xff; /* Get the data from the transfer buffer */ usb_stor_access_xfer_buf(buffer, len, us->srb, &sg, &sg_offset, FROM_XFER_BUF); /* ATA command 0x30 (WRITE SECTORS) */ usbat_pack_ata_sector_cmd(command, thistime, sector, 0x30); /* Write/execute ATA write command */ result = usbat_multiple_write(us, registers, command, 7); if (result != USB_STOR_TRANSPORT_GOOD) goto leave; /* Write the data */ result = usbat_write_blocks(us, buffer, len, 0); if (result != USB_STOR_TRANSPORT_GOOD) goto leave; sector += thistime; totallen -= len; } while (totallen > 0); kfree(buffer); return result; leave: kfree(buffer); return USB_STOR_TRANSPORT_ERROR; } /* * Squeeze a potentially huge (> 65535 byte) read10 command into * a little ( <= 65535 byte) ATAPI pipe */ static int usbat_hp8200e_handle_read10(struct us_data *us, unsigned char *registers, unsigned char *data, struct scsi_cmnd *srb) { int result = USB_STOR_TRANSPORT_GOOD; unsigned char *buffer; unsigned int len; unsigned int sector; unsigned int sg_offset = 0; struct scatterlist *sg = NULL; usb_stor_dbg(us, "transfersize %d\n", srb->transfersize); if (scsi_bufflen(srb) < 0x10000) { result = usbat_hp8200e_rw_block_test(us, USBAT_ATA, registers, data, 19, USBAT_ATA_DATA, USBAT_ATA_STATUS, 0xFD, (USBAT_QUAL_FCQ | USBAT_QUAL_ALQ), DMA_FROM_DEVICE, scsi_sglist(srb), scsi_bufflen(srb), scsi_sg_count(srb), 1); return result; } /* * Since we're requesting more data than we can handle in * a single read command (max is 64k-1), we will perform * multiple reads, but each read must be in multiples of * a sector. Luckily the sector size is in srb->transfersize * (see linux/drivers/scsi/sr.c). */ if (data[7+0] == GPCMD_READ_CD) { len = short_pack(data[7+9], data[7+8]); len <<= 16; len |= data[7+7]; usb_stor_dbg(us, "GPCMD_READ_CD: len %d\n", len); srb->transfersize = scsi_bufflen(srb)/len; } if (!srb->transfersize) { srb->transfersize = 2048; /* A guess */ usb_stor_dbg(us, "transfersize 0, forcing %d\n", srb->transfersize); } /* * Since we only read in one block at a time, we have to create * a bounce buffer and move the data a piece at a time between the * bounce buffer and the actual transfer buffer. */ len = (65535/srb->transfersize) * srb->transfersize; usb_stor_dbg(us, "Max read is %d bytes\n", len); len = min(len, scsi_bufflen(srb)); buffer = kmalloc(len, GFP_NOIO); if (buffer == NULL) /* bloody hell! */ return USB_STOR_TRANSPORT_FAILED; sector = short_pack(data[7+3], data[7+2]); sector <<= 16; sector |= short_pack(data[7+5], data[7+4]); transferred = 0; while (transferred != scsi_bufflen(srb)) { if (len > scsi_bufflen(srb) - transferred) len = scsi_bufflen(srb) - transferred; data[3] = len&0xFF; /* (cylL) = expected length (L) */ data[4] = (len>>8)&0xFF; /* (cylH) = expected length (H) */ /* Fix up the SCSI command sector and num sectors */ data[7+2] = MSB_of(sector>>16); /* SCSI command sector */ data[7+3] = LSB_of(sector>>16); data[7+4] = MSB_of(sector&0xFFFF); data[7+5] = LSB_of(sector&0xFFFF); if (data[7+0] == GPCMD_READ_CD) data[7+6] = 0; data[7+7] = MSB_of(len / srb->transfersize); /* SCSI command */ data[7+8] = LSB_of(len / srb->transfersize); /* num sectors */ result = usbat_hp8200e_rw_block_test(us, USBAT_ATA, registers, data, 19, USBAT_ATA_DATA, USBAT_ATA_STATUS, 0xFD, (USBAT_QUAL_FCQ | USBAT_QUAL_ALQ), DMA_FROM_DEVICE, buffer, len, 0, 1); if (result != USB_STOR_TRANSPORT_GOOD) break; /* Store the data in the transfer buffer */ usb_stor_access_xfer_buf(buffer, len, srb, &sg, &sg_offset, TO_XFER_BUF); /* Update the amount transferred and the sector number */ transferred += len; sector += len / srb->transfersize; } /* while transferred != scsi_bufflen(srb) */ kfree(buffer); return result; } static int usbat_select_and_test_registers(struct us_data *us) { int selector; unsigned char *status = us->iobuf; /* try device = master, then device = slave. */ for (selector = 0xA0; selector <= 0xB0; selector += 0x10) { if (usbat_write(us, USBAT_ATA, USBAT_ATA_DEVICE, selector) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_read(us, USBAT_ATA, USBAT_ATA_STATUS, status) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_read(us, USBAT_ATA, USBAT_ATA_DEVICE, status) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_read(us, USBAT_ATA, USBAT_ATA_LBA_ME, status) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_read(us, USBAT_ATA, USBAT_ATA_LBA_HI, status) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_write(us, USBAT_ATA, USBAT_ATA_LBA_ME, 0x55) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_write(us, USBAT_ATA, USBAT_ATA_LBA_HI, 0xAA) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_read(us, USBAT_ATA, USBAT_ATA_LBA_ME, status) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (usbat_read(us, USBAT_ATA, USBAT_ATA_LBA_ME, status) != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; } return USB_STOR_TRANSPORT_GOOD; } /* * Initialize the USBAT processor and the storage device */ static int init_usbat(struct us_data *us, int devicetype) { int rc; struct usbat_info *info; unsigned char subcountH = USBAT_ATA_LBA_HI; unsigned char subcountL = USBAT_ATA_LBA_ME; unsigned char *status = us->iobuf; us->extra = kzalloc(sizeof(struct usbat_info), GFP_NOIO); if (!us->extra) return -ENOMEM; info = (struct usbat_info *) (us->extra); /* Enable peripheral control signals */ rc = usbat_write_user_io(us, USBAT_UIO_OE1 | USBAT_UIO_OE0, USBAT_UIO_EPAD | USBAT_UIO_1); if (rc != USB_STOR_XFER_GOOD) return -EIO; usb_stor_dbg(us, "INIT 1\n"); msleep(2000); rc = usbat_read_user_io(us, status); if (rc != USB_STOR_TRANSPORT_GOOD) return -EIO; usb_stor_dbg(us, "INIT 2\n"); rc = usbat_read_user_io(us, status); if (rc != USB_STOR_XFER_GOOD) return -EIO; rc = usbat_read_user_io(us, status); if (rc != USB_STOR_XFER_GOOD) return -EIO; usb_stor_dbg(us, "INIT 3\n"); rc = usbat_select_and_test_registers(us); if (rc != USB_STOR_TRANSPORT_GOOD) return -EIO; usb_stor_dbg(us, "INIT 4\n"); rc = usbat_read_user_io(us, status); if (rc != USB_STOR_XFER_GOOD) return -EIO; usb_stor_dbg(us, "INIT 5\n"); /* Enable peripheral control signals and card detect */ rc = usbat_device_enable_cdt(us); if (rc != USB_STOR_TRANSPORT_GOOD) return -EIO; usb_stor_dbg(us, "INIT 6\n"); rc = usbat_read_user_io(us, status); if (rc != USB_STOR_XFER_GOOD) return -EIO; usb_stor_dbg(us, "INIT 7\n"); msleep(1400); rc = usbat_read_user_io(us, status); if (rc != USB_STOR_XFER_GOOD) return -EIO; usb_stor_dbg(us, "INIT 8\n"); rc = usbat_select_and_test_registers(us); if (rc != USB_STOR_TRANSPORT_GOOD) return -EIO; usb_stor_dbg(us, "INIT 9\n"); /* At this point, we need to detect which device we are using */ if (usbat_set_transport(us, info, devicetype)) return -EIO; usb_stor_dbg(us, "INIT 10\n"); if (usbat_get_device_type(us) == USBAT_DEV_FLASH) { subcountH = 0x02; subcountL = 0x00; } rc = usbat_set_shuttle_features(us, (USBAT_FEAT_ETEN | USBAT_FEAT_ET2 | USBAT_FEAT_ET1), 0x00, 0x88, 0x08, subcountH, subcountL); if (rc != USB_STOR_XFER_GOOD) return -EIO; usb_stor_dbg(us, "INIT 11\n"); return 0; } /* * Transport for the HP 8200e */ static int usbat_hp8200e_transport(struct scsi_cmnd *srb, struct us_data *us) { int result; unsigned char *status = us->iobuf; unsigned char registers[32]; unsigned char data[32]; unsigned int len; int i; len = scsi_bufflen(srb); /* * Send A0 (ATA PACKET COMMAND). * Note: I guess we're never going to get any of the ATA * commands... just ATA Packet Commands. */ registers[0] = USBAT_ATA_FEATURES; registers[1] = USBAT_ATA_SECCNT; registers[2] = USBAT_ATA_SECNUM; registers[3] = USBAT_ATA_LBA_ME; registers[4] = USBAT_ATA_LBA_HI; registers[5] = USBAT_ATA_DEVICE; registers[6] = USBAT_ATA_CMD; data[0] = 0x00; data[1] = 0x00; data[2] = 0x00; data[3] = len&0xFF; /* (cylL) = expected length (L) */ data[4] = (len>>8)&0xFF; /* (cylH) = expected length (H) */ data[5] = 0xB0; /* (device sel) = slave */ data[6] = 0xA0; /* (command) = ATA PACKET COMMAND */ for (i=7; i<19; i++) { registers[i] = 0x10; data[i] = (i-7 >= srb->cmd_len) ? 0 : srb->cmnd[i-7]; } result = usbat_get_status(us, status); usb_stor_dbg(us, "Status = %02X\n", *status); if (result != USB_STOR_XFER_GOOD) return USB_STOR_TRANSPORT_ERROR; if (srb->cmnd[0] == TEST_UNIT_READY) transferred = 0; if (srb->sc_data_direction == DMA_TO_DEVICE) { result = usbat_hp8200e_rw_block_test(us, USBAT_ATA, registers, data, 19, USBAT_ATA_DATA, USBAT_ATA_STATUS, 0xFD, (USBAT_QUAL_FCQ | USBAT_QUAL_ALQ), DMA_TO_DEVICE, scsi_sglist(srb), len, scsi_sg_count(srb), 10); if (result == USB_STOR_TRANSPORT_GOOD) { transferred += len; usb_stor_dbg(us, "Wrote %08X bytes\n", transferred); } return result; } else if (srb->cmnd[0] == READ_10 || srb->cmnd[0] == GPCMD_READ_CD) { return usbat_hp8200e_handle_read10(us, registers, data, srb); } if (len > 0xFFFF) { usb_stor_dbg(us, "Error: len = %08X... what do I do now?\n", len); return USB_STOR_TRANSPORT_ERROR; } result = usbat_multiple_write(us, registers, data, 7); if (result != USB_STOR_TRANSPORT_GOOD) return result; /* * Write the 12-byte command header. * * If the command is BLANK then set the timer for 75 minutes. * Otherwise set it for 10 minutes. * * NOTE: THE 8200 DOCUMENTATION STATES THAT BLANKING A CDRW * AT SPEED 4 IS UNRELIABLE!!! */ result = usbat_write_block(us, USBAT_ATA, srb->cmnd, 12, srb->cmnd[0] == GPCMD_BLANK ? 75 : 10, 0); if (result != USB_STOR_TRANSPORT_GOOD) return result; /* If there is response data to be read in then do it here. */ if (len != 0 && (srb->sc_data_direction == DMA_FROM_DEVICE)) { /* How many bytes to read in? Check cylL register */ if (usbat_read(us, USBAT_ATA, USBAT_ATA_LBA_ME, status) != USB_STOR_XFER_GOOD) { return USB_STOR_TRANSPORT_ERROR; } if (len > 0xFF) { /* need to read cylH also */ len = *status; if (usbat_read(us, USBAT_ATA, USBAT_ATA_LBA_HI, status) != USB_STOR_XFER_GOOD) { return USB_STOR_TRANSPORT_ERROR; } len += ((unsigned int) *status)<<8; } else len = *status; result = usbat_read_block(us, scsi_sglist(srb), len, scsi_sg_count(srb)); } return result; } /* * Transport for USBAT02-based CompactFlash and similar storage devices */ static int usbat_flash_transport(struct scsi_cmnd * srb, struct us_data *us) { int rc; struct usbat_info *info = (struct usbat_info *) (us->extra); unsigned long block, blocks; unsigned char *ptr = us->iobuf; static const unsigned char inquiry_response[36] = { 0x00, 0x80, 0x00, 0x01, 0x1F, 0x00, 0x00, 0x00 }; if (srb->cmnd[0] == INQUIRY) { usb_stor_dbg(us, "INQUIRY - Returning bogus response\n"); memcpy(ptr, inquiry_response, sizeof(inquiry_response)); fill_inquiry_response(us, ptr, 36); return USB_STOR_TRANSPORT_GOOD; } if (srb->cmnd[0] == READ_CAPACITY) { rc = usbat_flash_check_media(us, info); if (rc != USB_STOR_TRANSPORT_GOOD) return rc; rc = usbat_flash_get_sector_count(us, info); if (rc != USB_STOR_TRANSPORT_GOOD) return rc; /* hard coded 512 byte sectors as per ATA spec */ info->ssize = 0x200; usb_stor_dbg(us, "READ_CAPACITY: %ld sectors, %ld bytes per sector\n", info->sectors, info->ssize); /* * build the reply * note: must return the sector number of the last sector, * *not* the total number of sectors */ ((__be32 *) ptr)[0] = cpu_to_be32(info->sectors - 1); ((__be32 *) ptr)[1] = cpu_to_be32(info->ssize); usb_stor_set_xfer_buf(ptr, 8, srb); return USB_STOR_TRANSPORT_GOOD; } if (srb->cmnd[0] == MODE_SELECT_10) { usb_stor_dbg(us, "Gah! MODE_SELECT_10\n"); return USB_STOR_TRANSPORT_ERROR; } if (srb->cmnd[0] == READ_10) { block = ((u32)(srb->cmnd[2]) << 24) | ((u32)(srb->cmnd[3]) << 16) | ((u32)(srb->cmnd[4]) << 8) | ((u32)(srb->cmnd[5])); blocks = ((u32)(srb->cmnd[7]) << 8) | ((u32)(srb->cmnd[8])); usb_stor_dbg(us, "READ_10: read block 0x%04lx count %ld\n", block, blocks); return usbat_flash_read_data(us, info, block, blocks); } if (srb->cmnd[0] == READ_12) { /* * I don't think we'll ever see a READ_12 but support it anyway */ block = ((u32)(srb->cmnd[2]) << 24) | ((u32)(srb->cmnd[3]) << 16) | ((u32)(srb->cmnd[4]) << 8) | ((u32)(srb->cmnd[5])); blocks = ((u32)(srb->cmnd[6]) << 24) | ((u32)(srb->cmnd[7]) << 16) | ((u32)(srb->cmnd[8]) << 8) | ((u32)(srb->cmnd[9])); usb_stor_dbg(us, "READ_12: read block 0x%04lx count %ld\n", block, blocks); return usbat_flash_read_data(us, info, block, blocks); } if (srb->cmnd[0] == WRITE_10) { block = ((u32)(srb->cmnd[2]) << 24) | ((u32)(srb->cmnd[3]) << 16) | ((u32)(srb->cmnd[4]) << 8) | ((u32)(srb->cmnd[5])); blocks = ((u32)(srb->cmnd[7]) << 8) | ((u32)(srb->cmnd[8])); usb_stor_dbg(us, "WRITE_10: write block 0x%04lx count %ld\n", block, blocks); return usbat_flash_write_data(us, info, block, blocks); } if (srb->cmnd[0] == WRITE_12) { /* * I don't think we'll ever see a WRITE_12 but support it anyway */ block = ((u32)(srb->cmnd[2]) << 24) | ((u32)(srb->cmnd[3]) << 16) | ((u32)(srb->cmnd[4]) << 8) | ((u32)(srb->cmnd[5])); blocks = ((u32)(srb->cmnd[6]) << 24) | ((u32)(srb->cmnd[7]) << 16) | ((u32)(srb->cmnd[8]) << 8) | ((u32)(srb->cmnd[9])); usb_stor_dbg(us, "WRITE_12: write block 0x%04lx count %ld\n", block, blocks); return usbat_flash_write_data(us, info, block, blocks); } if (srb->cmnd[0] == TEST_UNIT_READY) { usb_stor_dbg(us, "TEST_UNIT_READY\n"); rc = usbat_flash_check_media(us, info); if (rc != USB_STOR_TRANSPORT_GOOD) return rc; return usbat_check_status(us); } if (srb->cmnd[0] == REQUEST_SENSE) { usb_stor_dbg(us, "REQUEST_SENSE\n"); memset(ptr, 0, 18); ptr[0] = 0xF0; ptr[2] = info->sense_key; ptr[7] = 11; ptr[12] = info->sense_asc; ptr[13] = info->sense_ascq; usb_stor_set_xfer_buf(ptr, 18, srb); return USB_STOR_TRANSPORT_GOOD; } if (srb->cmnd[0] == ALLOW_MEDIUM_REMOVAL) { /* * sure. whatever. not like we can stop the user from popping * the media out of the device (no locking doors, etc) */ return USB_STOR_TRANSPORT_GOOD; } usb_stor_dbg(us, "Gah! Unknown command: %d (0x%x)\n", srb->cmnd[0], srb->cmnd[0]); info->sense_key = 0x05; info->sense_asc = 0x20; info->sense_ascq = 0x00; return USB_STOR_TRANSPORT_FAILED; } static int init_usbat_cd(struct us_data *us) { return init_usbat(us, USBAT_DEV_HP8200); } static int init_usbat_flash(struct us_data *us) { return init_usbat(us, USBAT_DEV_FLASH); } static struct scsi_host_template usbat_host_template; static int usbat_probe(struct usb_interface *intf, const struct usb_device_id *id) { struct us_data *us; int result; result = usb_stor_probe1(&us, intf, id, (id - usbat_usb_ids) + usbat_unusual_dev_list, &usbat_host_template); if (result) return result; /* * The actual transport will be determined later by the * initialization routine; this is just a placeholder. */ us->transport_name = "Shuttle USBAT"; us->transport = usbat_flash_transport; us->transport_reset = usb_stor_CB_reset; us->max_lun = 0; result = usb_stor_probe2(us); return result; } static struct usb_driver usbat_driver = { .name = DRV_NAME, .probe = usbat_probe, .disconnect = usb_stor_disconnect, .suspend = usb_stor_suspend, .resume = usb_stor_resume, .reset_resume = usb_stor_reset_resume, .pre_reset = usb_stor_pre_reset, .post_reset = usb_stor_post_reset, .id_table = usbat_usb_ids, .soft_unbind = 1, .no_dynamic_id = 1, }; module_usb_stor_driver(usbat_driver, usbat_host_template, DRV_NAME); |
| 15 15 15 34 34 34 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 | // SPDX-License-Identifier: GPL-2.0-only /* * LCD Lowlevel Control Abstraction * * Copyright (C) 2003,2004 Hewlett-Packard Company * */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/module.h> #include <linux/init.h> #include <linux/device.h> #include <linux/lcd.h> #include <linux/notifier.h> #include <linux/ctype.h> #include <linux/err.h> #include <linux/slab.h> static DEFINE_MUTEX(lcd_dev_list_mutex); static LIST_HEAD(lcd_dev_list); static void lcd_notify_blank(struct lcd_device *ld, struct device *display_dev, int power) { guard(mutex)(&ld->ops_lock); if (!ld->ops || !ld->ops->set_power) return; if (ld->ops->controls_device && !ld->ops->controls_device(ld, display_dev)) return; ld->ops->set_power(ld, power); } void lcd_notify_blank_all(struct device *display_dev, int power) { struct lcd_device *ld; guard(mutex)(&lcd_dev_list_mutex); list_for_each_entry(ld, &lcd_dev_list, entry) lcd_notify_blank(ld, display_dev, power); } EXPORT_SYMBOL(lcd_notify_blank_all); static void lcd_notify_mode_change(struct lcd_device *ld, struct device *display_dev, unsigned int width, unsigned int height) { guard(mutex)(&ld->ops_lock); if (!ld->ops || !ld->ops->set_mode) return; if (ld->ops->controls_device && !ld->ops->controls_device(ld, display_dev)) return; ld->ops->set_mode(ld, width, height); } void lcd_notify_mode_change_all(struct device *display_dev, unsigned int width, unsigned int height) { struct lcd_device *ld; guard(mutex)(&lcd_dev_list_mutex); list_for_each_entry(ld, &lcd_dev_list, entry) lcd_notify_mode_change(ld, display_dev, width, height); } EXPORT_SYMBOL(lcd_notify_mode_change_all); static ssize_t lcd_power_show(struct device *dev, struct device_attribute *attr, char *buf) { int rc; struct lcd_device *ld = to_lcd_device(dev); mutex_lock(&ld->ops_lock); if (ld->ops && ld->ops->get_power) rc = sprintf(buf, "%d\n", ld->ops->get_power(ld)); else rc = -ENXIO; mutex_unlock(&ld->ops_lock); return rc; } static ssize_t lcd_power_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { int rc; struct lcd_device *ld = to_lcd_device(dev); unsigned long power; rc = kstrtoul(buf, 0, &power); if (rc) return rc; rc = -ENXIO; mutex_lock(&ld->ops_lock); if (ld->ops && ld->ops->set_power) { pr_debug("set power to %lu\n", power); ld->ops->set_power(ld, power); rc = count; } mutex_unlock(&ld->ops_lock); return rc; } static DEVICE_ATTR_RW(lcd_power); static ssize_t contrast_show(struct device *dev, struct device_attribute *attr, char *buf) { int rc = -ENXIO; struct lcd_device *ld = to_lcd_device(dev); mutex_lock(&ld->ops_lock); if (ld->ops && ld->ops->get_contrast) rc = sprintf(buf, "%d\n", ld->ops->get_contrast(ld)); mutex_unlock(&ld->ops_lock); return rc; } static ssize_t contrast_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { int rc; struct lcd_device *ld = to_lcd_device(dev); unsigned long contrast; rc = kstrtoul(buf, 0, &contrast); if (rc) return rc; rc = -ENXIO; mutex_lock(&ld->ops_lock); if (ld->ops && ld->ops->set_contrast) { pr_debug("set contrast to %lu\n", contrast); ld->ops->set_contrast(ld, contrast); rc = count; } mutex_unlock(&ld->ops_lock); return rc; } static DEVICE_ATTR_RW(contrast); static ssize_t max_contrast_show(struct device *dev, struct device_attribute *attr, char *buf) { struct lcd_device *ld = to_lcd_device(dev); return sprintf(buf, "%d\n", ld->props.max_contrast); } static DEVICE_ATTR_RO(max_contrast); static void lcd_device_release(struct device *dev) { struct lcd_device *ld = to_lcd_device(dev); kfree(ld); } static struct attribute *lcd_device_attrs[] = { &dev_attr_lcd_power.attr, &dev_attr_contrast.attr, &dev_attr_max_contrast.attr, NULL, }; ATTRIBUTE_GROUPS(lcd_device); static const struct class lcd_class = { .name = "lcd", .dev_groups = lcd_device_groups, }; /** * lcd_device_register - register a new object of lcd_device class. * @name: the name of the new object(must be the same as the name of the * respective framebuffer device). * @parent: pointer to the parent's struct device . * @devdata: an optional pointer to be stored in the device. The * methods may retrieve it by using lcd_get_data(ld). * @ops: the lcd operations structure. * * Creates and registers a new lcd device. Returns either an ERR_PTR() * or a pointer to the newly allocated device. */ struct lcd_device *lcd_device_register(const char *name, struct device *parent, void *devdata, const struct lcd_ops *ops) { struct lcd_device *new_ld; int rc; pr_debug("lcd_device_register: name=%s\n", name); new_ld = kzalloc(sizeof(struct lcd_device), GFP_KERNEL); if (!new_ld) return ERR_PTR(-ENOMEM); mutex_init(&new_ld->ops_lock); mutex_init(&new_ld->update_lock); new_ld->dev.class = &lcd_class; new_ld->dev.parent = parent; new_ld->dev.release = lcd_device_release; dev_set_name(&new_ld->dev, "%s", name); dev_set_drvdata(&new_ld->dev, devdata); new_ld->ops = ops; rc = device_register(&new_ld->dev); if (rc) { put_device(&new_ld->dev); return ERR_PTR(rc); } guard(mutex)(&lcd_dev_list_mutex); list_add(&new_ld->entry, &lcd_dev_list); return new_ld; } EXPORT_SYMBOL(lcd_device_register); /** * lcd_device_unregister - unregisters a object of lcd_device class. * @ld: the lcd device object to be unregistered and freed. * * Unregisters a previously registered via lcd_device_register object. */ void lcd_device_unregister(struct lcd_device *ld) { if (!ld) return; guard(mutex)(&lcd_dev_list_mutex); list_del(&ld->entry); mutex_lock(&ld->ops_lock); ld->ops = NULL; mutex_unlock(&ld->ops_lock); device_unregister(&ld->dev); } EXPORT_SYMBOL(lcd_device_unregister); static void devm_lcd_device_release(struct device *dev, void *res) { struct lcd_device *lcd = *(struct lcd_device **)res; lcd_device_unregister(lcd); } static int devm_lcd_device_match(struct device *dev, void *res, void *data) { struct lcd_device **r = res; return *r == data; } /** * devm_lcd_device_register - resource managed lcd_device_register() * @dev: the device to register * @name: the name of the device * @parent: a pointer to the parent device * @devdata: an optional pointer to be stored for private driver use * @ops: the lcd operations structure * * @return a struct lcd on success, or an ERR_PTR on error * * Managed lcd_device_register(). The lcd_device returned from this function * are automatically freed on driver detach. See lcd_device_register() * for more information. */ struct lcd_device *devm_lcd_device_register(struct device *dev, const char *name, struct device *parent, void *devdata, const struct lcd_ops *ops) { struct lcd_device **ptr, *lcd; ptr = devres_alloc(devm_lcd_device_release, sizeof(*ptr), GFP_KERNEL); if (!ptr) return ERR_PTR(-ENOMEM); lcd = lcd_device_register(name, parent, devdata, ops); if (!IS_ERR(lcd)) { *ptr = lcd; devres_add(dev, ptr); } else { devres_free(ptr); } return lcd; } EXPORT_SYMBOL(devm_lcd_device_register); /** * devm_lcd_device_unregister - resource managed lcd_device_unregister() * @dev: the device to unregister * @ld: the lcd device to unregister * * Deallocated a lcd allocated with devm_lcd_device_register(). Normally * this function will not need to be called and the resource management * code will ensure that the resource is freed. */ void devm_lcd_device_unregister(struct device *dev, struct lcd_device *ld) { int rc; rc = devres_release(dev, devm_lcd_device_release, devm_lcd_device_match, ld); WARN_ON(rc); } EXPORT_SYMBOL(devm_lcd_device_unregister); static void __exit lcd_class_exit(void) { class_unregister(&lcd_class); } static int __init lcd_class_init(void) { int ret; ret = class_register(&lcd_class); if (ret) { pr_warn("Unable to create backlight class; errno = %d\n", ret); return ret; } return 0; } /* * if this is compiled into the kernel, we need to ensure that the * class is registered before users of the class try to register lcd's */ postcore_initcall(lcd_class_init); module_exit(lcd_class_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Jamey Hicks <jamey.hicks@hp.com>, Andrew Zabolotny <zap@homelink.ru>"); MODULE_DESCRIPTION("LCD Lowlevel Control Abstraction"); |
| 8 1 1 1 1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 | /* SPDX-License-Identifier: GPL-2.0-only */ /* * IEEE 802.11 EHT definitions * * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen * <jkmaline@cc.hut.fi> * Copyright (c) 2002-2003, Jouni Malinen <jkmaline@cc.hut.fi> * Copyright (c) 2005, Devicescape Software, Inc. * Copyright (c) 2006, Michael Wu <flamingice@sourmilk.net> * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH * Copyright (c) 2016 - 2017 Intel Deutschland GmbH * Copyright (c) 2018 - 2025 Intel Corporation */ #ifndef LINUX_IEEE80211_EHT_H #define LINUX_IEEE80211_EHT_H #include <linux/types.h> #include <linux/if_ether.h> /* need HE definitions for the inlines here */ #include <linux/ieee80211-he.h> #define IEEE80211_TTLM_MAX_CNT 2 #define IEEE80211_TTLM_CONTROL_DIRECTION 0x03 #define IEEE80211_TTLM_CONTROL_DEF_LINK_MAP 0x04 #define IEEE80211_TTLM_CONTROL_SWITCH_TIME_PRESENT 0x08 #define IEEE80211_TTLM_CONTROL_EXPECTED_DUR_PRESENT 0x10 #define IEEE80211_TTLM_CONTROL_LINK_MAP_SIZE 0x20 #define IEEE80211_TTLM_DIRECTION_DOWN 0 #define IEEE80211_TTLM_DIRECTION_UP 1 #define IEEE80211_TTLM_DIRECTION_BOTH 2 /** * struct ieee80211_ttlm_elem - TID-To-Link Mapping element * * Defined in section 9.4.2.314 in P802.11be_D4 * * @control: the first part of control field * @optional: the second part of control field */ struct ieee80211_ttlm_elem { u8 control; u8 optional[]; } __packed; #define IEEE80211_EHT_MCS_NSS_RX 0x0f #define IEEE80211_EHT_MCS_NSS_TX 0xf0 /** * struct ieee80211_eht_mcs_nss_supp_20mhz_only - EHT 20MHz only station max * supported NSS for per MCS. * * For each field below, bits 0 - 3 indicate the maximal number of spatial * streams for Rx, and bits 4 - 7 indicate the maximal number of spatial streams * for Tx. * * @rx_tx_mcs7_max_nss: indicates the maximum number of spatial streams * supported for reception and the maximum number of spatial streams * supported for transmission for MCS 0 - 7. * @rx_tx_mcs9_max_nss: indicates the maximum number of spatial streams * supported for reception and the maximum number of spatial streams * supported for transmission for MCS 8 - 9. * @rx_tx_mcs11_max_nss: indicates the maximum number of spatial streams * supported for reception and the maximum number of spatial streams * supported for transmission for MCS 10 - 11. * @rx_tx_mcs13_max_nss: indicates the maximum number of spatial streams * supported for reception and the maximum number of spatial streams * supported for transmission for MCS 12 - 13. * @rx_tx_max_nss: array of the previous fields for easier loop access */ struct ieee80211_eht_mcs_nss_supp_20mhz_only { union { struct { u8 rx_tx_mcs7_max_nss; u8 rx_tx_mcs9_max_nss; u8 rx_tx_mcs11_max_nss; u8 rx_tx_mcs13_max_nss; }; u8 rx_tx_max_nss[4]; }; }; /** * struct ieee80211_eht_mcs_nss_supp_bw - EHT max supported NSS per MCS (except * 20MHz only stations). * * For each field below, bits 0 - 3 indicate the maximal number of spatial * streams for Rx, and bits 4 - 7 indicate the maximal number of spatial streams * for Tx. * * @rx_tx_mcs9_max_nss: indicates the maximum number of spatial streams * supported for reception and the maximum number of spatial streams * supported for transmission for MCS 0 - 9. * @rx_tx_mcs11_max_nss: indicates the maximum number of spatial streams * supported for reception and the maximum number of spatial streams * supported for transmission for MCS 10 - 11. * @rx_tx_mcs13_max_nss: indicates the maximum number of spatial streams * supported for reception and the maximum number of spatial streams * supported for transmission for MCS 12 - 13. * @rx_tx_max_nss: array of the previous fields for easier loop access */ struct ieee80211_eht_mcs_nss_supp_bw { union { struct { u8 rx_tx_mcs9_max_nss; u8 rx_tx_mcs11_max_nss; u8 rx_tx_mcs13_max_nss; }; u8 rx_tx_max_nss[3]; }; }; /** * struct ieee80211_eht_cap_elem_fixed - EHT capabilities fixed data * * This structure is the "EHT Capabilities element" fixed fields as * described in P802.11be_D2.0 section 9.4.2.313. * * @mac_cap_info: MAC capabilities, see IEEE80211_EHT_MAC_CAP* * @phy_cap_info: PHY capabilities, see IEEE80211_EHT_PHY_CAP* */ struct ieee80211_eht_cap_elem_fixed { u8 mac_cap_info[2]; u8 phy_cap_info[9]; } __packed; /** * struct ieee80211_eht_cap_elem - EHT capabilities element * @fixed: fixed parts, see &ieee80211_eht_cap_elem_fixed * @optional: optional parts */ struct ieee80211_eht_cap_elem { struct ieee80211_eht_cap_elem_fixed fixed; /* * Followed by: * Supported EHT-MCS And NSS Set field: 4, 3, 6 or 9 octets. * EHT PPE Thresholds field: variable length. */ u8 optional[]; } __packed; #define IEEE80211_EHT_OPER_INFO_PRESENT 0x01 #define IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT 0x02 #define IEEE80211_EHT_OPER_EHT_DEF_PE_DURATION 0x04 #define IEEE80211_EHT_OPER_GROUP_ADDRESSED_BU_IND_LIMIT 0x08 #define IEEE80211_EHT_OPER_GROUP_ADDRESSED_BU_IND_EXP_MASK 0x30 #define IEEE80211_EHT_OPER_MCS15_DISABLE 0x40 /** * struct ieee80211_eht_operation - eht operation element * * This structure is the "EHT Operation Element" fields as * described in P802.11be_D2.0 section 9.4.2.311 * * @params: EHT operation element parameters. See &IEEE80211_EHT_OPER_* * @basic_mcs_nss: indicates the EHT-MCSs for each number of spatial streams in * EHT PPDUs that are supported by all EHT STAs in the BSS in transmit and * receive. * @optional: optional parts */ struct ieee80211_eht_operation { u8 params; struct ieee80211_eht_mcs_nss_supp_20mhz_only basic_mcs_nss; u8 optional[]; } __packed; /** * struct ieee80211_eht_operation_info - eht operation information * * @control: EHT operation information control. * @ccfs0: defines a channel center frequency for a 20, 40, 80, 160, or 320 MHz * EHT BSS. * @ccfs1: defines a channel center frequency for a 160 or 320 MHz EHT BSS. * @optional: optional parts */ struct ieee80211_eht_operation_info { u8 control; u8 ccfs0; u8 ccfs1; u8 optional[]; } __packed; /* EHT MAC capabilities as defined in P802.11be_D2.0 section 9.4.2.313.2 */ #define IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS 0x01 #define IEEE80211_EHT_MAC_CAP0_OM_CONTROL 0x02 #define IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1 0x04 #define IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE2 0x08 #define IEEE80211_EHT_MAC_CAP0_RESTRICTED_TWT 0x10 #define IEEE80211_EHT_MAC_CAP0_SCS_TRAFFIC_DESC 0x20 #define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_MASK 0xc0 #define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_3895 0 #define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_7991 1 #define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_11454 2 #define IEEE80211_EHT_MAC_CAP1_MAX_AMPDU_LEN_MASK 0x01 #define IEEE80211_EHT_MAC_CAP1_EHT_TRS 0x02 #define IEEE80211_EHT_MAC_CAP1_TXOP_RET 0x04 #define IEEE80211_EHT_MAC_CAP1_TWO_BQRS 0x08 #define IEEE80211_EHT_MAC_CAP1_EHT_LINK_ADAPT_MASK 0x30 #define IEEE80211_EHT_MAC_CAP1_UNSOL_EPCS_PRIO_ACCESS 0x40 /* EHT PHY capabilities as defined in P802.11be_D2.0 section 9.4.2.313.3 */ #define IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ 0x02 #define IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ 0x04 #define IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI 0x08 #define IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO 0x10 #define IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER 0x20 #define IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE 0x40 /* EHT beamformee number of spatial streams <= 80MHz is split */ #define IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK 0x80 #define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK 0x03 #define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK 0x1c #define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_320MHZ_MASK 0xe0 #define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK 0x07 #define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK 0x38 /* EHT number of sounding dimensions for 320MHz is split */ #define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK 0xc0 #define IEEE80211_EHT_PHY_CAP3_SOUNDING_DIM_320MHZ_MASK 0x01 #define IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK 0x02 #define IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK 0x04 #define IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK 0x08 #define IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK 0x10 #define IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK 0x20 #define IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK 0x40 #define IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK 0x80 #define IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO 0x01 #define IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP 0x02 #define IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP 0x04 #define IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI 0x08 #define IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK 0xf0 #define IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK 0x01 #define IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP 0x02 #define IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP 0x04 #define IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT 0x08 #define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK 0x30 #define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_0US 0 #define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_8US 1 #define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_16US 2 #define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_20US 3 /* Maximum number of supported EHT LTF is split */ #define IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK 0xc0 #define IEEE80211_EHT_PHY_CAP5_SUPP_EXTRA_EHT_LTF 0x40 #define IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK 0x07 #define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_80MHZ 0x08 #define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_160MHZ 0x30 #define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_320MHZ 0x40 #define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK 0x78 #define IEEE80211_EHT_PHY_CAP6_EHT_DUP_6GHZ_SUPP 0x80 #define IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW 0x01 #define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ 0x02 #define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ 0x04 #define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_320MHZ 0x08 #define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ 0x10 #define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ 0x20 #define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ 0x40 #define IEEE80211_EHT_PHY_CAP7_TB_SOUNDING_FDBK_RATE_LIMIT 0x80 #define IEEE80211_EHT_PHY_CAP8_RX_1024QAM_WIDER_BW_DL_OFDMA 0x01 #define IEEE80211_EHT_PHY_CAP8_RX_4096QAM_WIDER_BW_DL_OFDMA 0x02 /* * EHT operation channel width as defined in P802.11be_D2.0 section 9.4.2.311 */ #define IEEE80211_EHT_OPER_CHAN_WIDTH 0x7 #define IEEE80211_EHT_OPER_CHAN_WIDTH_20MHZ 0 #define IEEE80211_EHT_OPER_CHAN_WIDTH_40MHZ 1 #define IEEE80211_EHT_OPER_CHAN_WIDTH_80MHZ 2 #define IEEE80211_EHT_OPER_CHAN_WIDTH_160MHZ 3 #define IEEE80211_EHT_OPER_CHAN_WIDTH_320MHZ 4 /* Calculate 802.11be EHT capabilities IE Tx/Rx EHT MCS NSS Support Field size */ static inline u8 ieee80211_eht_mcs_nss_size(const struct ieee80211_he_cap_elem *he_cap, const struct ieee80211_eht_cap_elem_fixed *eht_cap, bool from_ap) { u8 count = 0; /* on 2.4 GHz, if it supports 40 MHz, the result is 3 */ if (he_cap->phy_cap_info[0] & IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_IN_2G) return 3; /* on 2.4 GHz, these three bits are reserved, so should be 0 */ if (he_cap->phy_cap_info[0] & IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G) count += 3; if (he_cap->phy_cap_info[0] & IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G) count += 3; if (eht_cap->phy_cap_info[0] & IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ) count += 3; if (count) return count; return from_ap ? 3 : 4; } /* 802.11be EHT PPE Thresholds */ #define IEEE80211_EHT_PPE_THRES_NSS_POS 0 #define IEEE80211_EHT_PPE_THRES_NSS_MASK 0xf #define IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK 0x1f0 #define IEEE80211_EHT_PPE_THRES_INFO_PPET_SIZE 3 #define IEEE80211_EHT_PPE_THRES_INFO_HEADER_SIZE 9 /* * Calculate 802.11be EHT capabilities IE EHT field size */ static inline u8 ieee80211_eht_ppe_size(u16 ppe_thres_hdr, const u8 *phy_cap_info) { u32 n; if (!(phy_cap_info[5] & IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT)) return 0; n = hweight16(ppe_thres_hdr & IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK); n *= 1 + u16_get_bits(ppe_thres_hdr, IEEE80211_EHT_PPE_THRES_NSS_MASK); /* * Each pair is 6 bits, and we need to add the 9 "header" bits to the * total size. */ n = n * IEEE80211_EHT_PPE_THRES_INFO_PPET_SIZE * 2 + IEEE80211_EHT_PPE_THRES_INFO_HEADER_SIZE; return DIV_ROUND_UP(n, 8); } static inline bool ieee80211_eht_capa_size_ok(const u8 *he_capa, const u8 *data, u8 len, bool from_ap) { const struct ieee80211_eht_cap_elem_fixed *elem = (const void *)data; u8 needed = sizeof(struct ieee80211_eht_cap_elem_fixed); if (len < needed || !he_capa) return false; needed += ieee80211_eht_mcs_nss_size((const void *)he_capa, (const void *)data, from_ap); if (len < needed) return false; if (elem->phy_cap_info[5] & IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT) { u16 ppe_thres_hdr; if (len < needed + sizeof(ppe_thres_hdr)) return false; ppe_thres_hdr = get_unaligned_le16(data + needed); needed += ieee80211_eht_ppe_size(ppe_thres_hdr, elem->phy_cap_info); } return len >= needed; } static inline bool ieee80211_eht_oper_size_ok(const u8 *data, u8 len) { const struct ieee80211_eht_operation *elem = (const void *)data; u8 needed = sizeof(*elem); if (len < needed) return false; if (elem->params & IEEE80211_EHT_OPER_INFO_PRESENT) { needed += 3; if (elem->params & IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT) needed += 2; } return len >= needed; } /* must validate ieee80211_eht_oper_size_ok() first */ static inline u16 ieee80211_eht_oper_dis_subchan_bitmap(const struct ieee80211_eht_operation *eht_oper) { const struct ieee80211_eht_operation_info *info = (const void *)eht_oper->optional; if (!(eht_oper->params & IEEE80211_EHT_OPER_INFO_PRESENT)) return 0; if (!(eht_oper->params & IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT)) return 0; return get_unaligned_le16(info->optional); } #define IEEE80211_BW_IND_DIS_SUBCH_PRESENT BIT(1) struct ieee80211_bandwidth_indication { u8 params; struct ieee80211_eht_operation_info info; } __packed; static inline bool ieee80211_bandwidth_indication_size_ok(const u8 *data, u8 len) { const struct ieee80211_bandwidth_indication *bwi = (const void *)data; if (len < sizeof(*bwi)) return false; if (bwi->params & IEEE80211_BW_IND_DIS_SUBCH_PRESENT && len < sizeof(*bwi) + 2) return false; return true; } /* Protected EHT action codes */ enum ieee80211_protected_eht_actioncode { WLAN_PROTECTED_EHT_ACTION_TTLM_REQ = 0, WLAN_PROTECTED_EHT_ACTION_TTLM_RES = 1, WLAN_PROTECTED_EHT_ACTION_TTLM_TEARDOWN = 2, WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_REQ = 3, WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_RESP = 4, WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_TEARDOWN = 5, WLAN_PROTECTED_EHT_ACTION_EML_OP_MODE_NOTIF = 6, WLAN_PROTECTED_EHT_ACTION_LINK_RECOMMEND = 7, WLAN_PROTECTED_EHT_ACTION_ML_OP_UPDATE_REQ = 8, WLAN_PROTECTED_EHT_ACTION_ML_OP_UPDATE_RESP = 9, WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_NOTIF = 10, WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_REQ = 11, WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_RESP = 12, }; /* multi-link device */ #define IEEE80211_MLD_MAX_NUM_LINKS 15 #define IEEE80211_ML_CONTROL_TYPE 0x0007 #define IEEE80211_ML_CONTROL_TYPE_BASIC 0 #define IEEE80211_ML_CONTROL_TYPE_PREQ 1 #define IEEE80211_ML_CONTROL_TYPE_RECONF 2 #define IEEE80211_ML_CONTROL_TYPE_TDLS 3 #define IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS 4 #define IEEE80211_ML_CONTROL_PRESENCE_MASK 0xfff0 struct ieee80211_multi_link_elem { __le16 control; u8 variable[]; } __packed; #define IEEE80211_MLC_BASIC_PRES_LINK_ID 0x0010 #define IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT 0x0020 #define IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY 0x0040 #define IEEE80211_MLC_BASIC_PRES_EML_CAPA 0x0080 #define IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP 0x0100 #define IEEE80211_MLC_BASIC_PRES_MLD_ID 0x0200 #define IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP 0x0400 #define IEEE80211_MED_SYNC_DELAY_DURATION 0x00ff #define IEEE80211_MED_SYNC_DELAY_SYNC_OFDM_ED_THRESH 0x0f00 #define IEEE80211_MED_SYNC_DELAY_SYNC_MAX_NUM_TXOPS 0xf000 /* * Described in P802.11be_D3.0 * dot11MSDTimerDuration should default to 5484 (i.e. 171.375) * dot11MSDOFDMEDthreshold defaults to -72 (i.e. 0) * dot11MSDTXOPMAX defaults to 1 */ #define IEEE80211_MED_SYNC_DELAY_DEFAULT 0x10ac #define IEEE80211_EML_CAP_EMLSR_SUPP 0x0001 #define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY 0x000e #define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_0US 0 #define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_32US 1 #define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_64US 2 #define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_128US 3 #define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_256US 4 #define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY 0x0070 #define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_0US 0 #define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_16US 1 #define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_32US 2 #define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_64US 3 #define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_128US 4 #define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_256US 5 #define IEEE80211_EML_CAP_EMLMR_SUPPORT 0x0080 #define IEEE80211_EML_CAP_EMLMR_DELAY 0x0700 #define IEEE80211_EML_CAP_EMLMR_DELAY_0US 0 #define IEEE80211_EML_CAP_EMLMR_DELAY_32US 1 #define IEEE80211_EML_CAP_EMLMR_DELAY_64US 2 #define IEEE80211_EML_CAP_EMLMR_DELAY_128US 3 #define IEEE80211_EML_CAP_EMLMR_DELAY_256US 4 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT 0x7800 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_0 0 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128US 1 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_256US 2 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_512US 3 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_1TU 4 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_2TU 5 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_4TU 6 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_8TU 7 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_16TU 8 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_32TU 9 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_64TU 10 #define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128TU 11 #define IEEE80211_MLD_CAP_OP_MAX_SIMUL_LINKS 0x000f #define IEEE80211_MLD_CAP_OP_SRS_SUPPORT 0x0010 #define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP 0x0060 #define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_NO_SUPP 0 #define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP_SAME 1 #define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_RESERVED 2 #define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP_DIFF 3 #define IEEE80211_MLD_CAP_OP_FREQ_SEP_TYPE_IND 0x0f80 #define IEEE80211_MLD_CAP_OP_AAR_SUPPORT 0x1000 #define IEEE80211_MLD_CAP_OP_LINK_RECONF_SUPPORT 0x2000 #define IEEE80211_MLD_CAP_OP_ALIGNED_TWT_SUPPORT 0x4000 struct ieee80211_mle_basic_common_info { u8 len; u8 mld_mac_addr[ETH_ALEN]; u8 variable[]; } __packed; #define IEEE80211_MLC_PREQ_PRES_MLD_ID 0x0010 struct ieee80211_mle_preq_common_info { u8 len; u8 variable[]; } __packed; #define IEEE80211_MLC_RECONF_PRES_MLD_MAC_ADDR 0x0010 #define IEEE80211_MLC_RECONF_PRES_EML_CAPA 0x0020 #define IEEE80211_MLC_RECONF_PRES_MLD_CAPA_OP 0x0040 #define IEEE80211_MLC_RECONF_PRES_EXT_MLD_CAPA_OP 0x0080 /* no fixed fields in RECONF */ struct ieee80211_mle_tdls_common_info { u8 len; u8 ap_mld_mac_addr[ETH_ALEN]; } __packed; #define IEEE80211_MLC_PRIO_ACCESS_PRES_AP_MLD_MAC_ADDR 0x0010 #define IEEE80211_EML_CTRL_EMLSR_MODE BIT(0) #define IEEE80211_EML_CTRL_EMLMR_MODE BIT(1) #define IEEE80211_EML_CTRL_EMLSR_PARAM_UPDATE BIT(2) #define IEEE80211_EML_CTRL_INDEV_COEX_ACT BIT(3) #define IEEE80211_EML_EMLSR_PAD_DELAY 0x07 #define IEEE80211_EML_EMLSR_TRANS_DELAY 0x38 #define IEEE80211_EML_EMLMR_RX_MCS_MAP 0xf0 #define IEEE80211_EML_EMLMR_TX_MCS_MAP 0x0f /* no fixed fields in PRIO_ACCESS */ /** * ieee80211_mle_common_size - check multi-link element common size * @data: multi-link element, must already be checked for size using * ieee80211_mle_size_ok() * Return: the size of the multi-link element's "common" subfield */ static inline u8 ieee80211_mle_common_size(const u8 *data) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control = le16_to_cpu(mle->control); switch (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE)) { case IEEE80211_ML_CONTROL_TYPE_BASIC: case IEEE80211_ML_CONTROL_TYPE_PREQ: case IEEE80211_ML_CONTROL_TYPE_TDLS: case IEEE80211_ML_CONTROL_TYPE_RECONF: case IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS: /* * The length is the first octet pointed by mle->variable so no * need to add anything */ break; default: WARN_ON(1); return 0; } return sizeof(*mle) + mle->variable[0]; } /** * ieee80211_mle_get_link_id - returns the link ID * @data: the basic multi link element * Return: the link ID, or -1 if not present * * The element is assumed to be of the correct type (BASIC) and big enough, * this must be checked using ieee80211_mle_type_ok(). */ static inline int ieee80211_mle_get_link_id(const u8 *data) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control = le16_to_cpu(mle->control); const u8 *common = mle->variable; /* common points now at the beginning of ieee80211_mle_basic_common_info */ common += sizeof(struct ieee80211_mle_basic_common_info); if (!(control & IEEE80211_MLC_BASIC_PRES_LINK_ID)) return -1; return *common; } /** * ieee80211_mle_get_bss_param_ch_cnt - returns the BSS parameter change count * @data: pointer to the basic multi link element * Return: the BSS Parameter Change Count field value, or -1 if not present * * The element is assumed to be of the correct type (BASIC) and big enough, * this must be checked using ieee80211_mle_type_ok(). */ static inline int ieee80211_mle_get_bss_param_ch_cnt(const u8 *data) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control = le16_to_cpu(mle->control); const u8 *common = mle->variable; /* common points now at the beginning of ieee80211_mle_basic_common_info */ common += sizeof(struct ieee80211_mle_basic_common_info); if (!(control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT)) return -1; if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) common += 1; return *common; } /** * ieee80211_mle_get_eml_med_sync_delay - returns the medium sync delay * @data: pointer to the multi-link element * Return: the medium synchronization delay field value from the multi-link * element, or the default value (%IEEE80211_MED_SYNC_DELAY_DEFAULT) * if not present * * The element is assumed to be of the correct type (BASIC) and big enough, * this must be checked using ieee80211_mle_type_ok(). */ static inline u16 ieee80211_mle_get_eml_med_sync_delay(const u8 *data) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control = le16_to_cpu(mle->control); const u8 *common = mle->variable; /* common points now at the beginning of ieee80211_mle_basic_common_info */ common += sizeof(struct ieee80211_mle_basic_common_info); if (!(control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY)) return IEEE80211_MED_SYNC_DELAY_DEFAULT; if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) common += 1; return get_unaligned_le16(common); } /** * ieee80211_mle_get_eml_cap - returns the EML capability * @data: pointer to the multi-link element * Return: the EML capability field value from the multi-link element, * or 0 if not present * * The element is assumed to be of the correct type (BASIC) and big enough, * this must be checked using ieee80211_mle_type_ok(). */ static inline u16 ieee80211_mle_get_eml_cap(const u8 *data) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control = le16_to_cpu(mle->control); const u8 *common = mle->variable; /* common points now at the beginning of ieee80211_mle_basic_common_info */ common += sizeof(struct ieee80211_mle_basic_common_info); if (!(control & IEEE80211_MLC_BASIC_PRES_EML_CAPA)) return 0; if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) common += 2; return get_unaligned_le16(common); } /** * ieee80211_mle_get_mld_capa_op - returns the MLD capabilities and operations. * @data: pointer to the multi-link element * Return: the MLD capabilities and operations field value from the multi-link * element, or 0 if not present * * The element is assumed to be of the correct type (BASIC) and big enough, * this must be checked using ieee80211_mle_type_ok(). */ static inline u16 ieee80211_mle_get_mld_capa_op(const u8 *data) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control = le16_to_cpu(mle->control); const u8 *common = mle->variable; /* * common points now at the beginning of * ieee80211_mle_basic_common_info */ common += sizeof(struct ieee80211_mle_basic_common_info); if (!(control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP)) return 0; if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) common += 2; return get_unaligned_le16(common); } /* Defined in Figure 9-1074t in P802.11be_D7.0 */ #define IEEE80211_EHT_ML_EXT_MLD_CAPA_OP_PARAM_UPDATE 0x0001 #define IEEE80211_EHT_ML_EXT_MLD_CAPA_OP_RECO_MAX_LINKS_MASK 0x001e #define IEEE80211_EHT_ML_EXT_MLD_CAPA_NSTR_UPDATE 0x0020 #define IEEE80211_EHT_ML_EXT_MLD_CAPA_EMLSR_ENA_ON_ONE_LINK 0x0040 #define IEEE80211_EHT_ML_EXT_MLD_CAPA_BTM_MLD_RECO_MULTI_AP 0x0080 /** * ieee80211_mle_get_ext_mld_capa_op - returns the extended MLD capabilities * and operations. * @data: pointer to the multi-link element * Return: the extended MLD capabilities and operations field value from * the multi-link element, or 0 if not present * * The element is assumed to be of the correct type (BASIC) and big enough, * this must be checked using ieee80211_mle_type_ok(). */ static inline u16 ieee80211_mle_get_ext_mld_capa_op(const u8 *data) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control = le16_to_cpu(mle->control); const u8 *common = mle->variable; /* * common points now at the beginning of * ieee80211_mle_basic_common_info */ common += sizeof(struct ieee80211_mle_basic_common_info); if (!(control & IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP)) return 0; if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_MLD_ID) common += 1; return get_unaligned_le16(common); } /** * ieee80211_mle_get_mld_id - returns the MLD ID * @data: pointer to the multi-link element * Return: The MLD ID in the given multi-link element, or 0 if not present * * The element is assumed to be of the correct type (BASIC) and big enough, * this must be checked using ieee80211_mle_type_ok(). */ static inline u8 ieee80211_mle_get_mld_id(const u8 *data) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control = le16_to_cpu(mle->control); const u8 *common = mle->variable; /* * common points now at the beginning of * ieee80211_mle_basic_common_info */ common += sizeof(struct ieee80211_mle_basic_common_info); if (!(control & IEEE80211_MLC_BASIC_PRES_MLD_ID)) return 0; if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) common += 2; return *common; } /** * ieee80211_mle_size_ok - validate multi-link element size * @data: pointer to the element data * @len: length of the containing element * Return: whether or not the multi-link element size is OK */ static inline bool ieee80211_mle_size_ok(const u8 *data, size_t len) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u8 fixed = sizeof(*mle); u8 common = 0; bool check_common_len = false; u16 control; if (!data || len < fixed) return false; control = le16_to_cpu(mle->control); switch (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE)) { case IEEE80211_ML_CONTROL_TYPE_BASIC: common += sizeof(struct ieee80211_mle_basic_common_info); check_common_len = true; if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) common += 2; if (control & IEEE80211_MLC_BASIC_PRES_MLD_ID) common += 1; if (control & IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP) common += 2; break; case IEEE80211_ML_CONTROL_TYPE_PREQ: common += sizeof(struct ieee80211_mle_preq_common_info); if (control & IEEE80211_MLC_PREQ_PRES_MLD_ID) common += 1; check_common_len = true; break; case IEEE80211_ML_CONTROL_TYPE_RECONF: if (control & IEEE80211_MLC_RECONF_PRES_MLD_MAC_ADDR) common += ETH_ALEN; if (control & IEEE80211_MLC_RECONF_PRES_EML_CAPA) common += 2; if (control & IEEE80211_MLC_RECONF_PRES_MLD_CAPA_OP) common += 2; if (control & IEEE80211_MLC_RECONF_PRES_EXT_MLD_CAPA_OP) common += 2; break; case IEEE80211_ML_CONTROL_TYPE_TDLS: common += sizeof(struct ieee80211_mle_tdls_common_info); check_common_len = true; break; case IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS: common = ETH_ALEN + 1; break; default: /* we don't know this type */ return true; } if (len < fixed + common) return false; if (!check_common_len) return true; /* if present, common length is the first octet there */ return mle->variable[0] >= common; } /** * ieee80211_mle_type_ok - validate multi-link element type and size * @data: pointer to the element data * @type: expected type of the element * @len: length of the containing element * Return: whether or not the multi-link element type matches and size is OK */ static inline bool ieee80211_mle_type_ok(const u8 *data, u8 type, size_t len) { const struct ieee80211_multi_link_elem *mle = (const void *)data; u16 control; if (!ieee80211_mle_size_ok(data, len)) return false; control = le16_to_cpu(mle->control); if (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE) == type) return true; return false; } enum ieee80211_mle_subelems { IEEE80211_MLE_SUBELEM_PER_STA_PROFILE = 0, IEEE80211_MLE_SUBELEM_FRAGMENT = 254, }; #define IEEE80211_MLE_STA_CONTROL_LINK_ID 0x000f #define IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE 0x0010 #define IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT 0x0020 #define IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT 0x0040 #define IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT 0x0080 #define IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT 0x0100 #define IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT 0x0200 #define IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE 0x0400 #define IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT 0x0800 struct ieee80211_mle_per_sta_profile { __le16 control; u8 sta_info_len; u8 variable[]; } __packed; /** * ieee80211_mle_basic_sta_prof_size_ok - validate basic multi-link element sta * profile size * @data: pointer to the sub element data * @len: length of the containing sub element * Return: %true if the STA profile is large enough, %false otherwise */ static inline bool ieee80211_mle_basic_sta_prof_size_ok(const u8 *data, size_t len) { const struct ieee80211_mle_per_sta_profile *prof = (const void *)data; u16 control; u8 fixed = sizeof(*prof); u8 info_len = 1; if (len < fixed) return false; control = le16_to_cpu(prof->control); if (control & IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT) info_len += 6; if (control & IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT) info_len += 2; if (control & IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT) info_len += 8; if (control & IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT) info_len += 2; if (control & IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE && control & IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT) { if (control & IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE) info_len += 2; else info_len += 1; } if (control & IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT) info_len += 1; return prof->sta_info_len >= info_len && fixed + prof->sta_info_len - 1 <= len; } /** * ieee80211_mle_basic_sta_prof_bss_param_ch_cnt - get per-STA profile BSS * parameter change count * @prof: the per-STA profile, having been checked with * ieee80211_mle_basic_sta_prof_size_ok() for the correct length * * Return: The BSS parameter change count value if present, 0 otherwise. */ static inline u8 ieee80211_mle_basic_sta_prof_bss_param_ch_cnt(const struct ieee80211_mle_per_sta_profile *prof) { u16 control = le16_to_cpu(prof->control); const u8 *pos = prof->variable; if (!(control & IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT)) return 0; if (control & IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT) pos += 6; if (control & IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT) pos += 2; if (control & IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT) pos += 8; if (control & IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT) pos += 2; if (control & IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE && control & IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT) { if (control & IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE) pos += 2; else pos += 1; } return *pos; } #define IEEE80211_MLE_STA_RECONF_CONTROL_LINK_ID 0x000f #define IEEE80211_MLE_STA_RECONF_CONTROL_COMPLETE_PROFILE 0x0010 #define IEEE80211_MLE_STA_RECONF_CONTROL_STA_MAC_ADDR_PRESENT 0x0020 #define IEEE80211_MLE_STA_RECONF_CONTROL_AP_REM_TIMER_PRESENT 0x0040 #define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE 0x0780 #define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_AP_REM 0 #define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_OP_PARAM_UPDATE 1 #define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_ADD_LINK 2 #define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_DEL_LINK 3 #define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_NSTR_STATUS 4 #define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_PARAMS_PRESENT 0x0800 /** * ieee80211_mle_reconf_sta_prof_size_ok - validate reconfiguration multi-link * element sta profile size. * @data: pointer to the sub element data * @len: length of the containing sub element * Return: %true if the STA profile is large enough, %false otherwise */ static inline bool ieee80211_mle_reconf_sta_prof_size_ok(const u8 *data, size_t len) { const struct ieee80211_mle_per_sta_profile *prof = (const void *)data; u16 control; u8 fixed = sizeof(*prof); u8 info_len = 1; if (len < fixed) return false; control = le16_to_cpu(prof->control); if (control & IEEE80211_MLE_STA_RECONF_CONTROL_STA_MAC_ADDR_PRESENT) info_len += ETH_ALEN; if (control & IEEE80211_MLE_STA_RECONF_CONTROL_AP_REM_TIMER_PRESENT) info_len += 2; if (control & IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_PARAMS_PRESENT) info_len += 2; return prof->sta_info_len >= info_len && fixed + prof->sta_info_len - 1 <= len; } #define IEEE80211_MLE_STA_EPCS_CONTROL_LINK_ID 0x000f #define IEEE80211_EPCS_ENA_RESP_BODY_LEN 3 static inline bool ieee80211_tid_to_link_map_size_ok(const u8 *data, size_t len) { const struct ieee80211_ttlm_elem *t2l = (const void *)data; u8 control, fixed = sizeof(*t2l), elem_len = 0; if (len < fixed) return false; control = t2l->control; if (control & IEEE80211_TTLM_CONTROL_SWITCH_TIME_PRESENT) elem_len += 2; if (control & IEEE80211_TTLM_CONTROL_EXPECTED_DUR_PRESENT) elem_len += 3; if (!(control & IEEE80211_TTLM_CONTROL_DEF_LINK_MAP)) { u8 bm_size; elem_len += 1; if (len < fixed + elem_len) return false; if (control & IEEE80211_TTLM_CONTROL_LINK_MAP_SIZE) bm_size = 1; else bm_size = 2; elem_len += hweight8(t2l->optional[0]) * bm_size; } return len >= fixed + elem_len; } /** * ieee80211_emlsr_pad_delay_in_us - Fetch the EMLSR Padding delay * in microseconds * @eml_cap: EML capabilities field value from common info field of * the Multi-link element * Return: the EMLSR Padding delay (in microseconds) encoded in the * EML Capabilities field */ static inline u32 ieee80211_emlsr_pad_delay_in_us(u16 eml_cap) { /* IEEE Std 802.11be-2024 Table 9-417i—Encoding of the EMLSR * Padding Delay subfield. */ u32 pad_delay = u16_get_bits(eml_cap, IEEE80211_EML_CAP_EMLSR_PADDING_DELAY); if (!pad_delay || pad_delay > IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_256US) return 0; return 32 * (1 << (pad_delay - 1)); } /** * ieee80211_emlsr_trans_delay_in_us - Fetch the EMLSR Transition * delay in microseconds * @eml_cap: EML capabilities field value from common info field of * the Multi-link element * Return: the EMLSR Transition delay (in microseconds) encoded in the * EML Capabilities field */ static inline u32 ieee80211_emlsr_trans_delay_in_us(u16 eml_cap) { /* IEEE Std 802.11be-2024 Table 9-417j—Encoding of the EMLSR * Transition Delay subfield. */ u32 trans_delay = u16_get_bits(eml_cap, IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY); /* invalid values also just use 0 */ if (!trans_delay || trans_delay > IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_256US) return 0; return 16 * (1 << (trans_delay - 1)); } /** * ieee80211_eml_trans_timeout_in_us - Fetch the EMLSR Transition * timeout value in microseconds * @eml_cap: EML capabilities field value from common info field of * the Multi-link element * Return: the EMLSR Transition timeout (in microseconds) encoded in * the EML Capabilities field */ static inline u32 ieee80211_eml_trans_timeout_in_us(u16 eml_cap) { /* IEEE Std 802.11be-2024 Table 9-417m—Encoding of the * Transition Timeout subfield. */ u8 timeout = u16_get_bits(eml_cap, IEEE80211_EML_CAP_TRANSITION_TIMEOUT); /* invalid values also just use 0 */ if (!timeout || timeout > IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128TU) return 0; return 128 * (1 << (timeout - 1)); } #define for_each_mle_subelement(_elem, _data, _len) \ if (ieee80211_mle_size_ok(_data, _len)) \ for_each_element(_elem, \ _data + ieee80211_mle_common_size(_data),\ _len - ieee80211_mle_common_size(_data)) #endif /* LINUX_IEEE80211_EHT_H */ |
| 1 1 1 1 1 1 1 1 28 27 1 26 2 2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 | // SPDX-License-Identifier: GPL-2.0-or-later /* * Host Side support for RNDIS Networking Links * Copyright (C) 2005 by David Brownell */ #include <linux/module.h> #include <linux/netdevice.h> #include <linux/etherdevice.h> #include <linux/ethtool.h> #include <linux/workqueue.h> #include <linux/slab.h> #include <linux/mii.h> #include <linux/usb.h> #include <linux/usb/cdc.h> #include <linux/usb/usbnet.h> #include <linux/usb/rndis_host.h> /* * RNDIS is NDIS remoted over USB. It's a MSFT variant of CDC ACM ... of * course ACM was intended for modems, not Ethernet links! USB's standard * for Ethernet links is "CDC Ethernet", which is significantly simpler. * * NOTE that Microsoft's "RNDIS 1.0" specification is incomplete. Issues * include: * - Power management in particular relies on information that's scattered * through other documentation, and which is incomplete or incorrect even * there. * - There are various undocumented protocol requirements, such as the * need to send unused garbage in control-OUT messages. * - In some cases, MS-Windows will emit undocumented requests; this * matters more to peripheral implementations than host ones. * * Moreover there's a no-open-specs variant of RNDIS called "ActiveSync". * * For these reasons and others, ** USE OF RNDIS IS STRONGLY DISCOURAGED ** in * favor of such non-proprietary alternatives as CDC Ethernet or the newer (and * currently rare) "Ethernet Emulation Model" (EEM). */ /* * RNDIS notifications from device: command completion; "reverse" * keepalives; etc */ void rndis_status(struct usbnet *dev, struct urb *urb) { netdev_dbg(dev->net, "rndis status urb, len %d stat %d\n", urb->actual_length, urb->status); // FIXME for keepalives, respond immediately (asynchronously) // if not an RNDIS status, do like cdc_status(dev,urb) does } EXPORT_SYMBOL_GPL(rndis_status); /* * RNDIS indicate messages. */ static void rndis_msg_indicate(struct usbnet *dev, struct rndis_indicate *msg, int buflen) { struct cdc_state *info = (void *)&dev->data; struct device *udev = &info->control->dev; if (dev->driver_info->indication) { dev->driver_info->indication(dev, msg, buflen); } else { u32 status = le32_to_cpu(msg->status); switch (status) { case RNDIS_STATUS_MEDIA_CONNECT: dev_info(udev, "rndis media connect\n"); break; case RNDIS_STATUS_MEDIA_DISCONNECT: dev_info(udev, "rndis media disconnect\n"); break; default: dev_info(udev, "rndis indication: 0x%08x\n", status); } } } /* * RPC done RNDIS-style. Caller guarantees: * - message is properly byteswapped * - there's no other request pending * - buf can hold up to 1KB response (required by RNDIS spec) * On return, the first few entries are already byteswapped. * * Call context is likely probe(), before interface name is known, * which is why we won't try to use it in the diagnostics. */ int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf, int buflen) { struct cdc_state *info = (void *) &dev->data; struct usb_cdc_notification notification; int master_ifnum; int retval; int partial; unsigned count; u32 xid = 0, msg_len, request_id, msg_type, rsp, status; /* REVISIT when this gets called from contexts other than probe() or * disconnect(): either serialize, or dispatch responses on xid */ msg_type = le32_to_cpu(buf->msg_type); /* Issue the request; xid is unique, don't bother byteswapping it */ if (likely(msg_type != RNDIS_MSG_HALT && msg_type != RNDIS_MSG_RESET)) { xid = dev->xid++; if (!xid) xid = dev->xid++; buf->request_id = (__force __le32) xid; } master_ifnum = info->control->cur_altsetting->desc.bInterfaceNumber; retval = usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0), USB_CDC_SEND_ENCAPSULATED_COMMAND, USB_TYPE_CLASS | USB_RECIP_INTERFACE, 0, master_ifnum, buf, le32_to_cpu(buf->msg_len), RNDIS_CONTROL_TIMEOUT_MS); if (unlikely(retval < 0 || xid == 0)) return retval; /* Some devices don't respond on the control channel until * polled on the status channel, so do that first. */ if (dev->driver_info->data & RNDIS_DRIVER_DATA_POLL_STATUS) { retval = usb_interrupt_msg( dev->udev, usb_rcvintpipe(dev->udev, dev->status->desc.bEndpointAddress), ¬ification, sizeof(notification), &partial, RNDIS_CONTROL_TIMEOUT_MS); if (unlikely(retval < 0)) return retval; } /* Poll the control channel; the request probably completed immediately */ rsp = le32_to_cpu(buf->msg_type) | RNDIS_MSG_COMPLETION; for (count = 0; count < 10; count++) { memset(buf, 0, CONTROL_BUFFER_SIZE); retval = usb_control_msg(dev->udev, usb_rcvctrlpipe(dev->udev, 0), USB_CDC_GET_ENCAPSULATED_RESPONSE, USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE, 0, master_ifnum, buf, buflen, RNDIS_CONTROL_TIMEOUT_MS); if (likely(retval >= 8)) { msg_type = le32_to_cpu(buf->msg_type); msg_len = le32_to_cpu(buf->msg_len); status = le32_to_cpu(buf->status); request_id = (__force u32) buf->request_id; if (likely(msg_type == rsp)) { if (likely(request_id == xid)) { if (unlikely(rsp == RNDIS_MSG_RESET_C)) return 0; if (likely(RNDIS_STATUS_SUCCESS == status)) return 0; dev_dbg(&info->control->dev, "rndis reply status %08x\n", status); return -EL3RST; } dev_dbg(&info->control->dev, "rndis reply id %d expected %d\n", request_id, xid); /* then likely retry */ } else switch (msg_type) { case RNDIS_MSG_INDICATE: /* fault/event */ rndis_msg_indicate(dev, (void *)buf, buflen); break; case RNDIS_MSG_KEEPALIVE: { /* ping */ struct rndis_keepalive_c *msg = (void *)buf; msg->msg_type = cpu_to_le32(RNDIS_MSG_KEEPALIVE_C); msg->msg_len = cpu_to_le32(sizeof *msg); msg->status = cpu_to_le32(RNDIS_STATUS_SUCCESS); retval = usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0), USB_CDC_SEND_ENCAPSULATED_COMMAND, USB_TYPE_CLASS | USB_RECIP_INTERFACE, 0, master_ifnum, msg, sizeof *msg, RNDIS_CONTROL_TIMEOUT_MS); if (unlikely(retval < 0)) dev_dbg(&info->control->dev, "rndis keepalive err %d\n", retval); } break; default: dev_dbg(&info->control->dev, "unexpected rndis msg %08x len %d\n", le32_to_cpu(buf->msg_type), msg_len); } } else { /* device probably issued a protocol stall; ignore */ dev_dbg(&info->control->dev, "rndis response error, code %d\n", retval); } msleep(40); } dev_dbg(&info->control->dev, "rndis response timeout\n"); return -ETIMEDOUT; } EXPORT_SYMBOL_GPL(rndis_command); /* * rndis_query: * * Performs a query for @oid along with 0 or more bytes of payload as * specified by @in_len. If @reply_len is not set to -1 then the reply * length is checked against this value, resulting in an error if it * doesn't match. * * NOTE: Adding a payload exactly or greater than the size of the expected * response payload is an evident requirement MSFT added for ActiveSync. * * The only exception is for OIDs that return a variably sized response, * in which case no payload should be added. This undocumented (and * nonsensical!) issue was found by sniffing protocol requests from the * ActiveSync 4.1 Windows driver. */ static int rndis_query(struct usbnet *dev, struct usb_interface *intf, void *buf, u32 oid, u32 in_len, void **reply, int *reply_len) { int retval; union { void *buf; struct rndis_msg_hdr *header; struct rndis_query *get; struct rndis_query_c *get_c; } u; u32 off, len; u.buf = buf; memset(u.get, 0, sizeof *u.get + in_len); u.get->msg_type = cpu_to_le32(RNDIS_MSG_QUERY); u.get->msg_len = cpu_to_le32(sizeof *u.get + in_len); u.get->oid = cpu_to_le32(oid); u.get->len = cpu_to_le32(in_len); u.get->offset = cpu_to_le32(20); retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE); if (unlikely(retval < 0)) { dev_err(&intf->dev, "RNDIS_MSG_QUERY(0x%08x) failed, %d\n", oid, retval); return retval; } off = le32_to_cpu(u.get_c->offset); len = le32_to_cpu(u.get_c->len); if (unlikely((off > CONTROL_BUFFER_SIZE - 8) || (len > CONTROL_BUFFER_SIZE - 8 - off))) goto response_error; if (*reply_len != -1 && len != *reply_len) goto response_error; *reply = (unsigned char *) &u.get_c->request_id + off; *reply_len = len; return retval; response_error: dev_err(&intf->dev, "RNDIS_MSG_QUERY(0x%08x) " "invalid response - off %d len %d\n", oid, off, len); return -EDOM; } /* same as usbnet_netdev_ops but MTU change not allowed */ static const struct net_device_ops rndis_netdev_ops = { .ndo_open = usbnet_open, .ndo_stop = usbnet_stop, .ndo_start_xmit = usbnet_start_xmit, .ndo_tx_timeout = usbnet_tx_timeout, .ndo_get_stats64 = dev_get_tstats64, .ndo_set_mac_address = eth_mac_addr, .ndo_validate_addr = eth_validate_addr, }; int generic_rndis_bind(struct usbnet *dev, struct usb_interface *intf, int flags) { int retval; struct net_device *net = dev->net; struct cdc_state *info = (void *) &dev->data; union { void *buf; struct rndis_msg_hdr *header; struct rndis_init *init; struct rndis_init_c *init_c; struct rndis_query *get; struct rndis_query_c *get_c; struct rndis_set *set; struct rndis_set_c *set_c; struct rndis_halt *halt; } u; u32 tmp; __le32 phym_unspec, *phym; int reply_len; unsigned char *bp; /* we can't rely on i/o from stack working, or stack allocation */ u.buf = kmalloc(CONTROL_BUFFER_SIZE, GFP_KERNEL); if (!u.buf) return -ENOMEM; retval = usbnet_generic_cdc_bind(dev, intf); if (retval < 0) goto fail; u.init->msg_type = cpu_to_le32(RNDIS_MSG_INIT); u.init->msg_len = cpu_to_le32(sizeof *u.init); u.init->major_version = cpu_to_le32(1); u.init->minor_version = cpu_to_le32(0); /* max transfer (in spec) is 0x4000 at full speed, but for * TX we'll stick to one Ethernet packet plus RNDIS framing. * For RX we handle drivers that zero-pad to end-of-packet. * Don't let userspace change these settings. * * NOTE: there still seems to be weirdness here, as if we need * to do some more things to make sure WinCE targets accept this. * They default to jumbograms of 8KB or 16KB, which is absurd * for such low data rates and which is also more than Linux * can usually expect to allocate for SKB data... */ net->hard_header_len += sizeof (struct rndis_data_hdr); dev->hard_mtu = net->mtu + net->hard_header_len; dev->maxpacket = usb_maxpacket(dev->udev, dev->out); if (dev->maxpacket == 0) { netif_dbg(dev, probe, dev->net, "dev->maxpacket can't be 0\n"); retval = -EINVAL; goto fail_and_release; } dev->rx_urb_size = dev->hard_mtu + (dev->maxpacket + 1); dev->rx_urb_size &= ~(dev->maxpacket - 1); u.init->max_transfer_size = cpu_to_le32(dev->rx_urb_size); net->netdev_ops = &rndis_netdev_ops; retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE); if (unlikely(retval < 0)) { /* it might not even be an RNDIS device!! */ dev_err(&intf->dev, "RNDIS init failed, %d\n", retval); goto fail_and_release; } tmp = le32_to_cpu(u.init_c->max_transfer_size); if (tmp < dev->hard_mtu) { if (tmp <= net->hard_header_len) { dev_err(&intf->dev, "dev can't take %u byte packets (max %u)\n", dev->hard_mtu, tmp); retval = -EINVAL; goto halt_fail_and_release; } dev_warn(&intf->dev, "dev can't take %u byte packets (max %u), " "adjusting MTU to %u\n", dev->hard_mtu, tmp, tmp - net->hard_header_len); dev->hard_mtu = tmp; net->mtu = dev->hard_mtu - net->hard_header_len; } /* REVISIT: peripheral "alignment" request is ignored ... */ dev_dbg(&intf->dev, "hard mtu %u (%u from dev), rx buflen %zu, align %d\n", dev->hard_mtu, tmp, dev->rx_urb_size, 1 << le32_to_cpu(u.init_c->packet_alignment)); /* module has some device initialization code needs to be done right * after RNDIS_INIT */ if (dev->driver_info->early_init && dev->driver_info->early_init(dev) != 0) goto halt_fail_and_release; /* Check physical medium */ phym = NULL; reply_len = sizeof *phym; retval = rndis_query(dev, intf, u.buf, RNDIS_OID_GEN_PHYSICAL_MEDIUM, reply_len, (void **)&phym, &reply_len); if (retval != 0 || !phym) { /* OID is optional so don't fail here. */ phym_unspec = cpu_to_le32(RNDIS_PHYSICAL_MEDIUM_UNSPECIFIED); phym = &phym_unspec; } if ((flags & FLAG_RNDIS_PHYM_WIRELESS) && le32_to_cpup(phym) != RNDIS_PHYSICAL_MEDIUM_WIRELESS_LAN) { netif_dbg(dev, probe, dev->net, "driver requires wireless physical medium, but device is not\n"); retval = -ENODEV; goto halt_fail_and_release; } if ((flags & FLAG_RNDIS_PHYM_NOT_WIRELESS) && le32_to_cpup(phym) == RNDIS_PHYSICAL_MEDIUM_WIRELESS_LAN) { netif_dbg(dev, probe, dev->net, "driver requires non-wireless physical medium, but device is wireless.\n"); retval = -ENODEV; goto halt_fail_and_release; } /* Get designated host ethernet address */ reply_len = ETH_ALEN; retval = rndis_query(dev, intf, u.buf, RNDIS_OID_802_3_PERMANENT_ADDRESS, 48, (void **) &bp, &reply_len); if (unlikely(retval< 0)) { dev_err(&intf->dev, "rndis get ethaddr, %d\n", retval); goto halt_fail_and_release; } eth_hw_addr_set(net, bp); /* set a nonzero filter to enable data transfers */ memset(u.set, 0, sizeof *u.set); u.set->msg_type = cpu_to_le32(RNDIS_MSG_SET); u.set->msg_len = cpu_to_le32(4 + sizeof *u.set); u.set->oid = cpu_to_le32(RNDIS_OID_GEN_CURRENT_PACKET_FILTER); u.set->len = cpu_to_le32(4); u.set->offset = cpu_to_le32((sizeof *u.set) - 8); *(__le32 *)(u.buf + sizeof *u.set) = cpu_to_le32(RNDIS_DEFAULT_FILTER); retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE); if (unlikely(retval < 0)) { dev_err(&intf->dev, "rndis set packet filter, %d\n", retval); goto halt_fail_and_release; } retval = 0; kfree(u.buf); return retval; halt_fail_and_release: memset(u.halt, 0, sizeof *u.halt); u.halt->msg_type = cpu_to_le32(RNDIS_MSG_HALT); u.halt->msg_len = cpu_to_le32(sizeof *u.halt); (void) rndis_command(dev, (void *)u.halt, CONTROL_BUFFER_SIZE); fail_and_release: usb_set_intfdata(info->data, NULL); usb_driver_release_interface(driver_of(intf), info->data); info->data = NULL; fail: kfree(u.buf); return retval; } EXPORT_SYMBOL_GPL(generic_rndis_bind); static int rndis_bind(struct usbnet *dev, struct usb_interface *intf) { return generic_rndis_bind(dev, intf, FLAG_RNDIS_PHYM_NOT_WIRELESS); } static int zte_rndis_bind(struct usbnet *dev, struct usb_interface *intf) { int status = rndis_bind(dev, intf); if (!status && (dev->net->dev_addr[0] & 0x02)) eth_hw_addr_random(dev->net); return status; } void rndis_unbind(struct usbnet *dev, struct usb_interface *intf) { struct rndis_halt *halt; /* try to clear any rndis state/activity (no i/o from stack!) */ halt = kzalloc(CONTROL_BUFFER_SIZE, GFP_KERNEL); if (halt) { halt->msg_type = cpu_to_le32(RNDIS_MSG_HALT); halt->msg_len = cpu_to_le32(sizeof *halt); (void) rndis_command(dev, (void *)halt, CONTROL_BUFFER_SIZE); kfree(halt); } usbnet_cdc_unbind(dev, intf); } EXPORT_SYMBOL_GPL(rndis_unbind); /* * DATA -- host must not write zlps */ int rndis_rx_fixup(struct usbnet *dev, struct sk_buff *skb) { bool dst_mac_fixup; /* This check is no longer done by usbnet */ if (skb->len < dev->net->hard_header_len) return 0; dst_mac_fixup = !!(dev->driver_info->data & RNDIS_DRIVER_DATA_DST_MAC_FIXUP); /* peripheral may have batched packets to us... */ while (likely(skb->len)) { struct rndis_data_hdr *hdr = (void *)skb->data; struct sk_buff *skb2; u32 msg_type, msg_len, data_offset, data_len; msg_type = le32_to_cpu(hdr->msg_type); msg_len = le32_to_cpu(hdr->msg_len); data_offset = le32_to_cpu(hdr->data_offset); data_len = le32_to_cpu(hdr->data_len); /* don't choke if we see oob, per-packet data, etc */ if (unlikely(msg_type != RNDIS_MSG_PACKET || skb->len < msg_len || (data_offset + data_len + 8) > msg_len)) { dev->net->stats.rx_frame_errors++; netdev_dbg(dev->net, "bad rndis message %d/%d/%d/%d, len %d\n", le32_to_cpu(hdr->msg_type), msg_len, data_offset, data_len, skb->len); return 0; } skb_pull(skb, 8 + data_offset); /* at most one packet left? */ if (likely((data_len - skb->len) <= sizeof *hdr)) { skb_trim(skb, data_len); break; } /* try to return all the packets in the batch */ skb2 = skb_clone(skb, GFP_ATOMIC); if (unlikely(!skb2)) break; skb_pull(skb, msg_len - sizeof *hdr); skb_trim(skb2, data_len); if (unlikely(dst_mac_fixup)) usbnet_cdc_zte_rx_fixup(dev, skb2); usbnet_skb_return(dev, skb2); } /* caller will usbnet_skb_return the remaining packet */ if (unlikely(dst_mac_fixup)) usbnet_cdc_zte_rx_fixup(dev, skb); return 1; } EXPORT_SYMBOL_GPL(rndis_rx_fixup); struct sk_buff * rndis_tx_fixup(struct usbnet *dev, struct sk_buff *skb, gfp_t flags) { struct rndis_data_hdr *hdr; struct sk_buff *skb2; unsigned len = skb->len; if (likely(!skb_cloned(skb))) { int room = skb_headroom(skb); /* enough head room as-is? */ if (unlikely((sizeof *hdr) <= room)) goto fill; /* enough room, but needs to be readjusted? */ room += skb_tailroom(skb); if (likely((sizeof *hdr) <= room)) { skb->data = memmove(skb->head + sizeof *hdr, skb->data, len); skb_set_tail_pointer(skb, len); goto fill; } } /* create a new skb, with the correct size (and tailpad) */ skb2 = skb_copy_expand(skb, sizeof *hdr, 1, flags); dev_kfree_skb_any(skb); if (unlikely(!skb2)) return skb2; skb = skb2; /* fill out the RNDIS header. we won't bother trying to batch * packets; Linux minimizes wasted bandwidth through tx queues. */ fill: hdr = __skb_push(skb, sizeof *hdr); memset(hdr, 0, sizeof *hdr); hdr->msg_type = cpu_to_le32(RNDIS_MSG_PACKET); hdr->msg_len = cpu_to_le32(skb->len); hdr->data_offset = cpu_to_le32(sizeof(*hdr) - 8); hdr->data_len = cpu_to_le32(len); /* FIXME make the last packet always be short ... */ return skb; } EXPORT_SYMBOL_GPL(rndis_tx_fixup); static const struct driver_info rndis_info = { .description = "RNDIS device", .flags = FLAG_ETHER | FLAG_POINTTOPOINT | FLAG_FRAMING_RN | FLAG_NO_SETINT, .bind = rndis_bind, .unbind = rndis_unbind, .status = rndis_status, .rx_fixup = rndis_rx_fixup, .tx_fixup = rndis_tx_fixup, }; static const struct driver_info rndis_poll_status_info = { .description = "RNDIS device (poll status before control)", .flags = FLAG_ETHER | FLAG_POINTTOPOINT | FLAG_FRAMING_RN | FLAG_NO_SETINT, .data = RNDIS_DRIVER_DATA_POLL_STATUS, .bind = rndis_bind, .unbind = rndis_unbind, .status = rndis_status, .rx_fixup = rndis_rx_fixup, .tx_fixup = rndis_tx_fixup, }; static const struct driver_info zte_rndis_info = { .description = "ZTE RNDIS device", .flags = FLAG_ETHER | FLAG_POINTTOPOINT | FLAG_FRAMING_RN | FLAG_NO_SETINT, .data = RNDIS_DRIVER_DATA_DST_MAC_FIXUP, .bind = zte_rndis_bind, .unbind = rndis_unbind, .status = rndis_status, .rx_fixup = rndis_rx_fixup, .tx_fixup = rndis_tx_fixup, }; /*-------------------------------------------------------------------------*/ static const struct usb_device_id products [] = { { /* 2Wire HomePortal 1000SW */ USB_DEVICE_AND_INTERFACE_INFO(0x1630, 0x0042, USB_CLASS_COMM, 2 /* ACM */, 0x0ff), .driver_info = (unsigned long) &rndis_poll_status_info, }, { /* Hytera Communications DMR radios' "Radio to PC Network" */ USB_VENDOR_AND_INTERFACE_INFO(0x238b, USB_CLASS_COMM, 2 /* ACM */, 0x0ff), .driver_info = (unsigned long)&rndis_info, }, { /* ZTE WWAN modules */ USB_VENDOR_AND_INTERFACE_INFO(0x19d2, USB_CLASS_WIRELESS_CONTROLLER, 1, 3), .driver_info = (unsigned long)&zte_rndis_info, }, { /* ZTE WWAN modules, ACM flavour */ USB_VENDOR_AND_INTERFACE_INFO(0x19d2, USB_CLASS_COMM, 2 /* ACM */, 0x0ff), .driver_info = (unsigned long)&zte_rndis_info, }, { /* RNDIS is MSFT's un-official variant of CDC ACM */ USB_INTERFACE_INFO(USB_CLASS_COMM, 2 /* ACM */, 0x0ff), .driver_info = (unsigned long) &rndis_info, }, { /* "ActiveSync" is an undocumented variant of RNDIS, used in WM5 */ USB_INTERFACE_INFO(USB_CLASS_MISC, 1, 1), .driver_info = (unsigned long) &rndis_poll_status_info, }, { /* RNDIS for tethering */ USB_INTERFACE_INFO(USB_CLASS_WIRELESS_CONTROLLER, 1, 3), .driver_info = (unsigned long) &rndis_info, }, { /* Novatel Verizon USB730L */ USB_INTERFACE_INFO(USB_CLASS_MISC, 4, 1), .driver_info = (unsigned long) &rndis_info, }, { }, // END }; MODULE_DEVICE_TABLE(usb, products); static struct usb_driver rndis_driver = { .name = "rndis_host", .id_table = products, .probe = usbnet_probe, .disconnect = usbnet_disconnect, .suspend = usbnet_suspend, .resume = usbnet_resume, .disable_hub_initiated_lpm = 1, }; module_usb_driver(rndis_driver); MODULE_AUTHOR("David Brownell"); MODULE_DESCRIPTION("USB Host side RNDIS driver"); MODULE_LICENSE("GPL"); |
| 11 9 2 57 57 43 43 43 43 43 1 1 1 1 1 42 42 42 41 47 46 46 20 35 47 46 45 3 11 28 11 36 19 39 39 39 59 59 59 59 58 58 47 25 2 25 15 11 11 26 26 26 6 22 2 4 80 78 2 13 64 52 21 4 47 26 26 26 26 26 27 3 24 15 11 3 6 6 212 4 306 4 90 7 4 3 3 856 1 854 79 607 855 856 35 745 163 27 629 2 657 641 18 7 656 611 553 14 14 14 1 1 53 53 8 45 36 17 4 32 33 1 2 1 17 2 14 1 15 7 25 25 25 1 25 1 1 1 1 1 20 20 13 9 2 1 6 6 838 1 840 738 736 739 68 599 127 29 669 119 615 614 455 453 612 41 43 2 42 16 41 41 14 6 6 6 50 50 28 57 57 57 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 | // SPDX-License-Identifier: GPL-2.0-or-later /* * IPVS An implementation of the IP virtual server support for the * LINUX operating system. IPVS is now implemented as a module * over the Netfilter framework. IPVS can be used to build a * high-performance and highly available server based on a * cluster of servers. * * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> * Peter Kese <peter.kese@ijs.si> * Julian Anastasov <ja@ssi.bg> * * The IPVS code for kernel 2.2 was done by Wensong Zhang and Peter Kese, * with changes/fixes from Julian Anastasov, Lars Marowsky-Bree, Horms * and others. * * Changes: * Paul `Rusty' Russell properly handle non-linear skbs * Harald Welte don't use nfcache */ #define pr_fmt(fmt) "IPVS: " fmt #include <linux/module.h> #include <linux/kernel.h> #include <linux/ip.h> #include <linux/tcp.h> #include <linux/sctp.h> #include <linux/icmp.h> #include <linux/slab.h> #include <net/ip.h> #include <net/tcp.h> #include <net/udp.h> #include <net/icmp.h> /* for icmp_send */ #include <net/gue.h> #include <net/gre.h> #include <net/route.h> #include <net/ip6_checksum.h> #include <net/netns/generic.h> /* net_generic() */ #include <linux/netfilter.h> #include <linux/netfilter_ipv4.h> #ifdef CONFIG_IP_VS_IPV6 #include <net/ipv6.h> #include <linux/netfilter_ipv6.h> #include <net/ip6_route.h> #endif #include <net/ip_vs.h> #include <linux/indirect_call_wrapper.h> EXPORT_SYMBOL(register_ip_vs_scheduler); EXPORT_SYMBOL(unregister_ip_vs_scheduler); EXPORT_SYMBOL(ip_vs_proto_name); EXPORT_SYMBOL(ip_vs_conn_new); EXPORT_SYMBOL(ip_vs_conn_in_get); EXPORT_SYMBOL(ip_vs_conn_out_get); #ifdef CONFIG_IP_VS_PROTO_TCP EXPORT_SYMBOL(ip_vs_tcp_conn_listen); #endif EXPORT_SYMBOL(ip_vs_conn_put); #ifdef CONFIG_IP_VS_DEBUG EXPORT_SYMBOL(ip_vs_get_debug_level); #endif EXPORT_SYMBOL(ip_vs_new_conn_out); #if defined(CONFIG_IP_VS_PROTO_TCP) && defined(CONFIG_IP_VS_PROTO_UDP) #define SNAT_CALL(f, ...) \ INDIRECT_CALL_2(f, tcp_snat_handler, udp_snat_handler, __VA_ARGS__) #elif defined(CONFIG_IP_VS_PROTO_TCP) #define SNAT_CALL(f, ...) INDIRECT_CALL_1(f, tcp_snat_handler, __VA_ARGS__) #elif defined(CONFIG_IP_VS_PROTO_UDP) #define SNAT_CALL(f, ...) INDIRECT_CALL_1(f, udp_snat_handler, __VA_ARGS__) #else #define SNAT_CALL(f, ...) f(__VA_ARGS__) #endif static unsigned int ip_vs_net_id __read_mostly; /* netns cnt used for uniqueness */ static atomic_t ipvs_netns_cnt = ATOMIC_INIT(0); /* ID used in ICMP lookups */ #define icmp_id(icmph) (((icmph)->un).echo.id) #define icmpv6_id(icmph) (icmph->icmp6_dataun.u_echo.identifier) const char *ip_vs_proto_name(unsigned int proto) { static char buf[20]; switch (proto) { case IPPROTO_IP: return "IP"; case IPPROTO_UDP: return "UDP"; case IPPROTO_TCP: return "TCP"; case IPPROTO_SCTP: return "SCTP"; case IPPROTO_ICMP: return "ICMP"; #ifdef CONFIG_IP_VS_IPV6 case IPPROTO_ICMPV6: return "ICMPv6"; #endif default: sprintf(buf, "IP_%u", proto); return buf; } } void ip_vs_init_hash_table(struct list_head *table, int rows) { while (--rows >= 0) INIT_LIST_HEAD(&table[rows]); } static inline void ip_vs_in_stats(struct ip_vs_conn *cp, struct sk_buff *skb) { struct ip_vs_dest *dest = cp->dest; struct netns_ipvs *ipvs = cp->ipvs; if (dest && (dest->flags & IP_VS_DEST_F_AVAILABLE)) { struct ip_vs_cpu_stats *s; struct ip_vs_service *svc; local_bh_disable(); s = this_cpu_ptr(dest->stats.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.inpkts); u64_stats_add(&s->cnt.inbytes, skb->len); u64_stats_update_end(&s->syncp); svc = rcu_dereference(dest->svc); s = this_cpu_ptr(svc->stats.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.inpkts); u64_stats_add(&s->cnt.inbytes, skb->len); u64_stats_update_end(&s->syncp); s = this_cpu_ptr(ipvs->tot_stats->s.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.inpkts); u64_stats_add(&s->cnt.inbytes, skb->len); u64_stats_update_end(&s->syncp); local_bh_enable(); } } static inline void ip_vs_out_stats(struct ip_vs_conn *cp, struct sk_buff *skb) { struct ip_vs_dest *dest = cp->dest; struct netns_ipvs *ipvs = cp->ipvs; if (dest && (dest->flags & IP_VS_DEST_F_AVAILABLE)) { struct ip_vs_cpu_stats *s; struct ip_vs_service *svc; local_bh_disable(); s = this_cpu_ptr(dest->stats.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.outpkts); u64_stats_add(&s->cnt.outbytes, skb->len); u64_stats_update_end(&s->syncp); svc = rcu_dereference(dest->svc); s = this_cpu_ptr(svc->stats.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.outpkts); u64_stats_add(&s->cnt.outbytes, skb->len); u64_stats_update_end(&s->syncp); s = this_cpu_ptr(ipvs->tot_stats->s.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.outpkts); u64_stats_add(&s->cnt.outbytes, skb->len); u64_stats_update_end(&s->syncp); local_bh_enable(); } } static inline void ip_vs_conn_stats(struct ip_vs_conn *cp, struct ip_vs_service *svc) { struct netns_ipvs *ipvs = svc->ipvs; struct ip_vs_cpu_stats *s; local_bh_disable(); s = this_cpu_ptr(cp->dest->stats.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.conns); u64_stats_update_end(&s->syncp); s = this_cpu_ptr(svc->stats.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.conns); u64_stats_update_end(&s->syncp); s = this_cpu_ptr(ipvs->tot_stats->s.cpustats); u64_stats_update_begin(&s->syncp); u64_stats_inc(&s->cnt.conns); u64_stats_update_end(&s->syncp); local_bh_enable(); } static inline void ip_vs_set_state(struct ip_vs_conn *cp, int direction, const struct sk_buff *skb, struct ip_vs_proto_data *pd) { if (likely(pd->pp->state_transition)) pd->pp->state_transition(cp, direction, skb, pd); } static inline int ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc, struct sk_buff *skb, int protocol, const union nf_inet_addr *caddr, __be16 cport, const union nf_inet_addr *vaddr, __be16 vport, struct ip_vs_conn_param *p) { ip_vs_conn_fill_param(svc->ipvs, svc->af, protocol, caddr, cport, vaddr, vport, p); p->pe = rcu_dereference(svc->pe); if (p->pe && p->pe->fill_param) return p->pe->fill_param(p, skb); return 0; } /* * IPVS persistent scheduling function * It creates a connection entry according to its template if exists, * or selects a server and creates a connection entry plus a template. * Locking: we are svc user (svc->refcnt), so we hold all dests too * Protocols supported: TCP, UDP */ static struct ip_vs_conn * ip_vs_sched_persist(struct ip_vs_service *svc, struct sk_buff *skb, __be16 src_port, __be16 dst_port, int *ignored, struct ip_vs_iphdr *iph) { struct ip_vs_conn *cp = NULL; struct ip_vs_dest *dest; struct ip_vs_conn *ct; __be16 dport = 0; /* destination port to forward */ unsigned int flags; struct ip_vs_conn_param param; const union nf_inet_addr fwmark = { .ip = htonl(svc->fwmark) }; union nf_inet_addr snet; /* source network of the client, after masking */ const union nf_inet_addr *src_addr, *dst_addr; if (likely(!ip_vs_iph_inverse(iph))) { src_addr = &iph->saddr; dst_addr = &iph->daddr; } else { src_addr = &iph->daddr; dst_addr = &iph->saddr; } /* Mask saddr with the netmask to adjust template granularity */ #ifdef CONFIG_IP_VS_IPV6 if (svc->af == AF_INET6) ipv6_addr_prefix(&snet.in6, &src_addr->in6, (__force __u32) svc->netmask); else #endif snet.ip = src_addr->ip & svc->netmask; IP_VS_DBG_BUF(6, "p-schedule: src %s:%u dest %s:%u " "mnet %s\n", IP_VS_DBG_ADDR(svc->af, src_addr), ntohs(src_port), IP_VS_DBG_ADDR(svc->af, dst_addr), ntohs(dst_port), IP_VS_DBG_ADDR(svc->af, &snet)); /* * As far as we know, FTP is a very complicated network protocol, and * it uses control connection and data connections. For active FTP, * FTP server initialize data connection to the client, its source port * is often 20. For passive FTP, FTP server tells the clients the port * that it passively listens to, and the client issues the data * connection. In the tunneling or direct routing mode, the load * balancer is on the client-to-server half of connection, the port * number is unknown to the load balancer. So, a conn template like * <caddr, 0, vaddr, 0, daddr, 0> is created for persistent FTP * service, and a template like <caddr, 0, vaddr, vport, daddr, dport> * is created for other persistent services. */ { int protocol = iph->protocol; const union nf_inet_addr *vaddr = dst_addr; __be16 vport = 0; if (dst_port == svc->port) { /* non-FTP template: * <protocol, caddr, 0, vaddr, vport, daddr, dport> * FTP template: * <protocol, caddr, 0, vaddr, 0, daddr, 0> */ if (svc->port != FTPPORT) vport = dst_port; } else { /* Note: persistent fwmark-based services and * persistent port zero service are handled here. * fwmark template: * <IPPROTO_IP,caddr,0,fwmark,0,daddr,0> * port zero template: * <protocol,caddr,0,vaddr,0,daddr,0> */ if (svc->fwmark) { protocol = IPPROTO_IP; vaddr = &fwmark; } } /* return *ignored = -1 so NF_DROP can be used */ if (ip_vs_conn_fill_param_persist(svc, skb, protocol, &snet, 0, vaddr, vport, ¶m) < 0) { *ignored = -1; return NULL; } } /* Check if a template already exists */ ct = ip_vs_ct_in_get(¶m); if (!ct || !ip_vs_check_template(ct, NULL)) { struct ip_vs_scheduler *sched; /* * No template found or the dest of the connection * template is not available. * return *ignored=0 i.e. ICMP and NF_DROP */ sched = rcu_dereference(svc->scheduler); if (sched) { /* read svc->sched_data after svc->scheduler */ smp_rmb(); dest = sched->schedule(svc, skb, iph); } else { dest = NULL; } if (!dest) { IP_VS_DBG(1, "p-schedule: no dest found.\n"); kfree(param.pe_data); *ignored = 0; return NULL; } if (dst_port == svc->port && svc->port != FTPPORT) dport = dest->port; /* Create a template * This adds param.pe_data to the template, * and thus param.pe_data will be destroyed * when the template expires */ ct = ip_vs_conn_new(¶m, dest->af, &dest->addr, dport, IP_VS_CONN_F_TEMPLATE, dest, skb->mark); if (ct == NULL) { kfree(param.pe_data); *ignored = -1; return NULL; } ct->timeout = svc->timeout; } else { /* set destination with the found template */ dest = ct->dest; kfree(param.pe_data); } dport = dst_port; if (dport == svc->port && dest->port) dport = dest->port; flags = (svc->flags & IP_VS_SVC_F_ONEPACKET && iph->protocol == IPPROTO_UDP) ? IP_VS_CONN_F_ONE_PACKET : 0; /* * Create a new connection according to the template */ ip_vs_conn_fill_param(svc->ipvs, svc->af, iph->protocol, src_addr, src_port, dst_addr, dst_port, ¶m); cp = ip_vs_conn_new(¶m, dest->af, &dest->addr, dport, flags, dest, skb->mark); if (cp == NULL) { ip_vs_conn_put(ct); *ignored = -1; return NULL; } /* * Add its control */ ip_vs_control_add(cp, ct); ip_vs_conn_put(ct); ip_vs_conn_stats(cp, svc); return cp; } /* * IPVS main scheduling function * It selects a server according to the virtual service, and * creates a connection entry. * Protocols supported: TCP, UDP * * Usage of *ignored * * 1 : protocol tried to schedule (eg. on SYN), found svc but the * svc/scheduler decides that this packet should be accepted with * NF_ACCEPT because it must not be scheduled. * * 0 : scheduler can not find destination, so try bypass or * return ICMP and then NF_DROP (ip_vs_leave). * * -1 : scheduler tried to schedule but fatal error occurred, eg. * ip_vs_conn_new failure (ENOMEM) or ip_vs_sip_fill_param * failure such as missing Call-ID, ENOMEM on skb_linearize * or pe_data. In this case we should return NF_DROP without * any attempts to send ICMP with ip_vs_leave. */ struct ip_vs_conn * ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, struct ip_vs_proto_data *pd, int *ignored, struct ip_vs_iphdr *iph) { struct ip_vs_protocol *pp = pd->pp; struct ip_vs_conn *cp = NULL; struct ip_vs_scheduler *sched; struct ip_vs_dest *dest; __be16 _ports[2], *pptr, cport, vport; const void *caddr, *vaddr; unsigned int flags; *ignored = 1; /* * IPv6 frags, only the first hit here. */ pptr = frag_safe_skb_hp(skb, iph->len, sizeof(_ports), _ports); if (pptr == NULL) return NULL; if (likely(!ip_vs_iph_inverse(iph))) { cport = pptr[0]; caddr = &iph->saddr; vport = pptr[1]; vaddr = &iph->daddr; } else { cport = pptr[1]; caddr = &iph->daddr; vport = pptr[0]; vaddr = &iph->saddr; } /* * FTPDATA needs this check when using local real server. * Never schedule Active FTPDATA connections from real server. * For LVS-NAT they must be already created. For other methods * with persistence the connection is created on SYN+ACK. */ if (cport == FTPDATA) { IP_VS_DBG_PKT(12, svc->af, pp, skb, iph->off, "Not scheduling FTPDATA"); return NULL; } /* * Do not schedule replies from local real server. */ if ((!skb->dev || skb->dev->flags & IFF_LOOPBACK)) { iph->hdr_flags ^= IP_VS_HDR_INVERSE; cp = INDIRECT_CALL_1(pp->conn_in_get, ip_vs_conn_in_get_proto, svc->ipvs, svc->af, skb, iph); iph->hdr_flags ^= IP_VS_HDR_INVERSE; if (cp) { IP_VS_DBG_PKT(12, svc->af, pp, skb, iph->off, "Not scheduling reply for existing" " connection"); __ip_vs_conn_put(cp); return NULL; } } /* * Persistent service */ if (svc->flags & IP_VS_SVC_F_PERSISTENT) return ip_vs_sched_persist(svc, skb, cport, vport, ignored, iph); *ignored = 0; /* * Non-persistent service */ if (!svc->fwmark && vport != svc->port) { if (!svc->port) pr_err("Schedule: port zero only supported " "in persistent services, " "check your ipvs configuration\n"); return NULL; } sched = rcu_dereference(svc->scheduler); if (sched) { /* read svc->sched_data after svc->scheduler */ smp_rmb(); dest = sched->schedule(svc, skb, iph); } else { dest = NULL; } if (dest == NULL) { IP_VS_DBG(1, "Schedule: no dest found.\n"); return NULL; } flags = (svc->flags & IP_VS_SVC_F_ONEPACKET && iph->protocol == IPPROTO_UDP) ? IP_VS_CONN_F_ONE_PACKET : 0; /* * Create a connection entry. */ { struct ip_vs_conn_param p; ip_vs_conn_fill_param(svc->ipvs, svc->af, iph->protocol, caddr, cport, vaddr, vport, &p); cp = ip_vs_conn_new(&p, dest->af, &dest->addr, dest->port ? dest->port : vport, flags, dest, skb->mark); if (!cp) { *ignored = -1; return NULL; } } IP_VS_DBG_BUF(6, "Schedule fwd:%c c:%s:%u v:%s:%u " "d:%s:%u conn->flags:%X conn->refcnt:%d\n", ip_vs_fwd_tag(cp), IP_VS_DBG_ADDR(cp->af, &cp->caddr), ntohs(cp->cport), IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport), IP_VS_DBG_ADDR(cp->daf, &cp->daddr), ntohs(cp->dport), cp->flags, refcount_read(&cp->refcnt)); ip_vs_conn_stats(cp, svc); return cp; } static inline int ip_vs_addr_is_unicast(struct net *net, int af, union nf_inet_addr *addr) { #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) return ipv6_addr_type(&addr->in6) & IPV6_ADDR_UNICAST; #endif return (inet_addr_type(net, addr->ip) == RTN_UNICAST); } /* * Pass or drop the packet. * Called by ip_vs_in, when the virtual service is available but * no destination is available for a new connection. */ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, struct ip_vs_proto_data *pd, struct ip_vs_iphdr *iph) { __be16 _ports[2], *pptr, dport; struct netns_ipvs *ipvs = svc->ipvs; struct net *net = ipvs->net; pptr = frag_safe_skb_hp(skb, iph->len, sizeof(_ports), _ports); if (!pptr) return NF_DROP; dport = likely(!ip_vs_iph_inverse(iph)) ? pptr[1] : pptr[0]; /* if it is fwmark-based service, the cache_bypass sysctl is up and the destination is a non-local unicast, then create a cache_bypass connection entry */ if (sysctl_cache_bypass(ipvs) && svc->fwmark && !(iph->hdr_flags & (IP_VS_HDR_INVERSE | IP_VS_HDR_ICMP)) && ip_vs_addr_is_unicast(net, svc->af, &iph->daddr)) { int ret; struct ip_vs_conn *cp; unsigned int flags = (svc->flags & IP_VS_SVC_F_ONEPACKET && iph->protocol == IPPROTO_UDP) ? IP_VS_CONN_F_ONE_PACKET : 0; union nf_inet_addr daddr = { .all = { 0, 0, 0, 0 } }; /* create a new connection entry */ IP_VS_DBG(6, "%s(): create a cache_bypass entry\n", __func__); { struct ip_vs_conn_param p; ip_vs_conn_fill_param(svc->ipvs, svc->af, iph->protocol, &iph->saddr, pptr[0], &iph->daddr, pptr[1], &p); cp = ip_vs_conn_new(&p, svc->af, &daddr, 0, IP_VS_CONN_F_BYPASS | flags, NULL, skb->mark); if (!cp) return NF_DROP; } /* statistics */ ip_vs_in_stats(cp, skb); /* set state */ ip_vs_set_state(cp, IP_VS_DIR_INPUT, skb, pd); /* transmit the first SYN packet */ ret = cp->packet_xmit(skb, cp, pd->pp, iph); /* do not touch skb anymore */ if ((cp->flags & IP_VS_CONN_F_ONE_PACKET) && cp->control) atomic_inc(&cp->control->in_pkts); else atomic_inc(&cp->in_pkts); ip_vs_conn_put(cp); return ret; } /* * When the virtual ftp service is presented, packets destined * for other services on the VIP may get here (except services * listed in the ipvs table), pass the packets, because it is * not ipvs job to decide to drop the packets. */ if (svc->port == FTPPORT && dport != FTPPORT) return NF_ACCEPT; if (unlikely(ip_vs_iph_icmp(iph))) return NF_DROP; /* * Notify the client that the destination is unreachable, and * release the socket buffer. * Since it is in IP layer, the TCP socket is not actually * created, the TCP RST packet cannot be sent, instead that * ICMP_PORT_UNREACH is sent here no matter it is TCP/UDP. --WZ */ #ifdef CONFIG_IP_VS_IPV6 if (svc->af == AF_INET6) { if (!skb->dev) skb->dev = net->loopback_dev; icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0); } else #endif icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); return NF_DROP; } #ifdef CONFIG_SYSCTL static int sysctl_snat_reroute(struct netns_ipvs *ipvs) { return ipvs->sysctl_snat_reroute; } static int sysctl_nat_icmp_send(struct netns_ipvs *ipvs) { return ipvs->sysctl_nat_icmp_send; } #else static int sysctl_snat_reroute(struct netns_ipvs *ipvs) { return 0; } static int sysctl_nat_icmp_send(struct netns_ipvs *ipvs) { return 0; } #endif __sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset) { return csum_fold(skb_checksum(skb, offset, skb->len - offset, 0)); } static inline enum ip_defrag_users ip_vs_defrag_user(unsigned int hooknum) { if (NF_INET_LOCAL_IN == hooknum) return IP_DEFRAG_VS_IN; if (NF_INET_FORWARD == hooknum) return IP_DEFRAG_VS_FWD; return IP_DEFRAG_VS_OUT; } static inline int ip_vs_gather_frags(struct netns_ipvs *ipvs, struct sk_buff *skb, u_int32_t user) { int err; local_bh_disable(); err = ip_defrag(ipvs->net, skb, user); local_bh_enable(); if (!err) ip_send_check(ip_hdr(skb)); return err; } static int ip_vs_route_me_harder(struct netns_ipvs *ipvs, int af, struct sk_buff *skb, unsigned int hooknum) { if (!sysctl_snat_reroute(ipvs)) return 0; /* Reroute replies only to remote clients (FORWARD and LOCAL_OUT) */ if (NF_INET_LOCAL_IN == hooknum) return 0; #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { struct dst_entry *dst = skb_dst(skb); if (dst->dev && !(dst->dev->flags & IFF_LOOPBACK) && ip6_route_me_harder(ipvs->net, skb->sk, skb) != 0) return 1; } else #endif if (!(skb_rtable(skb)->rt_flags & RTCF_LOCAL) && ip_route_me_harder(ipvs->net, skb->sk, skb, RTN_LOCAL) != 0) return 1; return 0; } /* * Packet has been made sufficiently writable in caller * - inout: 1=in->out, 0=out->in */ void ip_vs_nat_icmp(struct sk_buff *skb, struct ip_vs_protocol *pp, struct ip_vs_conn *cp, int inout) { struct iphdr *iph = ip_hdr(skb); unsigned int icmp_offset = iph->ihl*4; struct icmphdr *icmph = (struct icmphdr *)(skb_network_header(skb) + icmp_offset); struct iphdr *ciph = (struct iphdr *)(icmph + 1); if (inout) { iph->saddr = cp->vaddr.ip; ip_send_check(iph); ciph->daddr = cp->vaddr.ip; ip_send_check(ciph); } else { iph->daddr = cp->daddr.ip; ip_send_check(iph); ciph->saddr = cp->daddr.ip; ip_send_check(ciph); } /* the TCP/UDP/SCTP port */ if (IPPROTO_TCP == ciph->protocol || IPPROTO_UDP == ciph->protocol || IPPROTO_SCTP == ciph->protocol) { __be16 *ports = (void *)ciph + ciph->ihl*4; if (inout) ports[1] = cp->vport; else ports[0] = cp->dport; } /* And finally the ICMP checksum */ icmph->checksum = 0; icmph->checksum = ip_vs_checksum_complete(skb, icmp_offset); skb->ip_summed = CHECKSUM_UNNECESSARY; if (inout) IP_VS_DBG_PKT(11, AF_INET, pp, skb, (void *)ciph - (void *)iph, "Forwarding altered outgoing ICMP"); else IP_VS_DBG_PKT(11, AF_INET, pp, skb, (void *)ciph - (void *)iph, "Forwarding altered incoming ICMP"); } #ifdef CONFIG_IP_VS_IPV6 void ip_vs_nat_icmp_v6(struct sk_buff *skb, struct ip_vs_protocol *pp, struct ip_vs_conn *cp, int inout) { struct ipv6hdr *iph = ipv6_hdr(skb); unsigned int icmp_offset = 0; unsigned int offs = 0; /* header offset*/ int protocol; struct icmp6hdr *icmph; struct ipv6hdr *ciph; unsigned short fragoffs; ipv6_find_hdr(skb, &icmp_offset, IPPROTO_ICMPV6, &fragoffs, NULL); icmph = (struct icmp6hdr *)(skb_network_header(skb) + icmp_offset); offs = icmp_offset + sizeof(struct icmp6hdr); ciph = (struct ipv6hdr *)(skb_network_header(skb) + offs); protocol = ipv6_find_hdr(skb, &offs, -1, &fragoffs, NULL); if (inout) { iph->saddr = cp->vaddr.in6; ciph->daddr = cp->vaddr.in6; } else { iph->daddr = cp->daddr.in6; ciph->saddr = cp->daddr.in6; } /* the TCP/UDP/SCTP port */ if (!fragoffs && (IPPROTO_TCP == protocol || IPPROTO_UDP == protocol || IPPROTO_SCTP == protocol)) { __be16 *ports = (void *)(skb_network_header(skb) + offs); IP_VS_DBG(11, "%s() changed port %d to %d\n", __func__, ntohs(inout ? ports[1] : ports[0]), ntohs(inout ? cp->vport : cp->dport)); if (inout) ports[1] = cp->vport; else ports[0] = cp->dport; } /* And finally the ICMP checksum */ icmph->icmp6_cksum = ~csum_ipv6_magic(&iph->saddr, &iph->daddr, skb->len - icmp_offset, IPPROTO_ICMPV6, 0); skb->csum_start = skb_network_header(skb) - skb->head + icmp_offset; skb->csum_offset = offsetof(struct icmp6hdr, icmp6_cksum); skb->ip_summed = CHECKSUM_PARTIAL; if (inout) IP_VS_DBG_PKT(11, AF_INET6, pp, skb, (void *)ciph - (void *)iph, "Forwarding altered outgoing ICMPv6"); else IP_VS_DBG_PKT(11, AF_INET6, pp, skb, (void *)ciph - (void *)iph, "Forwarding altered incoming ICMPv6"); } #endif /* Handle relevant response ICMP messages - forward to the right * destination host. */ static int handle_response_icmp(int af, struct sk_buff *skb, union nf_inet_addr *snet, __u8 protocol, struct ip_vs_conn *cp, struct ip_vs_protocol *pp, unsigned int offset, unsigned int ihl, unsigned int hooknum) { unsigned int verdict = NF_DROP; if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ) goto after_nat; /* Ensure the checksum is correct */ if (!skb_csum_unnecessary(skb) && ip_vs_checksum_complete(skb, ihl)) { /* Failed checksum! */ IP_VS_DBG_BUF(1, "Forward ICMP: failed checksum from %s!\n", IP_VS_DBG_ADDR(af, snet)); goto out; } if (IPPROTO_TCP == protocol || IPPROTO_UDP == protocol || IPPROTO_SCTP == protocol) offset += 2 * sizeof(__u16); if (skb_ensure_writable(skb, offset)) goto out; #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) ip_vs_nat_icmp_v6(skb, pp, cp, 1); else #endif ip_vs_nat_icmp(skb, pp, cp, 1); if (ip_vs_route_me_harder(cp->ipvs, af, skb, hooknum)) goto out; after_nat: /* do the statistics and put it back */ ip_vs_out_stats(cp, skb); skb->ipvs_property = 1; if (!(cp->flags & IP_VS_CONN_F_NFCT)) ip_vs_notrack(skb); else ip_vs_update_conntrack(skb, cp, 0); verdict = NF_ACCEPT; out: __ip_vs_conn_put(cp); return verdict; } /* * Handle ICMP messages in the inside-to-outside direction (outgoing). * Find any that might be relevant, check against existing connections. * Currently handles error types - unreachable, quench, ttl exceeded. */ static int ip_vs_out_icmp(struct netns_ipvs *ipvs, struct sk_buff *skb, int *related, unsigned int hooknum) { struct iphdr *iph; struct icmphdr _icmph, *ic; struct iphdr _ciph, *cih; /* The ip header contained within the ICMP */ struct ip_vs_iphdr ciph; struct ip_vs_conn *cp; struct ip_vs_protocol *pp; unsigned int offset, ihl; union nf_inet_addr snet; *related = 1; /* reassemble IP fragments */ if (ip_is_fragment(ip_hdr(skb))) { if (ip_vs_gather_frags(ipvs, skb, ip_vs_defrag_user(hooknum))) return NF_STOLEN; } iph = ip_hdr(skb); offset = ihl = iph->ihl * 4; ic = skb_header_pointer(skb, offset, sizeof(_icmph), &_icmph); if (ic == NULL) return NF_DROP; IP_VS_DBG(12, "Outgoing ICMP (%d,%d) %pI4->%pI4\n", ic->type, ntohs(icmp_id(ic)), &iph->saddr, &iph->daddr); /* * Work through seeing if this is for us. * These checks are supposed to be in an order that means easy * things are checked first to speed up processing.... however * this means that some packets will manage to get a long way * down this stack and then be rejected, but that's life. */ if ((ic->type != ICMP_DEST_UNREACH) && (ic->type != ICMP_SOURCE_QUENCH) && (ic->type != ICMP_TIME_EXCEEDED)) { *related = 0; return NF_ACCEPT; } /* Now find the contained IP header */ offset += sizeof(_icmph); cih = skb_header_pointer(skb, offset, sizeof(_ciph), &_ciph); if (cih == NULL) return NF_ACCEPT; /* The packet looks wrong, ignore */ pp = ip_vs_proto_get(cih->protocol); if (!pp) return NF_ACCEPT; /* Is the embedded protocol header present? */ if (unlikely(cih->frag_off & htons(IP_OFFSET) && pp->dont_defrag)) return NF_ACCEPT; IP_VS_DBG_PKT(11, AF_INET, pp, skb, offset, "Checking outgoing ICMP for"); ip_vs_fill_iph_skb_icmp(AF_INET, skb, offset, true, &ciph); /* The embedded headers contain source and dest in reverse order */ cp = INDIRECT_CALL_1(pp->conn_out_get, ip_vs_conn_out_get_proto, ipvs, AF_INET, skb, &ciph); if (!cp) return NF_ACCEPT; snet.ip = iph->saddr; return handle_response_icmp(AF_INET, skb, &snet, cih->protocol, cp, pp, ciph.len, ihl, hooknum); } #ifdef CONFIG_IP_VS_IPV6 static int ip_vs_out_icmp_v6(struct netns_ipvs *ipvs, struct sk_buff *skb, int *related, unsigned int hooknum, struct ip_vs_iphdr *ipvsh) { struct icmp6hdr _icmph, *ic; struct ip_vs_iphdr ciph = {.flags = 0, .fragoffs = 0};/*Contained IP */ struct ip_vs_conn *cp; struct ip_vs_protocol *pp; union nf_inet_addr snet; unsigned int offset; *related = 1; ic = frag_safe_skb_hp(skb, ipvsh->len, sizeof(_icmph), &_icmph); if (ic == NULL) return NF_DROP; /* * Work through seeing if this is for us. * These checks are supposed to be in an order that means easy * things are checked first to speed up processing.... however * this means that some packets will manage to get a long way * down this stack and then be rejected, but that's life. */ if (ic->icmp6_type & ICMPV6_INFOMSG_MASK) { *related = 0; return NF_ACCEPT; } /* Fragment header that is before ICMP header tells us that: * it's not an error message since they can't be fragmented. */ if (ipvsh->flags & IP6_FH_F_FRAG) return NF_DROP; IP_VS_DBG(8, "Outgoing ICMPv6 (%d,%d) %pI6c->%pI6c\n", ic->icmp6_type, ntohs(icmpv6_id(ic)), &ipvsh->saddr, &ipvsh->daddr); if (!ip_vs_fill_iph_skb_icmp(AF_INET6, skb, ipvsh->len + sizeof(_icmph), true, &ciph)) return NF_ACCEPT; /* The packet looks wrong, ignore */ pp = ip_vs_proto_get(ciph.protocol); if (!pp) return NF_ACCEPT; /* The embedded headers contain source and dest in reverse order */ cp = INDIRECT_CALL_1(pp->conn_out_get, ip_vs_conn_out_get_proto, ipvs, AF_INET6, skb, &ciph); if (!cp) return NF_ACCEPT; snet.in6 = ciph.saddr.in6; offset = ciph.len; return handle_response_icmp(AF_INET6, skb, &snet, ciph.protocol, cp, pp, offset, sizeof(struct ipv6hdr), hooknum); } #endif /* * Check if sctp chunc is ABORT chunk */ static inline int is_sctp_abort(const struct sk_buff *skb, int nh_len) { struct sctp_chunkhdr *sch, schunk; sch = skb_header_pointer(skb, nh_len + sizeof(struct sctphdr), sizeof(schunk), &schunk); if (sch == NULL) return 0; if (sch->type == SCTP_CID_ABORT) return 1; return 0; } static inline int is_tcp_reset(const struct sk_buff *skb, int nh_len) { struct tcphdr _tcph, *th; th = skb_header_pointer(skb, nh_len, sizeof(_tcph), &_tcph); if (th == NULL) return 0; return th->rst; } static inline bool is_new_conn(const struct sk_buff *skb, struct ip_vs_iphdr |