FuzzBench: 2023-06-22-libafl report

experiment summary

We show two different aggregate (cross-benchmark) rankings of fuzzers. The first is based on the average of per-benchmarks scores, where the score represents the percentage of the highest reached median code-coverage on a given benchmark (higher value is better). The second ranking shows the average rank of fuzzers, after we rank them on each benchmark according to their median reached code-covereges (lower value is better).

By avg. score

	average normalized score
fuzzer
aflplusplus	97.61
libafl	95.84
libafl_fuzzbench_grimoire	95.65
libfuzzer	89.01
mopt	84.23
eclipser	84.13
aflsmart	83.47
aflfast	81.82
honggfuzz	81.57
fairfuzz	79.29
centipede	67.91
afl	7.53

By avg. rank

	average rank
fuzzer
aflplusplus	2.39
libafl	4.43
libafl_fuzzbench_grimoire	4.74
aflsmart	5.48
eclipser	5.48
honggfuzz	5.48
libfuzzer	6.22
mopt	6.39
aflfast	8.09
centipede	8.22
fairfuzz	8.26
afl	11.04

Critical difference diagram

The diagram visualizes the average rank of fuzzers (second ranking above) while showing the significance of the differences as well. What is considered a "critical difference" (CD) is based on the Friedman/Nemenyi post-hoc test. See more in the documentation.

Note: If a fuzzer does not support all benchmarks, its ranking as shown in this diagram can be lower than it should be. So please check the list of supported benchmarks for the fuzzer(s) of your interest. The list could be specified in the fuzzer's README.md like this.

Median relative code-coverages on each benchmark

Note: The relative coverage summary table shows the median relative performance of each fuzzer to the experiment maximum. Thus the highest relative performance may not be 100%.
trial_relative_coverage = trial_coverage / experiment_max_coverage

	aflplusplus	libafl	libafl_fuzzbench_grimoire	honggfuzz	eclipser	libfuzzer	afl	mopt	aflsmart	aflfast	centipede	fairfuzz
FuzzerMedian	98.00	96.00	95.00	96.50	94.00	86.00	83.50	93.00	94.00	93.00	90.00	85.00
FuzzerMean	94.22	92.26	92.04	90.30	88.71	85.91	83.50	81.43	80.70	79.26	79.00	76.65
bloaty_fuzz_target	98.00	98.00	76.00	94.00	94.00	84.00	nan	96.00	95.00	93.00	nan	80.00
curl_curl_fuzzer_http	98.00	96.00	98.00	98.00	94.00	86.00	nan	94.00	94.00	93.00	nan	84.00
freetype2_ftfuzzer	91.00	85.00	79.00	87.00	71.00	67.00	nan	64.00	63.00	60.00	54.00	59.00
harfbuzz_hb-shape-fuzzer	98.00	99.00	83.00	96.00	97.00	86.00	nan	97.00	97.00	96.00	nan	87.00
jsoncpp_jsoncpp_fuzzer	99.00	98.00	98.00	99.00	98.00	100.00	nan	98.00	98.00	98.00	98.00	98.00
lcms_cms_transform_fuzzer	88.00	90.00	89.00	76.00	78.00	80.00	nan	62.00	40.00	28.00	38.00	50.00
libjpeg-turbo_libjpeg_turbo_fuzzer	99.00	82.00	98.00	82.00	nan	82.00	nan	82.00	82.00	82.00	82.00	82.00
libpcap_fuzz_both	93.00	85.00	88.00	nan	80.00	80.00	nan	1.00	1.00	1.00	90.00	1.00
libpng_libpng_read_fuzzer	96.00	95.00	94.00	96.00	95.00	96.00	nan	95.00	95.00	94.00	96.00	95.00
libxml2_xml	99.00	98.00	96.00	98.00	97.00	92.00	nan	97.00	97.00	97.00	93.00	89.00
libxslt_xpath	99.00	96.00	98.00	nan	96.00	78.00	nan	93.00	96.00	94.00	94.00	97.00
mbedtls_fuzz_dtlsclient	72.00	76.00	95.00	70.00	71.00	70.00	70.00	70.00	70.00	68.00	68.00	72.00
openh264_decoder_fuzzer	99.00	99.00	98.00	99.00	99.00	96.00	nan	99.00	99.00	99.00	nan	90.00
openssl_x509	99.00	99.00	99.00	99.00	99.00	99.00	nan	99.00	99.00	99.00	99.00	99.00
openthread_ot-ip6-send-fuzzer	67.00	71.00	78.00	66.00	66.00	68.00	nan	64.00	63.00	63.00	63.00	60.00
proj4_proj_crs_to_crs_fuzzer	87.00	87.00	93.00	97.00	59.00	90.00	nan	9.00	9.00	9.00	10.00	9.00
re2_fuzzer	99.00	98.00	99.00	98.00	99.00	99.00	nan	99.00	99.00	99.00	96.00	99.00
sqlite3_ossfuzz	98.00	89.00	79.00	68.00	93.00	58.00	nan	92.00	93.00	90.00	64.00	65.00
stb_stbi_read_fuzzer	96.00	95.00	95.00	91.00	91.00	85.00	nan	86.00	87.00	86.00	85.00	86.00
systemd_fuzz-link-parser	99.00	98.00	98.00	nan	91.00	97.00	nan	90.00	90.00	91.00	97.00	85.00
vorbis_decode_fuzzer	99.00	98.00	98.00	98.00	98.00	99.00	nan	98.00	98.00	98.00	90.00	98.00
woff2_convert_woff2ttf_fuzzer	97.00	96.00	93.00	97.00	nan	86.00	nan	92.00	94.00	90.00	88.00	81.00
zlib_zlib_uncompress_fuzzer	97.00	94.00	95.00	97.00	97.00	98.00	97.00	96.00	97.00	95.00	96.00	97.00

Fuzzers are sorted by "FuzzerMean" (average median relative coverage), highest on the left.
Green background = highest relative median coverage.
Blue gradient background = greater than 95% relative median coverage.

bloaty_fuzz_target summary

Ranking by median reached code coverage

Mean code coverage growth over time

* The error bands show the 95% confidence interval around the mean code coverage.

Sample statistics and statistical significance (code coverage)

Code coverage sample statistics

		count	mean	std	min	25%	median	75%	max
fuzzer	time
libafl	82800	20.0	6353.80	71.997515	6140.0	6344.75	6371.5	6388.00	6442.0
aflplusplus	82800	20.0	6349.55	80.970739	6190.0	6275.25	6371.0	6406.25	6481.0
mopt	82800	20.0	6236.05	71.516266	6071.0	6226.50	6243.5	6268.50	6354.0
aflsmart	82800	20.0	6198.45	127.984981	5937.0	6102.25	6205.0	6295.25	6394.0
honggfuzz	82800	20.0	6103.85	168.969719	5769.0	5961.25	6121.0	6260.00	6373.0
eclipser	82800	20.0	6110.20	114.487232	5900.0	6066.75	6107.0	6144.50	6360.0
aflfast	82800	20.0	6080.10	119.230824	5858.0	6033.50	6059.0	6136.50	6350.0
libfuzzer	82800	20.0	5482.00	108.839331	5188.0	5405.25	5501.0	5545.25	5689.0
fairfuzz	82800	20.0	5217.75	121.813868	5060.0	5145.00	5194.5	5272.75	5570.0
libafl_fuzzbench_grimoire	82800	20.0	4984.15	34.891222	4898.0	4968.75	4980.0	4996.75	5065.0

Vargha-Delaney A12 measure

The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.

Mann-Whitney U test

The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.

Unique code coverage plots

Ranking by unique code branches covered
Each bar shows the total number of code branches found by a given fuzzer. The colored area shows the number of unique code branches (i.e., branches that were not covered by any other fuzzers).

Pairwise unique code coverage
Each cell represents the number of code branches covered by the fuzzer of the column but not by the fuzzer of the row

Code coverage reports for each fuzzer on this benchmark

afl aflfast aflplusplus aflsmart centipede eclipser fairfuzz honggfuzz libafl libafl_fuzzbench_grimoire libfuzzer mopt