FuzzBench: 2023-01-15-latest-cov 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
libafl	99.12
libfuzzer	93.92
honggfuzz	91.78
afl	88.09
aflfast	86.76
fairfuzz	85.14
mopt	83.60
eclipser	71.52
centipede	62.55
aflsmart	22.01

By avg. rank

	average rank
fuzzer
libafl	2.76
honggfuzz	3.57
libfuzzer	3.71
afl	4.86
mopt	5.05
eclipser	5.24
fairfuzz	6.48
aflfast	6.76
centipede	7.19
aflsmart	7.86

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

	libafl	honggfuzz	aflsmart	libfuzzer	eclipser	afl	mopt	aflfast	centipede	fairfuzz
FuzzerMedian	98.00	96.00	96.00	93.00	94.50	95.00	95.50	93.00	89.00	89.00
FuzzerMean	94.81	92.05	90.40	89.71	89.12	84.86	84.35	83.57	82.47	81.90
bloaty_fuzz_target	99.00	94.00	96.00	89.00	94.00	94.00	96.00	93.00	nan	79.00
curl_curl_fuzzer_http	98.00	99.00	94.00	91.00	94.00	94.00	94.00	93.00	nan	84.00
freetype2_ftfuzzer	91.00	95.00	69.00	78.00	77.00	69.00	68.00	66.00	59.00	65.00
harfbuzz_hb-shape-fuzzer	99.00	96.00	96.00	94.00	96.00	96.00	96.00	95.00	nan	90.00
jsoncpp_jsoncpp_fuzzer	98.00	99.00	nan	100.00	98.00	98.00	98.00	98.00	98.00	98.00
lcms_cms_transform_fuzzer	95.00	80.00	nan	86.00	75.00	40.00	40.00	28.00	36.00	69.00
libjpeg-turbo_libjpeg_turbo_fuzzer	99.00	99.00	nan	99.00	nan	99.00	99.00	99.00	95.00	99.00
libpcap_fuzz_both	84.00	84.00	nan	76.00	73.00	1.00	1.00	1.00	85.00	1.00
libpng_libpng_read_fuzzer	94.00	96.00	nan	96.00	95.00	95.00	95.00	93.00	96.00	94.00
libxml2_xml	99.00	99.00	nan	98.00	97.00	97.00	97.00	97.00	93.00	90.00
libxslt_xpath	97.00	99.00	nan	93.00	96.00	95.00	nan	95.00	nan	97.00
mbedtls_fuzz_dtlsclient	81.00	71.00	nan	71.00	71.00	70.00	70.00	68.00	69.00	73.00
openssl_x509	99.00	99.00	nan	99.00	99.00	99.00	99.00	99.00	99.00	99.00
openthread_ot-ip6-send-fuzzer	77.00	77.00	nan	77.00	75.00	73.00	73.00	72.00	72.00	69.00
re2_fuzzer	99.00	98.00	nan	99.00	nan	99.00	98.00	99.00	95.00	99.00
sqlite3_ossfuzz	95.00	72.00	97.00	84.00	96.00	97.00	97.00	96.00	67.00	67.00
stb_stbi_read_fuzzer	98.00	94.00	nan	90.00	nan	89.00	89.00	89.00	88.00	89.00
systemd_fuzz-link-parser	99.00	97.00	nan	72.00	92.00	91.00	91.00	91.00	nan	84.00
vorbis_decode_fuzzer	98.00	nan	nan	99.00	98.00	98.00	98.00	98.00	nan	95.00
woff2_convert_woff2ttf_fuzzer	97.00	96.00	nan	95.00	nan	91.00	92.00	90.00	89.00	82.00
zlib_zlib_uncompress_fuzzer	95.00	97.00	nan	98.00	nan	97.00	96.00	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.

The following fuzzers do not have enough samples: honggfuzz, libfuzzer, mopt, eclipser, libafl.

Sample statistics and statistical significance (code coverage)

Code coverage sample statistics

		count	mean	std	min	25%	median	75%	max
fuzzer	time
libafl	82800	11.0	6388.090909	42.117584	6315.0	6364.00	6390.0	6408.00	6454.0
mopt	82800	12.0	6255.083333	117.571223	6057.0	6211.00	6253.0	6334.00	6450.0
aflsmart	82800	19.0	6174.684211	126.515283	5952.0	6056.00	6200.0	6261.00	6396.0
honggfuzz	82800	15.0	6089.333333	153.275973	5878.0	5932.50	6120.0	6177.50	6359.0
afl	82800	17.0	6113.470588	67.120337	6043.0	6068.00	6088.0	6115.00	6268.0
eclipser	82800	11.0	6053.181818	68.930136	5914.0	6022.50	6078.0	6095.50	6135.0
aflfast	82800	18.0	6039.277778	126.970031	5799.0	5992.50	6025.5	6078.00	6355.0
libfuzzer	82800	12.0	5792.166667	150.381233	5524.0	5735.75	5805.0	5893.25	6018.0
fairfuzz	82800	19.0	5153.947368	65.586646	5007.0	5133.00	5148.0	5195.00	5279.0

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 aflsmart centipede eclipser fairfuzz honggfuzz libafl libfuzzer mopt