FuzzBench: 2023-10-05-fafuzz 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_407	95.23
libafl	95.12
aflplusplus	94.44
fafuzz	94.11
honggfuzz	92.09
mopt	90.77
afl	90.59
aflsmart	89.95
libfuzzer	89.88
aflfast	88.62
fairfuzz	83.83
eclipser	77.73
centipede	65.42

By avg. rank

	average rank
fuzzer
aflplusplus_407	2.57
aflplusplus	3.07
libafl	4.36
aflsmart	5.86
libfuzzer	5.93
eclipser	6.43
mopt	7.07
honggfuzz	7.14
afl	7.43
fafuzz	7.93
aflfast	9.50
fairfuzz	9.93
centipede	10.64

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_407	libafl	aflplusplus	fafuzz	honggfuzz	mopt	afl	eclipser	aflsmart	libfuzzer	aflfast	fairfuzz	centipede
FuzzerMedian	98.00	97.50	98.00	94.00	95.50	93.00	93.00	93.50	93.00	92.00	92.00	82.50	85.00
FuzzerMean	92.29	91.86	91.43	91.43	89.21	88.00	87.79	87.75	87.21	87.00	85.86	81.14	80.27
bloaty_fuzz_target	97.00	98.00	98.00	95.00	94.00	96.00	94.00	94.00	96.00	91.00	93.00	80.00	nan
curl_curl_fuzzer_http	98.00	97.00	98.00	92.00	98.00	93.00	93.00	94.00	94.00	91.00	93.00	83.00	nan
freetype2_ftfuzzer	94.00	92.00	90.00	67.00	91.00	66.00	66.00	74.00	66.00	78.00	63.00	62.00	57.00
harfbuzz_hb-shape-fuzzer	98.00	99.00	98.00	96.00	95.00	96.00	96.00	97.00	97.00	94.00	95.00	87.00	nan
jsoncpp_jsoncpp_fuzzer	99.00	98.00	99.00	98.00	99.00	98.00	98.00	98.00	98.00	100.00	98.00	98.00	98.00
libjpeg-turbo_libjpeg_turbo_fuzzer	99.00	99.00	99.00	99.00	99.00	99.00	99.00	nan	82.00	99.00	82.00	82.00	82.00
mbedtls_fuzz_dtlsclient	74.00	83.00	75.00	72.00	72.00	72.00	72.00	73.00	73.00	72.00	71.00	75.00	71.00
openssl_x509	99.00	99.00	99.00	99.00	99.00	99.00	99.00	99.00	99.00	99.00	99.00	99.00	99.00
re2_fuzzer	99.00	98.00	99.00	98.00	98.00	98.00	98.00	99.00	99.00	99.00	99.00	99.00	95.00
sqlite3_ossfuzz	98.00	90.00	89.00	90.00	71.00	92.00	92.00	93.00	93.00	80.00	90.00	65.00	64.00
systemd_fuzz-link-parser	99.00	98.00	99.00	91.00	97.00	91.00	91.00	92.00	91.00	80.00	91.00	85.00	97.00
vorbis_decode_fuzzer	94.00	93.00	94.00	93.00	93.00	93.00	93.00	93.00	93.00	94.00	93.00	93.00	85.00
woff2_convert_woff2ttf_fuzzer	97.00	96.00	96.00	91.00	96.00	92.00	91.00	nan	93.00	93.00	89.00	81.00	88.00
zlib_zlib_uncompress_fuzzer	47.00	46.00	47.00	99.00	47.00	47.00	47.00	47.00	47.00	48.00	46.00	47.00	47.00

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	18.0	6370.000000	76.211007	6139.0	6333.25	6382.0	6413.00	6460.0
aflplusplus	82800	20.0	6356.000000	57.764495	6145.0	6345.50	6358.5	6393.00	6415.0
aflplusplus_407	82800	20.0	6310.800000	97.262315	6115.0	6245.75	6320.5	6393.75	6449.0
mopt	82800	20.0	6197.850000	59.793746	6022.0	6164.75	6211.0	6240.25	6277.0
aflsmart	82800	20.0	6198.450000	127.984981	5937.0	6102.25	6205.0	6295.25	6394.0
fafuzz	82800	18.0	6162.388889	82.742790	5985.0	6120.75	6146.0	6209.50	6344.0
honggfuzz	82800	20.0	6100.500000	209.354522	5744.0	5948.25	6131.5	6295.50	6391.0
eclipser	82800	20.0	6110.200000	114.487232	5900.0	6066.75	6107.0	6144.50	6360.0
afl	82800	19.0	6072.631579	136.266981	5854.0	5987.00	6073.0	6128.00	6337.0
aflfast	82800	20.0	6080.100000	119.230824	5858.0	6033.50	6059.0	6136.50	6350.0
libfuzzer	82800	20.0	5897.050000	169.163414	5559.0	5805.00	5890.5	5988.00	6182.0
fairfuzz	82800	20.0	5217.750000	121.813868	5060.0	5145.00	5194.5	5272.75	5570.0

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 aflplusplus_407 aflsmart centipede eclipser fafuzz fairfuzz honggfuzz libafl libfuzzer mopt