FuzzBench: 2023-09-04-libafl-2 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_fuzzbench_ngram4	85.90
aflplusplus	81.27
libafl_fuzzbench_naive	80.27
libafl_fuzzbench_ngram8	79.45
libafl_fuzzbench_naive_ctx	79.19
libfuzzer	17.37
afl	12.11
honggfuzz	11.76
centipede	0.32
aflfast	0.00
aflsmart	0.00
eclipser	0.00
fairfuzz	0.00
mopt	0.00

By avg. rank

	average rank
fuzzer
aflplusplus	2.06
libafl_fuzzbench_ngram4	2.76
libafl_fuzzbench_ngram8	3.00
libafl_fuzzbench_naive	3.18
libafl_fuzzbench_naive_ctx	3.94
honggfuzz	5.24
libfuzzer	5.24
afl	5.53
centipede	5.65
aflfast	5.71
aflsmart	5.71
eclipser	5.71
fairfuzz	5.71
mopt	5.71

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_fuzzbench_ngram4	libafl_fuzzbench_naive	aflplusplus	libafl_fuzzbench_ngram8	libafl_fuzzbench_naive_ctx	libfuzzer	afl	honggfuzz	centipede	aflfast	aflsmart	eclipser	fairfuzz	mopt
FuzzerMedian	98.00	98.50	99.00	98.50	97.00	95.00	47.00	48.00	5.00	0.00	0.00	0.00	0.00	0.00
FuzzerMean	93.80	93.71	93.57	93.00	92.50	80.67	50.67	48.67	5.00	0.00	0.00	0.00	0.00	0.00
arduinojson_json_fuzzer	99.00	99.00	100.00	100.00	99.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
assimp_assimp_fuzzer	68.00	67.00	nan	61.00	65.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
astc-encoder_fuzz_astc_physical_to_symbolic	95.00	95.00	nan	95.00	95.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
botan_tls_server	nan	nan	48.00	nan	nan	48.00	47.00	48.00	nan	nan	nan	nan	nan	nan
brotli_decode_fuzzer	99.00	99.00	nan	99.00	nan	nan	nan	nan	nan	nan	nan	nan	nan	nan
double-conversion_string_to_double_fuzzer	97.00	97.00	99.00	98.00	96.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
draco_draco_pc_decoder_fuzzer	70.00	75.00	92.00	72.00	70.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
dropbear_fuzzer-postauth_nomaths	nan	nan	96.00	nan	nan	95.00	78.00	98.00	nan	nan	nan	nan	nan	nan
firestore_firestore_serializer_fuzzer	100.00	100.00	100.00	100.00	100.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
fmt_chrono-duration-fuzzer	98.00	98.00	99.00	94.00	97.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
guetzli_guetzli_fuzzer	97.00	99.00	99.00	96.00	97.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
icu_unicode_string_codepage_create_fuzzer	99.00	99.00	99.00	99.00	99.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
jansson_json_load_dump_fuzzer	99.00	99.00	99.00	99.00	99.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
libaom_av1_dec_fuzzer	96.00	96.00	97.00	nan	91.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
libcoap_pdu_parse_fuzzer	91.00	90.00	99.00	91.00	89.00	nan	nan	nan	nan	nan	nan	nan	nan	nan
libhevc_hevc_dec_fuzzer	99.00	99.00	99.00	99.00	99.00	nan	nan	nan	5.00	0.00	0.00	0.00	0.00	0.00
librdkafka_fuzz_regex	100.00	nan	84.00	99.00	99.00	99.00	27.00	0.00	nan	nan	nan	nan	nan	nan

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.

arduinojson_json_fuzzer 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
aflplusplus	82800	20.0	401.60	0.502625	401.0	401.00	402.0	402.00	402.0
libafl_fuzzbench_ngram8	82800	20.0	401.75	0.444262	401.0	401.75	402.0	402.00	402.0
libafl_fuzzbench_ngram4	82800	20.0	400.75	0.910465	399.0	400.00	401.0	401.25	402.0
libafl_fuzzbench_naive_ctx	82800	20.0	400.05	1.145931	398.0	399.00	400.0	401.00	402.0
libafl_fuzzbench_naive	82800	20.0	398.45	1.099043	397.0	398.00	398.0	399.00	401.0

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 aflplusplus libafl_fuzzbench_naive libafl_fuzzbench_naive_ctx libafl_fuzzbench_ngram4 libafl_fuzzbench_ngram8 libfuzzer