Timeout in mediasource_WEBM_OPUS_VP9_pipeline_integration_fuzzer |
||||||
Issue descriptionDetailed report: https://clusterfuzz.com/testcase?key=5349030104924160 Fuzzer: libFuzzer_mediasource_WEBM_OPUS_VP9_pipeline_integration_fuzzer Job Type: libfuzzer_chrome_msan Platform Id: linux Crash Type: Timeout (exceeds 25 secs) Crash Address: Crash State: mediasource_WEBM_OPUS_VP9_pipeline_integration_fuzzer Sanitizer: memory (MSAN) Regressed: https://clusterfuzz.com/revisions?job=libfuzzer_chrome_msan&range=519029:519040 Reproducer Testcase: https://clusterfuzz.com/download?testcase_id=5349030104924160 Issue filed automatically. See https://chromium.googlesource.com/chromium/src/+/master/testing/libfuzzer/reference.md for more information.
,
Jul 19
,
Jul 19
,
Jul 24
ClusterFuzz testcase 5349030104924160 appears to be flaky, updating reproducibility label.
,
Jul 27
,
Jul 27
Nothing to really do here, there are 12000 frames and my desktop manages ~1000fps, probably bots may be slower at some point. +mmoroz if he'd rather we abort the test for clips beyond a certain duration.
,
Jul 29
If there is an easy way to reject such inputs, it would be nice, but WontFix is also fine, as this seems to be a fairly rare problem as per crash stats.
,
Jul 30
I think we can check duration before starting preroll, but if the fuzzer ends up fuzzing something outside of our allowed duration, that's something an external reporter might find. Is that really an okay compromise?
,
Jul 30
Good point! If you see a potential of finding security bugs that aren't reachable with a constrained duration, then we definitely shouldn't do that. Thanks for clarifying that, Dale! |
||||||
►
Sign in to add a comment |
||||||
Comment 1 by kkaluri@chromium.org
, Jul 19Components: Internals>Media>Source
Labels: M-68 Test-Predator-Wrong CF-NeedsTriage