CHECK failure: false. Non-finite (i.e. NaN or positive/negative infinity) values cannot be repr |
|||||
Issue descriptionDetailed report: https://clusterfuzz.com/testcase?key=6448633629900800 Fuzzer: inferno_flicker Job Type: linux_debug_chrome Platform Id: linux Crash Type: CHECK failure Crash Address: Crash State: false. Non-finite (i.e. NaN or positive/negative infinity) values cannot be repr base::Value::Value media::MediaLog::CreateSeekEvent Sanitizer: address (ASAN) Reproducer Testcase: https://clusterfuzz.com/download?testcase_id=6448633629900800 Issue filed automatically. See https://github.com/google/clusterfuzz-tools for more information.
,
Aug 13
At first glance, this isn't MSE; it's using multibuffer datasource, so src=. Seems the WMPI::Seek() might be called with an out-of-range double. I'll take a closer look to see if I can find the problem.
,
Aug 13
Internally, WMPI::Seek(+Infinity) is leading to this DCHECK failure. Looks like a DCHECK in values.cc on attempt to MediaLog a non-finite double value, the seek target time. +Mounir: Is this allowed per spec? If so, what should the practical resulting behavior be? HTMLMediaElement.currentTime = +Infinity; /* or -Infinity or NaN */
,
Aug 13
There was a bug on file for this that was auto-fixed because ClusterFuzz couldn't repro onymore. The Web API wouldn't allow Infinity. The only place where WMPI::Seek() is called from HTMLMediaElement is from HTMLMediaElemnet::Seek(). There are no obvious situations in which Infinity in used in this code. I will see later if I can repro locally.
,
Aug 14
ClusterFuzz testcase 6448633629900800 appears to be flaky, updating reproducibility label.
,
Aug 27
ClusterFuzz testcase 6448633629900800 is flaky and no longer crashes, so closing issue. If this is incorrect, please add ClusterFuzz-Wrong label and re-open the issue. |
|||||
►
Sign in to add a comment |
|||||
Comment 1 by kkaluri@chromium.org
, Aug 13Components: Internals>Media>Source
Labels: M-69 Test-Predator-Wrong CF-NeedsTriage