CHECK failure: false. GL error NUMBER was unhandled in error_state.cc |
||||||
Issue descriptionDetailed report: https://clusterfuzz.com/testcase?key=6356001052950528 Fuzzer: libFuzzer_gpu_angle_fuzzer Job Type: libfuzzer_chrome_asan_debug Platform Id: linux Crash Type: CHECK failure Crash Address: Crash State: false. GL error NUMBER was unhandled in error_state.cc gpu::gles2::ErrorStateImpl::ClearRealGLErrors gpu::gles2::ScopedGLErrorSuppressor::~ScopedGLErrorSuppressor Sanitizer: address (ASAN) Reproducer Testcase: https://clusterfuzz.com/download?testcase_id=6356001052950528 Issue manually filed by: piman See https://chromium.googlesource.com/chromium/src/+/master/testing/libfuzzer/reference.md for more information. Note: This crash might not be reproducible with the provided testcase. That said, for the past 14 days we've been seeing this crash frequently. If you are unable to reproduce this, please try a speculative fix based on the crash stacktrace in the report. The fix can be verified by looking at the crash statistics in the report, a day after the fix is deployed. We will auto-close the bug if the crash is not seen for 14 days.
,
Nov 10 2017
Redo Task has been performed for regression range. Thank You.
,
Nov 10 2017
Looks like boliu@ last modified GLES2DecoderImpl::ClearFramebufferForWorkaround. Any thoughts on this?
,
Nov 13 2017
Current log says the error is GL_INVALID_OPERATION, from the GL clear workaround blit. It's crashing on a NOTREACHED: https://chromium.googlesource.com/chromium/src/+/1af4b9e0047dd97ba32c26c111d631c1566e0715/gpu/command_buffer/service/error_state.cc#215 meaning asan bot enabled DCHECKs Assign this to kbr since this is desktop. A few things to consider: * should asan bot even be building with DCHECKs enabled? seems pretty brittle * should that NOTREACHED just be removed? doesn't seem to be service much since error is already logged * clusterfuzz added Unreproducible label, so presumably this was transient, and will automatically closed in awhile anyway
,
Nov 14 2017
Yes, we should build with DCHECKs enabled everywhere we can, to catch errors earlier. This is happening with a test case from the libFuzzer_gpu_angle_fuzzer. Geoff, can you help diagnose this? Theoretically it's happening on Linux.
,
Nov 14 2017
I'll be able to take a look at this when I get back to my office. There have been some recent changes with fuzzer GL error handling so this may not be relevant anymore.
,
Nov 22 2017
ClusterFuzz testcase 6356001052950528 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 ClusterFuzz
, Nov 10 2017Labels: Test-Predator-Auto-Components