New issue
Advanced search Search tips

Issue 908754 link

Starred by 2 users

Issue metadata

Status: Assigned
Owner:
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 2
Type: Bug



Sign in to add a comment

Timeout in dawn_spirv_cross_hlsl_fast_fuzzer

Project Member Reported by ClusterFuzz, Nov 27

Issue description

Detailed report: https://clusterfuzz.com/testcase?key=5987250992840704

Fuzzer: libFuzzer_dawn_spirv_cross_hlsl_fast_fuzzer
Job Type: windows_libfuzzer_chrome_asan
Platform Id: windows

Crash Type: Timeout (exceeds 25 secs)
Crash Address: 
Crash State:
  dawn_spirv_cross_hlsl_fast_fuzzer
  
Sanitizer: address (ASAN)

Regressed: https://clusterfuzz.com/revisions?job=windows_libfuzzer_chrome_asan&range=609745:609746

Reproducer Testcase: https://clusterfuzz.com/download?testcase_id=5987250992840704

Issue filed automatically.

See https://chromium.googlesource.com/chromium/src/+/master/testing/libfuzzer/reproducing_on_windows.md for more information.
 
Cc: kkaluri@chromium.org
Labels: M-72 Test-Predator-Wrong CF-NeedsTriage
Unable to find actual suspect through code search and also observing no CL's under regression range, hence adding appropriate label and requesting someone from dev team to look in to this issue.

Thanks!
Components: Internals>GPU>Dawn
Labels: -CF-NeedsTriage
Owner: rharrison@chromium.org
Status: Assigned (was: Untriaged)
rharrison@, just wondering do you have any inputs here?
Project Member

Comment 3 by ClusterFuzz, Dec 1

Labels: -Reproducible Unreproducible
ClusterFuzz testcase 5987250992840704 appears to be flaky, updating reproducibility label.
Labels: -Unreproducible Reproducible
Please ignore the last comment about testcase being unreproducible. The testcase is still reproducible. This happened due to a code refactoring on ClusterFuzz side, and the underlying root cause is now fixed. Resetting the label back to Reproducible. Sorry about the inconvenience caused from these incorrect notifications.
Labels: -Pri-1 -M-72 Pri-2
This is another issue related to the fact SPIRV-cross allocates storage for all potential ids during initialization. This normally expresses itself as a memory usage limit issue, but in this case it is timing out before it can overrun its memory limit.

This will either be resolved by performing some sort of optimization in the new shaderc wrapper for this code or directly resolving https://github.com/KhronosGroup/SPIRV-Cross/issues/781.
The documentation for reproducing on Windows has been moved to https://chromium.googlesource.com/chromium/src/+/master/testing/libfuzzer/reproducing.md

Sign in to add a comment