New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 707474 link

Starred by 3 users

Issue metadata

Status: Fixed
Owner:
Closed: Apr 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Mac
Pri: 1
Type: Bug-Regression

Blocked on:
issue 705156
issue 707473

Blocking:
issue 531673



Sign in to add a comment

Random WebGL 2.0 conformance failures on Mac Intel

Project Member Reported by kbr@chromium.org, Apr 1 2017

Issue description

Random WebGL 2.0 conformance failures are being seen on the Mac Intel bots.

The failures are all over the place. On these two consecutive builds:

https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Release%20%28Intel%29/builds/1697
https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Release%20%28Intel%29/builds/1698

the following failures were seen:
WebglConformance_conformance2_textures_canvas_tex_2d_srgb8_rgb_unsigned_byte
WebglConformance_conformance2_textures_misc_integer_cubemap_texture_sampling
WebglConformance_deqp_functional_gles3_transformfeedback_array_interleaved_points
WebglConformance_deqp_functional_gles3_transformfeedback_basic_types_interleaved_points
WebglConformance_deqp_functional_gles3_transformfeedback_random_separate_points

and:
WebglConformance_deqp_functional_gles3_fbomultisample_4_samples
WebglConformance_deqp_functional_gles3_shadertexturefunction_texturelod
WebglConformance_deqp_functional_gles3_transformfeedback_basic_types_separate_points

Failures are seen here:
https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%20Release%20(Intel)

and on the associated bot on the V8 waterfall:
https://build.chromium.org/p/client.v8.fyi/builders/Mac%20Release%20%28Intel%29?numbuilds=200

The tests were running reliably up till about March 24, when a regression crept in that turned the bots red for a weekend ( Issue 705156 ). Then the tests started flaking not long afterward.

It does look like there might be at least one hardware failure:
https://chromium-swarm.appspot.com/bot?id=build603-m4&sort_stats=total%3Adesc

looks like it's returning 0x0 for the GPU vendor and device ID. I'm taking it out of the Swarming pool and have filed Issue 707473 about it.

John, since you're pixel wrangler next week could you triage some of the other failures and see whether they seem to be a similar failure mode? Thanks.

 
Cc: wittman@chromium.org jbau...@chromium.org
Labels: -Pri-2 Pri-1
Owner: a...@chromium.org
Callstacks all seem to be related to NativeStackSamplerMac, so probably r461145.

Thread 2 (crashed)
   0  libunwind.dylib + 0x1195
      rax = 0x00007fffe4ea7195   rdx = 0x00000000000058d1
      rcx = 0x000070000c178c38   rbx = 0x000070000c178be8
      rsi = 0x00000000f39afbdc   rdi = 0x0000000000000000
      rbp = 0x000070000c178ba0   rsp = 0x000070000c178b98
       r8 = 0x00000000f39afc04    r9 = 0x00007fffe4ea7214
      r10 = 0xffffffffffffffff   r11 = 0x0000000000012068
      r12 = 0x00007fffe4df14be   r13 = 0x00007f819b109aa0
      r14 = 0x0000000000000000   r15 = 0x000070000c178be8
      rip = 0x00007fffe4ea7195
      Found by: given as instruction pointer in context
   1  libunwind.dylib + 0x10d1
      rbp = 0x000070000c178bd0   rsp = 0x000070000c178bb0
      rip = 0x00007fffe4ea70d1
      Found by: previous frame's frame pointer
   2  Chromium Framework!__ZN4base12_GLOBAL__N_120WalkStackFromContextIZNS0_21NativeStackSamplerMac27SuspendThreadAndRecordStackEPNS_21StackSamplingProfiler6SampleEE3$_0EEbP13unw_context_tPmRKT_ + 0x36b
      rbp = 0x000070000c179240   rsp = 0x000070000c178be0
      rip = 0x0000000111c9b8bb
      Found by: previous frame's frame pointer
   3  Chromium Framework!__ZN4base12_GLOBAL__N_121NativeStackSamplerMac17RecordStackSampleEPNS_21StackSamplingProfiler6SampleE + 0x5c8
      rbp = 0x000070000c179c00   rsp = 0x000070000c179250
      rip = 0x0000000111c9b4f8
      Found by: previous frame's frame pointer
   4  Chromium Framework!__ZN4base21StackSamplingProfiler14SamplingThread14CollectProfileEPNS0_16CallStackProfileEPNS_9TimeDeltaEPb + 0x2fc
      rbp = 0x000070000c179c90   rsp = 0x000070000c179c10
      rip = 0x0000000111c9d72c
      Found by: previous frame's frame pointer
   5  Chromium Framework!__ZN4base21StackSamplingProfiler14SamplingThread15CollectProfilesEPNSt3__16vectorINS0_16CallStackProfileENS2_9allocatorIS4_EEEE + 0xba
      rbp = 0x000070000c179d30   rsp = 0x000070000c179ca0
      rip = 0x0000000111c9d32a
      Found by: previous frame's frame pointer
   6  Chromium Framework!__ZN4base21StackSamplingProfiler14SamplingThread10ThreadMainEv + 0xcf
      rbp = 0x000070000c179ec0   rsp = 0x000070000c179d40
      rip = 0x0000000111c9d14f
      Found by: previous frame's frame pointer
   7  Chromium Framework!__ZN4base12_GLOBAL__N_110ThreadFuncEPv + 0x5f
      rbp = 0x000070000c179ef0   rsp = 0x000070000c179ed0
      rip = 0x0000000111cdda4f
      Found by: previous frame's frame pointer
   8  libsystem_pthread.dylib + 0x3aab
      rbp = 0x000070000c179f10   rsp = 0x000070000c179f00
      rip = 0x00007fffe4e6faab
      Found by: previous frame's frame pointer
   9  libsystem_pthread.dylib + 0x39f7
      rbp = 0x000070000c179f50   rsp = 0x000070000c179f20
      rip = 0x00007fffe4e6f9f7
      Found by: previous frame's frame pointer
  10  libsystem_pthread.dylib + 0x31fd
      rbp = 0x000070000c179f78   rsp = 0x000070000c179f60
      rip = 0x00007fffe4e6f1fd
      Found by: previous frame's frame pointer
  11  Chromium Framework!__ZN4base14PlatformThread6DetachENS_20PlatformThreadHandleE + 0x70
      rsp = 0x000070000c17a028   rip = 0x0000000111cdd9f0
      Found by: stack scanning

Comment 2 by kbr@chromium.org, Apr 3 2017

Blocking: 531673
Labels: Hotlist-PixelWrangler
We're also seeing this issue on the ASAN bot:
https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.gpu.fyi%2FMac_GPU_ASAN_Release%2F2238%2F%2B%2Frecipes%2Fsteps%2Fwebgl2_conformance_tests_on_Intel_GPU_on_Mac_on_Mac-10.12%2F0%2Fstdout

==10709==ERROR: AddressSanitizer: SEGV on unknown address 0xffffffffffffffd8 (pc 0x7fffb37e1195 bp 0x700003661080 sp 0x700003661078 T4)
==10709==The signal is caused by a READ memory access.
==10709==WARNING: invalid path to external symbolizer!
==10709==WARNING: Failed to use and restart external symbolizer!
    #0 0x7fffb37e1194 in libunwind::CompactUnwinder_x86_64<libunwind::LocalAddressSpace>::stepWithCompactEncodingRBPFrame(unsigned int, unsigned long long, libunwind::LocalAddressSpace&, libunwind::Registers_x86_64&) (/usr/lib/system/libunwind.dylib:x86_64+0x1194)
    #1 0x7fffb37e10d0 in libunwind::UnwindCursor<libunwind::LocalAddressSpace, libunwind::Registers_x86_64>::step() (/usr/lib/system/libunwind.dylib:x86_64+0x10d0)
    #2 0x10d09329f in bool base::(anonymous namespace)::WalkStackFromContext<base::(anonymous namespace)::NativeStackSamplerMac::SuspendThreadAndRecordStack(base::StackSamplingProfiler::Sample*)::$_0>(unw_context_t*, unsigned long*, base::(anonymous namespace)::NativeStackSamplerMac::SuspendThreadAndRecordStack(base::StackSamplingProfiler::Sample*)::$_0 const&) (/b/s/w/irwENaSa/out/Release/Chromium.app/Contents/Versions/59.0.3062.0/Chromium Framework.framework/Versions/A/Chromium Framework:x86_64+0x651d29f)
    #3 0x10d09222f in base::(anonymous namespace)::NativeStackSamplerMac::RecordStackSample(base::StackSamplingProfiler::Sample*) (/b/s/w/irwENaSa/out/Release/Chromium.app/Contents/Versions/59.0.3062.0/Chromium Framework.framework/Versions/A/Chromium Framework:x86_64+0x651c22f)
    #4 0x10d095f88 in base::StackSamplingProfiler::SamplingThread::CollectProfile(base::StackSamplingProfiler::CallStackProfile*, base::TimeDelta*, bool*) (/b/s/w/irwENaSa/out/Release/Chromium.app/Contents/Versions/59.0.3062.0/Chromium Framework.framework/Versions/A/Chromium Framework:x86_64+0x651ff88)
    #5 0x10d095608 in base::StackSamplingProfiler::SamplingThread::CollectProfiles(std::__1::vector<base::StackSamplingProfiler::CallStackProfile, std::__1::allocator<base::StackSamplingProfiler::CallStackProfile> >*) (/b/s/w/irwENaSa/out/Release/Chromium.app/Contents/Versions/59.0.3062.0/Chromium Framework.framework/Versions/A/Chromium Framework:x86_64+0x651f608)
    #6 0x10d095018 in base::StackSamplingProfiler::SamplingThread::ThreadMain() (/b/s/w/irwENaSa/out/Release/Chromium.app/Contents/Versions/59.0.3062.0/Chromium Framework.framework/Versions/A/Chromium Framework:x86_64+0x651f018)
    #7 0x10d12d91d in base::(anonymous namespace)::ThreadFunc(void*) (/b/s/w/irwENaSa/out/Release/Chromium.app/Contents/Versions/59.0.3062.0/Chromium Framework.framework/Versions/A/Chromium Framework:x86_64+0x65b791d)
    #8 0x7fffb37a9aaa in _pthread_body (/usr/lib/system/libsystem_pthread.dylib:x86_64+0x3aaa)
    #9 0x7fffb37a99f6 in _pthread_start (/usr/lib/system/libsystem_pthread.dylib:x86_64+0x39f6)
    #10 0x7fffb37a91fc in thread_start (/usr/lib/system/libsystem_pthread.dylib:x86_64+0x31fc)

Comment 4 by a...@chromium.org, Apr 4 2017

Status: Fixed (was: Assigned)
Stack sampler reverted.

Comment 5 by kbr@chromium.org, Apr 4 2017

Thanks Avi.

Sign in to add a comment