Thanks for filing. Digging into the log from the failing shard:
https://chromium-swarm.appspot.com/task?id=41d7934292eaa110&refresh=10&show_raw=1

the symbolized minidump looks like there's a bug in the tracing subsystem:

  Operating system: Linux
                    0.0.0 Linux 3.13.0-61-generic #100-Ubuntu SMP Wed Jul 29 11:21:34 UTC 2015 x86_64
  CPU: amd64
       family 6 model 60 stepping 3
       1 CPU
  
  GPU: UNKNOWN
  
  Crash reason:  SIGSEGV
  Crash address: 0x0
  Process uptime: not available
  
  Thread 7 (crashed)
   0  chrome!GetRef [scoped_refptr.h : 238 + 0x0]
      rax = 0x00007f58b7478450   rdx = 0x00007f58b5772e53
      rcx = 0x00000000000001b5   rbx = 0x0000000000000000
      rsi = 0x0000000000000000   rdi = 0x00007f58a3c99cf0
      rbp = 0x00007f58a3c99cd0   rsp = 0x00007f58a3c99cb0
       r8 = 0x0000000000000000    r9 = 0x0000000000000276
      r10 = 0x0000000000000000   r11 = 0x000000001a311521
      r12 = 0x00000fa27df13540   r13 = 0x00007f58a3c99cf0
      r14 = 0x00007f58a3c99cf0   r15 = 0x0000000000000000
      rip = 0x00007f58baa715b1
      Found by: given as instruction pointer in context
   1  chrome!SendRecorder [weak_ptr.h : 359 + 0xb]
      rbx = 0x0000000000000000   rbp = 0x00007f58a3c99ea0
      rsp = 0x00007f58a3c99ce0   r12 = 0x00000fa27df13540
      r13 = 0x00007f58a3c99cf0   r14 = 0x00007f58a3c99ec8
      r15 = 0x00000fa27dfd4aa0   rip = 0x00007f58b8bb9675
      Found by: call frame info
   2  chrome!Invoke<void (tracing::Coordinator::*)(base::WeakPtr<tracing::AgentRegistry::AgentEntry>, mojo::InterfacePtr<tracing::mojom::Recorder>), base::WeakPtr<tracing::Coordinator>, base::WeakPtr<tracing::AgentRegistry::AgentEntry>, mojo::InterfacePtr<tracing::mojom::Recorder> > [bind_internal.h : 516 + 0xc]
      rbx = 0x00000fa27df13540   rbp = 0x00007f58a3c9a050
      rsp = 0x00007f58a3c99eb0   r12 = 0x00007f58a3c99eb8
      r13 = 0x00000fa2827afed0   r14 = 0x00007f58b8bb9530
      r15 = 0x00007f58a3c99ec8   rip = 0x00007f58b8bbb97f
      Found by: call frame info
   3  chrome!RunTask [callback.h : 99 + 0x3]
      rbx = 0x00007f58a3c9a290   rbp = 0x00007f58a3c9a210
      rsp = 0x00007f58a3c9a060   r12 = 0x00000fa27dbd4a00
      r13 = 0x00007f58bffbfcd0   r14 = 0x0000000000000000
      r15 = 0x00007f58c0689058   rip = 0x00007f58baa75839
      Found by: call frame info
   4  chrome!DoWorkImpl [thread_controller_with_message_pump_impl.cc : 244 + 0x13]
      rbx = 0x00007f58bffbfcd0   rbp = 0x00007f58a3c9a450
      rsp = 0x00007f58a3c9a220   r12 = 0x00000fa27dbd4a00
      r13 = 0x00007f58b5772e53   r14 = 0x0000000000000000
      r15 = 0x0000000000000000   rip = 0x00007f58baace2f6
      Found by: call frame info
   5  chrome!Run [message_pump_libevent.cc : 210 + 0x9]
      rbx = 0x00000fa27dc2ed90   rbp = 0x00007f58a3c9a4b0
      rsp = 0x00007f58a3c9a460   r12 = 0x00000fa27dc2eda0
      r13 = 0x000000007dbd4a01   r14 = 0x00000fa27dbd4a08
      r15 = 0x0000000000000001   rip = 0x00007f58bab496a9
      Found by: call frame info
   6  chrome!Run [thread_controller_with_message_pump_impl.cc : 342 + 0x6]
      rbx = 0x00000fa27dbd4a00   rbp = 0x00007f58a3c9a610
      rsp = 0x00007f58a3c9a4c0   r12 = 0x00007f58a3c9aa80
      r13 = 0x00000fa27db81dc0   r14 = 0x0000000000000001
      r15 = 0x00000fa27db62300   rip = 0x00007f58baaceae9
      Found by: call frame info
   7  chrome!Run [run_loop.cc : 102 + 0x9]
      rbx = 0x00007f58a3c9aa80   rbp = 0x00007f58a3c9a770
      rsp = 0x00007f58a3c9a620   r12 = 0x00007f58a3c9aa80
      r13 = 0x00000fa27db81dc0   r14 = 0x00007f58a3c9aa98
      r15 = 0x00000fa27db62300   rip = 0x00007f58baa9e446
      Found by: call frame info
   8  chrome!Run [thread.cc : 250 + 0x8]
      rbx = 0x000000000000140e   rbp = 0x00007f58a3c9a8d0
      rsp = 0x00007f58a3c9a780   r12 = 0x00007f58a3c9aa80
      r13 = 0x00000fa27db81dc0   r14 = 0x00007f58a3c9aa80
      r15 = 0x00000fa27db62300   rip = 0x00007f58baaf411a
      Found by: call frame info
   9  chrome!content::BrowserProcessSubThread::IOThreadRun(base::RunLoop*) + 0x14
      rbx = 0x00000fa27db81dc0   rbp = 0x00007f58a3c9a8f0
      rsp = 0x00007f58a3c9a8e0   r12 = 0x00007f58a3c9aa80
      r13 = 0x00000fa27db81dc0   r14 = 0x00007f58a3c9aa80
      r15 = 0x00000fa27db62300   rip = 0x00007f58b84bbad4
      Found by: call frame info
  10  chrome!Run [browser_process_sub_thread.cc : 126 + 0xb]
      rbx = 0x00000fa27db81dc0   rbp = 0x00007f58a3c9aa50
      rsp = 0x00007f58a3c9a900   r12 = 0x00007f58a3c9aa80
      r13 = 0x00000fa27db81dc0   r14 = 0x00007f58a3c9aa80
      r15 = 0x00000fa27db62300   rip = 0x00007f58b84bba1f
      Found by: call frame info
  11  chrome!ThreadMain [thread.cc : 332 + 0xd]
      rbx = 0x00000fa27db81dd8   rbp = 0x00007f58a3c9abe0
      rsp = 0x00007f58a3c9aa60   r12 = 0x00007f58a3c9aa80
      r13 = 0x00000fa27db81dc0   r14 = 0x00000fa27db81dd0
      r15 = 0x00000fa27db62300   rip = 0x00007f58baaf44a8
      Found by: call frame info
  12  chrome!ThreadFunc [platform_thread_posix.cc : 81 + 0x8]
      rbx = 0x00000fa27dc27f30   rbp = 0x00007f58a3c9ac10
      rsp = 0x00007f58a3c9abf0   r12 = 0x00007f58a3c9b700
      r13 = 0x0000000000000000   r14 = 0x00000fa27db81dc0
      r15 = 0x00000fa27db603f0   rip = 0x00007f58bab3bb38
      Found by: call frame info
  13  libpthread-2.19.so + 0x8182
      rbx = 0x00007f58a3c9b700   rbp = 0x0000000000000000
      rsp = 0x00007f58a3c9ac20   r12 = 0x0000000000000000
      r13 = 0x0000000000000000   r14 = 0x00007f58a3c9b9c0
      r15 = 0x00007f58a3c9b700   rip = 0x00007f58b40c5182
      Found by: call frame info
  14  libc-2.19.so + 0xfa47d
      rsp = 0x00007f58a3c9acc0   rip = 0x00007f58ae2f347d
      Found by: stack scanning


Could someone from the DevTools or Speed teams take a look at this and see if there's anything actionable? The source code for these tests is here:
https://cs.chromium.org/chromium/src/content/test/gpu/gpu_tests/trace_integration_test.py?q=trace_inte&sq=package:chromium&g=0&l=1

and can be run via these instructions:
https://chromium.googlesource.com/chromium/src/+/master/docs/gpu/gpu_testing.md#Running-the-GPU-Tests-Locally

Thanks.

@alph could you please take a look?

GPU Triage: Assigning to alph@ per #2

This is not DevTools related.

If this isn't actionable from the devtools side, not sure there's much we can do. Closing this out.