Skip to content
This repository was archived by the owner on Apr 18, 2023. It is now read-only.
This repository was archived by the owner on Apr 18, 2023. It is now read-only.

[IE-MKLDNN/IE-clDNN] Chromium crashed when running the quantized models #851

@Christywl

Description

@Christywl

Test Env:
Chromium Version: nightly build 75.0.3739.0 (1a2331d)
Platform: Windows

Expected Result:
Chromium shouldn't crash, should throw an error if IE doesn't support TENSOR_QUANT8_ASYMM type.

Actual Result:
Chromium crashed when running the quantized models:

Received signal 11 SEGV_MAPERR 7f82a80f5000
#0 0x5642b916ede9 <unknown>
#1 0x5642b90d6083 <unknown>
#2 0x5642b916e971 <unknown>
#3 0x7f82bdbf1390 <unknown>
#4 0x5642b7d1efb3 <unknown>
#5 0x5642b7d1e455 <unknown>
#6 0x5642b7d19b47 <unknown>
#7 0x5642b7d17c8b <unknown>
#8 0x5642b7d11855 <unknown>
#9 0x5642b7c8c3ff <unknown>
#10 0x5642b7d16d76 <unknown>
#11 0x5642b91c2268 <unknown>
#12 0x5642b91c6dce <unknown>
#13 0x5642b91c6638 <unknown>
#14 0x5642b91c013d <unknown>
#15 0x5642b91c0a9e <unknown>
#16 0x5642b91d40ce <unknown>
#17 0x5642b912bb95 <unknown>
#18 0x5642b913bf72 <unknown>
#19 0x5642b913bc2a <unknown>
#20 0x5642b917df6b <unknown>
#21 0x5642b913b3bd <unknown>
#22 0x5642b9137f40 <unknown>
#23 0x5642b9137d64 <unknown>
#24 0x5642b917e71e <unknown>
#25 0x7f82bdbe76ba start_thread
#26 0x7f82b7b6041d clone
  r8: 00000000000001e9  r9: 00000000000281e9 r10: 0000000000000000 r11: 00000000000281e9
 r12: 0000000000000000 r13: 00000000000281e9 r14: 00007f82a805485c r15: 00000000000281e9
  di: 00000000000281e9  si: 0000000000000001  bp: 00007f82a78724c0  bx: 00003d40ba469000
  dx: 0000000000000200  ax: 00000000000281e9  cx: 0000000000000001  sp: 00007f82a7872320
  ip: 00005642b7d1efb3 efl: 0000000000010202 cgf: 002b000000000033 erf: 0000000000000004
 trp: 000000000000000e msk: 0000000000000000 cr2: 00007f82a80f5000
[end of stack trace]
Calling _exit(1). Core file will not be generated.

How to Reproduce:

  1. Setup test server and start
  2. Visit http://localhost:8080/examples/image_classification
  3. Select FAST_SINGLE_ANSWER or SUSTAINED_SPEED
  4. Select one quantized model

The quantized models in Object Detection also have this issue.

Metadata

Metadata

Labels

bugSomething isn't workingp1Highest priority

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions