Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

MemorySanitizer/check_device_global.cpp failing on Gen12 pre-commit on unrelated changes #16978

Open
YuriPlyakhin opened this issue Feb 11, 2025 · 4 comments
Labels
bug Something isn't working

Comments

@YuriPlyakhin
Copy link
Contributor

Describe the bug

https://github.com/intel/llvm/actions/runs/13270134353/job/37052546827?pr=13590

FAIL: SYCL :: MemorySanitizer/check_device_global.cpp (221 of 2274)
******************** TEST 'SYCL :: MemorySanitizer/check_device_global.cpp' FAILED ********************
Exit Code: 1

Command Output (stderr):
--
RUN: at line 3: env ONEAPI_DEVICE_SELECTOR=opencl:cpu  not /__w/llvm/llvm/build-e2e/MemorySanitizer/Output/check_device_global.cpp.tmp1.out 2>&1 | /__w/llvm/llvm/toolchain/bin/FileCheck /__w/llvm/llvm/llvm/sycl/test-e2e/MemorySanitizer/check_device_global.cpp
+ /__w/llvm/llvm/toolchain/bin/FileCheck /__w/llvm/llvm/llvm/sycl/test-e2e/MemorySanitizer/check_device_global.cpp
+ env ONEAPI_DEVICE_SELECTOR=opencl:cpu not /__w/llvm/llvm/build-e2e/MemorySanitizer/Output/check_device_global.cpp.tmp1.out
/__w/llvm/llvm/llvm/sycl/test-e2e/MemorySanitizer/check_device_global.cpp:55:12: error: CHECK: expected string not found in input
 // CHECK: use-of-uninitialized-value
           ^
<stdin>:1:1: note: scanning from here
==== DeviceSanitizer: MSAN
^
<stdin>:1:6: note: possible intended match here
==== DeviceSanitizer: MSAN
     ^

Input file: <stdin>
Check file: /__w/llvm/llvm/llvm/sycl/test-e2e/MemorySanitizer/check_device_global.cpp

-dump-input=help explains the following input dump.

Input was:
<<<<<<
            1: ==== DeviceSanitizer: MSAN 
check:55'0     X~~~~~~~~~~~~~~~~~~~~~~~~~~ error: no match found
check:55'1          ?                      possible intended match
            2: terminate called after throwing an instance of 'sycl::_V1::exception' 
check:55'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
            3:  what(): Native API failed. Native API returns: 7 (UR_RESULT_ERROR_INVALID_BINARY) 
check:55'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
            4: error: Aborted (core dumped) 
check:55'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>>>

--

To reproduce

No response

Environment

No response

Additional context

No response

@wenju-he
Copy link
Contributor

@zhaomaosu
Copy link
Contributor

looks like the test failed on OpenCL CPU device, and I cannot reproduce this issue on my local machine. Not sure if I missed any detail.

@AlexeySachkov
Copy link
Contributor

looks like the test failed on OpenCL CPU device, and I cannot reproduce this issue on my local machine. Not sure if I missed any detail.

Note that at intel/llvm we use older OCL CPU RT (we only use publicly released once, meaning that right now we use version released with oneAPI 2025.0) - it could be that an issue has already been fixed internally and we just need to wait for a OCL CPU RT update. To confirm/disprove that idea you could try reproducing with older OCL CPU RT locally

@againull
Copy link
Contributor

It looks like test fails flakily on L0 gpu as well.
see here: https://github.com/intel/llvm/actions/runs/13320726315/job/37205124861

2025-02-14T01:58:48.6199905Z FAIL: SYCL :: MemorySanitizer/check_device_global.cpp (175 of 2282)
2025-02-14T01:58:48.6200463Z ******************** TEST 'SYCL :: MemorySanitizer/check_device_global.cpp' FAILED ********************
2025-02-14T01:58:48.6200918Z Exit Code: 1
2025-02-14T01:58:48.6201069Z 
2025-02-14T01:58:48.6201205Z Command Output (stderr):
2025-02-14T01:58:48.6201493Z --
2025-02-14T01:58:48.6202455Z RUN: at line 3: env ONEAPI_DEVICE_SELECTOR=level_zero:gpu  not /__w/llvm/llvm/build-e2e/MemorySanitizer/Output/check_device_global.cpp.tmp1.out 2>&1 | /__w/llvm/llvm/toolchain/bin/FileCheck /__w/llvm/llvm/llvm/sycl/test-e2e/MemorySanitizer/check_device_global.cpp
2025-02-14T01:58:48.6203768Z + env ONEAPI_DEVICE_SELECTOR=level_zero:gpu not /__w/llvm/llvm/build-e2e/MemorySanitizer/Output/check_device_global.cpp.tmp1.out
2025-02-14T01:58:48.6204631Z + /__w/llvm/llvm/toolchain/bin/FileCheck /__w/llvm/llvm/llvm/sycl/test-e2e/MemorySanitizer/check_device_global.cpp
2025-02-14T01:58:48.6205513Z /__w/llvm/llvm/llvm/sycl/test-e2e/MemorySanitizer/check_device_global.cpp:59:12: error: CHECK: expected string not found in input
2025-02-14T01:58:48.6206149Z  // CHECK: use-of-uninitialized-value
2025-02-14T01:58:48.6206467Z            ^
2025-02-14T01:58:48.6206720Z <stdin>:1:1: note: scanning from here
2025-02-14T01:58:48.6207050Z ==== DeviceSanitizer: MSAN
2025-02-14T01:58:48.6207330Z ^
2025-02-14T01:58:48.6207611Z <stdin>:1:6: note: possible intended match here
2025-02-14T01:58:48.6207956Z ==== DeviceSanitizer: MSAN
2025-02-14T01:58:48.6208218Z      ^
2025-02-14T01:58:48.6208357Z 
2025-02-14T01:58:48.6208466Z Input file: <stdin>
2025-02-14T01:58:48.6208951Z Check file: /__w/llvm/llvm/llvm/sycl/test-e2e/MemorySanitizer/check_device_global.cpp
2025-02-14T01:58:48.6209310Z 
2025-02-14T01:58:48.6209473Z -dump-input=help explains the following input dump.
2025-02-14T01:58:48.6209720Z 
2025-02-14T01:58:48.6209837Z Input was:
2025-02-14T01:58:48.6210073Z <<<<<<
2025-02-14T01:58:48.6210318Z             1: ==== DeviceSanitizer: MSAN 
2025-02-14T01:58:48.6210704Z check:59'0     X~~~~~~~~~~~~~~~~~~~~~~~~~~ error: no match found
2025-02-14T01:58:48.6211120Z check:59'1          ?                      possible intended match
2025-02-14T01:58:48.6211606Z             2: terminate called after throwing an instance of 'sycl::_V1::exception' 
2025-02-14T01:58:48.6212160Z check:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2025-02-14T01:58:48.6212654Z             3:  what(): Native API failed. Native API returns: 7 (UR_RESULT_ERROR_INVALID_BINARY) 
2025-02-14T01:58:48.6213170Z check:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2025-02-14T01:58:48.6213579Z             4: error: Aborted (core dumped) 
2025-02-14T01:58:48.6213924Z check:59'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2025-02-14T01:58:48.6214230Z >>>>>>
2025-02-14T01:58:48.6214353Z 
2025-02-14T01:58:48.6214461Z --
2025-02-14T01:58:48.6214583Z 
2025-02-14T01:58:48.6214686Z ********************

For now I've disabled test on all platforms here: #17014

againull added a commit that referenced this issue Feb 14, 2025
…pp (#17014)

Temporarily disable on all required platforms because of
#16978
Fail shows up on unrelated PRs.

---------

Co-authored-by: Chris Perkins <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

5 participants