Skip to content

Conversation

@nrspruit
Copy link
Contributor

@nrspruit nrspruit commented Sep 5, 2024

-pre-commit PR for oneapi-src/unified-runtime#2063

@aarongreig aarongreig force-pushed the refactor_l0_default_init branch from a2cca10 to 561226d Compare October 17, 2024 13:14
@aarongreig aarongreig marked this pull request as ready for review October 17, 2024 13:14
@aarongreig aarongreig requested a review from a team as a code owner October 17, 2024 13:14
@aarongreig
Copy link
Contributor

looks like the nvidia runner is in a bad state:

error: Unable to create '/__w/repo_cache/intel/llvm/.git/index.lock': File exists.
Updating e94cfda4fb08..17911151ad75

Another git process seems to be running in this repository, e.g.
an editor opened by 'git commit'. Please make sure all processes
are terminated then try again. If it still fails, a git process
may have crashed in this repository earlier:
remove the file manually to continue.

this tag update only affects the level zero adapter so we should be ok to merge without that job

@aarongreig
Copy link
Contributor

@intel/llvm-gatekeepers please merge if you're ok doing so without the cuda build (given the above)

@sarnex
Copy link
Contributor

sarnex commented Oct 17, 2024

sorry i just fixed the runner, ill restart the job and merge once it passes, sorry for the trouble

@sarnex sarnex merged commit e19e557 into intel:sycl Oct 17, 2024
12 checks passed
@steffenlarsen
Copy link
Contributor

steffenlarsen commented Oct 18, 2024

@nrspruit - This PR is causing problems on local builds, for example sycl-ls fail with

SYCL Exception encountered: Native API failed. Native API returns: 2147483646 (UR_RESULT_ERROR_UNKNOWN)

UR trace:

   ---> urAdapterGet
   <--- urAdapterGet(.NumEntries = 0, .phAdapters = {}, .pNumAdapters = 0x7ffdf96559fc (2)) -> UR_RESULT_SUCCESS;
   ---> urAdapterGet
   <--- urAdapterGet(.NumEntries = 2, .phAdapters = {0x5613e34c2350, 0x5613e34cd910}, .pNumAdapters = nullptr) -> UR_RESULT_SUCCESS;
   ---> urAdapterGetInfo
   <--- urAdapterGetInfo(.hAdapter = 0x5613e34c2350, .propName = UR_ADAPTER_INFO_BACKEND, .propSize = 4, .pPropValue = 0x7ffdf9655a08 (UR_ADAPTER_BACKEND_LEVEL_ZERO), .pPropSizeRet = nullptr) -> UR_RESULT_SUCCESS;
   ---> urAdapterGetInfo
   <--- urAdapterGetInfo(.hAdapter = 0x5613e34cd910, .propName = UR_ADAPTER_INFO_BACKEND, .propSize = 4, .pPropValue = 0x7ffdf9655a08 (UR_ADAPTER_BACKEND_OPENCL), .pPropSizeRet = nullptr) -> UR_RESULT_SUCCESS;
   ---> urPlatformGet
   <--- urPlatformGet(.phAdapters = {0x5613e34c2350}, .NumAdapters = 1, .NumEntries = 0, .phPlatforms = {}, .pNumPlatforms = 0x7ffdf9655a84 (0)) -> UR_RESULT_ERROR_UNKNOWN;
SYCL Exception encountered: Native API failed. Native API returns: 2147483646 (UR_RESULT_ERROR_UNKNOWN)

   ---> urAdapterRelease
   <--- urAdapterRelease(.hAdapter = 0x5613e34c2350) -> UR_RESULT_SUCCESS;
   ---> urAdapterRelease
   <--- urAdapterRelease(.hAdapter = 0x5613e34cd910) -> UR_RESULT_SUCCESS;

@kbenzie | @aarongreig - Is it safe to revert this while this gets investigated? (#15753)

@kbenzie
Copy link
Contributor

kbenzie commented Oct 18, 2024

@nrspruit - This PR is causing problems on local builds, for example sycl-ls fail with

SYCL Exception encountered: Native API failed. Native API returns: 2147483646 (UR_RESULT_ERROR_UNKNOWN)

UR trace:

   ---> urAdapterGet
   <--- urAdapterGet(.NumEntries = 0, .phAdapters = {}, .pNumAdapters = 0x7ffdf96559fc (2)) -> UR_RESULT_SUCCESS;
   ---> urAdapterGet
   <--- urAdapterGet(.NumEntries = 2, .phAdapters = {0x5613e34c2350, 0x5613e34cd910}, .pNumAdapters = nullptr) -> UR_RESULT_SUCCESS;
   ---> urAdapterGetInfo
   <--- urAdapterGetInfo(.hAdapter = 0x5613e34c2350, .propName = UR_ADAPTER_INFO_BACKEND, .propSize = 4, .pPropValue = 0x7ffdf9655a08 (UR_ADAPTER_BACKEND_LEVEL_ZERO), .pPropSizeRet = nullptr) -> UR_RESULT_SUCCESS;
   ---> urAdapterGetInfo
   <--- urAdapterGetInfo(.hAdapter = 0x5613e34cd910, .propName = UR_ADAPTER_INFO_BACKEND, .propSize = 4, .pPropValue = 0x7ffdf9655a08 (UR_ADAPTER_BACKEND_OPENCL), .pPropSizeRet = nullptr) -> UR_RESULT_SUCCESS;
   ---> urPlatformGet
   <--- urPlatformGet(.phAdapters = {0x5613e34c2350}, .NumAdapters = 1, .NumEntries = 0, .phPlatforms = {}, .pNumPlatforms = 0x7ffdf9655a84 (0)) -> UR_RESULT_ERROR_UNKNOWN;
SYCL Exception encountered: Native API failed. Native API returns: 2147483646 (UR_RESULT_ERROR_UNKNOWN)

   ---> urAdapterRelease
   <--- urAdapterRelease(.hAdapter = 0x5613e34c2350) -> UR_RESULT_SUCCESS;
   ---> urAdapterRelease
   <--- urAdapterRelease(.hAdapter = 0x5613e34cd910) -> UR_RESULT_SUCCESS;

@kbenzie | @aarongreig - Is it safe to revert this while this gets investigated? (#15753)

I've approved the revert @steffenlarsen, I'll need to do the same in UR until a fix is found.

@nrspruit
Copy link
Contributor Author

@nrspruit - This PR is causing problems on local builds, for example sycl-ls fail with

SYCL Exception encountered: Native API failed. Native API returns: 2147483646 (UR_RESULT_ERROR_UNKNOWN)

UR trace:

   ---> urAdapterGet
   <--- urAdapterGet(.NumEntries = 0, .phAdapters = {}, .pNumAdapters = 0x7ffdf96559fc (2)) -> UR_RESULT_SUCCESS;
   ---> urAdapterGet
   <--- urAdapterGet(.NumEntries = 2, .phAdapters = {0x5613e34c2350, 0x5613e34cd910}, .pNumAdapters = nullptr) -> UR_RESULT_SUCCESS;
   ---> urAdapterGetInfo
   <--- urAdapterGetInfo(.hAdapter = 0x5613e34c2350, .propName = UR_ADAPTER_INFO_BACKEND, .propSize = 4, .pPropValue = 0x7ffdf9655a08 (UR_ADAPTER_BACKEND_LEVEL_ZERO), .pPropSizeRet = nullptr) -> UR_RESULT_SUCCESS;
   ---> urAdapterGetInfo
   <--- urAdapterGetInfo(.hAdapter = 0x5613e34cd910, .propName = UR_ADAPTER_INFO_BACKEND, .propSize = 4, .pPropValue = 0x7ffdf9655a08 (UR_ADAPTER_BACKEND_OPENCL), .pPropSizeRet = nullptr) -> UR_RESULT_SUCCESS;
   ---> urPlatformGet
   <--- urPlatformGet(.phAdapters = {0x5613e34c2350}, .NumAdapters = 1, .NumEntries = 0, .phPlatforms = {}, .pNumPlatforms = 0x7ffdf9655a84 (0)) -> UR_RESULT_ERROR_UNKNOWN;
SYCL Exception encountered: Native API failed. Native API returns: 2147483646 (UR_RESULT_ERROR_UNKNOWN)

   ---> urAdapterRelease
   <--- urAdapterRelease(.hAdapter = 0x5613e34c2350) -> UR_RESULT_SUCCESS;
   ---> urAdapterRelease
   <--- urAdapterRelease(.hAdapter = 0x5613e34cd910) -> UR_RESULT_SUCCESS;

@kbenzie | @aarongreig - Is it safe to revert this while this gets investigated? (#15753)

I've approved the revert @steffenlarsen, I'll need to do the same in UR until a fix is found.

This patch weas authored a while back now, it is possible the code changed enough causing an issue, I will rebase and update the change to verify no issues before re-posting thank you!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants