[CUDA] Use appropriate return code for out of registers kernel launch #1318
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov ReportAll modified and coverable lines are covered by tests ✅
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## main #1318 +/- ##
=======================================
Coverage 15.39% 15.39%
=======================================
Files 240 240
Lines 34099 34122 +23
Branches 3775 3779 +4
=======================================
+ Hits 5250 5254 +4
- Misses 28798 28818 +20
+ Partials 51 50 -1 ☔ View full report in Codecov by Sentry. |
b85dbe2 to
301767f
Compare
ldrumm
requested changes
Feb 6, 2024
301767f to
a12bc66
Compare
Good shout. Reworded now. |
a12bc66 to
eec7aa4
Compare
ldrumm
approved these changes
Feb 6, 2024
Change the returned error code for exiting the kernel launch entry point in CUDA when exceeding the maximum available registers for execution on the SM. Previously we were returning a misleading error code.
eec7aa4 to
78a71f3
Compare
sommerlukas
pushed a commit
to intel/llvm
that referenced
this pull request
Jun 3, 2024
#12604) This PR improves the handling of errors by specializing `PI_ERROR_OUT_OF_RESOURCES`. Previously, in the CUDA backend we handled the out of resources launch error (for exceeded registers) as invalid work group size error. Now pairing the new specialized handling with the UR adapter change oneapi-src/unified-runtime#1318 to return the correct error code, we no longer output a misleading error message to users. Also, added a fallback message for the generic out of resources error codes returned from APIs (e.g. for kernel launch). Fixes issue: oneapi-src/unified-runtime#1308
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR changes the returned error code for exiting the kernel launch entry point in CUDA when exceeding the maximum available registers for execution on the SM. Previously we were returning a misleading error code.
This change requires DPC++ to adapt the specific-error handling, see PR: intel/llvm#12604.