Support NVIDIA 50 series GPU #6120
Closed
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
masahi
reviewed
Mar 6, 2025
| versionsSupported = {3, 2}; | ||
| } else if (computeCapability < 110) { | ||
| } else if (computeCapability < 130) { | ||
| versionsSupported = {5, 2}; |
There was a problem hiding this comment.
Consumer cards only support Ampere-style mma. So only the version 2 is supported.
There was a problem hiding this comment.
If you still get an error go to AccelerateMatmul.cpp and just delete the 5 and leave it like:
} else if (computeCapability < 130) {
versionsSupported = {2};
As @masahi pointed, consumer RTX 5090 only supports version 2.
There was a problem hiding this comment.
Deleting the 110 branch is wrong. Need to add a 130 branch with only version {2}.
ThomasRaoux
reviewed
Mar 6, 2025
| static int getMMAVersionSafe(int computeCapability, DotOp op) { | ||
| // List supported mma version in order of preference. | ||
| SmallVector<int> versionsSupported; | ||
| op.emitRemark() << "Your computeCapability is " << computeCapability; |
There was a problem hiding this comment.
please remove this to avoid noise.
6 tasks
|
Close as it has been addressed in #6131 (review) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
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Outdated suggestions cannot be applied.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 fixes the issue that it failed to compile with NVIDIA 5090. As the
computeCapabilityof some 50 series GPU is 120 which is greater than 110, this PR updates the condition to avoid exception.New contributor declaration
I am not making a trivial change, such as fixing a typo in a comment.
I have written a PR description following these
rules.
I have run
pre-commit run --from-ref origin/main --to-ref HEAD.Select one of the following.
/testforlittests/unittestfor C++ tests/python/testfor end-to-end testsFILL THIS IN.Select one of the following.
littests.littests I have added follow these best practices,including the "tests should be minimal" section. (Usually running Python code
and using the instructions it generates is not minimal.)