CUDA: fix should_use_mmvf for ne11 == 1 #17085
Merged
+15
−2
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
Collaborator
Author
|
Sorry, I just realized that I misdiagnosed the problem, I'll push another version. |
Collaborator
|
I tested this on your server |
JohannesGaessler
force-pushed
the
cuda-fix-should_use_mmvf
branch
from
7d65657 to
9779b58
Compare
JohannesGaessler
force-pushed
the
cuda-fix-should_use_mmvf
branch
from
9779b58 to
2bd5465
Compare
am17an
approved these changes
Nov 7, 2025
| if (src0_ne[0] % 2 != 0) { | ||
| return false; | ||
| } | ||
|
|
| return false; | ||
| } | ||
| } | ||
|
|
Collaborator
Author
|
Okay, so the problem should be fixed now. We were checking that the stride for dimension 0 is divisible by the sizes of e.g. |
Co-authored-by: Aman Gupta <[email protected]>
github-actions
bot
added
Nvidia GPU
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
See #16988 (comment) .
The logic for preventing misaligned pointers on strides not divisible by the size of the data type is stricter than necessary,
for a single column we only need to check the stride of dimension 0the strides in question are forsrc0rather thansrc1, strictly speaking we would need to be checking that tensor too. @am17an can you share the model and command line you had used to test this?