Vulkan: Fix mmq int dot float cache size #12722
Merged
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
jeffbolznv
approved these changes
Apr 2, 2025
github-actions
bot
added
Vulkan
|
For some reason this change removed a large chunk of the performance increase of the shader, I'm not sure how that happened. It added 8 bytes of register use, maybe that crossed some occupancy limit, but that is very weird. I hope it can be fixed. |
This was referenced Apr 4, 2025
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Outdated suggestions cannot be applied.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.
I don't know how I (and everyone else) missed this, considering it means models are completely incoherent when using the new int dot shaders, but here's the fix. The cache buffer for the quant dm values was too small and overflowed, leading to NaN results.