Conversation
jeffbolznv
approved these changes
Sep 14, 2025
github-actions
Bot
added
Vulkan
blime4
referenced
this pull request
in blime4/llama.cpp
Feb 5, 2026
* vulkan: move mul_mm dequantization steps into a separate file and functions * improve mul_mm vector load code * fix debug mode issues and warnings
Seunghhon
pushed a commit
to Seunghhon/llama.cpp
that referenced
this pull request
Apr 26, 2026
* vulkan: move mul_mm dequantization steps into a separate file and functions * improve mul_mm vector load code * fix debug mode issues and warnings
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The mul_mm was the first shader I wrote and many of the readability options in GLSL weren't known to me at the time, so in this PR I move all of the dequantization stuff outside of the main file. This makes the shader much easier to work with.
I also did some minor optimizations that I spotted on the way.
It's basically a port of the good changes in #12515, while leaving out the ones that didn't work well.