vulkan: fix top_k bug when there are ties in the input#17659
Merged
0cc4m merged 3 commits intoggml-org:masterfrom Dec 5, 2025
Merged
vulkan: fix top_k bug when there are ties in the input#176590cc4m merged 3 commits intoggml-org:masterfrom
0cc4m merged 3 commits intoggml-org:masterfrom
Conversation
- Compute row size for the temp buffer based on the output of the first pass. - Update shader addressing math to use the output row size - Pass the output row size as "ncols_output", what used to be "ncols_output" is now "k" For the common case of K=40 and src0=(200000,1,1,1), this reduces the temporary buffer from about 3.2MB to 500KB.
I noticed by inspection a bug in the vulkan top_k shader where if the least value in the top_k appears multiple times we could end up writing those extra copies out rather than some larger values (if the larger values are on higher numbered threads). I rewrote the test verification to handle this case, where the final index set is not necessarily the same.
ggerganov
approved these changes
Dec 2, 2025
Member
ggerganov
left a comment
There was a problem hiding this comment.
Ack on the test-backend-ops changes
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
0cc4m
approved these changes
Dec 5, 2025
JayZenith
pushed a commit
to JayZenith/llama.cpp
that referenced
this pull request
Dec 7, 2025
* vulkan: Reduce temporary memory usage for TOP_K - Compute row size for the temp buffer based on the output of the first pass. - Update shader addressing math to use the output row size - Pass the output row size as "ncols_output", what used to be "ncols_output" is now "k" For the common case of K=40 and src0=(200000,1,1,1), this reduces the temporary buffer from about 3.2MB to 500KB. * vulkan: fix top_k bug when there are ties in the input I noticed by inspection a bug in the vulkan top_k shader where if the least value in the top_k appears multiple times we could end up writing those extra copies out rather than some larger values (if the larger values are on higher numbered threads). I rewrote the test verification to handle this case, where the final index set is not necessarily the same. * Update tests/test-backend-ops.cpp Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
0Marble
pushed a commit
to 0Marble/llama.cpp
that referenced
this pull request
Dec 18, 2025
* vulkan: Reduce temporary memory usage for TOP_K - Compute row size for the temp buffer based on the output of the first pass. - Update shader addressing math to use the output row size - Pass the output row size as "ncols_output", what used to be "ncols_output" is now "k" For the common case of K=40 and src0=(200000,1,1,1), this reduces the temporary buffer from about 3.2MB to 500KB. * vulkan: fix top_k bug when there are ties in the input I noticed by inspection a bug in the vulkan top_k shader where if the least value in the top_k appears multiple times we could end up writing those extra copies out rather than some larger values (if the larger values are on higher numbered threads). I rewrote the test verification to handle this case, where the final index set is not necessarily the same. * Update tests/test-backend-ops.cpp Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Anico2
added a commit
to Anico2/llama.cpp
that referenced
this pull request
Jan 15, 2026
* vulkan: Reduce temporary memory usage for TOP_K - Compute row size for the temp buffer based on the output of the first pass. - Update shader addressing math to use the output row size - Pass the output row size as "ncols_output", what used to be "ncols_output" is now "k" For the common case of K=40 and src0=(200000,1,1,1), this reduces the temporary buffer from about 3.2MB to 500KB. * vulkan: fix top_k bug when there are ties in the input I noticed by inspection a bug in the vulkan top_k shader where if the least value in the top_k appears multiple times we could end up writing those extra copies out rather than some larger values (if the larger values are on higher numbered threads). I rewrote the test verification to handle this case, where the final index set is not necessarily the same. * Update tests/test-backend-ops.cpp Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
blime4
referenced
this pull request
in blime4/llama.cpp
Feb 5, 2026
* vulkan: Reduce temporary memory usage for TOP_K - Compute row size for the temp buffer based on the output of the first pass. - Update shader addressing math to use the output row size - Pass the output row size as "ncols_output", what used to be "ncols_output" is now "k" For the common case of K=40 and src0=(200000,1,1,1), this reduces the temporary buffer from about 3.2MB to 500KB. * vulkan: fix top_k bug when there are ties in the input I noticed by inspection a bug in the vulkan top_k shader where if the least value in the top_k appears multiple times we could end up writing those extra copies out rather than some larger values (if the larger values are on higher numbered threads). I rewrote the test verification to handle this case, where the final index set is not necessarily the same. * Update tests/test-backend-ops.cpp Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
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
3 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.
And update tests to exercise this case.
This is stacked on #17623.