cuda : relax tensor contiguity requirements for quantized concat - #25678
Merged
Conversation
github-actions
Bot
added
testing
am17an
approved these changes
Jul 15, 2026
ggerganov
approved these changes
Jul 15, 2026
zengde
pushed a commit
to zengde/llama.cpp
that referenced
this pull request
Jul 16, 2026
…l-org#25678) * cuda : relax tensor contiguity requirements for quantized concat * tests : add test cases for non-contiguous quantized concat * ggml : relax contiguity requirements for quantized concat --------- Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
CowboyTim
pushed a commit
to aardbeiplantje/llama.cpp
that referenced
this pull request
Jul 21, 2026
…l-org#25678) * cuda : relax tensor contiguity requirements for quantized concat * tests : add test cases for non-contiguous quantized concat * ggml : relax contiguity requirements for quantized concat --------- Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
smalinin
pushed a commit
to smalinin/llama.cpp
that referenced
this pull request
Aug 4, 2026
…l-org#25678) * cuda : relax tensor contiguity requirements for quantized concat * tests : add test cases for non-contiguous quantized concat * ggml : relax contiguity requirements for quantized concat --------- Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
satindergrewal
pushed a commit
to satindergrewal/llama.cpp
that referenced
this pull request
Aug 12, 2026
…l-org#25678) * cuda : relax tensor contiguity requirements for quantized concat * tests : add test cases for non-contiguous quantized concat * ggml : relax contiguity requirements for quantized concat --------- Co-authored-by: Stanisław Szymczyk <sszymczy@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
4 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.
Overview
This PR allows concatenation of non-contiguous quantized tensors in CUDA backend for one specific case where dim is not 3 and tensors are contiguous in dims < 3. We need this to handle multi-stream quantized KV cache tensor concatenation in DeepSeek V4 implementation.
Fixes #25530
Additional information
The existing implementation has a loop iterating over dim 3, so it handles this case without any changes. Contiguity requirement in CPU backend was also relaxed so that test cases could work.
Requirements