kv-cache : better estimate of n_kv for multi-sequence batches #15610
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
gabe-l-hart
added a commit
to gabe-l-hart/llama.cpp
that referenced
this pull request
Aug 28, 2025
…upport * origin/master: (61 commits) scripts: add sqlite3 check for compare-commits.sh (ggml-org#15633) kv-cache : remove LLAMA_SET_ROWS checks (ggml-org#15505) gguf-py: byteswapping improvements (ggml-org#12851) cli : change log to warning to explain reason for stopping (ggml-org#15604) model-conversion : add mmproj conversion target (ggml-org#15628) cuda: Add cublasLt_static linking when GGML_STATIC is enabled (ggml-org#15622) server: higher timeout for tests (ggml-org#15621) presets : add qwen3-30B-a3b FIM (ggml-org#15616) HIP: Enable support for ggml_backend_cuda_register_host_buffer (ggml-org#15615) kv-cache : better estimate of n_kv for multi-sequence batches (ggml-org#15610) CANN: refactor mask handling and improve performance in FA (ggml-org#15561) ggml-cpu : add basic RVV support for vector f32 ops (ggml-org#15057) common : add -m to bash completion for --model [no ci] (ggml-org#15591) OpenCL: add fused group_norm/norm, mul, add (ggml-org#15314) tests : fix test-opt with GGML_BACKEND_DL (ggml-org#15599) SYCL: fix rms_norm_mul_add for tensor dim not a multiple of sg_size (ggml-org#15592) mtmd : fix mtmd ios build (ggml-org#15579) tests: add performance test for mul mat id (ggml-org#15543) llamafile: PowerPC Sgemm Optimization (ggml-org#15558) graph : fix assert in memory-less build_attn (ggml-org#15590) ...
gabe-l-hart
added a commit
to gabe-l-hart/llama.cpp
that referenced
this pull request
Aug 28, 2025
…g-model-disabled-agent-prefill * origin/master: (76 commits) scripts: add sqlite3 check for compare-commits.sh (ggml-org#15633) kv-cache : remove LLAMA_SET_ROWS checks (ggml-org#15505) gguf-py: byteswapping improvements (ggml-org#12851) cli : change log to warning to explain reason for stopping (ggml-org#15604) model-conversion : add mmproj conversion target (ggml-org#15628) cuda: Add cublasLt_static linking when GGML_STATIC is enabled (ggml-org#15622) server: higher timeout for tests (ggml-org#15621) presets : add qwen3-30B-a3b FIM (ggml-org#15616) HIP: Enable support for ggml_backend_cuda_register_host_buffer (ggml-org#15615) kv-cache : better estimate of n_kv for multi-sequence batches (ggml-org#15610) CANN: refactor mask handling and improve performance in FA (ggml-org#15561) ggml-cpu : add basic RVV support for vector f32 ops (ggml-org#15057) common : add -m to bash completion for --model [no ci] (ggml-org#15591) OpenCL: add fused group_norm/norm, mul, add (ggml-org#15314) tests : fix test-opt with GGML_BACKEND_DL (ggml-org#15599) SYCL: fix rms_norm_mul_add for tensor dim not a multiple of sg_size (ggml-org#15592) mtmd : fix mtmd ios build (ggml-org#15579) tests: add performance test for mul mat id (ggml-org#15543) llamafile: PowerPC Sgemm Optimization (ggml-org#15558) graph : fix assert in memory-less build_attn (ggml-org#15590) ...
gabe-l-hart
added a commit
to gabe-l-hart/llama.cpp
that referenced
this pull request
Aug 28, 2025
…nemotron-nano-15409 * origin/master: (59 commits) scripts: add sqlite3 check for compare-commits.sh (ggml-org#15633) kv-cache : remove LLAMA_SET_ROWS checks (ggml-org#15505) gguf-py: byteswapping improvements (ggml-org#12851) cli : change log to warning to explain reason for stopping (ggml-org#15604) model-conversion : add mmproj conversion target (ggml-org#15628) cuda: Add cublasLt_static linking when GGML_STATIC is enabled (ggml-org#15622) server: higher timeout for tests (ggml-org#15621) presets : add qwen3-30B-a3b FIM (ggml-org#15616) HIP: Enable support for ggml_backend_cuda_register_host_buffer (ggml-org#15615) kv-cache : better estimate of n_kv for multi-sequence batches (ggml-org#15610) CANN: refactor mask handling and improve performance in FA (ggml-org#15561) ggml-cpu : add basic RVV support for vector f32 ops (ggml-org#15057) common : add -m to bash completion for --model [no ci] (ggml-org#15591) OpenCL: add fused group_norm/norm, mul, add (ggml-org#15314) tests : fix test-opt with GGML_BACKEND_DL (ggml-org#15599) SYCL: fix rms_norm_mul_add for tensor dim not a multiple of sg_size (ggml-org#15592) mtmd : fix mtmd ios build (ggml-org#15579) tests: add performance test for mul mat id (ggml-org#15543) llamafile: PowerPC Sgemm Optimization (ggml-org#15558) graph : fix assert in memory-less build_attn (ggml-org#15590) ...
Minh141120
pushed a commit
to janhq/llama.cpp
that referenced
this pull request
Aug 29, 2025
Nexesenex
added a commit
to Nexesenex/croco.cpp
that referenced
this pull request
Oct 6, 2025
…ggml-org#15610)" This reverts commit 1bded5a.
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.
The logic for creating the KQ mask tensor was over-estimating the size of
n_kv. Instead of considering only the sequence lengths in the current ubatch, it was considering all sequence lengths in the KV cache, which leads to allocating a much larger mask than necessary. This resulted in performance overhead when computing a short sequence in the presence of a large sequence in the cache. The difference is quite noticeable withllama-serverwith parallel sequences.