Always export idle slots to RAM#24190
Merged
Merged
Conversation
Without this, a slot's VRAM cache may not be written to RAM. If this slot happens to be busy then later on, this triggers needless preprocessing in another slot.
ggerganov
reviewed
Jun 5, 2026
ggerganov
left a comment
Member
There was a problem hiding this comment.
Yes, this seems like an improvement.
ggerganov
reviewed
Jun 8, 2026
ggerganov
left a comment
Member
There was a problem hiding this comment.
I think the proposed logic is incorrect because it will keep saving idle slots to prompt cache over and over again.
Contributor
Author
Does this https://github.com/ggml-org/llama.cpp/blob/master/tools/server/server-task.cpp#L2006 not keep it from being rewritten? |
Jcfunk
added a commit
to Jcfunk/llama.cpp
that referenced
this pull request
Jun 11, 2026
* upstream/HEAD: (329 commits) vendor : update LibreSSL to 4.3.2 (ggml-org#24397) Remove padding and multiple D2D copies for MTP (ggml-org#24086) chat: fix LFM2/LFM2.5 ignoring json_schema (ggml-org#24377) CUDA: Fix ssm_scan_f32 data-races (ggml-org#24360) ci : bump komac version (ggml-org#24396) speculative : fix "ngram-map-k4v" name in logging (ggml-org#24253) webui: implement pinned conversations support (ggml-org#21387) graph: Fix granite speech model inference by applying embedding scale when deepstack is not used (ggml-org#24357) ci : fix windows release (ggml-org#24369) ui: add opt-in run_javascript frontend tool (ggml-org#24244) mtmd: build_vit batching (ggml-org#24352) vulkan: reduce iq1 shared memory usage for mul_mm (ggml-org#24287) vulkan: add `v_dot2_f32_f16` support in matrix-matrix multiplication and Flash Attention (ggml-org#24123) ui: Fix excessive style recalculation on hover (ggml-org#24243) mtmd: refactor video subproc handling (ggml-org#24316) server: log prompts to directory (ggml-org#22031) ui: fix mobile chat form overflow and bust stale bundle cache (ggml-org#24158) ggml : add GGML_OP_COL2IM_1D (ggml-org#24206) server : do not clear slots without unified KV cache (ggml-org#24190) models : fix plamo2 attention_key/value_length regression (ggml-org#24317) ...
turbo-tan
pushed a commit
to turbo-tan/llama.cpp-tq3
that referenced
this pull request
Jun 11, 2026
* Always export idle slots to RAM Without this, a slot's VRAM cache may not be written to RAM. If this slot happens to be busy then later on, this triggers needless preprocessing in another slot. * cont : clean-up --------- Co-authored-by: Christoph Weiss <weiss@wsoptics.de> Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> (cherry picked from commit 961e9a3)
turbo-tan
pushed a commit
to turbo-tan/llama.cpp-tq3
that referenced
this pull request
Jun 12, 2026
* Always export idle slots to RAM Without this, a slot's VRAM cache may not be written to RAM. If this slot happens to be busy then later on, this triggers needless preprocessing in another slot. * cont : clean-up --------- Co-authored-by: Christoph Weiss <weiss@wsoptics.de> Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> (cherry picked from commit 961e9a3)
turbo-tan
pushed a commit
to turbo-tan/llama.cpp-tq3
that referenced
this pull request
Jun 12, 2026
* Always export idle slots to RAM Without this, a slot's VRAM cache may not be written to RAM. If this slot happens to be busy then later on, this triggers needless preprocessing in another slot. * cont : clean-up --------- Co-authored-by: Christoph Weiss <weiss@wsoptics.de> Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> (cherry picked from commit 961e9a3)
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.
Without this, a slot's VRAM cache may not be written to RAM. If this slot happens to be busy then later on, this triggers needless preprocessing in another slot. It seems to me the correct behavior is what's being implemented here: respect
--cache-idle-slotsalso without unified KV cache, but not evict anything from VRAM since that wouldn't serve any purpose.Overview
Make
--cache-idle-slotsexport to RAM also without unified KV cache, but not evict from VRAM.Additional information
This closes #22942 which observes the problem of seemingly superfluous preprocessing with parallel slots without unified KV cache.
Requirements