hexagon: HMX quantized matmul rework#23368
Merged
max-krasnyansky merged 10 commits intoMay 20, 2026
Merged
Conversation
It seems that we don't reall need non-pipelined version
Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>
lhez
approved these changes
May 20, 2026
github-actions
Bot
added
script
Member
Author
|
@ggml-org/maintainers can I get the second approval please |
CISC
approved these changes
May 20, 2026
ProTekk
pushed a commit
to ProTekk/buun-llama-cpp
that referenced
this pull request
May 20, 2026
* hmx-mm: update debug logging in hmx-mm * hmx-mm: update dequant logic to use HVX_vector_x2/4 * hmx-mm: remove non-pipelined version of the quantize matmul It seems that we don't reall need non-pipelined version * hmx-mm: use activation depth mode and update naming Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com> * hex-mm: minor hmx matmul naming updates * hmx-mm: remove unused vars * snapdragon: scripts bump default ubatch-size to 1K * hexagon: combine HMX and power and clock settings into a single set_power call * hmx-mm: remove leftover of the scale repl helper * hexagon: fix editconf error --------- Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>
max-krasnyansky
added a commit
to qualcomm/llama.cpp
that referenced
this pull request
May 20, 2026
* hmx-mm: update debug logging in hmx-mm * hmx-mm: update dequant logic to use HVX_vector_x2/4 * hmx-mm: remove non-pipelined version of the quantize matmul It seems that we don't reall need non-pipelined version * hmx-mm: use activation depth mode and update naming Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com> * hex-mm: minor hmx matmul naming updates * hmx-mm: remove unused vars * snapdragon: scripts bump default ubatch-size to 1K * hexagon: combine HMX and power and clock settings into a single set_power call * hmx-mm: remove leftover of the scale repl helper * hexagon: fix editconf error --------- Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>
dbrain
pushed a commit
to dbrain/hbd-llama-cpp-turboquant
that referenced
this pull request
May 21, 2026
* hmx-mm: update debug logging in hmx-mm * hmx-mm: update dequant logic to use HVX_vector_x2/4 * hmx-mm: remove non-pipelined version of the quantize matmul It seems that we don't reall need non-pipelined version * hmx-mm: use activation depth mode and update naming Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com> * hex-mm: minor hmx matmul naming updates * hmx-mm: remove unused vars * snapdragon: scripts bump default ubatch-size to 1K * hexagon: combine HMX and power and clock settings into a single set_power call * hmx-mm: remove leftover of the scale repl helper * hexagon: fix editconf error --------- Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>
4 tasks
baramofme
pushed a commit
to baramofme/llama-cpp-turboquant
that referenced
this pull request
May 23, 2026
* hmx-mm: update debug logging in hmx-mm * hmx-mm: update dequant logic to use HVX_vector_x2/4 * hmx-mm: remove non-pipelined version of the quantize matmul It seems that we don't reall need non-pipelined version * hmx-mm: use activation depth mode and update naming Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com> * hex-mm: minor hmx matmul naming updates * hmx-mm: remove unused vars * snapdragon: scripts bump default ubatch-size to 1K * hexagon: combine HMX and power and clock settings into a single set_power call * hmx-mm: remove leftover of the scale repl helper * hexagon: fix editconf error --------- Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>
Jcfunk
added a commit
to Jcfunk/llama.cpp
that referenced
this pull request
May 23, 2026
* upstream/HEAD: (38 commits) vocab : add Carbon-3B (HybridDNATokenizer) support (ggml-org#23410) doc: fix spec mtp typo (ggml-org#23435) ui: Improve Git Hooks for UI development (ggml-org#23403) ggml : Check the right iface method before using the fallback 2d get (ggml-org#23306) llama-graph: fix null-buffer crash in llm_graph_input_attn_kv_iswa for SWA-only models (ggml-org#23131) hexagon: ssm-conv fix for large prompts (ggml-org#23307) app : show version (ggml-org#23426) mtmd, model : merge HunyuanOCR into HunyuanVL and fix OCR vision precision (ggml-org#23329) ui: Add max image size option (ggml-org#22849) Move to backend sampling for MTP draft path (ggml-org#23287) opencl: refactor backend initilization (ggml-org#23318) common/speculative : fix nullptr crash in get_devices_str (ggml-org#23386) mtmd : DeepSeek-OCR image processing fixes, img_tool::resize padding refactor (ggml-org#23345) vulkan: optimize operations in the IM2COL shader (ggml-org#22685) feat: Add WAV MIME type variants and improve audio format detection (ggml-org#23396) hexagon: HMX quantized matmul rework (ggml-org#23368) Programmatic Dependent Launch (PDL) for more performance on newer NVIDIA GPUs (Hopper+) (ggml-org#22522) app : introduce the llama unified executable (ggml-org#23296) refactor: Move text attachments up before the message content in chat completions payload (ggml-org#23406) mtmd: fit_params now take into account mmproj (ggml-org#21489) ...
srossitto79
pushed a commit
to srossitto79/llama.cpp
that referenced
this pull request
May 23, 2026
* hmx-mm: update debug logging in hmx-mm * hmx-mm: update dequant logic to use HVX_vector_x2/4 * hmx-mm: remove non-pipelined version of the quantize matmul It seems that we don't reall need non-pipelined version * hmx-mm: use activation depth mode and update naming Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com> * hex-mm: minor hmx matmul naming updates * hmx-mm: remove unused vars * snapdragon: scripts bump default ubatch-size to 1K * hexagon: combine HMX and power and clock settings into a single set_power call * hmx-mm: remove leftover of the scale repl helper * hexagon: fix editconf error --------- Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>
7 tasks
fewtarius
pushed a commit
to fewtarius/llama.cpp
that referenced
this pull request
May 30, 2026
* hmx-mm: update debug logging in hmx-mm * hmx-mm: update dequant logic to use HVX_vector_x2/4 * hmx-mm: remove non-pipelined version of the quantize matmul It seems that we don't reall need non-pipelined version * hmx-mm: use activation depth mode and update naming Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com> * hex-mm: minor hmx matmul naming updates * hmx-mm: remove unused vars * snapdragon: scripts bump default ubatch-size to 1K * hexagon: combine HMX and power and clock settings into a single set_power call * hmx-mm: remove leftover of the scale repl helper * hexagon: fix editconf error --------- Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com>
turbo-tan
pushed a commit
to turbo-tan/llama.cpp-tq3
that referenced
this pull request
Jun 2, 2026
* hmx-mm: update debug logging in hmx-mm * hmx-mm: update dequant logic to use HVX_vector_x2/4 * hmx-mm: remove non-pipelined version of the quantize matmul It seems that we don't reall need non-pipelined version * hmx-mm: use activation depth mode and update naming Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.com> * hex-mm: minor hmx matmul naming updates * hmx-mm: remove unused vars * snapdragon: scripts bump default ubatch-size to 1K * hexagon: combine HMX and power and clock settings into a single set_power call * hmx-mm: remove leftover of the scale repl helper * hexagon: fix editconf error --------- Co-authored-by: Kim-Chyan Gan <kgan@qti.qualcomm.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.
Overview
This PR updates the HMX matmul to use activation depth mode, and simplifies quantized HMX matmul implementation.
Based on testing with latest models (see the sweep below) we do not really need non-pipelined kernel flavors any more.
Perhaps, at some point those provided benefits but after all the recent updates and fixes they do not.
Additional information
Details
Requirements