ggml-cpu : add basic RVV support for vector f32 ops #15057
Merged
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
ggerganov
approved these changes
Aug 27, 2025
There was a problem hiding this comment.
A bit hesitant about these changes because the GGML_SIMD abstraction was quite simple at unifying the various SIMD implementations. And now with SVE and RVV that do not fit in GGML_SIMD we add extra branches which complicates the code, while AFAIK the SVE and RVV hardware is not very ubiquitous today. Hence we increase the maintenance effort quite a bit disproportionately from the benefits.
Would be nice to think about some refactoring here to make the implementation more manageable in the future.
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) ...
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 menloresearch/llama.cpp
that referenced
this pull request
Aug 29, 2025
* ggml-cpu : add basic RVV support for vector f32 ops * ggml-cpu : add RVV support for f32 softmax
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.
This PR introduces RVV support for several f32 vector kernels.
The implementation required refactoring the vectorization logic. Due to RVV's flexible vector length, its intrinsic types are sizeless, which prevents the compiler from creating arrays of vector registers (a similar limitation also can be found in Arm's SVE). This makes traditional loop unrolling techniques incompatible, necessitating a rewrite of the code to support RVV's architecture.