ggml-cpu: extend RVV quantization vec dot to higher VLENs#22754
Merged
ggerganov merged 3 commits intoJun 4, 2026
Conversation
rehan-10xengineer
changed the title
github-actions
Bot
added
the
ggml
taimur-10x
force-pushed
the
10x/riscv-quant-vec-dot-vlens
branch
2 times, most recently
from
1e611d6 to
3db486a
Compare
Contributor
Author
|
@xctan @ggerganov Could you please review this PR? |
xctan
approved these changes
May 23, 2026
taimur-10x
force-pushed
the
10x/riscv-quant-vec-dot-vlens
branch
2 times, most recently
from
a3bddb3 to
92d3a50
Compare
…s, iq2_s, iq2_xs, iq2_xxs improve iq2_xs impl for rvv 256 Co-authored-by: Rehan Qasim <rehan.qasim@10xengineers.ai>
taimur-10x
force-pushed
the
10x/riscv-quant-vec-dot-vlens
branch
from
92d3a50 to
a29a5ff
Compare
ggerganov
added
the
merge ready
Member
|
Merging on green CI |
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
5 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
This PR adds RVV implementations for quantized vector dot kernels (for VLENs 512-bit and 1024-bit).
Key Changes
ggml_vec_dot_tq1_0_q8_K_vl512.Testing
Kernels were functionally tested through
test-quantize-fnsfor VLENs512-bitand1024-bitwith QEMU, usingtest-quantize-fns.