ggml-cpu: extend RVV quantization vec dot to higher VLENs (llama/22754)

Browse Source

* ggml-cpu: add rvv 512b,1024b impls for iq4_xs

* ggml-cpu: refactor; add rvv 512b, 1024b impls for q6_K, i-quants

* ggml-cpu: refactor; add 512 and 1024 implementations of tq3_s, iq3_xxs, iq2_s, iq2_xs, iq2_xxs

improve iq2_xs impl for rvv 256

Co-authored-by: Rehan Qasim <rehan.qasim@10xengineers.ai>

---------

Co-authored-by: taimur-10x <taimur.ahmad@10xengineers.ai>
Co-authored-by: Rehan Qasim <rehan.qasim@10xengineers.ai>

...

This commit is contained in:

rehan-10xengineer

2026-06-08 14:36:36 +03:00

committed by Georgi Gerganov

co-authored by Rehan Qasim taimur-10x

parent 00a9728de3

commit a1a3186887

1 changed files with 3025 additions and 982 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

ggml/src/ggml-cpu/arch/riscv/quants.c

+3025 -982

File diff suppressed because it is too large Load Diff