Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Vulkan: MMVQ Integer Dot K-Quant and MUL_MAT_ID support (#16900)

Browse Source 
* vulkan: split mul_mmq_funcs for mul_mat_vecq use

* add mxfp4 mmvq

* add q2_k mmvq

* add q3_k mmvq

* add q4_k and q5_k mmvq

* add q6_k mmvq

* handle 4x4 quants per mmvq thread

* enable MUL_MAT_ID mmvq support

* enable subgroup optimizations for mul_mat_vec_id shaders

* device tuning

* request prealloc_y sync after quantization

* fix indentation

* fix llvmpipe test failures

* fix mul_mat_id mmvq condition

* fix unused variable warning
This commit is contained in:
Ruben Ortlam
2025-11-29 09:37:22 +01:00
committed by GitHub
parent 59d8d4e963
commit 47a268ea50
12 changed files with 682 additions and 288 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-vulkan/vulkan-shaders/dequant_funcs.glsl
-7
View File
@@ -4,13 +4,6 @@
#include "types.glsl"
#if defined(A_TYPE_PACKED16)
layout (binding = 0) readonly buffer A_PACKED16 {A_TYPE_PACKED16 data_a_packed16[];};
#endif
#if defined(A_TYPE_PACKED32)
layout (binding = 0) readonly buffer A_PACKED32 {A_TYPE_PACKED32 data_a_packed32[];};
#endif
#if defined(DATA_A_F32)
vec2 dequantize(uint ib, uint iqs, uint a_offset) {
return vec2(data_a[a_offset + ib], data_a[a_offset + ib + 1]);