Ruben Ortlam
47a268ea50
* 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
36 lines
1.1 KiB
GLSL
36 lines
1.1 KiB
GLSL
#include "types.glsl"
|
|
|
|
#define MAT_VEC_FUSION_FLAGS_BIAS0 0x1
|
|
#define MAT_VEC_FUSION_FLAGS_BIAS1 0x2
|
|
#define MAT_VEC_FUSION_FLAGS_SCALE0 0x4
|
|
#define MAT_VEC_FUSION_FLAGS_SCALE1 0x8
|
|
|
|
layout (binding = 0) readonly buffer A {A_TYPE data_a[];};
|
|
#if defined(A_TYPE_VEC4)
|
|
layout (binding = 0) readonly buffer AV4 {A_TYPE_VEC4 data_a_v4[];};
|
|
#endif
|
|
#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
|
|
|
|
layout (binding = 1) readonly buffer B {B_TYPE data_b[];};
|
|
#ifdef B_TYPE_VEC2
|
|
layout (binding = 1) readonly buffer BV2 {B_TYPE_VEC2 data_b_v2[];};
|
|
#endif
|
|
#ifdef B_TYPE_VEC4
|
|
layout (binding = 1) readonly buffer BV4 {B_TYPE_VEC4 data_b_v4[];};
|
|
#endif
|
|
|
|
layout (binding = 2) writeonly buffer D {D_TYPE data_d[];};
|
|
|
|
layout (binding = 3) readonly buffer Fuse0 {D_TYPE data_fuse0[];};
|
|
layout (binding = 4) readonly buffer Fuse1 {D_TYPE data_fuse1[];};
|
|
|
|
#ifdef MUL_MAT_ID
|
|
layout (binding = 5) readonly buffer IDS {int data_ids[];};
|
|
#endif
|
|
|