tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 04:35:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

vulkan: add specific MMV kernels for IQ2 and IQ3 quants + optimizations (#11595)

Browse Source 
* vulkan: implement specialized MMV kernels for IQ2 quantizations

* vulkan: add MMV kernels for IQ3 quants

* vulkan: Increase MMV batch size and unroll IQ LUT setup

* vulkan: fix init_iq_shmem for WG sizes larger than tables

* vulkan: common batch size for all I-quants
This commit is contained in:
Rémy O
2025-02-28 09:42:52 +01:00
committed by GitHub
parent 9c42b1718c
commit 438a83926a
9 changed files with 509 additions and 42 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp
View File

@ -402,7 +402,7 @@ void process_shaders() {
for (const auto& tname : type_names) {
// mul mat vec
std::string data_a_key = "DATA_A_" + to_uppercase(tname);
std::string shader = (string_ends_with(tname, "_k") || string_starts_with(tname, "iq1_")) ? "mul_mat_vec_" + tname + ".comp" : "mul_mat_vec.comp";
std::string shader = (string_ends_with(tname, "_k") || string_starts_with(tname, "iq1_") || string_starts_with(tname, "iq2_") || string_starts_with(tname, "iq3_")) ? "mul_mat_vec_" + tname + ".comp" : "mul_mat_vec.comp";
string_to_spv("mul_mat_vec_" + tname + "_f32_f32", shader, merge_maps(base_dict, {{data_a_key, "1"}, {"B_TYPE", "float"}, {"B_TYPE_VEC2", "vec2"}, {"B_TYPE_VEC4", "vec4"}, {"D_TYPE", "float"}}));
string_to_spv("mul_mat_vec_" + tname + "_f16_f32", shader, merge_maps(base_dict, {{data_a_key, "1"}, {"B_TYPE", "float16_t"}, {"B_TYPE_VEC2", "f16vec2"}, {"B_TYPE_VEC4", "f16vec4"}, {"D_TYPE", "float"}}));