vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-14 02:49:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	common/CMakeLists.txt
#	docs/backend/SYCL.md
#	ggml/CMakeLists.txt
#	ggml/src/ggml-sycl/CMakeLists.txt
#	ggml/src/ggml-sycl/binbcast.cpp
#	ggml/src/ggml-sycl/convert.cpp
#	ggml/src/ggml-sycl/dequantize.hpp
#	ggml/src/ggml-sycl/dmmv.cpp
#	ggml/src/ggml-sycl/gemm.hpp
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	ggml/src/ggml-sycl/mmvq.cpp
#	ggml/src/ggml-sycl/vecdotq.hpp
#	ggml/src/ggml-vulkan/CMakeLists.txt
#	ggml/src/ggml-vulkan/vulkan-shaders/CMakeLists.txt
#	ggml/src/gguf.cpp
#	scripts/compare-llama-bench.py
#	tests/CMakeLists.txt
#	tests/test-chat.cpp
#	tools/llama-bench/llama-bench.cpp
#	tools/server/README.md
This commit is contained in:
Concedo 2025-05-16 15:30:31 +08:00
commit e5d26a2356
47 changed files with 2671 additions and 504 deletions
Download patch file Download diff file Expand all files Collapse all files

13
ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp
View file

@ -227,7 +227,7 @@ static std::mutex compile_count_mutex;
static std::condition_variable compile_count_cond;
void string_to_spv_func(const std::string& _name, const std::string& in_fname, const std::map<std::string, std::string>& defines, bool fp16 = true, bool coopmat = false, bool coopmat2 = false, bool f16acc = false) {
std::string name = _name + (f16acc ? "_f16acc" : "") + (coopmat ? "_coopmat" : "") + (coopmat2 ? "_cm2" : (fp16 ? "" : "_fp32"));
std::string name = _name + (f16acc ? "_f16acc" : "") + (coopmat ? "_cm1" : "") + (coopmat2 ? "_cm2" : (fp16 ? "" : "_fp32"));
std::string out_fname = join_paths(output_dir, name + ".spv");
std::string in_path = join_paths(input_dir, in_fname);
@ -438,6 +438,7 @@ void process_shaders() {
// flash attention
for (const auto& f16acc : {false, true}) {
std::string acctype = f16acc ? "float16_t" : "float";
std::string acctypev4 = f16acc ? "f16vec4" : "vec4";
for (const auto& tname : type_names) {
if (tname == "f32") {
@ -454,6 +455,16 @@ void process_shaders() {
string_to_spv("flash_attn_f32_f16_" + tname, "flash_attn_cm2.comp",
merge_maps(base_dict, {{data_a_key, "1"}, {"Q_TYPE", "float"}, {"D_TYPE", "float"}, {"ACC_TYPE", acctype}, {"DEQUANTFUNC", "dequantFunc"+to_uppercase(tname) }, {"BLOCK_SIZE", "QUANT_K_"+to_uppercase(tname) }}), true, false, true, f16acc);
}
#endif
#if defined(GGML_VULKAN_COOPMAT_GLSLC_SUPPORT)
if (tname == "f16") {
string_to_spv("flash_attn_f32_f16_" + tname, "flash_attn_cm1.comp",
merge_maps(base_dict, {{"Q_TYPE", "float"}, {"D_TYPE", "float"}, {"ACC_TYPE", acctype}, {"ACC_TYPEV4", acctypev4}, {"COOPMAT", "1"}}), true, true, false, f16acc);
} else if (tname == "q4_0" || tname == "q8_0") {
std::string data_a_key = "DATA_A_" + to_uppercase(tname);
string_to_spv("flash_attn_f32_f16_" + tname, "flash_attn_cm1.comp",
merge_maps(base_dict, {{data_a_key, "1"}, {"Q_TYPE", "float"}, {"D_TYPE", "float"}, {"ACC_TYPE", acctype}, {"ACC_TYPEV4", acctypev4}, {"BLOCK_SIZE", "QUANT_K_"+to_uppercase(tname)}, {"COOPMAT", "1"}}), true, true, false, f16acc);
}
#endif
if (tname == "f16") {
string_to_spv("flash_attn_f32_f16_" + tname, "flash_attn.comp",