vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-13 02:19:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	.github/workflows/build.yml
#	CMakeLists.txt
#	Makefile
#	README.md
#	common/CMakeLists.txt
#	docs/backend/SYCL.md
#	docs/build.md
#	docs/docker.md
#	examples/export-lora/export-lora.cpp
#	examples/main/README.md
#	examples/main/main.cpp
#	examples/run/README.md
#	examples/run/run.cpp
#	examples/server/README.md
#	examples/simple-chat/simple-chat.cpp
#	ggml/CMakeLists.txt
#	ggml/src/ggml-hip/CMakeLists.txt
#	src/CMakeLists.txt
#	tests/test-backend-ops.cpp
#	tests/test-chat-template.cpp
This commit is contained in:
Concedo 2025-01-25 14:16:50 +08:00
commit bec231422a
46 changed files with 4305 additions and 578 deletions
Download patch file Download diff file Expand all files Collapse all files

10
ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp
View file

@ -17,6 +17,7 @@
#include <cstring>
#include <cstdlib>
#include <cassert>
#include <algorithm>
#include <sys/stat.h>
#include <sys/types.h>
#include <algorithm>
@ -24,7 +25,6 @@
#ifdef _WIN32
#include <windows.h>
#include <direct.h> // For _mkdir on Windows
#include <algorithm> // For std::replace on w64devkit
#else
#include <unistd.h>
#include <sys/wait.h>
@ -324,8 +324,11 @@ void matmul_shaders(bool fp16, bool matmul_id, bool coopmat, bool coopmat2, bool
// For aligned matmul loads
std::string load_vec_a = (coopmat2 || tname == "f32" || tname == "f16") ? load_vec : "2";
string_to_spv(shader_name + "_" + tname + "_f32", source_name, merge_maps(base_dict, {{data_a_key, "1"}, {"LOAD_VEC_A", load_vec_a_unaligned}, {"B_TYPE", "float"}, {"D_TYPE", "float"}, {"B_IS_FLOAT", "1"}}), fp16, coopmat, coopmat2, f16acc);
string_to_spv(shader_name + "_" + tname + "_f32_aligned", source_name, merge_maps(base_dict, {{data_a_key, "1"}, {"LOAD_VEC_A", load_vec_a}, {"LOAD_VEC_B", load_vec}, {"B_TYPE", aligned_b_type_f32}, {"D_TYPE", "float"}, {"B_IS_FLOAT", "1"}, {"ALIGNED", "1"}}), fp16, coopmat, coopmat2, f16acc);
// don't generate f32 variants for coopmat2
if (!coopmat2) {
string_to_spv(shader_name + "_" + tname + "_f32", source_name, merge_maps(base_dict, {{data_a_key, "1"}, {"LOAD_VEC_A", load_vec_a_unaligned}, {"B_TYPE", "float"}, {"D_TYPE", "float"}, {"B_IS_FLOAT", "1"}}), fp16, coopmat, coopmat2, f16acc);
string_to_spv(shader_name + "_" + tname + "_f32_aligned", source_name, merge_maps(base_dict, {{data_a_key, "1"}, {"LOAD_VEC_A", load_vec_a}, {"LOAD_VEC_B", load_vec}, {"B_TYPE", aligned_b_type_f32}, {"D_TYPE", "float"}, {"B_IS_FLOAT", "1"}, {"ALIGNED", "1"}}), fp16, coopmat, coopmat2, f16acc);
}
if (tname != "f16" && tname != "f32") {
string_to_spv(shader_name + "_" + tname + "_f16", source_name, merge_maps(base_dict, {{data_a_key, "1"}, {"LOAD_VEC_A", load_vec_a_unaligned}, {"B_TYPE", "float16_t"}, {"D_TYPE", "float"}, {"B_IS_FLOAT", "1"}}), fp16, coopmat, coopmat2, f16acc);
@ -507,6 +510,7 @@ void write_output_files() {
fprintf(hdr, "#include <cstdint>\n\n");
fprintf(src, "#include \"%s\"\n\n", basename(target_hpp).c_str());
//std::sort(shader_fnames.begin(), shader_fnames.end()); //remove sort for kcpp for now
for (const auto& pair : shader_fnames) {
const std::string& name = pair.first;
#ifdef _WIN32