CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

metal : extend mat-mat multiplication support (#16225)

Browse Source 
* metal : support mul_mm with src1->type == GGML_TYPE_F16

* metal : support mul_mm_id with src1->type == GGML_TYPE_F16

[no ci]

* metal : mul_mm support ne00 % 32 != 0

* metal : support mul_mm_id with ne00 % 32 != 0

* cont : remove unnecessary unrolls

* cont : simplify data loading

* metal : optimize mul_mm when output bounds checks are not needed
This commit is contained in:
Georgi Gerganov
2025-09-28 09:34:44 +03:00
committed by GitHub
parent 3b53634fe3
commit 6a2c6145a0
7 changed files with 247 additions and 120 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
tests/test-backend-ops.cpp
View File

@@ -6301,6 +6301,22 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F32, GGML_TYPE_F32, 16, 32, 32, { 1, 1}, {1, 1}, {0, 1, 2, 3}, true, 3));
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F32, GGML_TYPE_F32, 64, 77, 77, {12,1}, {1,1}));
#if 0
// test the mat-mat path for Metal
for (int k = 1; k < 512; ++k) {
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F16, GGML_TYPE_F32, 64, 127, k, {12,1}, {1,1}));
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F32, GGML_TYPE_F32, 64, 127, k, {12,1}, {1,1}));
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F16, GGML_TYPE_F32, 64, 77, k, {12,1}, {1,1}));
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F32, GGML_TYPE_F32, 64, 77, k, {12,1}, {1,1}));
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F16, GGML_TYPE_F32, 64, 128, k, {12,1}, {1,1}));
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F32, GGML_TYPE_F32, 64, 128, k, {12,1}, {1,1}));
test_cases.emplace_back(new test_mul_mat_id(GGML_TYPE_F16, GGML_TYPE_F32, 16, 16, false, 50, 200, k));
test_cases.emplace_back(new test_mul_mat_id(GGML_TYPE_F16, GGML_TYPE_F32, 16, 16, true, 50, 200, k));
test_cases.emplace_back(new test_mul_mat_id(GGML_TYPE_F32, GGML_TYPE_F32, 16, 16, false, 50, 200, k));
test_cases.emplace_back(new test_mul_mat_id(GGML_TYPE_F32, GGML_TYPE_F32, 16, 16, true, 50, 200, k));
}
#endif
for (auto bs2 : {1,3}) {
for (auto bs : {1,2,4,8}) {
for (auto nr : {1,4}) {