thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

SOLVE_TRI CUDA kernel for small matrices (#17457)

Browse source
This commit is contained in:
Piotr Wilkin (ilintar) 2025-11-28 05:15:32 +01:00 • committed by GitHub
parent efaaccdd69
commit cd0e3a7a3b
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 215 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

3
tests/test-backend-ops.cpp
View file

@ -7935,6 +7935,9 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_perf() {
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F16, GGML_TYPE_F32, 16416, 1, 128, {8, 1}, {4, 1}, {0, 2, 1, 3}));
test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F16, GGML_TYPE_F32, 128, 1, 16416, {8, 1}, {4, 1}, {0, 1, 2, 3}, 2*16416));
test_cases.emplace_back(new test_solve_tri(GGML_TYPE_F32, { 64, 64, 4, 2 }, { 6, 64, 4, 2 }));
test_cases.emplace_back(new test_solve_tri(GGML_TYPE_F32, { 128, 128, 4, 1 }, { 8, 128, 4, 1 }));
for (int bs : {1, 2, 3, 4, 5, 8, 512}) {
for (ggml_type type_a : all_types) {
for (ggml_type type_b : {GGML_TYPE_F32}) {