Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

vulkan: Change mul_mat_id to pad K rather than N (#27925)

Browse Source 
The N padding is needed for mul_mat, but not mul_mat_id. For mul_mat_id,
we indirect the row index through a shared memory lookup table which avoids
any OOB row coordinate. But that callback doesn't bounds check K, so we
actually need K padding instead.
This commit is contained in:
Jeff Bolz
2026-08-29 10:09:24 +03:00
committed by GitHub
parent d7bd3bfcad
commit 77f132cb1d
5 changed files with 68 additions and 52 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/test-backend-ops.cpp
+6 -1
View File
@@ -9367,7 +9367,12 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
test_cases.emplace_back(new test_mul_mat_id(GGML_TYPE_BF16, GGML_TYPE_F32, 16, 16, b, 50, 200, 64));
}
test_cases.emplace_back(new test_mul_mat_id(GGML_TYPE_F16, GGML_TYPE_F32, 1, 1, false, 8, 16, 1));
// For issue 27873
test_cases.emplace_back(new test_mul_mat_id(GGML_TYPE_IQ2_XXS, GGML_TYPE_F32, 1, 1, false, 1, 8192, 4096));
for (int k : {1, 63, 65}) {
test_cases.emplace_back(new test_mul_mat_id(GGML_TYPE_F16, GGML_TYPE_F32, 1, 1, false, 8, 16, k));
}
test_cases.emplace_back(new test_mul_mat_id_fusion(GGML_TYPE_F16, GGML_TYPE_F32, 16, 16, false, 32, 32, 32, 3));
// gpt-oss issue with Vulkan mmq_id