Skip to content

sycl: fix mul_mat selection #15092

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Aug 5, 2025
Merged

sycl: fix mul_mat selection #15092

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion ggml/src/ggml-sycl/ggml-sycl.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2609,6 +2609,8 @@ static void ggml_sycl_mul_mat_vec_nc(ggml_backend_sycl_context & ctx, const ggml
GGML_ASSERT(!ggml_backend_buffer_is_sycl_split(src0->buffer));
GGML_ASSERT(src0->type == GGML_TYPE_F16);
GGML_ASSERT(src1->type == GGML_TYPE_F32);
GGML_ASSERT(src1->ne[1] == 1);
GGML_ASSERT(src1->ne[3] == 1);

const int64_t ne00 = src0->ne[0];
const int64_t ne01 = src0->ne[1];
Expand Down Expand Up @@ -3196,7 +3198,7 @@ static void ggml_sycl_mul_mat(ggml_backend_sycl_context & ctx, const ggml_tensor
// The kernel from the if path is faster for that specific case, but does not support all mul mats.
ggml_sycl_mul_mat_batched_sycl(ctx, src0, src1, dst);
}
} else if (!split && src0->type == GGML_TYPE_F16 && !ggml_is_contiguous(src0) && !ggml_is_transposed(src1) && src1->ne[1] == 1) {
} else if (!split && src0->type == GGML_TYPE_F16 && !ggml_is_contiguous(src0) && !ggml_is_transposed(src1) && src1->ne[1] == 1 && src1->ne[3] == 1) {
// KQV single-batch
ggml_sycl_mul_mat_vec_nc(ctx, src0, src1, dst);
} else if (!split && src0->type == GGML_TYPE_F16 && !ggml_is_transposed(src0) && !ggml_is_transposed(src1) && src1->ne[2] * src1->ne[3] > 1) {
Expand Down
Loading