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

opencl: fix leak in Adreno q8_0 path (#21212)

Browse source
This commit is contained in:
lhez 2026-04-01 12:54:58 -07:00 • committed by GitHub
parent 12dbf1da95
commit 95a6ebabb2
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 9 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

12
ggml/src/ggml-opencl/ggml-opencl.cpp
View file

@ -9612,6 +9612,9 @@ static void ggml_cl_mul_mat_q8_0_f32_adreno(ggml_backend_t backend, const ggml_t
cl_mem B_image1d; cl_mem B_image1d;
cl_mem B_sub_buffer; cl_mem B_sub_buffer;
cl_mem S_image1d; cl_mem S_image1d;
// for B transpose
cl_mem B_image1d_trans = nullptr;
cl_mem B_d = nullptr;
cl_mem D_image1d; cl_mem D_image1d;
cl_mem D_sub_buffer; cl_mem D_sub_buffer;
@ -9703,9 +9706,6 @@ static void ggml_cl_mul_mat_q8_0_f32_adreno(ggml_backend_t backend, const ggml_t
global_work_size[2] = 1; global_work_size[2] = 1;
} else { } else {
cl_ulong offsetd = extrad->offset + dst->view_offs; cl_ulong offsetd = extrad->offset + dst->view_offs;
cl_mem B_image1d_trans = nullptr;
// for B transpose
cl_mem B_d = nullptr;
int padding; int padding;
//how many extra elements beyond multiple of 8 //how many extra elements beyond multiple of 8
@ -9800,6 +9800,12 @@ static void ggml_cl_mul_mat_q8_0_f32_adreno(ggml_backend_t backend, const ggml_t
CL_CHECK(clReleaseMemObject(S_image1d)); CL_CHECK(clReleaseMemObject(S_image1d));
CL_CHECK(clReleaseMemObject(D_sub_buffer)); CL_CHECK(clReleaseMemObject(D_sub_buffer));
CL_CHECK(clReleaseMemObject(D_image1d)); CL_CHECK(clReleaseMemObject(D_image1d));
if (B_image1d_trans) {
CL_CHECK(clReleaseMemObject(B_image1d_trans));
}
if (B_d) {
CL_CHECK(clReleaseMemObject(B_d));
}
#else #else
GGML_UNUSED(backend); GGML_UNUSED(backend);
GGML_UNUSED(src0); GGML_UNUSED(src0);