root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-09-22 21:16:20 +00:00
Code Issues Actions 15 Packages Projects Releases Wiki Activity

CUDA/HIP: fix tests/test-backend-ops (#8896)

Browse Source
This commit is contained in:
Johannes Gäßler 2024-08-07 09:07:52 +02:00 committed by GitHub
parent 506122d854
commit a8dbc6f753
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
ggml/src/ggml-cuda.cu
View File

@ -2742,11 +2742,12 @@ GGML_CALL static bool ggml_backend_cuda_supports_op(ggml_backend_t backend, cons
case GGML_OP_MUL_MAT_ID:
{
struct ggml_tensor * a = op->src[0];
if (op->op == GGML_OP_MUL_MAT) {
struct ggml_tensor * b = op->src[1];
if (a->ne[3] != b->ne[3]) {
if (b->type == GGML_TYPE_F16 && a->type != GGML_TYPE_F16) {
return false;
}
if (op->op == GGML_OP_MUL_MAT && a->ne[3] != b->ne[3]) {
return false;
}
switch (a->type) {
case GGML_TYPE_F32:
@ -2877,7 +2878,7 @@ GGML_CALL static bool ggml_backend_cuda_supports_op(ggml_backend_t backend, cons
return true;
case GGML_OP_FLASH_ATTN_EXT:
#if defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__)
return op->src[0]->ne[0] == 64 || op->src[0]->ne[0] == 128;
return (op->src[0]->ne[0] == 64 && op->src[1]->type == GGML_TYPE_F16) || op->src[0]->ne[0] == 128;
#else
if (op->src[0]->ne[0] == 128) {
return true;