Public/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-20 01:31:31 +02:00
Code Issues Packages Projects Releases Wiki Activity

rm cuda supports_op guard, fix webgpu clang-format

Browse Source
This commit is contained in:
Xuan Son Nguyen
2026-08-15 17:01:02 +02:00
parent 8f75594531
commit 5f4001e047
2 changed files with 2 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/ggml-cuda.cu
-3
View File
@@ -5233,9 +5233,6 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
return false;
case GGML_OP_ROPE:
case GGML_OP_ROPE_BACK: {
if (((const int32_t *) op->op_params)[15] != 0) {
return false; // FIXME: support ggml_rope_set_offset
}
return op->src[0]->nb[0] == ggml_type_size(op->src[0]->type) && ggml_is_contiguous_2(op->src[0]);
}
case GGML_OP_IM2COL:
ggml/src/ggml-webgpu/ggml-webgpu.cpp
+2 -2
View File
@@ -4456,8 +4456,8 @@ static bool ggml_backend_webgpu_device_supports_op(ggml_backend_dev_t dev, const
break;
case GGML_OP_ROPE:
// FIXME: support ggml_rope_set_offset
supports_op = (op->type == GGML_TYPE_F32 || op->type == GGML_TYPE_F16) &&
((const int32_t *) op->op_params)[15] == 0;
supports_op =
(op->type == GGML_TYPE_F32 || op->type == GGML_TYPE_F16) && ((const int32_t *) op->op_params)[15] == 0;
break;
case GGML_OP_GLU:
switch (ggml_get_glu_op(op)) {