CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

CANN: Fix failed test cases (#12708)

Browse Source 
* CANN: Fix memory waste in aclnn_tensor

* CANN: fix backend ops fail

* CANN: fix acl_tensor memory alloc.

* CANN: format

* CANN: remove trailing whitespace
This commit is contained in:
hipudding
2025-04-03 08:49:51 +08:00
committed by GitHub
parent 97a20c012b
commit 2a0dc97e56
5 changed files with 30 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
ggml/src/ggml-cann/aclnn_ops.cpp
View File

@@ -358,8 +358,6 @@ void ggml_cann_sqr(ggml_backend_cann_context& ctx, ggml_tensor* dst) {
void ggml_cann_clamp(ggml_backend_cann_context& ctx, ggml_tensor* dst) {
ggml_tensor* src = dst->src[0];
GGML_ASSERT(src->type == GGML_TYPE_F32);
GGML_ASSERT(dst->type == GGML_TYPE_F32);
float min;
float max;
@@ -1090,8 +1088,6 @@ void ggml_cann_rms_norm(ggml_backend_cann_context& ctx, ggml_tensor* dst) {
float eps;
memcpy(&eps, dst->op_params, sizeof(float));
GGML_ASSERT(eps > 0.0f);
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
void* workspaceAddr = nullptr;
@@ -3152,7 +3148,7 @@ void ggml_cann_rope(ggml_backend_cann_context& ctx, ggml_tensor* dst) {
// TODO: use ascendc
// Only test with LLAMA model.
ggml_tensor* src0 = dst->src[0]; // input
ggml_tensor* src2 = dst->src[2]; // freq_factors
// ggml_tensor* src2 = dst->src[2]; // freq_factors, not used now.
// param
float freq_base, freq_scale, ext_factor, attn_factor, beta_fast, beta_slow;