happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

cuda : add error checking for cudaMemcpyAsync in argsort (#17599) 

* cuda : add error checking for cudaMemcpyAsync in argsort (#12836)

* fix indentation
This commit is contained in:
Mahekk Shaikh 2025-11-29 19:16:28 -05:00 committed by GitHub
parent 385c3da5e6
commit 00425e2ed1
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/ggml-cuda/argsort.cu
View File

@ -44,7 +44,7 @@ static void argsort_f32_i32_cuda_cub(ggml_cuda_pool & pool,
const dim3 offset_grid((nrows + block_size - 1) / block_size); const dim3 offset_grid((nrows + block_size - 1) / block_size);
init_offsets<<<offset_grid, block_size, 0, stream>>>(d_offsets, ncols, nrows); init_offsets<<<offset_grid, block_size, 0, stream>>>(d_offsets, ncols, nrows);
cudaMemcpyAsync(temp_keys, x, ncols * nrows * sizeof(float), cudaMemcpyDeviceToDevice, stream); CUDA_CHECK(cudaMemcpyAsync(temp_keys, x, ncols * nrows * sizeof(float), cudaMemcpyDeviceToDevice, stream));
size_t temp_storage_bytes = 0; size_t temp_storage_bytes = 0;