CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-13 10:57:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

vulkan : support ggml_mean (#15393)

Browse Source 
* vulkan : support ggml_mean

* vulkan : support sum, sum_rows and mean with non-contiguous tensors

* vulkan : fix subbuffer size not accounting for misalign offset

* tests : add backend-op tests for non-contiguous sum_rows

* cuda : require contiguous src for SUM_ROWS, MEAN support
* sycl : require contiguous src for SUM, SUM_ROWS, ARGSORT support

* require ggml_contiguous_rows in supports_op and expect nb00=1 in the shader
This commit is contained in:
Acly
2025-08-23 08:35:21 +02:00
committed by GitHub
parent 330c3d2d21
commit 0a9b43e507
5 changed files with 135 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
ggml/src/ggml-cuda/ggml-cuda.cu
View File

@@ -3485,11 +3485,11 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
case GGML_OP_CONV_TRANSPOSE_2D:
case GGML_OP_POOL_2D:
case GGML_OP_SUM:
case GGML_OP_SUM_ROWS:
case GGML_OP_MEAN:
case GGML_OP_ARGSORT:
case GGML_OP_ACC:
return true;
case GGML_OP_SUM_ROWS:
case GGML_OP_MEAN:
case GGML_OP_GROUP_NORM:
return ggml_is_contiguous(op->src[0]);
case GGML_OP_UPSCALE: