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

CUDA: add conv_2d_dw (#14265)

Browse Source 
* CUDA: add conv_2d_dw

* better naming

* simplify using template

* Review: fix operation ordering in ggml-cuda, use __forceinline__, use more const
This commit is contained in:
Aman Gupta
2025-06-20 09:50:24 +08:00
committed by GitHub
parent 8f71d0f3e8
commit 9eaa51e7f0
3 changed files with 171 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
ggml/src/ggml-cuda/ggml-cuda.cu
View File

@@ -11,6 +11,7 @@
#include "ggml-cuda/clamp.cuh"
#include "ggml-cuda/concat.cuh"
#include "ggml-cuda/conv-transpose-1d.cuh"
#include "ggml-cuda/conv2d-dw.cuh"
#include "ggml-cuda/convert.cuh"
#include "ggml-cuda/count-equal.cuh"
#include "ggml-cuda/cpy.cuh"
@@ -2310,6 +2311,9 @@ static bool ggml_cuda_compute_forward(ggml_backend_cuda_context & ctx, struct gg
case GGML_OP_IM2COL:
ggml_cuda_op_im2col(ctx, dst);
break;
case GGML_OP_CONV_2D_DW:
ggml_cuda_op_conv2d_dw(ctx, dst);
break;
case GGML_OP_CONV_TRANSPOSE_1D:
ggml_cuda_op_conv_transpose_1d(ctx,dst);
break;
@@ -3209,6 +3213,7 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
return op->src[0]->nb[0] == ggml_type_size(op->src[0]->type) && ggml_is_contiguous_2(op->src[0]);
}
case GGML_OP_IM2COL:
case GGML_OP_CONV_2D_DW:
case GGML_OP_POOL_2D:
case GGML_OP_SUM:
case GGML_OP_SUM_ROWS: