Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

metal : add CONV_2D_DW (depthwise convolution) support (#21565)

Browse Source 
* metal : add CONV_2D_DW (depthwise 2D convolution) support

* test : add perf cases for CONV_2D_DW

* metal : use 3D dispatch for CONV_2D_DW kernel

* metal : add channel-tiled CONV_2D_DW kernel for non-contiguous layouts

* metal : simplify CONV_2D_DW dispatch and trim comments

* metal : merge duplicate CONV_2D_DW pipeline getters

* tests : add F16 CONV2D_DW tests

* cpu : fix F16 kernel support for CONV_2D_DW

* tests : remove commented-out CONV_2D_DW test block

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Sou-ly
2026-07-09 12:29:15 +03:00
committed by GitHub
co-authored by Georgi Gerganov
parent ccb0c34223
commit 92b187c97e
9 changed files with 381 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-metal/ggml-metal-device.cpp
+23
View File
@@ -1869,6 +1869,29 @@ ggml_metal_pipeline_with_params ggml_metal_library_get_pipeline_conv_2d(ggml_met
return res;
}
ggml_metal_pipeline_with_params ggml_metal_library_get_pipeline_conv_2d_dw(ggml_metal_library_t lib, const ggml_tensor * op, bool tiled) {
assert(op->op == GGML_OP_CONV_2D_DW);
GGML_ASSERT(op->src[0]->type == GGML_TYPE_F16 || op->src[0]->type == GGML_TYPE_F32);
GGML_ASSERT(op->src[1]->type == GGML_TYPE_F32);
GGML_ASSERT(op->type == GGML_TYPE_F32);
char base[256];
char name[256];
snprintf(base, 256, "kernel_conv_2d_dw%s_%s_%s",
tiled ? "_tiled" : "",
ggml_type_name(op->src[0]->type), ggml_type_name(op->src[1]->type));
snprintf(name, 256, "%s", base);
ggml_metal_pipeline_with_params res = ggml_metal_library_get_pipeline(lib, name);
if (!res.pipeline) {
res = ggml_metal_library_compile_pipeline(lib, base, name, nullptr);
}
return res;
}
ggml_metal_pipeline_with_params ggml_metal_library_get_pipeline_conv_3d(ggml_metal_library_t lib, const ggml_tensor * op) {
assert(op->op == GGML_OP_CONV_3D);