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

refactor: simplify kernel argument setting in ggml_cl_div function

This commit is contained in:
chraac 2026-01-05 11:46:06 +08:00
parent 220a33afe8
commit e1ac6411d3
1 changed files with 12 additions and 55 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

67
ggml/src/ggml-opencl/ggml-opencl.cpp
View File

@ -5183,38 +5183,13 @@ static void ggml_cl_div(ggml_backend_t backend, const ggml_tensor * src0, const
const int ne02 = src0->ne[2];
const int ne03 = src0->ne[3];
const cl_ulong nb00 = src0->nb[0];
const cl_ulong nb01 = src0->nb[1];
const cl_ulong nb02 = src0->nb[2];
const cl_ulong nb03 = src0->nb[3];
const int ne10 = src1->ne[0];
const int ne11 = src1->ne[1];
const int ne12 = src1->ne[2];
const int ne13 = src1->ne[3];
const cl_ulong nb10 = src1->nb[0];
const cl_ulong nb11 = src1->nb[1];
const cl_ulong nb12 = src1->nb[2];
const cl_ulong nb13 = src1->nb[3];
const int ne0 = dst->ne[0];
const cl_ulong nb0 = dst->nb[0];
const cl_ulong nb1 = dst->nb[1];
const cl_ulong nb2 = dst->nb[2];
const cl_ulong nb3 = dst->nb[3];
ggml_backend_opencl_context *backend_ctx = (ggml_backend_opencl_context *)backend->context;
ggml_tensor_extra_cl * extra0 = (ggml_tensor_extra_cl *)src0->extra;
ggml_tensor_extra_cl * extra1 = (ggml_tensor_extra_cl *)src1->extra;
ggml_tensor_extra_cl * extrad = (ggml_tensor_extra_cl *)dst->extra;
cl_ulong offset0 = extra0->offset + src0->view_offs;
cl_ulong offset1 = extra1->offset + src1->view_offs;
cl_ulong offsetd = extrad->offset + dst->view_offs;
bool bcast_row = false;
cl_kernel kernel;
@ -5233,13 +5208,7 @@ static void ggml_cl_div(ggml_backend_t backend, const ggml_tensor * src0, const
kernel = backend_ctx->kernel_div_row_f16;
}
CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), &extra0->data_device));
CL_CHECK(clSetKernelArg(kernel, 1, sizeof(cl_ulong), &offset0));
CL_CHECK(clSetKernelArg(kernel, 2, sizeof(cl_mem), &extra1->data_device));
CL_CHECK(clSetKernelArg(kernel, 3, sizeof(cl_ulong), &offset1));
CL_CHECK(clSetKernelArg(kernel, 4, sizeof(cl_mem), &extrad->data_device));
CL_CHECK(clSetKernelArg(kernel, 5, sizeof(cl_ulong), &offsetd));
CL_CHECK(clSetKernelArg(kernel, 6, sizeof(int), &ne));
cl_set_kernel_args(kernel, src0, src1, dst, ne);
} else {
if (src0->type == GGML_TYPE_F32) {
kernel = backend_ctx->kernel_div;
@ -5247,29 +5216,17 @@ static void ggml_cl_div(ggml_backend_t backend, const ggml_tensor * src0, const
kernel = backend_ctx->kernel_div_f16;
}
CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), &extra0->data_device));
CL_CHECK(clSetKernelArg(kernel, 1, sizeof(cl_ulong), &offset0));
CL_CHECK(clSetKernelArg(kernel, 2, sizeof(cl_mem), &extra1->data_device));
CL_CHECK(clSetKernelArg(kernel, 3, sizeof(cl_ulong), &offset1));
CL_CHECK(clSetKernelArg(kernel, 4, sizeof(cl_mem), &extrad->data_device));
CL_CHECK(clSetKernelArg(kernel, 5, sizeof(cl_ulong), &offsetd));
CL_CHECK(clSetKernelArg(kernel, 6, sizeof(cl_ulong), &nb00));
CL_CHECK(clSetKernelArg(kernel, 7, sizeof(cl_ulong), &nb01));
CL_CHECK(clSetKernelArg(kernel, 8, sizeof(cl_ulong), &nb02));
CL_CHECK(clSetKernelArg(kernel, 9, sizeof(cl_ulong), &nb03));
CL_CHECK(clSetKernelArg(kernel, 10, sizeof(int), &ne10));
CL_CHECK(clSetKernelArg(kernel, 11, sizeof(int), &ne11));
CL_CHECK(clSetKernelArg(kernel, 12, sizeof(int), &ne12));
CL_CHECK(clSetKernelArg(kernel, 13, sizeof(int), &ne13));
CL_CHECK(clSetKernelArg(kernel, 14, sizeof(cl_ulong), &nb10));
CL_CHECK(clSetKernelArg(kernel, 15, sizeof(cl_ulong), &nb11));
CL_CHECK(clSetKernelArg(kernel, 16, sizeof(cl_ulong), &nb12));
CL_CHECK(clSetKernelArg(kernel, 17, sizeof(cl_ulong), &nb13));
CL_CHECK(clSetKernelArg(kernel, 18, sizeof(int), &ne0));
CL_CHECK(clSetKernelArg(kernel, 19, sizeof(cl_ulong), &nb0));
CL_CHECK(clSetKernelArg(kernel, 20, sizeof(cl_ulong), &nb1));
CL_CHECK(clSetKernelArg(kernel, 21, sizeof(cl_ulong), &nb2));
CL_CHECK(clSetKernelArg(kernel, 22, sizeof(cl_ulong), &nb3));
cl_set_kernel_args(
kernel,
src0,
src1,
dst,
src0->nb,
src1->ne,
src1->nb,
ne0,
dst->nb
);
}
if (bcast_row) {