vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

cuda : refactor into multiple files (#6269)

Browse source
This commit is contained in:
slaren 2024-03-25 13:50:23 +01:00 • committed by GitHub
parent ad3a0505e3
commit ae1f211ce2
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
59 changed files with 9154 additions and 8987 deletions
Download patch file Download diff file Expand all files Collapse all files

32
ggml-cuda/scale.cu Normal file
View file

@ -0,0 +1,32 @@
#include "scale.cuh"
static __global__ void scale_f32(const float * x, float * dst, const float scale, const int k) {
const int i = blockDim.x*blockIdx.x + threadIdx.x;
if (i >= k) {
return;
}
dst[i] = scale * x[i];
}
static void scale_f32_cuda(const float * x, float * dst, const float scale, const int k, cudaStream_t stream) {
const int num_blocks = (k + CUDA_SCALE_BLOCK_SIZE - 1) / CUDA_SCALE_BLOCK_SIZE;
scale_f32<<<num_blocks, CUDA_SCALE_BLOCK_SIZE, 0, stream>>>(x, dst, scale, k);
}
void ggml_cuda_op_scale(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
const ggml_tensor * src0 = dst->src[0];
const float * src0_d = (const float *)src0->data;
float * dst_d = (float *)dst->data;
cudaStream_t stream = ctx.stream();
GGML_ASSERT(src0->type == GGML_TYPE_F32);
GGML_ASSERT( dst->type == GGML_TYPE_F32);
float scale;
memcpy(&scale, dst->op_params, sizeof(float));
scale_f32_cuda(src0_d, dst_d, scale, ggml_nelements(src0), stream);
CUDA_CHECK(cudaGetLastError());
}