cuda : add FILL op support (#17851)

* cuda : add FILL op support

* cuda : add missing FILL op files
This commit is contained in:
Jay Zenith 2025-12-08 05:10:12 -08:00 committed by GitHub
parent 37a4f63244
commit 51e0c2d917
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 45 additions and 0 deletions

View File

@ -0,0 +1,37 @@
#include "fill.cuh"
#include "convert.cuh"
#define CUDA_FILL_BLOCK_SIZE 256
template <typename T>
static __global__ void fill_kernel(T * __restrict__ dst, const int64_t k, const T value) {
const int64_t i = (int64_t)blockDim.x * blockIdx.x + threadIdx.x;
if (i >= k) {
return;
}
dst[i] = value;
}
void ggml_cuda_op_fill(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
void * dst_d = dst->data;
cudaStream_t stream = ctx.stream();
GGML_ASSERT(ggml_is_contiguous(dst));
float value;
memcpy(&value, dst->op_params, sizeof(float));
const int64_t k = ggml_nelements(dst);
const int64_t num_blocks = (k + CUDA_FILL_BLOCK_SIZE - 1) / CUDA_FILL_BLOCK_SIZE;
switch (dst->type) {
case GGML_TYPE_F32:
fill_kernel<<<num_blocks, CUDA_FILL_BLOCK_SIZE, 0, stream>>>((float *)dst_d, k, value);
break;
case GGML_TYPE_F16:
fill_kernel<<<num_blocks, CUDA_FILL_BLOCK_SIZE, 0, stream>>>((half *)dst_d, k, ggml_cuda_cast<half>(value));
break;
default:
GGML_ABORT("unsupported type");
}
}

View File

@ -0,0 +1,3 @@
#include "common.cuh"
void ggml_cuda_op_fill(ggml_backend_cuda_context & ctx, ggml_tensor * dst);

View File

@ -56,6 +56,7 @@
#include "ggml-cuda/solve_tri.cuh"
#include "ggml-cuda/tri.cuh"
#include "ggml-cuda/cumsum.cuh"
#include "ggml-cuda/fill.cuh"
#include "ggml.h"
#include <algorithm>
@ -2730,6 +2731,9 @@ static bool ggml_cuda_compute_forward(ggml_backend_cuda_context & ctx, struct gg
case GGML_OP_SOLVE_TRI:
ggml_cuda_op_solve_tri(ctx, dst);
break;
case GGML_OP_FILL:
ggml_cuda_op_fill(ctx, dst);
break;
default:
return false;
}
@ -4617,6 +4621,7 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
case GGML_OP_CROSS_ENTROPY_LOSS_BACK:
case GGML_OP_OPT_STEP_ADAMW:
case GGML_OP_OPT_STEP_SGD:
case GGML_OP_FILL:
case GGML_OP_CUMSUM:
case GGML_OP_TRI:
return true;