gahow/xserv
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

kernels/cuda: paged-attention kernel, dispatch, pinned host memory

Browse Source 
CUDA layer for the paged-KV + swap work:
- csrc: new paged_attention.cu plus updates across attention/gemm/norm/
  activation/embedding/reduce kernels and common.cuh.
- xserv-kernels: new dispatch module and kernel-binding updates.
- xserv-cuda: cudaMallocHost/FreeHost bindings + PinnedBuffer (host swap
  pool backing) and offset-aware D2H/H2D copies used to move KV blocks
  between the GPU pool and pinned host memory.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
Gahow Wang
2026-05-28 19:58:36 +08:00
parent 3f1c3d429a
commit 4c3f914459
27 changed files with 581 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
csrc/activation/activations.cu
View File

@@ -1,5 +1,6 @@
#include <cuda_bf16.h>
#include <math.h>
#include "../common.cuh"
// GELU (tanh approximation):
// gelu(x) = 0.5 * x * (1 + tanh(sqrt(2/pi) * (x + 0.044715 * x^3)))
@@ -83,6 +84,7 @@ void launch_gelu_f32(const void* x, void* out, int n, void* stream) {
int block = 256;
int grid = (n + block - 1) / block;
gelu_f32<<<grid, block, 0, (cudaStream_t)stream>>>((const float*)x, (float*)out, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_gelu_bf16(const void* x, void* out, int n, void* stream) {
@@ -90,12 +92,14 @@ void launch_gelu_bf16(const void* x, void* out, int n, void* stream) {
int grid = (n + block - 1) / block;
gelu_bf16<<<grid, block, 0, (cudaStream_t)stream>>>(
(const __nv_bfloat16*)x, (__nv_bfloat16*)out, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_silu_f32(const void* x, void* out, int n, void* stream) {
int block = 256;
int grid = (n + block - 1) / block;
silu_f32<<<grid, block, 0, (cudaStream_t)stream>>>((const float*)x, (float*)out, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_silu_bf16(const void* x, void* out, int n, void* stream) {
@@ -103,6 +107,7 @@ void launch_silu_bf16(const void* x, void* out, int n, void* stream) {
int grid = (n + block - 1) / block;
silu_bf16<<<grid, block, 0, (cudaStream_t)stream>>>(
(const __nv_bfloat16*)x, (__nv_bfloat16*)out, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_scale_f32(const void* x, void* out, float scale, int n, void* stream) {
@@ -110,6 +115,7 @@ void launch_scale_f32(const void* x, void* out, float scale, int n, void* stream
int grid = (n + block - 1) / block;
scale_f32_kernel<<<grid, block, 0, (cudaStream_t)stream>>>(
(const float*)x, (float*)out, scale, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_scale_bf16(const void* x, void* out, float scale, int n, void* stream) {
@@ -117,6 +123,7 @@ void launch_scale_bf16(const void* x, void* out, float scale, int n, void* strea
int grid = (n + block - 1) / block;
scale_bf16_kernel<<<grid, block, 0, (cudaStream_t)stream>>>(
(const __nv_bfloat16*)x, (__nv_bfloat16*)out, scale, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_add_f32(const void* a, const void* b, void* out, int n, void* stream) {
@@ -124,24 +131,28 @@ void launch_add_f32(const void* a, const void* b, void* out, int n, void* stream
int grid = (n + block - 1) / block;
add_f32_kernel<<<grid, block, 0, (cudaStream_t)stream>>>(
(const float*)a, (const float*)b, (float*)out, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_add_bf16(const void* a, const void* b, void* out, int n, void* stream) {
int block = 256;
int grid = (n + block - 1) / block;
add_bf16_kernel<<<grid, block, 0, (cudaStream_t)stream>>>(
(const __nv_bfloat16*)a, (const __nv_bfloat16*)b, (__nv_bfloat16*)out, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_mul_f32(const void* a, const void* b, void* out, int n, void* stream) {
int block = 256;
int grid = (n + block - 1) / block;
mul_f32_kernel<<<grid, block, 0, (cudaStream_t)stream>>>(
(const float*)a, (const float*)b, (float*)out, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_mul_bf16(const void* a, const void* b, void* out, int n, void* stream) {
int block = 256;
int grid = (n + block - 1) / block;
mul_bf16_kernel<<<grid, block, 0, (cudaStream_t)stream>>>(
(const __nv_bfloat16*)a, (const __nv_bfloat16*)b, (__nv_bfloat16*)out, n);
CUDA_CHECK_LAST_ERROR();
}
void launch_silu_mul_bf16(const void* gate, const void* up, void* out, int n, void* stream) {
@@ -149,6 +160,7 @@ void launch_silu_mul_bf16(const void* gate, const void* up, void* out, int n, vo
int grid = (n + block - 1) / block;
silu_mul_bf16_kernel<<<grid, block, 0, (cudaStream_t)stream>>>(
(const __nv_bfloat16*)gate, (const __nv_bfloat16*)up, (__nv_bfloat16*)out, n);
CUDA_CHECK_LAST_ERROR();
}
}