biondizzle
7c79fb4ee7
CUDA 13 changed cudaMemAdvise to take cudaMemLocation struct instead of int. Updated to use cudaMemLocation with type=cudaMemLocationTypeDevice.
54 lines
1.8 KiB
Plaintext
54 lines
1.8 KiB
Plaintext
// managed_alloc.cu - cudaMallocManaged allocator for PyTorch
|
|
// Compile: nvcc -shared -o libmanaged_alloc.so managed_alloc.cu -Xcompiler -fPIC
|
|
// Compatible with CUDA 13+ (uses cudaMemLocation API)
|
|
#include <cuda_runtime.h>
|
|
#include <stdio.h>
|
|
|
|
extern "C" {
|
|
|
|
// PyTorch pluggable allocator signature: void*(size_t, int, cudaStream_t)
|
|
void* managed_malloc(size_t size, int device, cudaStream_t stream) {
|
|
void* ptr = nullptr;
|
|
|
|
// Set the device before allocating
|
|
cudaError_t err = cudaSetDevice(device);
|
|
if (err != cudaSuccess) {
|
|
fprintf(stderr, "[managed_alloc] cudaSetDevice(%d) failed: %s\n",
|
|
device, cudaGetErrorString(err));
|
|
return nullptr;
|
|
}
|
|
|
|
// Use cudaMallocManaged - this is the key: allocations can page-fault
|
|
// across HBM and LPDDR on GH200 with EGM enabled
|
|
err = cudaMallocManaged(&ptr, size, cudaMemAttachGlobal);
|
|
if (err != cudaSuccess) {
|
|
fprintf(stderr, "[managed_alloc] cudaMallocManaged failed: %s "
|
|
"(size=%zu bytes / %.2f GiB)\n",
|
|
cudaGetErrorString(err), size, (double)size / (1024.0*1024.0*1024.0));
|
|
return nullptr;
|
|
}
|
|
|
|
// Advise the driver to prefer GPU placement initially.
|
|
// On GH200 with EGM, the hardware will migrate pages as needed.
|
|
// CUDA 13+ uses cudaMemLocation struct instead of int for device
|
|
cudaMemLocation location;
|
|
location.type = cudaMemLocationTypeDevice;
|
|
location.id = device;
|
|
cudaMemAdvise(ptr, size, cudaMemAdviseSetPreferredLocation, location);
|
|
|
|
return ptr;
|
|
}
|
|
|
|
// PyTorch pluggable allocator signature: void(void*, size_t, int, cudaStream_t)
|
|
void managed_free(void* ptr, size_t size, int device, cudaStream_t stream) {
|
|
if (ptr != nullptr) {
|
|
// Sync the stream before freeing to avoid use-after-free
|
|
if (stream != nullptr) {
|
|
cudaStreamSynchronize(stream);
|
|
}
|
|
cudaFree(ptr);
|
|
}
|
|
}
|
|
|
|
} // extern "C"
|