Skip to content

Make a local copy of CUB's CachingDeviceAllocator and patch it for the race condition #240

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
fwyzard merged 2 commits into from
Jan 8, 2019
Merged

Make a local copy of CUB's CachingDeviceAllocator and patch it for the race condition #240

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
68f9fda
Include local copy of CachingDeviceAllocator
makortel Jan 7, 2019
839d23a
Move mutex unlock after the cudaEventRecord() in DeviceFree()
makortel Jan 7, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 5 additions & 6 deletions HeterogeneousCore/CUDAServices/src/CUDAService.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,15 +5,14 @@
#include <cuda.h>
#include <cuda/api_wrappers.h>

#include <cub/util_allocator.cuh>

#include "FWCore/MessageLogger/interface/MessageLogger.h"
#include "FWCore/ParameterSet/interface/ConfigurationDescriptions.h"
#include "FWCore/ParameterSet/interface/ParameterSet.h"
#include "FWCore/ParameterSet/interface/ParameterSetDescription.h"
#include "HeterogeneousCore/CUDAServices/interface/CUDAService.h"
#include "HeterogeneousCore/CUDAUtilities/interface/cudaCheck.h"

#include "CachingDeviceAllocator.h"
#include "CachingHostAllocator.h"

void setCudaLimit(cudaLimit limit, const char* name, size_t request) {
Expand Down Expand Up @@ -311,7 +310,7 @@ CUDAService::CUDAService(edm::ParameterSet const& config, edm::ActivityRegistry&
<< " max bin " << maxBin << "\n"
<< " resulting bins:\n";
for (auto bin = minBin; bin <= maxBin; ++bin) {
auto binSize = cub::CachingDeviceAllocator::IntPow(binGrowth, bin);
auto binSize = notcub::CachingDeviceAllocator::IntPow(binGrowth, bin);
if (binSize >= (1<<30) and binSize % (1<<30) == 0) {
log << " " << std::setw(8) << (binSize >> 30) << " GB\n";
} else if (binSize >= (1<<20) and binSize % (1<<20) == 0) {
Expand All @@ -324,7 +323,7 @@ CUDAService::CUDAService(edm::ParameterSet const& config, edm::ActivityRegistry&
}
log << " maximum amount of cached memory: " << (minCachedBytes >> 20) << " MB\n";

allocator_ = std::make_unique<Allocator>(cub::CachingDeviceAllocator::IntPow(binGrowth, maxBin),
allocator_ = std::make_unique<Allocator>(notcub::CachingDeviceAllocator::IntPow(binGrowth, maxBin),
binGrowth, minBin, maxBin, minCachedBytes,
false, // do not skip cleanup
debug
Expand Down Expand Up @@ -427,8 +426,8 @@ struct CUDAService::Allocator {
void hostPreallocate(int numberOfDevices, const std::vector<unsigned int>& bytes);

size_t maxAllocation;
cub::CachingDeviceAllocator deviceAllocator;
cub::CachingHostAllocator hostAllocator;
notcub::CachingDeviceAllocator deviceAllocator;
notcub::CachingHostAllocator hostAllocator;
};

void *CUDAService::allocate_device(int dev, size_t nbytes, cuda::stream_t<>& stream) {
Expand Down
Loading