Skip to content

Move cudaIpcMemHandle_t hash and equality to custom namespace #700

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
Binyang2014 merged 4 commits into from
Dec 4, 2025
Merged

Move cudaIpcMemHandle_t hash and equality to custom namespace #700

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
ff70b05
Initial plan
Copilot Dec 4, 2025
10970ce
Refactor handle cache implementation to use custom hash and equality
Copilot Dec 4, 2025
b6fff56
Update progress - all tasks completed
Copilot Dec 4, 2025
b22066b
Delete _codeql_detected_source_root
Binyang2014 Dec 4, 2025
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
20 changes: 11 additions & 9 deletions src/registered_memory.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,21 +29,22 @@
} \
} while (false)

namespace std {
template <>
struct hash<cudaIpcMemHandle_t> {
namespace {

// Custom hash and equality for cudaIpcMemHandle_t
struct CudaIpcMemHandleHash {
size_t operator()(const cudaIpcMemHandle_t& handle) const {
std::string_view view(handle.reserved, sizeof(handle.reserved));
return std::hash<std::string_view>{}(view);
}
};
} // namespace std

inline bool operator==(const cudaIpcMemHandle_t& lhs, const cudaIpcMemHandle_t& rhs) {
return std::memcmp(lhs.reserved, rhs.reserved, sizeof(lhs.reserved)) == 0;
}
struct CudaIpcMemHandleEqual {
bool operator()(const cudaIpcMemHandle_t& lhs, const cudaIpcMemHandle_t& rhs) const noexcept {
return std::memcmp(lhs.reserved, rhs.reserved, sizeof(lhs.reserved)) == 0;
}
};

namespace {

CUmemAllocationHandleType getNvlsMemHandleType() {
#if (CUDA_NVLS_API_AVAILABLE)
Expand All @@ -68,7 +69,8 @@ std::shared_ptr<void> getPeerMemoryHandle(cudaIpcMemHandle_t ipcHandle) {
}
};
#if defined(__HIP_PLATFORM_AMD__)
static std::unordered_map<cudaIpcMemHandle_t, std::weak_ptr<void>> peerMemoryHandleMap;
static std::unordered_map<cudaIpcMemHandle_t, std::weak_ptr<void>, CudaIpcMemHandleHash, CudaIpcMemHandleEqual>
peerMemoryHandleMap;
static std::mutex mutex;
std::lock_guard<std::mutex> lock(mutex);
auto it = peerMemoryHandleMap.find(ipcHandle);
Expand Down
Loading