Skip to content

Fix PR #449 #453

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
chhwang merged 1 commit into from
Jan 15, 2025
Merged

Fix PR #449 #453

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
6 changes: 3 additions & 3 deletions src/connection.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76,7 +76,7 @@ void CudaIpcConnection::write(RegisteredMemory dst, uint64_t dstOffset, Register
char* dstPtr = (char*)dst.data();
char* srcPtr = (char*)src.data();

if (!env().cudaIpcUseDefaultStream && stream_->empty()) stream_->set(cudaStreamNonBlocking);
if (!env()->cudaIpcUseDefaultStream && stream_->empty()) stream_->set(cudaStreamNonBlocking);

MSCCLPP_CUDATHROW(cudaMemcpyAsync(dstPtr + dstOffset, srcPtr + srcOffset, size, cudaMemcpyDeviceToDevice, *stream_));
INFO(MSCCLPP_P2P, "CudaIpcConnection write: from %p to %p, size %lu", srcPtr + srcOffset, dstPtr + dstOffset, size);
Expand All @@ -96,7 +96,7 @@ void CudaIpcConnection::updateAndSync(RegisteredMemory dst, uint64_t dstOffset,
*src = newValue;
uint64_t* dstPtr = reinterpret_cast<uint64_t*>(reinterpret_cast<char*>(dst.data()) + dstOffset);

if (!env().cudaIpcUseDefaultStream && stream_->empty()) stream_->set(cudaStreamNonBlocking);
if (!env()->cudaIpcUseDefaultStream && stream_->empty()) stream_->set(cudaStreamNonBlocking);

MSCCLPP_CUDATHROW(cudaMemcpyAsync(dstPtr, src, sizeof(uint64_t), cudaMemcpyHostToDevice, *stream_));
INFO(MSCCLPP_P2P, "CudaIpcConnection atomic write: from %p to %p, %lu -> %lu", src, dstPtr + dstOffset, oldValue,
Expand All @@ -116,7 +116,7 @@ void CudaIpcConnection::flush(int64_t timeoutUsec) {
INFO(MSCCLPP_P2P, "CudaIpcConnection flush: timeout is not supported, ignored");
}

if (!env().cudaIpcUseDefaultStream && stream_->empty()) stream_->set(cudaStreamNonBlocking);
if (!env()->cudaIpcUseDefaultStream && stream_->empty()) stream_->set(cudaStreamNonBlocking);

AvoidCudaGraphCaptureGuard guard;
MSCCLPP_CUDATHROW(cudaStreamSynchronize(*stream_));
Expand Down