NASP/cuda-samples
35
0
Fork 0
mirror of https://github.com/NVIDIA/cuda-samples.git synced 2024-11-24 20:29:15 +08:00
Code Issues Packages Projects Releases Wiki Activity

add cudaDeviceReset() into p2pBandwidthLatencyTest to free gpu memory after test

Browse Source
This commit is contained in:
Konnase Lee 2019-08-10 16:42:49 +08:00
parent 337815dbee
commit f53f159886
1 changed files with 6 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
Samples/p2pBandwidthLatencyTest/p2pBandwidthLatencyTest.cu
View File

@ -277,11 +277,10 @@ void outputBandwidthMatrix(int numGPUs, bool p2p, P2PDataTransfer p2p_method) {
cudaCheckError(); cudaCheckError();
cudaStreamDestroy(stream[d]); cudaStreamDestroy(stream[d]);
cudaCheckError(); cudaCheckError();
} cudaDeviceReset();
cudaFreeHost((void *)flag);
cudaCheckError(); cudaCheckError();
} }
}
void outputBidirectionalBandwidthMatrix(int numGPUs, bool p2p) { void outputBidirectionalBandwidthMatrix(int numGPUs, bool p2p) {
int numElems = 10000000; int numElems = 10000000;
@ -433,11 +432,10 @@ void outputBidirectionalBandwidthMatrix(int numGPUs, bool p2p) {
cudaCheckError(); cudaCheckError();
cudaStreamDestroy(stream1[d]); cudaStreamDestroy(stream1[d]);
cudaCheckError(); cudaCheckError();
} cudaDeviceReset();
cudaFreeHost((void *)flag);
cudaCheckError(); cudaCheckError();
} }
}
void outputLatencyMatrix(int numGPUs, bool p2p, P2PDataTransfer p2p_method) { void outputLatencyMatrix(int numGPUs, bool p2p, P2PDataTransfer p2p_method) {
int repeat = 100; int repeat = 100;
@ -586,12 +584,11 @@ void outputLatencyMatrix(int numGPUs, bool p2p, P2PDataTransfer p2p_method) {
cudaCheckError(); cudaCheckError();
cudaStreamDestroy(stream[d]); cudaStreamDestroy(stream[d]);
cudaCheckError(); cudaCheckError();
cudaDeviceReset();
cudaCheckError();
} }
sdkDeleteTimer(&stopWatch); sdkDeleteTimer(&stopWatch);
cudaFreeHost((void *)flag);
cudaCheckError();
} }
int main(int argc, char **argv) { int main(int argc, char **argv) {