mmr/tkDNN
Archived
1
0
Fork 0
Code Issues 46 Pull Requests 12 Releases 4 Activity

memcpyasync

Browse Source
This commit is contained in:
Francesco Gatti
2017-08-11 13:56:36 +00:00
parent 3124f86878
commit 04f96048b6
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/pluginsRT/RegionRT.cpp
+1 -1
View File
@@ -47,7 +47,7 @@ public:
dnnType *srcData = (dnnType*)reinterpret_cast<const dnnType*>(inputs[0]); dnnType *srcData = (dnnType*)reinterpret_cast<const dnnType*>(inputs[0]);
dnnType *dstData = reinterpret_cast<dnnType*>(outputs[0]); dnnType *dstData = reinterpret_cast<dnnType*>(outputs[0]);
checkCuda( cudaMemcpy(dstData, srcData, batchSize*c*h*w*sizeof(dnnType), cudaMemcpyDeviceToDevice)); checkCuda( cudaMemcpyAsync(dstData, srcData, batchSize*c*h*w*sizeof(dnnType), cudaMemcpyDeviceToDevice, stream));
for (int b = 0; b < batchSize; ++b){ for (int b = 0; b < batchSize; ++b){
for(int n = 0; n < num; ++n){ for(int n = 0; n < num; ++n){