microsoft/mscclpp
1
0
Fork 0
mirror of https://github.com/microsoft/mscclpp.git synced 2026-04-19 22:39:11 +00:00
Code Issues Packages Projects Releases Wiki Activity

Allow binding allocated memory to NVLS multicast pointer (#290)

Browse Source 
And change NVLS multimem instructions to static functions
This commit is contained in:
Roshan Dathathri
2024-04-18 17:11:31 -07:00
committed by GitHub
parent 64d837f9ab
commit 41e0964d93
5 changed files with 26 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
python/test/nvls_test.cu
View File

@@ -41,8 +41,8 @@ extern "C" __global__ void __launch_bounds__(1024, 1)
for (int idx = my_st + my_offset; idx < my_en; idx += my_step) {
uint4 val;
nvlsPtrs.multimemLoad(val, mc_ptr + idx);
nvlsPtrs.multimemStore(val, mc_ptr + idx);
DeviceMulticastPointerDeviceHandle::multimemLoad(val, mc_ptr + idx);
DeviceMulticastPointerDeviceHandle::multimemStore(val, mc_ptr + idx);
}
deviceSyncer.sync(gridDim.x);