AI/rocm-systems
2
0
Vork 0
Code Kwesties Pull-aanvragen Actions Paketten Projecten Publicaties Wiki Activiteit

SWDEV-343921 - added Max stack size

Bladeren bron 
Change-Id: I5c1a088e05215ca951afc9d92f8d298c5e3a65f1
This commit is contained in:
Sarbojit Sarkar
2022-07-18 07:37:16 +00:00
gecommit door Sarbojit Sarkar
bovenliggende 2eeb00ba8a
commit 27a08a132f
3 gewijzigde bestanden met toevoegingen van 7 en 8 verwijderingen
Download Patch-bestand Download Diff-bestand Expand all files Collapse all files
rocclr/device/rocm/rocvirtual.cpp
+4 -4
Bestand weergeven
@@ -2972,10 +2972,10 @@ bool VirtualGPU::submitKernelInternal(const amd::NDRangeContainer& sizes,
dispatchPacket.private_segment_size = devKernel->workGroupInfo()->privateMemSize_;
if ((devKernel->workGroupInfo()->usedStackSize_ & 0x1) == 0x1) {
dispatchPacket.private_segment_size += dev().StackSize();
uint32_t maxMemPerThread = device().info().localMemSizePerCU_ / device().info().maxThreadsPerCU_;
if (dispatchPacket.private_segment_size > maxMemPerThread) {
dispatchPacket.private_segment_size = maxMemPerThread;
dispatchPacket.private_segment_size =
std::max<uint64_t>(dev().StackSize(), dispatchPacket.private_segment_size);
if (dispatchPacket.private_segment_size > 16 * Ki) {
dispatchPacket.private_segment_size = 16 * Ki;
}
}