AI/rocm-systems
2
0
Forka 0
Codice Problemi Pull Request Actions Pacchetti Progetti Rilasci Wiki Attività

Correct HIP_FUNC_ATTRIBUTE_MAX_THREADS_PER_BLOCK query

Sfoglia il codice sorgente 
We should be returning the max workgroup size calculated by the compiler.

Change-Id: If86590efbb9b291f470bdbe87e5df992e661c539
This commit is contained in:
Vlad Sytchenko
2020-05-08 11:08:08 -04:00
committato da Vladislav Sytchenko
parent 276bfc9667
commit b5f9d2f818
2 ha cambiato i file con 2 aggiunte e 3 eliminazioni
Scarica il file Patch Scarica il file Diff Expand all files Collapse all files
hipamd/rocclr/hip_platform.cpp
+1 -1
Vedi File
@@ -313,7 +313,7 @@ bool ihipGetFuncAttributes(const char* func_name, amd::Program* program, hipFunc
func_attr->localSizeBytes = wginfo->privateMemSize_;
func_attr->sharedSizeBytes = wginfo->localMemSize_;
func_attr->maxDynamicSharedSizeBytes = wginfo->availableLDSSize_ - wginfo->localMemSize_;
func_attr->maxThreadsPerBlock = wginfo->wavefrontSize_;
func_attr->maxThreadsPerBlock = wginfo->size_;
func_attr->numRegs = wginfo->usedVGPRs_;
return true;