AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

P4 to Git Change 1092066 by xcui@merged_opencl_jxcwin on 2014/10/29 18:11:54

Browse Source 
EPR #406328 - modified the opencl runtime so that SVM allocation is done for every SVM capable devices, not just one device. This is the part of changes for SVM multiple device support.

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_svm.cpp#7 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/context.cpp#32 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/context.hpp#22 edit


[ROCm/clr commit: ffd485bda2]
This commit is contained in:
foreman
2014-10-29 18:17:14 -04:00
parent 2186a93d3c
commit ba934e930a
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
projects/clr/opencl/api/opencl/amdocl/cl_svm.cpp
+1 -1
View File
@@ -133,7 +133,7 @@ RUNTIME_ENTRY_RET_NOERRCODE(void*, clSVMAlloc, (
return NULL;
}
const std::vector<amd::Device*>& devices = as_amd(context)->devices();
const std::vector<amd::Device*>& devices = as_amd(context)->svmDevices();
bool sizePass = false;
cl_device_svm_capabilities combinedSvmCapabilities = 0;
const cl_uint hostAddressBits = LP64_SWITCH(32, 64);