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

P4 to Git Change 1599157 by gandryey@gera-ocl-lc on 2018/08/28 17:11:04

Browse Source 
SWDEV-79445 - OCL generic changes and code clean-up
	- Add devkerenle.cpp/hpp files for device::Kernel object
	- Move generic code for the arguments setup from the device layer to the abstraction layer
	- Update ROCr and PAL paths to utilize the generic logic for the arguments setup

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#226 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#313 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devkernel.cpp#1 add
... //depot/stg/opencl/drivers/opencl/runtime/device/devkernel.hpp#1 add
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpukernel.cpp#328 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpukernel.hpp#130 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.cpp#61 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.hpp#20 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#40 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.hpp#24 edit
... //depot/stg/opencl/drivers/opencl/runtime/runtimedefs#51 edit
This commit is contained in:
foreman
2018-08-28 17:30:29 -04:00
parent 0c5f415b5f
commit 5cfeb120ed
10 changed files with 1052 additions and 1832 deletions
Download Patch File Download Diff File Expand all files Collapse all files
rocclr/runtime/device/device.cpp
-80
View File
@@ -585,86 +585,6 @@ Settings::Settings() {
//!< concurrent Virtual GPUs for default
}
bool Kernel::createSignature(
const parameters_t& params, uint32_t numParameters,
uint32_t version) {
std::stringstream attribs;
if (workGroupInfo_.compileSize_[0] != 0) {
attribs << "reqd_work_group_size(";
for (size_t i = 0; i < 3; ++i) {
if (i != 0) {
attribs << ",";
}
attribs << workGroupInfo_.compileSize_[i];
}
attribs << ")";
}
if (workGroupInfo_.compileSizeHint_[0] != 0) {
attribs << " work_group_size_hint(";
for (size_t i = 0; i < 3; ++i) {
if (i != 0) {
attribs << ",";
}
attribs << workGroupInfo_.compileSizeHint_[i];
}
attribs << ")";
}
if (!workGroupInfo_.compileVecTypeHint_.empty()) {
attribs << " vec_type_hint(" << workGroupInfo_.compileVecTypeHint_ << ")";
}
// Destroy old signature if it was allocated before
// (offline devices path)
delete signature_;
signature_ = new amd::KernelSignature(params, attribs.str(), numParameters, version);
if (NULL != signature_) {
return true;
}
return false;
}
Kernel::~Kernel() { delete signature_; }
std::string Kernel::openclMangledName(const std::string& name) {
const oclBIFSymbolStruct* bifSym = findBIF30SymStruct(symOpenclKernel);
assert(bifSym && "symbol not found");
return std::string("&") + bifSym->str[bif::PRE] + name + bifSym->str[bif::POST];
}
void Memory::saveMapInfo(const void* mapAddress, const amd::Coord3D origin,
const amd::Coord3D region, uint mapFlags, bool entire,
amd::Image* baseMip) {
// Map/Unmap must be serialized.
amd::ScopedLock lock(owner()->lockMemoryOps());
WriteMapInfo info = {};
WriteMapInfo* pInfo = &info;
auto it = writeMapInfo_.find(mapAddress);
if (it != writeMapInfo_.end()) {
LogWarning("Double map of the same or overlapped region!");
pInfo = &it->second;
}
if (mapFlags & (CL_MAP_WRITE | CL_MAP_WRITE_INVALIDATE_REGION)) {
pInfo->origin_ = origin;
pInfo->region_ = region;
pInfo->entire_ = entire;
pInfo->unmapWrite_ = true;
}
if (mapFlags & CL_MAP_READ) {
pInfo->unmapRead_ = true;
}
pInfo->baseMip_ = baseMip;
// Insert into the map if it's the first region
if (++pInfo->count_ == 1) {
writeMapInfo_.insert({mapAddress, info});
}
}
Program::Program(amd::Device& device)
: device_(device),
type_(TYPE_NONE),