AI/rocm-systems
2
0
Bifurcation 0
Code Tickets Demandes d'ajout Actions Paquets Projets Publications Wiki Activité

P4 to Git Change 1701409 by gandryey@gera-w8 on 2018/11/01 17:43:16

Parcourir la source 
SWDEV-79445 - OCL generic changes and code clean-up
	- Remove obsolete settings
	- Simplify some logic when runtime checks dimension override in the dispatch critical path

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_execute.cpp#29 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#233 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#323 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devkernel.cpp#9 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#599 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpukernel.cpp#333 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuprogram.cpp#246 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#246 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusettings.cpp#361 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusettings.hpp#103 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#425 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#184 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palsettings.cpp#60 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palsettings.hpp#20 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#39 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.hpp#18 edit
... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#300 edit
Cette révision appartient à :
foreman
2018-11-01 17:49:35 -04:00
Parent 320c9a6375
révision 21f6e4337a
16 fichiers modifiés avec 62 ajouts et 172 suppressions
Télécharger le Fichier Patch Télécharger le Fichier des Différences Développer tous les fichiers Réduire tous les fichiers
rocclr/runtime/device/gpu/gpuvirtual.cpp
+7 -11
Voir le fichier
@@ -1429,11 +1429,9 @@ void VirtualGPU::findIterations(const amd::NDRangeContainer& sizes, const amd::N
// Find the total amount of all groups
groups = sizes.global() / local;
if (dev().settings().partialDispatch_) {
for (uint j = 0; j < dimensions; ++j) {
if ((sizes.global()[j] % local[j]) != 0) {
groups[j]++;
}
for (uint j = 0; j < dimensions; ++j) {
if ((sizes.global()[j] % local[j]) != 0) {
groups[j]++;
}
}
@@ -1481,12 +1479,10 @@ void VirtualGPU::setupIteration(uint iteration, const amd::NDRangeContainer& siz
}
global = groups * local;
if (dev().settings().partialDispatch_) {
for (uint j = 0; j < dimensions; ++j) {
size_t offset = groupOffset[j] * local[j];
if ((offset + global[j]) > sizes.global()[j]) {
global[j] = sizes.global()[j] - offset;
}
for (uint j = 0; j < dimensions; ++j) {
size_t offset = groupOffset[j] * local[j];
if ((offset + global[j]) > sizes.global()[j]) {
global[j] = sizes.global()[j] - offset;
}
}