Commit Graph

74730 Commits

Author SHA1 Message Date
foreman 1a5cdb9116 P4 to Git Change 1088957 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/18 03:00:11
EPR #010002 - Change OpenCL version number from 1661 to 1662.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1408 edit


[ROCm/clr commit: 69de913e4b]
2014-10-18 03:12:08 -04:00
foreman 69de913e4b P4 to Git Change 1088957 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/18 03:00:11
EPR #010002 - Change OpenCL version number from 1661 to 1662.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1408 edit
2014-10-18 03:12:08 -04:00
foreman 06950ca748 P4 to Git Change 1088793 by bdhanase@bala_workpc_ocl on 2014/10/17 14:36:17
EPR #405824 - Back out changelist 1088461

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#470 edit


[ROCm/clr commit: 8770a7ca06]
2014-10-17 14:49:15 -04:00
foreman 8770a7ca06 P4 to Git Change 1088793 by bdhanase@bala_workpc_ocl on 2014/10/17 14:36:17
EPR #405824 - Back out changelist 1088461

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#470 edit
2014-10-17 14:49:15 -04:00
foreman ee15667629 P4 to Git Change 1088508 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/17 03:00:11
EPR #010002 - Change OpenCL version number from 1660 to 1661.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1407 edit


[ROCm/clr commit: a4865b25bd]
2014-10-17 03:10:11 -04:00
foreman a4865b25bd P4 to Git Change 1088508 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/17 03:00:11
EPR #010002 - Change OpenCL version number from 1660 to 1661.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1407 edit
2014-10-17 03:10:11 -04:00
foreman 66f34b4c46 P4 to Git Change 1088461 by bdhanase@bala_workpc_ocl on 2014/10/16 23:15:14
EPR #405824 - On apus, if we run out of local memory to allocate cl_mem objects, ocl runtime will use remote (system) memory.  Update maxMemAllocSize_ to include that.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#469 edit


[ROCm/clr commit: 1dfb720281]
2014-10-16 23:25:14 -04:00
foreman 1dfb720281 P4 to Git Change 1088461 by bdhanase@bala_workpc_ocl on 2014/10/16 23:15:14
EPR #405824 - On apus, if we run out of local memory to allocate cl_mem objects, ocl runtime will use remote (system) memory.  Update maxMemAllocSize_ to include that.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#469 edit
2014-10-16 23:25:14 -04:00
foreman a57aa10099 P4 to Git Change 1088224 by gandryey@gera-dev-w7 on 2014/10/16 14:05:07
EPR #403341 - [Regression][OCL] P2P SDI to GPU (RGBA) and P2P SDI to GPU (RGB) show corruption on sdi output
	- Remove CALResGLBufferTypeEnum.
	- Code clean-up
	- Add a workaround for conformance fence_sync failures

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#195 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.hpp#77 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/include/cal/calcl.h#30 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.h#42 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDeviceGL.cpp#19 edit


[ROCm/clr commit: 9b0402343e]
2014-10-16 14:25:54 -04:00
foreman 9b0402343e P4 to Git Change 1088224 by gandryey@gera-dev-w7 on 2014/10/16 14:05:07
EPR #403341 - [Regression][OCL] P2P SDI to GPU (RGBA) and P2P SDI to GPU (RGB) show corruption on sdi output
	- Remove CALResGLBufferTypeEnum.
	- Code clean-up
	- Add a workaround for conformance fence_sync failures

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#195 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.hpp#77 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/include/cal/calcl.h#30 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.h#42 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDeviceGL.cpp#19 edit
2014-10-16 14:25:54 -04:00
foreman 1b8726a5e9 P4 to Git Change 1087965 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/16 03:00:13
EPR #010002 - Change OpenCL version number from 1659 to 1660.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1406 edit


[ROCm/clr commit: b124361187]
2014-10-16 03:13:32 -04:00
foreman b124361187 P4 to Git Change 1087965 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/16 03:00:13
EPR #010002 - Change OpenCL version number from 1659 to 1660.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1406 edit
2014-10-16 03:13:32 -04:00
foreman 1ab80d4b9b P4 to Git Change 1087847 by lmoriche@lmoriche_opencl_dev on 2014/10/15 17:33:42
EPR #407144 - Fixes an issue introduced by CL#1082950

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/thread/semaphore.cpp#8 edit


[ROCm/clr commit: 42ae4367c2]
2014-10-15 17:49:44 -04:00
foreman 42ae4367c2 P4 to Git Change 1087847 by lmoriche@lmoriche_opencl_dev on 2014/10/15 17:33:42
EPR #407144 - Fixes an issue introduced by CL#1082950

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/thread/semaphore.cpp#8 edit
2014-10-15 17:49:44 -04:00
foreman c945a7f207 P4 to Git Change 1087846 by bdhanase@bala_workpc_ocl on 2014/10/15 17:33:32
EPR #407358 - Enable remoteAlloc_ by default if the frame buffer is less than 128 MB on an APU system

	Reviewed by: German

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusettings.cpp#290 edit


[ROCm/clr commit: e00c0d6992]
2014-10-15 17:44:44 -04:00
foreman e00c0d6992 P4 to Git Change 1087846 by bdhanase@bala_workpc_ocl on 2014/10/15 17:33:32
EPR #407358 - Enable remoteAlloc_ by default if the frame buffer is less than 128 MB on an APU system

	Reviewed by: German

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusettings.cpp#290 edit
2014-10-15 17:44:44 -04:00
foreman 9b7df3602c P4 to Git Change 1087805 by yaxunl@yaxunl_stg_win50 on 2014/10/15 16:07:03
EPR #407587 - Fix llvm -stats not printing out statistics.

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/codegen.cpp#59 edit


[ROCm/clr commit: f10df9ffcd]
2014-10-15 17:02:40 -04:00
foreman f10df9ffcd P4 to Git Change 1087805 by yaxunl@yaxunl_stg_win50 on 2014/10/15 16:07:03
EPR #407587 - Fix llvm -stats not printing out statistics.

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/codegen.cpp#59 edit
2014-10-15 17:02:40 -04:00
foreman 73e46f4b61 P4 to Git Change 1087568 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/15 03:00:12
EPR #010002 - Change OpenCL version number from 1658 to 1659.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1405 edit


[ROCm/clr commit: f4582fd447]
2014-10-15 03:13:41 -04:00
foreman f4582fd447 P4 to Git Change 1087568 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/15 03:00:12
EPR #010002 - Change OpenCL version number from 1658 to 1659.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1405 edit
2014-10-15 03:13:41 -04:00
foreman 2703a2edee P4 to Git Change 1087490 by gandryey@gera-dev-w7 on 2014/10/14 19:07:04
EPR #403341 - [Regression][OCL] P2P SDI to GPU (RGBA) and P2P SDI to GPU (RGB) show corruption on sdi output
	- Enable GL Acquire/Release calls for more resource types. GL can enable color compression, but compute doesn't support it. Thus decompression can be required.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#194 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.h#41 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDeviceGL.cpp#18 edit


[ROCm/clr commit: b4f7a4274b]
2014-10-14 19:15:37 -04:00
foreman b4f7a4274b P4 to Git Change 1087490 by gandryey@gera-dev-w7 on 2014/10/14 19:07:04
EPR #403341 - [Regression][OCL] P2P SDI to GPU (RGBA) and P2P SDI to GPU (RGB) show corruption on sdi output
	- Enable GL Acquire/Release calls for more resource types. GL can enable color compression, but compute doesn't support it. Thus decompression can be required.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#194 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.h#41 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDeviceGL.cpp#18 edit
2014-10-14 19:15:37 -04:00
foreman dcf2e47687 P4 to Git Change 1087396 by bsumner@bsumner-lin-opencl on 2014/10/14 15:21:38
ECR #304775 - move majority of rutime kernels into library

Affected files ...

... //depot/stg/opencl/drivers/opencl/library/common.hsa/src/blitKernels.cl#3 add
... //depot/stg/opencl/drivers/opencl/library/common/src/blitKernels.cl#3 add
... //depot/stg/opencl/drivers/opencl/library/hsa/hsail/src/devenq/devenq.h#6 edit
... //depot/stg/opencl/drivers/opencl/library/hsa/hsail/src/devenq/schedule.cl#3 add
... //depot/stg/opencl/drivers/opencl/library/x86/gen/build/Makefile.gen#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/blitcl.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#32 edit


[ROCm/clr commit: af02d04050]
2014-10-14 15:57:52 -04:00
foreman af02d04050 P4 to Git Change 1087396 by bsumner@bsumner-lin-opencl on 2014/10/14 15:21:38
ECR #304775 - move majority of rutime kernels into library

Affected files ...

... //depot/stg/opencl/drivers/opencl/library/common.hsa/src/blitKernels.cl#3 add
... //depot/stg/opencl/drivers/opencl/library/common/src/blitKernels.cl#3 add
... //depot/stg/opencl/drivers/opencl/library/hsa/hsail/src/devenq/devenq.h#6 edit
... //depot/stg/opencl/drivers/opencl/library/hsa/hsail/src/devenq/schedule.cl#3 add
... //depot/stg/opencl/drivers/opencl/library/x86/gen/build/Makefile.gen#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/blitcl.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#32 edit
2014-10-14 15:57:52 -04:00
foreman cacb80e9c6 P4 to Git Change 1087364 by mbareghe@mbareghe_staging_win30 on 2014/10/14 14:49:19
ECR #304775 - clp re-implementation - Refactored SPIR loader and Added replaceTrivialConversionFunc to linker and offline linker. clp Moved to prelinker.

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/clc/clpSrc/Makefile#2 delete
... //depot/stg/opencl/drivers/opencl/compiler/clc/clpSrc/build/Makefile#3 delete
... //depot/stg/opencl/drivers/opencl/compiler/clc/clpSrc/clpMain.cpp#4 delete
... //depot/stg/opencl/drivers/opencl/compiler/clc/clpSrc/clpVectorExpansion.cpp#28 delete
... //depot/stg/opencl/drivers/opencl/compiler/clc/src/build/Makefile.clc#14 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/linker.cpp#115 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/include/llvm/AMDPrelinkOpt.h#3 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Linker/AMDPrelinkOpt.cpp#3 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Linker/clpVectorExpansion.cpp#1 add
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Transforms/SPIR/AMDSPIRLoader.cpp#87 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/tools/llvm-link/llvm-link.cpp#49 edit
... //depot/stg/opencl/drivers/opencl/library/common/src/commonConversions.cl#17 edit
... //depot/stg/opencl/drivers/opencl/library/common/src/commonVectorExpansions.cl#22 edit
... //depot/stg/opencl/drivers/opencl/library/hsa/hsail/build/Makefile.hsail#36 edit


[ROCm/clr commit: 19a31b5fe9]
2014-10-14 15:09:17 -04:00
foreman 19a31b5fe9 P4 to Git Change 1087364 by mbareghe@mbareghe_staging_win30 on 2014/10/14 14:49:19
ECR #304775 - clp re-implementation - Refactored SPIR loader and Added replaceTrivialConversionFunc to linker and offline linker. clp Moved to prelinker.

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/clc/clpSrc/Makefile#2 delete
... //depot/stg/opencl/drivers/opencl/compiler/clc/clpSrc/build/Makefile#3 delete
... //depot/stg/opencl/drivers/opencl/compiler/clc/clpSrc/clpMain.cpp#4 delete
... //depot/stg/opencl/drivers/opencl/compiler/clc/clpSrc/clpVectorExpansion.cpp#28 delete
... //depot/stg/opencl/drivers/opencl/compiler/clc/src/build/Makefile.clc#14 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/linker.cpp#115 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/include/llvm/AMDPrelinkOpt.h#3 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Linker/AMDPrelinkOpt.cpp#3 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Linker/clpVectorExpansion.cpp#1 add
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Transforms/SPIR/AMDSPIRLoader.cpp#87 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/tools/llvm-link/llvm-link.cpp#49 edit
... //depot/stg/opencl/drivers/opencl/library/common/src/commonConversions.cl#17 edit
... //depot/stg/opencl/drivers/opencl/library/common/src/commonVectorExpansions.cl#22 edit
... //depot/stg/opencl/drivers/opencl/library/hsa/hsail/build/Makefile.hsail#36 edit
2014-10-14 15:09:17 -04:00
foreman 2bd682d396 P4 to Git Change 1087092 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/14 03:00:11
EPR #010002 - Change OpenCL version number from 1657 to 1658.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1404 edit


[ROCm/clr commit: 41d7aee04a]
2014-10-14 04:05:48 -04:00
foreman 41d7aee04a P4 to Git Change 1087092 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/14 03:00:11
EPR #010002 - Change OpenCL version number from 1657 to 1658.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1404 edit
2014-10-14 04:05:48 -04:00
Alexey Skidanov 5d82c65602 Add GPU local memory module
Add graphic interop support

Local memory rework

Signed-off-by: Alexey Skidanov <Alexey.Skidanov@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@amd.com>

Adjust to new ioctl format

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-10-13 11:29:39 +03:00
Oded Gabbay 7972bb00c5 kfd_ioctl: interop
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:02:34 +02:00
Oded Gabbay d995344ae7 Add new definitions of HSA functions and types
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-10-20 08:48:28 +03:00
Oded Gabbay d14eb7aa00 kfd_ioctl: local memory
Add two new IOCTLs

Remove old IOCTLs

Signed-off-by: Alexey Skidanov <Alexey.Skidanov@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:02:27 +02:00
Alexey Skidanov a7bdd0fdff Add GPU local memory module
Add graphic interop support

Local memory rework

Signed-off-by: Alexey Skidanov <Alexey.Skidanov@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@amd.com>

Adjust to new ioctl format

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>


[ROCm/ROCR-Runtime commit: 5d82c65602]
2014-10-13 11:29:39 +03:00
Oded Gabbay ec3149aeb7 kfd_ioctl: interop
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>


[ROCm/ROCR-Runtime commit: 7972bb00c5]
2014-12-08 10:02:34 +02:00
Oded Gabbay cb29071e77 Add new definitions of HSA functions and types
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>


[ROCm/ROCR-Runtime commit: d995344ae7]
2014-10-20 08:48:28 +03:00
Oded Gabbay 06d3f9c375 kfd_ioctl: local memory
Add two new IOCTLs

Remove old IOCTLs

Signed-off-by: Alexey Skidanov <Alexey.Skidanov@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@amd.com>


[ROCm/ROCR-Runtime commit: d14eb7aa00]
2014-12-08 10:02:27 +02:00
Yair Shachar 39f693aec0 Add H/W debug module
Signed-off-by: Yair Shachar <yair.shachar@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>

Adjust to new ioctl format

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-10-13 11:29:03 +03:00
Oded Gabbay f7c3c3bd1b kfd_ioctl: debug
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:02:12 +02:00
Andrew Lewycky f66e7d71b1 Add HSA Signals module
Signed-off-by: Andrew Lewycky <Andrew.Lewycky@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>

Adjust to new ioctl format

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-10-13 11:30:18 +03:00
Oded Gabbay 03604633ca kfd_ioctl: events
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:01:26 +02:00
Yair Shachar c146ef74c2 Add H/W debug module
Signed-off-by: Yair Shachar <yair.shachar@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>

Adjust to new ioctl format

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>


[ROCm/ROCR-Runtime commit: 39f693aec0]
2014-10-13 11:29:03 +03:00
Oded Gabbay 8a4fd73eaf kfd_ioctl: debug
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>


[ROCm/ROCR-Runtime commit: f7c3c3bd1b]
2014-12-08 10:02:12 +02:00
Andrew Lewycky 82c89fc12b Add HSA Signals module
Signed-off-by: Andrew Lewycky <Andrew.Lewycky@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>

Adjust to new ioctl format

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>


[ROCm/ROCR-Runtime commit: f66e7d71b1]
2014-10-13 11:30:18 +03:00
Oded Gabbay b4df6bc3aa kfd_ioctl: events
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>


[ROCm/ROCR-Runtime commit: 03604633ca]
2014-12-08 10:01:26 +02:00
foreman ae33bb1078 P4 to Git Change 1086680 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/11 03:00:14
EPR #010002 - Change OpenCL version number from 1656 to 1657.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1403 edit


[ROCm/clr commit: 971db82561]
2014-10-11 03:10:01 -04:00
foreman 971db82561 P4 to Git Change 1086680 by johtaylo@johtaylo-JTBUILDER03-increment on 2014/10/11 03:00:14
EPR #010002 - Change OpenCL version number from 1656 to 1657.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1403 edit
2014-10-11 03:10:01 -04:00
foreman 184a8b03de P4 to Git Change 1086362 by emankov@em-hsa-amd on 2014/10/10 10:40:16
ECR #333753 - Performance: Stop obligatory BRIG disassembling to HSAIL

	[Important]: HSAIL is not being disassembled from BRIG and not being inserted into BIF anymore by default.

	Testing: pre check-in, smoke_clang

	Reviewers: Stanislav Mekhanoshin, Brian Sumner, Artem Tamazov

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/v0_8/if_acl.cpp#55 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/hsail_be.cpp#33 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/hsail_be.hpp#11 edit
... //depot/stg/opencl/drivers/opencl/compiler/tools/aoc2/aoc2.cpp#60 edit


[ROCm/clr commit: 7f8d708ca9]
2014-10-10 11:10:24 -04:00
foreman 7f8d708ca9 P4 to Git Change 1086362 by emankov@em-hsa-amd on 2014/10/10 10:40:16
ECR #333753 - Performance: Stop obligatory BRIG disassembling to HSAIL

	[Important]: HSAIL is not being disassembled from BRIG and not being inserted into BIF anymore by default.

	Testing: pre check-in, smoke_clang

	Reviewers: Stanislav Mekhanoshin, Brian Sumner, Artem Tamazov

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/v0_8/if_acl.cpp#55 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/hsail_be.cpp#33 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/hsail_be.hpp#11 edit
... //depot/stg/opencl/drivers/opencl/compiler/tools/aoc2/aoc2.cpp#60 edit
2014-10-10 11:10:24 -04:00
foreman bf319c8d1a P4 to Git Change 1086296 by emankov@em-hsa-amd on 2014/10/10 07:46:11
ECR #333753 - HSA RT/Performance: Elimination of HSAIL text usage in RT (part 2)

	(as it's already done for gpuprogram.cpp)

	Extracting HSAIL from the binary and parsing it for the kernel names in RT were replaced with aclQueryInfo call for RT_KERNEL_NAMES. Kernel names are obtained now from the corresponding metadata symbols names, which are already presented in BIF at kernel finalization stage.

	Side effect: performance improvement

	Next Step: Performance: Stop obligatory BRIG disassembling to HSAIL and insertion into BIF (previously was needed only by RT).

	testing: pre check-in

	Reviewers: Stanislav Mekhanoshin, German Andryeyev, Brian Sumner

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/hsa/hsaprogram.cpp#37 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/hsa_foundation/hsaprogram.cpp#6 edit


[ROCm/clr commit: 3e74ee1f88]
2014-10-10 07:57:34 -04:00
foreman 3e74ee1f88 P4 to Git Change 1086296 by emankov@em-hsa-amd on 2014/10/10 07:46:11
ECR #333753 - HSA RT/Performance: Elimination of HSAIL text usage in RT (part 2)

	(as it's already done for gpuprogram.cpp)

	Extracting HSAIL from the binary and parsing it for the kernel names in RT were replaced with aclQueryInfo call for RT_KERNEL_NAMES. Kernel names are obtained now from the corresponding metadata symbols names, which are already presented in BIF at kernel finalization stage.

	Side effect: performance improvement

	Next Step: Performance: Stop obligatory BRIG disassembling to HSAIL and insertion into BIF (previously was needed only by RT).

	testing: pre check-in

	Reviewers: Stanislav Mekhanoshin, German Andryeyev, Brian Sumner

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/hsa/hsaprogram.cpp#37 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/hsa_foundation/hsaprogram.cpp#6 edit
2014-10-10 07:57:34 -04:00