AI/rocm-systems
2
0 Fork 0
Koodi Ongelmat Pull-pyynnöt Actions Paketit Projektit Julkaisut Wiki Toiminta
76 333 Commitit 3 Branchit 0 Tagit
develop
Commit Graph
76333 Commitit
This Branch
This Branch
Kaikki haarat
Tekijä SHA1 Viesti Päivämäärä
Evgeny Mankov 0ac55b1a9d [HIP] Introduce library_types.h as a common header for libs (#1509) 
* [HIP] Introduce library_types.h as a common header for libs

[Reason]
Currently, hipFFT, hipBLAS and other HIP libs use their own data types, prefixed with HIPFFT or HIPBLAS, whereas in CUDA those types are common and declared in library_types.h

[TODO]
Switch hipFFT, hipBLAS and other HIP libs to use common library_types.h.

* [HIP] Move include for library_types.h to hip_runtime.h

[Reason]
Repeat CUDA's behaviour, where library_types.h is included in cuda_runtime.h


[ROCm/hip commit: 94eb4155dd]
 2019-10-10 19:57:28 +05:30
Evgeny Mankov 94eb4155dd [HIP] Introduce library_types.h as a common header for libs (#1509) 
* [HIP] Introduce library_types.h as a common header for libs

[Reason]
Currently, hipFFT, hipBLAS and other HIP libs use their own data types, prefixed with HIPFFT or HIPBLAS, whereas in CUDA those types are common and declared in library_types.h

[TODO]
Switch hipFFT, hipBLAS and other HIP libs to use common library_types.h.

* [HIP] Move include for library_types.h to hip_runtime.h

[Reason]
Repeat CUDA's behaviour, where library_types.h is included in cuda_runtime.h
 2019-10-10 19:57:28 +05:30
srinivamd ad8e82ca0b remove dependencies on /opt/rocm path (#1379) 
* remove hard coded dependencies on /opt/rocm path

[ROCm/clr commit: 706e4498e3]
 2019-10-10 19:56:55 +05:30
srinivamd 706e4498e3 remove dependencies on /opt/rocm path (#1379) 
* remove hard coded dependencies on /opt/rocm path
 2019-10-10 19:56:55 +05:30
srinivamd fe7c030148 remove dependencies on /opt/rocm path (#1379) 
* remove hard coded dependencies on /opt/rocm path

[ROCm/hip commit: 908e14d176]
 2019-10-10 19:56:55 +05:30
srinivamd 908e14d176 remove dependencies on /opt/rocm path (#1379) 
* remove hard coded dependencies on /opt/rocm path
 2019-10-10 19:56:55 +05:30
Evgeny Mankov bb329f29bc Merge pull request #1519 from emankov/hipify-clang 
[HIPIFY][test] Update allocators.cu test

[ROCm/clr commit: 656c0979bb]
 2019-10-10 17:22:33 +03:00
Evgeny Mankov 656c0979bb Merge pull request #1519 from emankov/hipify-clang 
[HIPIFY][test] Update allocators.cu test
 2019-10-10 17:22:33 +03:00
Evgeny Mankov 459e5dac4d Merge pull request #1519 from emankov/hipify-clang 
[HIPIFY][test] Update allocators.cu test

[ROCm/hip commit: 1ee62f0744]
 2019-10-10 17:22:33 +03:00
Evgeny Mankov 1ee62f0744 Merge pull request #1519 from emankov/hipify-clang 
[HIPIFY][test] Update allocators.cu test
 2019-10-10 17:22:33 +03:00
Evgeny Mankov 0219b7c05d [HIPIFY][test] Update allocators.cu test 
[ROCm/clr commit: e3319acbf8]
 2019-10-10 17:20:41 +03:00
Evgeny Mankov e3319acbf8 [HIPIFY][test] Update allocators.cu test 2019-10-10 17:20:41 +03:00
Evgeny Mankov b4d99584dc [HIPIFY][test] Update allocators.cu test 
[ROCm/hip commit: 88cb3504ff]
 2019-10-10 17:20:41 +03:00
Evgeny Mankov 88cb3504ff [HIPIFY][test] Update allocators.cu test 2019-10-10 17:20:41 +03:00
rkebichi a3f63dc67f Delete genOstreamOps.py 
[ROCm/roctracer commit: 9b1bcd5b01]
 2019-10-10 10:00:56 -04:00
rkebichi 9b1bcd5b01 Delete genOstreamOps.py 2019-10-10 10:00:56 -04:00
rkebichi 8576ede945 Create gen_ostream_ops.py 
[ROCm/roctracer commit: 013cbe59db]
 2019-10-10 10:00:27 -04:00
rkebichi 013cbe59db Create gen_ostream_ops.py 2019-10-10 10:00:27 -04:00
Jenkins c98300818b Merge 'master' into 'amd-master' 
Change-Id: Ib9198a176626c8a328ba5ee26925882425b2a035


[ROCm/clr commit: c4be4cca79]
 2019-10-10 06:23:20 -04:00
Jenkins c4be4cca79 Merge 'master' into 'amd-master' 
Change-Id: Ib9198a176626c8a328ba5ee26925882425b2a035
 2019-10-10 06:23:20 -04:00
Jenkins 695e2d1a08 Merge 'master' into 'amd-master' 
Change-Id: Ib9198a176626c8a328ba5ee26925882425b2a035


[ROCm/hip commit: afb9a7db62]
 2019-10-10 06:23:20 -04:00
Jenkins afb9a7db62 Merge 'master' into 'amd-master' 
Change-Id: Ib9198a176626c8a328ba5ee26925882425b2a035
 2019-10-10 06:23:20 -04:00
foreman 6d7d9c4f41 P4 to Git Change 2011330 by chui@ocl-promo-incrementor on 2019/10/10 03:00:46 
SWDEV-2 - Change OpenCL version number from 3016 to 3017.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2764 edit


[ROCm/clr commit: 8dfb60f2c7]
 2019-10-10 03:02:35 -04:00
foreman 8dfb60f2c7 P4 to Git Change 2011330 by chui@ocl-promo-incrementor on 2019/10/10 03:00:46 
SWDEV-2 - Change OpenCL version number from 3016 to 3017.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2764 edit
 2019-10-10 03:02:35 -04:00
Evgeny 8ebf2fc5fa roctx rocprof integration 
Change-Id: Ia5e3a60c8d89ef0b04a83f09be361454a95e9689
 2019-10-09 23:39:59 -05:00
Evgeny 3e6441d470 roctx rocprof integration 
Change-Id: Ia5e3a60c8d89ef0b04a83f09be361454a95e9689


[ROCm/rocprofiler commit: 8ebf2fc5fa]
 2019-10-09 23:39:59 -05:00
Evgeny d7591ee8aa roctx tool support integration 
[ROCm/roctracer commit: eb232e7e3c]
 2019-10-09 23:32:01 -05:00
Evgeny eb232e7e3c roctx tool support integration 2019-10-09 23:32:01 -05:00
Evgeny 484a8402e1 custom metrics file rocprof option 
Change-Id: I3914740a7640a291558cbe70ca1bdb0400ff7cee
 2019-10-09 18:03:04 -05:00
Evgeny a80892e8f3 custom metrics file rocprof option 
Change-Id: I3914740a7640a291558cbe70ca1bdb0400ff7cee


[ROCm/rocprofiler commit: 484a8402e1]
 2019-10-09 18:03:04 -05:00
Evgeny c783e62230 --trace-period option 
Change-Id: I29bce7763e62659d22d6abe144b8d970ab4f41c3
 2019-10-09 18:34:48 -04:00
Evgeny 6ae534323b --trace-period option 
Change-Id: I29bce7763e62659d22d6abe144b8d970ab4f41c3


[ROCm/rocprofiler commit: c783e62230]
 2019-10-09 18:34:48 -04:00
rkebichi 9d05cf6bc0 Update genOstreamOps.py 
[ROCm/roctracer commit: 14c933bedb]
 2019-10-09 15:28:59 -04:00
rkebichi 14c933bedb Update genOstreamOps.py 2019-10-09 15:28:59 -04:00
rkebichi 2945337ddf Update CMakeLists.txt 
[ROCm/roctracer commit: c342215416]
 2019-10-09 15:27:43 -04:00
rkebichi c342215416 Update CMakeLists.txt 2019-10-09 15:27:43 -04:00
Evgeny 8e998410b7 output cosmetic fix 
[ROCm/roctracer commit: 2a480667ae]
 2019-10-09 13:00:27 -05:00
Evgeny 2a480667ae output cosmetic fix 2019-10-09 13:00:27 -05:00
Wenkai Du fbcdfd8348 Merge pull request #136 from wenkaidu/tree 
Enable tree kernels in build

[ROCm/rccl commit: 062c798c86]
 2019-10-09 10:58:52 -07:00
Wenkai Du 062c798c86 Merge pull request #136 from wenkaidu/tree 
Enable tree kernels in build
 2019-10-09 10:58:52 -07:00
Wenkai Du c4ed3d2e08 Merge pull request #134 from changpeng/master 
Tuning the inline and unroll to reduce the scratch usage

[ROCm/rccl commit: 662281e599]
 2019-10-09 10:58:38 -07:00
Wenkai Du 662281e599 Merge pull request #134 from changpeng/master 
Tuning the inline and unroll to reduce the scratch usage
 2019-10-09 10:58:38 -07:00
rkebichi eb3102077a Create genOstreamOps.py 
[ROCm/roctracer commit: 3e8f5f2058]
 2019-10-09 13:50:55 -04:00
rkebichi 3e8f5f2058 Create genOstreamOps.py 2019-10-09 13:50:55 -04:00
foreman 59396fa27e P4 to Git Change 2010800 by axie@axie-hip-vdi-pal2 on 2019/10/09 12:09:28 
SWDEV-198862 - Options for hip-clang-vdi path to provide the chicken bits, or functional equivalents to HCC_OPT_FLUSH
	Add HCC_OPT_FLUSH flag to use fence scope agent when possible for HIP VDI. The flag is defaulted to turn on, similiar to HIP HCC.
	Add AMD_OCL_OPT_FLUSH to use fence scope agent when possible for OpenCL. This was tested in Windows and PAL. Default is off.
	This flag can be used for future OpenCL test.

	Tests:
	1. http://ocltc.amd.com:8111/viewModification.html?modId=127189&personal=true&tab=vcsModificationBuilds
	The teamcity test includes HIP - VDI - Rocm tests.
	2. VEGA10 , Windows, HIP, 110 hiptests PASS.
	3. VEGA10 , Linux AMDGPU PRO, HIP - PAL, 110 hiptests PASS.

	Newer:
	http://ocltc.amd.com:8111/viewModification.html?modId=127193&personal=true&tab=vcsModificationBuilds

	Reviewboard:  http://ocltc.amd.com/reviews/r/18092/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#247 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#342 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#89 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#29 edit
... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#321 edit


[ROCm/clr commit: d43f2b6372]
 2019-10-09 12:24:11 -04:00
foreman d43f2b6372 P4 to Git Change 2010800 by axie@axie-hip-vdi-pal2 on 2019/10/09 12:09:28 
SWDEV-198862 - Options for hip-clang-vdi path to provide the chicken bits, or functional equivalents to HCC_OPT_FLUSH
	Add HCC_OPT_FLUSH flag to use fence scope agent when possible for HIP VDI. The flag is defaulted to turn on, similiar to HIP HCC.
	Add AMD_OCL_OPT_FLUSH to use fence scope agent when possible for OpenCL. This was tested in Windows and PAL. Default is off.
	This flag can be used for future OpenCL test.

	Tests:
	1. http://ocltc.amd.com:8111/viewModification.html?modId=127189&personal=true&tab=vcsModificationBuilds
	The teamcity test includes HIP - VDI - Rocm tests.
	2. VEGA10 , Windows, HIP, 110 hiptests PASS.
	3. VEGA10 , Linux AMDGPU PRO, HIP - PAL, 110 hiptests PASS.

	Newer:
	http://ocltc.amd.com:8111/viewModification.html?modId=127193&personal=true&tab=vcsModificationBuilds

	Reviewboard:  http://ocltc.amd.com/reviews/r/18092/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#247 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#342 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#89 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#29 edit
... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#321 edit
 2019-10-09 12:24:11 -04:00
foreman 90f7210123 P4 to Git Change 2010785 by gandryey@gera-win10 on 2019/10/09 12:00:14 
SWDEV-205994 - [CQE OCL][NAVI10][DTB-BLOCKER] ~ 10% -50% performance drop observed while running IndigoBench Benchmark | Faulty CL#2007647
	- PAL changed the value reported in numAvailableVgprs on Navi10. Runtime has to switch to vgprsPerSimd for scratch buffer size calculation.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#172 edit


[ROCm/clr commit: a65bdb6d6d]
 2019-10-09 12:18:24 -04:00
foreman a65bdb6d6d P4 to Git Change 2010785 by gandryey@gera-win10 on 2019/10/09 12:00:14 
SWDEV-205994 - [CQE OCL][NAVI10][DTB-BLOCKER] ~ 10% -50% performance drop observed while running IndigoBench Benchmark | Faulty CL#2007647
	- PAL changed the value reported in numAvailableVgprs on Navi10. Runtime has to switch to vgprsPerSimd for scratch buffer size calculation.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#172 edit
 2019-10-09 12:18:24 -04:00
Evgeny a88c2c7f93 tool tracing control integration 
[ROCm/roctracer commit: 9a5a603440]
 2019-10-09 10:56:19 -05:00
Evgeny 9a5a603440 tool tracing control integration 2019-10-09 10:56:19 -05:00