AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
54,851 Commits 3 Branches 0 Tags
a2cedb7fe5ccb4c0bbd911a661672e7d581e95e4
Commit Graph

54851 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Ben Sander 882dab4536 Refactor hipStreamWaitEvent 
- Null streams use same flow as non-null.
- Add HIP_SYNC_STREAM_WAIT
- Resolve null stream.
 2017-08-31 03:00:14 +00:00
Chris Freehill 92e46584f8 Async mem. copy test with NUMA awareness 
Change-Id: If655ac4c087be2d379e868aad83812f2437d78b9
 2017-08-30 21:35:37 -04:00
Chris Freehill af02d06972 Async mem. copy test with NUMA awareness 
Change-Id: If655ac4c087be2d379e868aad83812f2437d78b9


[ROCm/ROCR-Runtime commit: 92e46584f8]
 2017-08-30 21:35:37 -04:00
foreman 77256aebbe P4 to Git Change 1453366 by asalmanp@asalmanp-ocl-stg on 2017/08/30 15:16:37 
SWDEV-128844 - OpenCL conformance test 2.0/gl/test_gl.exe fails to run on Vega10

	Before switching to PAL on gfx9, we didn’t need to pass any argument for running Conformance tests and the first GPU device was selected automatically to run the tests.
	But after switching to PAL, we need to explicitly pass either “gpu” or “GPU_DEVICE_TYPE_GPU” as an argument to run Conformance tests otherwise the tests are failing (clGetDeviceIDs failure)

	The reason for this issue is that the logic for creating device and null devices (offline devices) changed after switching to PAL for gfx9+ devices.

	To fix this issue, don’t assign an offline device as "default" device.

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/13360/diff/2/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#212 edit


[ROCm/clr commit: 7335e429cb]
 2017-08-30 15:37:01 -04:00
foreman 7335e429cb P4 to Git Change 1453366 by asalmanp@asalmanp-ocl-stg on 2017/08/30 15:16:37 
SWDEV-128844 - OpenCL conformance test 2.0/gl/test_gl.exe fails to run on Vega10

	Before switching to PAL on gfx9, we didn’t need to pass any argument for running Conformance tests and the first GPU device was selected automatically to run the tests.
	But after switching to PAL, we need to explicitly pass either “gpu” or “GPU_DEVICE_TYPE_GPU” as an argument to run Conformance tests otherwise the tests are failing (clGetDeviceIDs failure)

	The reason for this issue is that the logic for creating device and null devices (offline devices) changed after switching to PAL for gfx9+ devices.

	To fix this issue, don’t assign an offline device as "default" device.

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/13360/diff/2/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#212 edit
 2017-08-30 15:37:01 -04:00
foreman 11234a1ffd P4 to Git Change 1453266 by fdaniil@rocm_spec on 2017/08/30 12:27:36 
SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment

	Update from git http://git.amd.com:8080/#/c/103512/

	1. fixed ninja warning on multiple targets generating .inc files
	2. added libamdocl64.so and libOpenCL.so.1.2 libraries installation into install/lib
	3. added Khronos OpenCL headers installation into install/include
	4. added clinfo application installation int install/bin
	5. removed installation of dozens of LLVM libs and tools
	6. fixed minor misspellings in llvm options setup

Affected files ...

... //depot/stg/opencl/drivers/opencl/CMakeLists.txt#15 edit
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/CMakeLists.txt#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/CMakeLists.txt#9 edit
... //depot/stg/opencl/drivers/opencl/tools/clinfo/CMakeLists.txt#3 edit


[ROCm/clr commit: bd0e789f7c]
 2017-08-30 12:44:23 -04:00
foreman bd0e789f7c P4 to Git Change 1453266 by fdaniil@rocm_spec on 2017/08/30 12:27:36 
SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment

	Update from git http://git.amd.com:8080/#/c/103512/

	1. fixed ninja warning on multiple targets generating .inc files
	2. added libamdocl64.so and libOpenCL.so.1.2 libraries installation into install/lib
	3. added Khronos OpenCL headers installation into install/include
	4. added clinfo application installation int install/bin
	5. removed installation of dozens of LLVM libs and tools
	6. fixed minor misspellings in llvm options setup

Affected files ...

... //depot/stg/opencl/drivers/opencl/CMakeLists.txt#15 edit
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/CMakeLists.txt#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/CMakeLists.txt#9 edit
... //depot/stg/opencl/drivers/opencl/tools/clinfo/CMakeLists.txt#3 edit
 2017-08-30 12:44:23 -04:00
foreman 6be88c5152 P4 to Git Change 1453266 by fdaniil@rocm_spec on 2017/08/30 12:27:36 
SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment

	Update from git http://git.amd.com:8080/#/c/103512/

	1. fixed ninja warning on multiple targets generating .inc files
	2. added libamdocl64.so and libOpenCL.so.1.2 libraries installation into install/lib
	3. added Khronos OpenCL headers installation into install/include
	4. added clinfo application installation int install/bin
	5. removed installation of dozens of LLVM libs and tools
	6. fixed minor misspellings in llvm options setup

Affected files ...

... //depot/stg/opencl/drivers/opencl/CMakeLists.txt#15 edit
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/CMakeLists.txt#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/CMakeLists.txt#9 edit
... //depot/stg/opencl/drivers/opencl/tools/clinfo/CMakeLists.txt#3 edit


[ROCm/clr commit: 81a57cc378]
 2017-08-30 12:44:23 -04:00
foreman 81a57cc378 P4 to Git Change 1453266 by fdaniil@rocm_spec on 2017/08/30 12:27:36 
SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment

	Update from git http://git.amd.com:8080/#/c/103512/

	1. fixed ninja warning on multiple targets generating .inc files
	2. added libamdocl64.so and libOpenCL.so.1.2 libraries installation into install/lib
	3. added Khronos OpenCL headers installation into install/include
	4. added clinfo application installation int install/bin
	5. removed installation of dozens of LLVM libs and tools
	6. fixed minor misspellings in llvm options setup

Affected files ...

... //depot/stg/opencl/drivers/opencl/CMakeLists.txt#15 edit
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/CMakeLists.txt#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/CMakeLists.txt#9 edit
... //depot/stg/opencl/drivers/opencl/tools/clinfo/CMakeLists.txt#3 edit
 2017-08-30 12:44:23 -04:00
Amber Lin a74f6896ea Revert "Set guard page as disabled as default" 
This reverts commit 65d680c035.

Change-Id: I09b7e7915ec4759cab57d5863089a2c4a44dfacd
Signed-off-by: Amber Lin <Amber.Lin@amd.com>
 2017-08-30 10:48:23 -04:00
Amber Lin 2506a31248 Revert "Set guard page as disabled as default" 
This reverts commit 4d839c4dc6.

Change-Id: I09b7e7915ec4759cab57d5863089a2c4a44dfacd
Signed-off-by: Amber Lin <Amber.Lin@amd.com>


[ROCm/ROCR-Runtime commit: a74f6896ea]
 2017-08-30 10:48:23 -04:00
foreman 523163bf32 P4 to Git Change 1453046 by johtaylo@johtaylo-jtincrementor-increment on 2017/08/30 03:00:04 
SWDEV-2 - Change OpenCL version number from 2487 to 2488.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2235 edit


[ROCm/clr commit: 2a0e896048]
 2017-08-30 03:33:56 -04:00
foreman 2a0e896048 P4 to Git Change 1453046 by johtaylo@johtaylo-jtincrementor-increment on 2017/08/30 03:00:04 
SWDEV-2 - Change OpenCL version number from 2487 to 2488.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2235 edit
 2017-08-30 03:33:56 -04:00
Ben Sander 13dabd677c Merge pull request #167 from bensander/event_safety 
Event safety

[ROCm/clr commit: 8add022539]
 2017-08-29 13:55:46 -05:00
Ben Sander 8add022539 Merge pull request #167 from bensander/event_safety 
Event safety
 2017-08-29 13:55:46 -05:00
Ben Sander cc08b36647 Merge pull request #167 from bensander/event_safety 
Event safety
 2017-08-29 13:55:46 -05:00
Ben Sander f39e87de0e Merge pull request #166 from gargrahul/input_ptr_null_check 
Null check on input pointer arguments

[ROCm/clr commit: ed3325f952]
 2017-08-29 13:55:38 -05:00
Ben Sander ed3325f952 Merge pull request #166 from gargrahul/input_ptr_null_check 
Null check on input pointer arguments
 2017-08-29 13:55:38 -05:00
Ben Sander 3c172aabe7 Merge pull request #166 from gargrahul/input_ptr_null_check 
Null check on input pointer arguments
 2017-08-29 13:55:38 -05:00
foreman 8fe66c7ee2 P4 to Git Change 1452726 by asalmanp@asalmanp-ocl-stg on 2017/08/29 11:55:13 
SWDEV-79278 - adding missing Vega12 IDs
	ReviewBoardURL = http://ocltc.amd.com/reviews/r/13357/

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/libUtils.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings_hsail.h#24 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings_hsail64.h#24 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/libUtils.cpp#28 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail.h#49 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail64.h#44 edit


[ROCm/clr commit: 037a4a70cc]
 2017-08-29 12:04:09 -04:00
foreman 037a4a70cc P4 to Git Change 1452726 by asalmanp@asalmanp-ocl-stg on 2017/08/29 11:55:13 
SWDEV-79278 - adding missing Vega12 IDs
	ReviewBoardURL = http://ocltc.amd.com/reviews/r/13357/

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/libUtils.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings_hsail.h#24 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings_hsail64.h#24 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/libUtils.cpp#28 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail.h#49 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail64.h#44 edit
 2017-08-29 12:04:09 -04:00
Ben Sander 72dcf7d488 Merge pull request #164 from mangupta/add_install_script 
Add a basic build+install script

[ROCm/clr commit: f39d7929e9]
 2017-08-29 11:01:14 -05:00
Ben Sander f39d7929e9 Merge pull request #164 from mangupta/add_install_script 
Add a basic build+install script
 2017-08-29 11:01:14 -05:00
Ben Sander e069b26e29 Merge pull request #164 from mangupta/add_install_script 
Add a basic build+install script
 2017-08-29 11:01:14 -05:00
Amber Lin a81b29890c Fix mmap when the count reaches the max 
Applications may try to allocate lots host memory and reaches the mmap
limit (/proc/sys/vm/max_map_count). When Applications fails to allocate
memory and calls hsaKmtFreeMemory to release the memory, Thunk fails to
reduce the maps count so the following hsaKmtAllocMemory calls continue
to fail, which doesn't make sense to the application. This patch checks
the mmap to NORESERVE return value. If it fails and the error number is
ENOMEM, reduce the map count by munmap and map it again immediately.



Change-Id: I127cb479dfd86b199172eef269d59426f23859ea
Signed-off-by: Amber Lin <Amber.Lin@amd.com>
 2017-08-29 11:47:29 -04:00
Amber Lin 13282b9115 Fix mmap when the count reaches the max 
Applications may try to allocate lots host memory and reaches the mmap
limit (/proc/sys/vm/max_map_count). When Applications fails to allocate
memory and calls hsaKmtFreeMemory to release the memory, Thunk fails to
reduce the maps count so the following hsaKmtAllocMemory calls continue
to fail, which doesn't make sense to the application. This patch checks
the mmap to NORESERVE return value. If it fails and the error number is
ENOMEM, reduce the map count by munmap and map it again immediately.



Change-Id: I127cb479dfd86b199172eef269d59426f23859ea
Signed-off-by: Amber Lin <Amber.Lin@amd.com>


[ROCm/ROCR-Runtime commit: a81b29890c]
 2017-08-29 11:47:29 -04:00
Jenkins d9a316bee9 Merge 'master' into 'amd-master' 
Change-Id: I4ae1f3ea1158d8fc499f7473065b0cb243dda0d0


[ROCm/clr commit: f77f5997d8]
 2017-08-29 04:11:15 -05:00
Jenkins f77f5997d8 Merge 'master' into 'amd-master' 
Change-Id: I4ae1f3ea1158d8fc499f7473065b0cb243dda0d0
 2017-08-29 04:11:15 -05:00
Jenkins 2687611fb1 Merge 'master' into 'amd-master' 
Change-Id: I4ae1f3ea1158d8fc499f7473065b0cb243dda0d0
 2017-08-29 04:11:15 -05:00
foreman 327eb8ca2e P4 to Git Change 1452576 by johtaylo@johtaylo-jtincrementor-increment on 2017/08/29 03:00:04 
SWDEV-2 - Change OpenCL version number from 2486 to 2487.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2234 edit


[ROCm/clr commit: 4d445db266]
 2017-08-29 03:18:36 -04:00
foreman 4d445db266 P4 to Git Change 1452576 by johtaylo@johtaylo-jtincrementor-increment on 2017/08/29 03:00:04 
SWDEV-2 - Change OpenCL version number from 2486 to 2487.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2234 edit
 2017-08-29 03:18:36 -04:00
Maneesh Gupta 73ab902694 Merge pull request #168 from kknox/update-nvidia-volume 
Update the nvidia volume driver version

[ROCm/clr commit: d152ebe161]
 2017-08-29 09:37:39 +05:30
Maneesh Gupta d152ebe161 Merge pull request #168 from kknox/update-nvidia-volume 
Update the nvidia volume driver version
 2017-08-29 09:37:39 +05:30
Maneesh Gupta a5b1aab349 Merge pull request #168 from kknox/update-nvidia-volume 
Update the nvidia volume driver version
 2017-08-29 09:37:39 +05:30
foreman 7b1fdee2c5 P4 to Git Change 1452486 by skudchad@skudchad_test_win_opencl2 on 2017/08/28 20:01:08 
SWDEV-116136 - Support -Og for Clang
	- Fix assert for -cl-disable-opt

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/options.cpp#7 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/options.cpp#44 edit


[ROCm/clr commit: c6a3aed5f8]
 2017-08-28 20:15:30 -04:00
foreman c6a3aed5f8 P4 to Git Change 1452486 by skudchad@skudchad_test_win_opencl2 on 2017/08/28 20:01:08 
SWDEV-116136 - Support -Og for Clang
	- Fix assert for -cl-disable-opt

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/options.cpp#7 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/options.cpp#44 edit
 2017-08-28 20:15:30 -04:00
Ben Sander a1cdc40b86 Merge pull request #169 from bensander/doc_update 
Update docs.

[ROCm/clr commit: f275d267f6]
 2017-08-28 18:47:47 -05:00
Ben Sander f275d267f6 Merge pull request #169 from bensander/doc_update 
Update docs.
 2017-08-28 18:47:47 -05:00
Ben Sander b87ed391df Merge pull request #169 from bensander/doc_update 
Update docs.
 2017-08-28 18:47:47 -05:00
Ben Sander 070bdb3736 Lock streams when waiting on event completion or querying event safety. 
[ROCm/clr commit: bc9ba7cd81]
 2017-08-28 18:40:16 -05:00
Ben Sander bc9ba7cd81 Lock streams when waiting on event completion or querying event safety. 2017-08-28 18:40:16 -05:00
Ben Sander 6ff74d0e97 Lock streams when waiting on event completion or querying event safety. 2017-08-28 18:40:16 -05:00
Ben Sander ef278f9709 Merge pull request #165 from gargrahul/enable_strmsync_rcrdevnt_dtests 
Enable hipStreamSync2 and record_event dtests

[ROCm/clr commit: e2a25fcab4]
 2017-08-28 15:51:35 -05:00
Ben Sander e2a25fcab4 Merge pull request #165 from gargrahul/enable_strmsync_rcrdevnt_dtests 
Enable hipStreamSync2 and record_event dtests
 2017-08-28 15:51:35 -05:00
Ben Sander 3ffe5c9189 Merge pull request #165 from gargrahul/enable_strmsync_rcrdevnt_dtests 
Enable hipStreamSync2 and record_event dtests
 2017-08-28 15:51:35 -05:00
Ben Sander dcd8961a9d Update docs. 
- Review unsupported comments and refresh.
- Review experimental tag in API.


[ROCm/clr commit: a6d1778fff]
 2017-08-28 15:44:38 -05:00
Ben Sander a6d1778fff Update docs. 
- Review unsupported comments and refresh.
- Review experimental tag in API.
 2017-08-28 15:44:38 -05:00
Ben Sander 1a9e588932 Update docs. 
- Review unsupported comments and refresh.
- Review experimental tag in API.
 2017-08-28 15:44:38 -05:00
Felix Kuehling 52598cf37e Generalize size-dependent virtual address alignment 
Support all fragment sizes up to 2MB by aligning buffers according
to their size.

Change-Id: I82b7ef8be6f1507d941e5c97edb6618adf8c66de
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
 2017-08-28 15:01:33 -04:00
Felix Kuehling 7e59d7d5de Generalize size-dependent virtual address alignment 
Support all fragment sizes up to 2MB by aligning buffers according
to their size.

Change-Id: I82b7ef8be6f1507d941e5c97edb6618adf8c66de
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>


[ROCm/ROCR-Runtime commit: 52598cf37e]
 2017-08-28 15:01:33 -04:00