e9da934ac6
Fix hipExtLaunchMultiKernelMultiDevice refactoring. ( #1714 )
...
- Use the correct condition for HIP VDI runtime.
2019-12-06 09:49:17 -08:00
655f16bdb1
Merge 'master' into 'amd-master'
...
Change-Id: I11a48b1198f902fa74e9c0cf7e234c81fb704968
2019-12-06 10:51:37 +00:00
c1d63fbdbb
P4 to Git Change 2041756 by cpaquot@cpaquot-ocl-lc-lnx on 2019/12/06 00:27:07
...
SWDEV-213526 - [hip] OOM issue
Restore lazy kernel uploads for 2.10 release.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#54 edit
2019-12-06 00:33:55 -05:00
e53fc316f1
Revert - Changes related to hipMemcpyWithStream ( #1718 )
...
Reverting #1673 , #1697 and #1707 .
Support for hipMemcpyWithStream and memcpy optimizations, will be brought in again once issues seen with these are resolved independently.
2019-12-06 09:51:53 +05:30
65fd472364
P4 to Git Change 2040890 by skudchad@skudchad_test2_win_opencl on 2019/12/04 14:16:51
...
SWDEV-203814 - HIPRTC Inprocess runtime changes(part2)
ReviewBoardURL = http://ocltc.amd.com/reviews/r/18342/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#37 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.map.in#35 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#50 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_rtc.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devprogram.cpp#73 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devprogram.hpp#38 edit
2019-12-04 14:22:03 -05:00
188a357527
P4 to Git Change 2040884 by jujiang@JJ-HIP on 2019/12/04 13:59:15
...
SWDEV-214490 - Update HIP RT for texture3D in HIP/PAL on Windows
-Update hipMemcpy3D function
http://ocltc.amd.com/reviews/r/18346/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#90 edit
2019-12-04 14:06:53 -05:00
a077140ae2
P4 to Git Change 2040867 by kjayapra@0_HIPWS_LNX1_ROCM on 2019/12/04 13:32:27
...
SWDEV-145570 - Adding back the lazy kernel changes because the OOM issue is because of KFD/RocR.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#53 edit
2019-12-04 13:34:03 -05:00
496f0c53b3
Merge pull request #1712 from emankov/hipify
...
[HIPIFY] CUDA 10.2 (functions only)
2019-12-04 18:39:10 +03:00
ec9b6a95a3
[HIPIFY] CUDA 10.2 (functions only)
...
+ Both Driver API and RT API are supported and synced with each other
+ Update *.md docs and hipify-perl accordingly
+ Add new conversion type "virtual_memory", introduced in Dirver API
2019-12-04 18:37:31 +03:00
c518929fbc
Merge pull request #1710 from emankov/hipify
...
[HIPIFY] CUDA 10.2 fp_16 support
2019-12-04 13:28:43 +03:00
e3c3c6d86e
[HIPIFY] CUDA 10.2 fp_16 support
2019-12-04 13:27:46 +03:00
1f6b3e2aa5
[packaging] Rename hip packages to conform to debian package renaming rules ( #1708 )
2019-12-04 15:55:37 +05:30
09998838e4
Merge pull request #1709 from emankov/hipify
...
[HIPIFY] CUDA 10.2 support (data types only)
2019-12-04 12:26:51 +03:00
5a03e8ccb9
[HIPIFY] CUDA 10.2 support (data types only)
...
+ Both Driver API and RT API are supported and synced with each other
+ Update *.md docs and hipify-perl accordingly
2019-12-04 12:25:20 +03:00
892d2a562e
Fix hipMemcpyWithStream ( #1707 )
...
* Fix hipMemcpyWithStream
* Add a dtest for the same.
2019-12-04 11:56:23 +05:30
68cc787781
[hip] refactoring hipExtLaunchMultiKernelMultiDevice API ( #1698 )
...
[Background] it was found that if lazy linking used for a library that calls hipExtLaunchMultiKernelMultiDevice API then this API can get the wrong program_state object for looking up device kernels leading to a "No device code available" error in this API.
To fix this issue, the API was refactored to be inline and get and pass the correct program_state to an internal hip API to request a multi-device kernel launch.
2019-12-04 11:50:51 +05:30
69dcd2b862
P4 to Git Change 2039536 by jujiang@JJ-HIP on 2019/12/02 14:31:13
...
SWDEV-214490 - Update HIP RT for texture3D in HIP/PAL on Windows
-Update ihipBindTexture
http://ocltc.amd.com/reviews/r/18333/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#89 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_texture.cpp#28 edit
2019-12-02 13:37:35 -06:00
355b722246
Merge 'master' into 'amd-master'
...
Change-Id: I79c383bed37cf6278ddb733ddf3bf3661e3dad35
2019-12-02 01:27:13 -05:00
4585b7f6f3
Revert "[packaging] Conform to debian package naming rules ( #1647 )" ( #1688 )
...
This reverts commit 2ca1a82036
2019-12-01 14:47:43 +05:30
1e9ac70f3b
Merge pull request #1702 from emankov/hipify
...
[HIPIFY] Sync cudaError/CUresult between CUDA Driver API/CUDA RT API/HIP
2019-11-29 19:31:28 +03:00
d33dd71ed3
[HIPIFY] Sync cudaError/CUresult between CUDA Driver API/CUDA RT API/HIP
...
+ Update *.md docs and hipify-perl accordingly
[Reason]
Starting with CUDA 10.1 all error codes are merged between Driver and RT APIs
[ToDo]
Do the same merge in HIP API as there is no need in distinguishing return codes by API
2019-11-29 19:29:11 +03:00
5349bd8036
P4 to Git Change 2038688 by jujiang@JJ-HIP on 2019/11/29 10:14:49
...
SWDEV-214490 - Update HIP RT for texture3D in HIP/PAL on Windows
- Update function hipMemcpy3D for Texture Array
- Add hipArrayCubemap support in hipMalloc3DArray
http://ocltc.amd.com/reviews/r/18328/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#88 edit
2019-11-29 10:17:28 -05:00
32442c6506
Revert changes for atomic FADD support when address is in LDS ( #1701 )
...
This reverts PR #1591 and follow-on PR #1695
2019-11-29 11:58:12 +05:30
6ecd303d94
add error checking for code object loading
2019-11-28 14:00:48 -05:00
b6514fffb9
Uniform is_shared query. ( #1695 )
2019-11-28 13:39:05 +05:30
601b88ad18
P4 to Git Change 2037147 by kjayapra@1_HIPWS_LNX1_PAL on 2019/11/26 16:58:07
...
SWDEV-144570 - Adding hipMemcpyWithStream API
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#36 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.map.in#34 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#87 edit
2019-11-26 17:03:41 -05:00
299bbe2325
fix hipMempcy precondition, test for zero size first ( #1697 )
2019-11-25 19:03:03 -08:00
edf29b8673
replace array designator C99
2019-11-25 16:51:49 +05:30
5c5588bf20
P4 to Git Change 2035631 by cpaquot@cpaquot-ocl-lc-lnx on 2019/11/22 17:52:20
...
SWDEV-213526 - pytorch tests fail with hipErrorOutofMemory
There's a bug in ROCr when loading a lot of kernels and not syncing
So for now, if an allocation fails, sync devices and retry before
returning hipErrorOutOfMemory error.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#86 edit
2019-11-22 17:56:07 -05:00
90546b742e
P4 to Git Change 2035286 by kjayapra@0_HIPWS_LNX1_ROCM on 2019/11/22 11:19:57
...
SWDEV-145570 - Revering CL 2033170
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#52 edit
2019-11-22 11:29:49 -05:00
aaf31b6b96
Unary operators were too restrictive in the type of their argument. ( #1683 )
2019-11-22 07:54:53 +05:30
70f457d7ea
[HIPIFY] fix typo ( #1681 )
2019-11-22 07:54:37 +05:30
e60dec51da
Fix rocBLAS compilation failure ( #1677 )
...
SWDEV-212749:
o Recent changes to “add support for extended launch” require hip_runtime.h to be include in hip_ext.h
o Order in which external applications include hip_hcc.h/hip_runtime.h causes compilation failure
2019-11-22 07:54:17 +05:30
d597e7ca20
Use native support for atomic FADD when address is in LDS ( #1591 )
2019-11-22 07:53:48 +05:30
de09e11c0a
P4 to Git Change 2034910 by kjayapra@0_HIPWS_LNX1_ROCM on 2019/11/21 19:28:10
...
SWDEV-145570 - Revering CL#2033779 for now.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_internal.hpp#48 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#49 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#51 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devprogram.cpp#70 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devprogram.hpp#36 edit
2019-11-21 19:33:05 -05:00
6b06911ef1
fixed directed tests fail when hcc bumped to 3.0 ( #1678 )
...
Handled the HCC version check appropriately as few of the directed tests (SWDEV-212161) were failing when hcc was bumped to 3.0.
2019-11-20 21:37:52 +05:30
5a1f823739
General sync memcpy improvements. Add hipMemcpyWithStream ( #1673 )
...
* General sync memcpy improvements. Add `hipMemcpyWithStream`
* Update hip_memory.cpp
2019-11-20 21:36:37 +05:30
b3161e9fa0
Update error codes for hipGetDevice for doxygen and move up null check ( #1668 )
...
* [docs] Update error codes for hipGetDevice
* Move up out ptr check
2019-11-20 21:35:27 +05:30
8519a1411c
Fix helper header when using c++17 ( #1666 )
...
This will fix issue #1621 . It also adds tests for is_callable with c++11, c++14, and c++17.
The fallback implementation was completely broken so I rewrote it so it pass the tests as well. This should be used instead of PR #1631 .
2019-11-20 21:33:42 +05:30
b5b3d1bbaa
Extend vector type capabilities and add tests to reflect it. ( #1656 )
2019-11-20 21:32:32 +05:30
3705a045f0
P4 to Git Change 2033779 by kjayapra@0_HIPWS_LNX1_ROCM on 2019/11/20 07:17:56
...
SWDEV-144570 - Update function global during hipModuleLoad.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_internal.hpp#47 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#48 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#50 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devprogram.cpp#69 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devprogram.hpp#35 edit
2019-11-20 07:24:20 -05:00
153a959280
Revert [HIP] Fixed hipStreamAddCallback ( #1674 )
...
This reverts commit 45613311d7
2019-11-20 11:55:46 +05:30
02f8c57eb5
P4 to Git Change 2033170 by cpaquot@cpaquot-ocl-lc-lnx on 2019/11/19 13:26:16
...
SWDEV-212215 - HIP-VDI runtime should not lazy load device code objects
Temporarly disable lazy load of code objects for debugger.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#49 edit
2019-11-19 13:34:00 -05:00
3a7eb694f5
hipStreamSynchronize can skip marker if stream is empty ( #1667 )
2019-11-19 09:42:43 -08:00
188752b2cb
[dtest] Rename hipExtLaunchMultiKernelMultiDevice dtest ( #1671 )
2019-11-19 09:41:13 -08:00
363caf29c6
hostcall: add tests that are only used by the VDI build ( #1664 )
2019-11-18 19:52:40 -08:00
ebe0c56f4f
Fix mathlib and app builds with hip-clang. ( #1665 )
2019-11-18 08:18:20 -08:00
e39d7497ec
Fix gcc build on NVCC path ( #1661 )
...
* Fix gcc build on NVCC path
* Fix CI build errors
* [dtest] Fix texture and surface obj2D tests
2019-11-18 12:19:22 +05:30
3f2316086f
fix complex conjugate for double-complex ( #1659 )
...
The sign in the y component returned from hipConj incorrect for double-complex. Fix to match as in hipConjf above.
2019-11-18 12:19:12 +05:30
b865a50e44
Added null check in hipEventSynchronize ( #1627 )
...
* Added missing null check in hipEventSynchronize
* Minor correction in the Event API description
2019-11-18 12:18:55 +05:30
mhbliao
Jenkins
foreman
Rahul Garg
Evgeny Mankov
Maneesh Gupta
Aryan Salmanpour
Siu Chi Chan
Alex Voicu
Jeff Daily
Anusha Godavarthy Surya
Shyam Saladi
ansurya
satyanveshd
Paul Fultz II
Sarbojit2019
Sameer Sahasrabuddhe
Nick Curtis