Jenkins
50b3ddee6c
Merge 'master' into 'amd-master'
...
Change-Id: I0f3e131eb9415f0111526828bb1010a0c2c0be2a
[ROCm/clr commit: da0af7844f ]
2019-08-09 05:53:35 -04:00
foreman
5a0cdd92cb
P4 to Git Change 1980493 by cpaquot@cpaquot-ocl-lc-lnx on 2019/08/08 18:57:21
...
SWDEV-198556 - [HIP] Override queue if src/dst's device doesn't match the queue's device.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#66 edit
[ROCm/clr commit: ab33c14089 ]
2019-08-08 19:03:46 -04:00
Evgeny Mankov
d79d523dc9
Merge pull request #1300 from amd-lthakur/windows-Single-Double-Precision-Math-Host-fixes
...
Fixed compilation errors being observed on windows
[ROCm/clr commit: 90c4dbe84a ]
2019-08-09 00:15:48 +03:00
Evgeny Mankov
2764438b2b
Merge pull request #1307 from rwvo/fix_hipify_perl
...
[HIPIFY] Add missing hipHostRegister flags to hipify-perl
[ROCm/clr commit: 3aeff1276f ]
2019-08-08 18:59:32 +03:00
Rene van Oostrum
7246f9eb29
Add missing hipHostRegister flags to hipify-perl for flags that are defined and supported in hip_runtime_api.h
...
[ROCm/clr commit: 3e5fd98fc3 ]
2019-08-08 10:46:45 -05:00
Siu Chi Chan
1436897785
Compile HIP runtime with hidden visibility by default ( #1303 )
...
* add default visibility to most APIs in program_state
* remove unwanted C++ headers
* Add symbol visibility pragmas and compiler flags
* Add visibility attribute to APIs in channel_descriptor and hip_hcc
* remove unused headers
* simplify build flags with hcc
* add pragma visibility hidden to functional_grid_launch
* [CMake] add gfx908 back
[ROCm/clr commit: 3cf2d7d385 ]
2019-08-08 08:33:04 +00:00
Maneesh Gupta
af006dcf0e
Merge pull request #1305 from ROCm-Developer-Tools/fix-hip-vdi-include-clang
...
Fix HIP-VDI using incorrect include files
[ROCm/clr commit: 268d82aefd ]
2019-08-08 08:28:20 +00:00
Maneesh Gupta
19b92e5a12
Merge pull request #1302 from davidrohr/master
...
Treat .cxx files like .cpp files
[ROCm/clr commit: 520f97b920 ]
2019-08-08 08:28:01 +00:00
Rahul Garg
6641c0dc94
Add support for hipFuncGetAttribute ( #1279 )
...
* Add support for hipFunGetAttribute
* Support NVCC path
* Test using sample module_api_global
* Try fixing CI build failure due to hip_prof_gen scan
* Fix for CI build issue
* Resolve conflict
* Rebase and resolve conflicts with master
* Fix build error
* Fix NVCC path build error
[ROCm/clr commit: 8b6317d041 ]
2019-08-08 08:27:41 +00:00
Evgeny Mankov
f2b3d63e24
Merge pull request #1290 from emankov/master
...
[HIP][doc] Populate README.md with the supported CUDA APIs links
[ROCm/clr commit: 19a929f070 ]
2019-08-08 11:17:50 +03:00
lthakur
b62254608c
Resubmitting the fix so as to address the changed format in the last submission
...
[ROCm/clr commit: 80acd13ba7 ]
2019-08-08 11:26:26 +05:30
Aaron Enye Shi
92a468a647
Fix HIP-VDI using incorrect include files
...
When HIP-VDI depends on an llvm-amdgpu package, the include clang directory is not properly assessed.
[ROCm/clr commit: 3750b2cf42 ]
2019-08-07 16:59:17 -04:00
David Rohr
324995e472
Treat .cxx files like .cpp files
...
[ROCm/clr commit: a22119d381 ]
2019-08-07 19:11:35 +02:00
foreman
a2b8eca8da
P4 to Git Change 1979518 by kjayapra@3_HIPWS_TXHDR_ROCM on 2019/08/07 12:40:50
...
SWDEV-198194 - Adding lock in Unregistervar.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#36 edit
[ROCm/clr commit: 24e2e8a6b2 ]
2019-08-07 12:45:50 -04:00
foreman
4aca868dba
P4 to Git Change 1979482 by zukhan@zukhan-ocl-win10 on 2019/08/07 11:30:33
...
SWDEV-196326 - [hipclang-vdi-rocm]: [FBA-80]: Runtime error when all GPUs are hidden by ROCR_VISIBLE_DEVICES
- Adjusted hipGetDeviceCount to return "hipErrorNoDevice".
- This was done to match the same behaviour as HIP-HCC, and API spec for when no devices are found. Rather then return "count = 0" as it did.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/17789/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_device.cpp#22 edit
[ROCm/clr commit: bf395ba54a ]
2019-08-07 11:44:31 -04:00
foreman
9a8f0c273a
P4 to Git Change 1979479 by cpaquot@cpaquot-ocl-lc-lnx on 2019/08/07 11:25:59
...
SWDEV-198424 - [HIP] Implemented missing fields for hipGetDeviceAttribute.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_device.cpp#21 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_device_runtime.cpp#17 edit
[ROCm/clr commit: 2f761998c3 ]
2019-08-07 11:30:59 -04:00
foreman
833fa0b6eb
P4 to Git Change 1979460 by kjayapra@3_HIPWS_TXHDR_ROCM on 2019/08/07 11:05:27
...
SWDEV-145570 - Code changes after header update.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_texture.cpp#13 edit
[ROCm/clr commit: 534fcf8222 ]
2019-08-07 11:07:16 -04:00
lthakur
6a15871e91
Fixed compilation errors being observed on windows
...
[ROCm/clr commit: c4f6d58354 ]
2019-08-07 15:54:46 +05:30
Rahul Garg
b75cd71dd4
Enable temporarily disabled device properties on HIP/VDI
...
[ROCm/clr commit: 971a64c6b9 ]
2019-08-06 22:03:19 -04:00
foreman
1b567c9a54
P4 to Git Change 1979123 by cpaquot@cpaquot-ocl-lc-lnx on 2019/08/06 17:15:10
...
SWDEV-198556 - [HIP] Gnarly bug due to macros:
HIP_RETURN(ret) duplicates ret twice first by setting the last error
then via LogDebugInfo. So if HIP_RETURN has a function as a parameter,
the function would get called twice. So ihipMalloc and ihipMemcpy were
being called twice (and perhaps more functions).
Also logging the pointer returned by ihipMalloc so we can track memory
in logs more easily.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_internal.hpp#33 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#65 edit
[ROCm/clr commit: 1ee23f0f43 ]
2019-08-06 17:19:28 -04:00
Jenkins
3f5380ed51
Merge 'master' into 'amd-master'
...
Change-Id: Ifbc64e6a162790ce7e2e949904e9d53ef1f6540d
[ROCm/clr commit: 93ab8116b3 ]
2019-08-06 05:10:19 -04:00
Evgeny Mankov
a525c9ae2a
[HIP][doc] Populate README.md with the supported CUDA APIs links
...
[ROCm/clr commit: 5bcff6f081 ]
2019-08-05 16:05:39 +03:00
Evgeny Mankov
8eff77f75b
Merge pull request #1289 from emankov/master
...
[HIPIFY][doc] Update README.md
[ROCm/clr commit: 034ac34844 ]
2019-08-05 14:50:43 +03:00
Evgeny Mankov
764aef78a9
[HIPIFY][doc] Update README.md
...
+ cuDNN, Python, and VS supported versions update;
+ testing output update.
[ROCm/clr commit: 00ac238921 ]
2019-08-05 14:49:01 +03:00
Maneesh Gupta
306f93fba3
Merge pull request #1286 from mangupta/swdev-198753
...
Partial revert of 'Add HSA_PATH to hip_Includes in cmake and hipconfig (#1260 )'
[ROCm/clr commit: 8bc4ae6936 ]
2019-08-05 09:52:47 +00:00
Alex Voicu
a59924ae98
Fix hip_throw. ( #1285 )
...
* Fix hip_throw.
* Fix typo
* No, really fix typo
[ROCm/clr commit: e24df75cc1 ]
2019-08-05 09:52:22 +00:00
Maneesh Gupta
21e5fc4e33
Merge pull request #1280 from ROCm-Developer-Tools/fix_dont_break_hcc_just_because
...
This difference makes absolutely no sense.
[ROCm/clr commit: 48a775373c ]
2019-08-05 09:51:53 +00:00
Sarbojit2019
55fa6a83c6
Enabled gcc for hip host code ( #1214 )
...
* Enabled gcc for hip host code
* Adding tests for hip code + (gcc & g++), without kernels
* Excluding nvcc platforms for gcc and g++ tests + Addressing review comments
* minor code clean-up
* Add rocm include path
* Added relative path for library
* Hiding non supported functions for gcc
* Incorporating review comments
[ROCm/clr commit: f23c1a1499 ]
2019-08-05 09:51:36 +00:00
Jeff Daily
68f674205e
consolidate thread local storage ( #915 )
...
* all thread local access now through single struct
* clean up old commented-out code, more use of GET_TLS()
* fewer calls to GET_TLS by passing tls as a funtion argument
* revert unnecessary change to printf
* fix failing tests due to TLS change
* fix merge conflicts in ihipOccupancyMaxActiveBlocksPerMultiprocessor
[ROCm/clr commit: f337ae1edb ]
2019-08-05 09:51:02 +00:00
Maneesh Gupta
c48cc34059
Partial revert of 'Add HSA_PATH to hip_Includes in cmake and hipconfig ( #1260 )'
...
Change-Id: I35173f2defe446c0d1e5511e582e9c8722841381
[ROCm/clr commit: fcb7f2a219 ]
2019-08-05 13:32:46 +05:30
Evgeny Mankov
12804fe223
Merge pull request #1282 from emankov/hipify-clang
...
[HIPIFY][fix][#211 ] Taking into account include guard controlling macro
[ROCm/clr commit: bb7cfaf91a ]
2019-08-04 16:31:53 +03:00
foreman
36f4a1ce78
P4 to Git Change 1978068 by cpaquot@cpaquot-ocl-lc-lnx on 2019/08/03 12:16:57
...
SWDEV-198546 - [HIP] Implemented basic Ctx APIs that are deprecated.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_context.cpp#20 edit
[ROCm/clr commit: 897669d6a4 ]
2019-08-03 12:23:25 -04:00
Evgeny Mankov
e278fbf2f9
[HIPIFY][fix][ #211 ] Taking into account include guard controlling macro
...
...while including HIP main header file, which is inserted now after #indef controlling macro, or after #pragma once, if it's occurred earlier.
+ Add a couple of unit tests.
ToDo: Check backward compatibility on older clang versions.
[ROCm/clr commit: fedef02c37 ]
2019-08-02 16:46:45 +03:00
Maneesh Gupta
f36dced5d7
Merge pull request #1278 from gargrahul/fix_hipfuncGetAttribute_logstatus
...
Fix missing logstatus in hipFuncGetAttributes
[ROCm/clr commit: a489877153 ]
2019-08-02 10:00:38 +00:00
wkwchau
7b0f478767
Added CooperativeLaunch and CooperativeMultiDeviceLaunch flag and property for hipDeviceGetAttribute() and hipGetDeviceProperties() ( #1247 )
...
[ROCm/clr commit: ed04e96e2d ]
2019-08-02 10:00:25 +00:00
Alex Voicu
6547d7f57c
This difference makes absolutely no sense.
...
[ROCm/clr commit: 0e3d967656 ]
2019-08-02 12:30:03 +03:00
Jenkins
a9713a67ca
Merge 'master' into 'amd-master'
...
Change-Id: I1e1c4dd98314f65c5d9a53537eafee324c855718
[ROCm/clr commit: 7165fde23e ]
2019-08-02 05:10:20 -04:00
Rahul Garg
b064d7cab2
Fix missing logstatus in hipFuncGetAttributes
...
[ROCm/clr commit: 20e9aba94e ]
2019-08-02 11:51:34 +05:30
foreman
31001101cd
P4 to Git Change 1977240 by wchau@wc_hip_vdi on 2019/08/01 16:35:47
...
SWDEV-180872 - Runtime support changes for Cooperative Group Features
- Taking into account of SGPRs usage to determine the block size
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#35 edit
[ROCm/clr commit: 3d842e88ea ]
2019-08-01 16:40:24 -04:00
foreman
abd12f5d33
P4 to Git Change 1977018 by cpaquot@cpaquot-ocl-lc-lnx on 2019/08/01 11:58:59
...
SWDEV-197462 - [HIP] Add HIP_INIT_API to hipFree to initialize current thread.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#64 edit
[ROCm/clr commit: 678ee6de51 ]
2019-08-01 12:06:04 -04:00
wkwchau
75f1bb21b3
Added query of hipDeviceAttributeHdpMemFlushCntl and hipDeviceAttribu… ( #1238 )
...
* Added query of hipDeviceAttributeHdpMemFlushCntl and hipDeviceAttributeHdpRegFlushCntl
* Added NVCC blocker for the hip*FlushCntl test cases
[ROCm/clr commit: abe6776677 ]
2019-08-01 16:03:35 +00:00
Maneesh Gupta
98669d1867
Merge pull request #1277 from mangupta/nvcc_devprop
...
[nvcc] Populate missing fields in hipGetDeviceProperties
[ROCm/clr commit: d5a3202a47 ]
2019-08-01 08:59:58 +00:00
Maneesh Gupta
969eddc258
Merge pull request #1276 from vsytch/SWDEV-197675
...
[hip][tests] Don't use a hardcoded warp size, since it can be dynamically changed.…
[ROCm/clr commit: 6ee5fcc07c ]
2019-08-01 08:59:43 +00:00
Maneesh Gupta
b458d72079
Merge pull request #1275 from yxsamliu/fix-std
...
Fix -std=c++14 for windows
[ROCm/clr commit: d87e243c51 ]
2019-08-01 08:59:27 +00:00
Maneesh Gupta
60210d987c
Merge pull request #1243 from jeffdaily/master-stream-lock-fix
...
remove stream locks where it is safe to do so
[ROCm/clr commit: 3de3f57468 ]
2019-08-01 08:59:13 +00:00
wkwchau
a19b4fbd8b
Added support of hipOccupancyMaxActiveBlocksPerMultiprocessor & hipOc… ( #1240 )
...
* Added support of hipOccupancyMaxActiveBlocksPerMultiprocessor & hipOccupancyMaxActiveBlocksPerMultiprocessorWithFlags APIs
* Taking into account of SGPR usage to determine the max active blocks in hipOccupancyMaxActiveBlocksPerMultiprocessor()
[ROCm/clr commit: 7b9801fe9a ]
2019-08-01 08:58:48 +00:00
Maneesh Gupta
d5f73d22aa
[nvcc] Populate missing fields in hipGetDeviceProperties
...
Change-Id: Ie90e02674d503e385f144f1ead3d53ff7b49cecc
[ROCm/clr commit: b24a4000f8 ]
2019-08-01 13:16:39 +05:30
Vladislav Sytchenko
5b201fe854
Don't use a hardcoded warp size, since it can be dynamically changed. Query it from the runtime instead.
...
[ROCm/clr commit: 9a1835ddc3 ]
2019-07-31 17:04:31 -04:00
Yaxun (Sam) Liu
05b71135e9
Fix -std=c++14 for windows
...
[ROCm/clr commit: c1dc675e3d ]
2019-07-31 16:36:47 -04:00
Evgeny Mankov
f99c7e2bb6
Merge pull request #1274 from emankov/cuDNN
...
[HIP][doc] Populate CUDA Runtime API doc with CUDA version field
[ROCm/clr commit: a6dcaf4bcd ]
2019-07-31 23:01:32 +03:00