Evgeny Mankov
513cbab184
Merge pull request #1431 from emankov/hipify-clang
...
[HIPIFY][#1430 ] Build fix for LLVM < 8.0.0
2019-09-18 11:56:16 +03:00
Evgeny Mankov
3011e41420
[HIPIFY][ #1430 ] Build fix for LLVM < 8.0.0
2019-09-18 11:53:23 +03:00
Evgeny Mankov
47362154c9
Merge pull request #1426 from emankov/master
...
[HIPIFY] Add supported device functions (from math_functions.h)
2019-09-18 09:43:31 +03:00
foreman
1955dc1f99
P4 to Git Change 1999915 by cpaquot@cpaquot-ocl-lc-lnx on 2019/09/17 19:43:34
...
SWDEV-201128 - [HIP] test_snli_cuda failure
Default to sync packet
Make sure GPU_NUM_MEM_DEPENDENCY is 0 for HIP
No sync packet is only used when there are mem dependency check
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_context.cpp#22 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#86 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#28 edit
2019-09-17 19:47:11 -04:00
Evgeny Mankov
042c125601
[HIPIFY] Add supported device functions (from math_functions.h)
...
+ [perl] Add generation of sub countSupportedDeviceFunctions
+ [FIX] Do not perform any replacement for supported device functions
+ [perl] Sync hipify-perl accordingly
2019-09-17 16:17:28 +03:00
Evgeny Mankov
313eb28379
Merge pull request #1423 from emankov/master
...
[HIPIFY][perl][fix] Treat ::device_function as a device function
2019-09-17 10:06:56 +03:00
foreman
d5181ce77a
P4 to Git Change 1998982 by cpaquot@cpaquot-ocl-lc-lnx on 2019/09/16 17:49:10
...
SWDEV-145570 - [HIP] Fix regression from new headers.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_device.cpp#23 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_device_runtime.cpp#18 edit
2019-09-16 17:58:06 -04:00
Evgeny Mankov
4f59ec25fe
[HIPIFY][perl][fix] Treat ::device_function as a device function
...
+ Do not treat somenamespace::device_function_name as a device function
+ Fix generation of warnUnsupportedDeviceFunctions function in hipify-clang
+ Update hipify-perl based on hipify-clang -perl generation
+ Update device test math_functions.cu for hipify-perl
[Restrictions]
- hipify-perl is yet unable to handle function declarations in user namespaces
- hipify-perl is yet unable to handle using directive
2019-09-16 17:36:55 +03:00
Evgeny Mankov
a8bf3c04f1
Merge pull request #1422 from emankov/master
...
[HIPIFY] Fix build failure due to typo in [#1377 ]
2019-09-16 14:30:01 +03:00
Evgeny Mankov
3f66e7b0b9
[HIPIFY] Fix build failure due to typo in [ #1377 ]
2019-09-16 14:28:52 +03:00
mhbliao
1f8c3bbd3b
[HIP] Remove a circular including. ( #1418 )
2019-09-16 08:32:47 +00:00
satyanveshd
cfbd0e77fe
Fix checks in hipMemcpy[D/H]to[D/H]() ( #1397 )
...
* SWDEV-202043 hipMemcpyDtoD() issue
2019-09-16 08:32:38 +00:00
AlexBinXie
5ed1f3e2c8
[hip]Skip test when hipHostMallocCoherent is not supported by implementation ( #1380 )
2019-09-16 08:31:43 +00:00
ansurya
ceb734b917
Added new device attributes ( #1377 )
...
* Added new device attributes
* updated comment
* updated with new device attributes supported
2019-09-16 08:31:30 +00:00
Aryan Salmanpour
bac52d3729
[hip] add initial support for hipLaunchCooperativeKernelMultiDevice API ( #1368 )
...
* [hip] add initial support for hipLaunchCooperativeKernelMultiDevice API
* fix formatting
2019-09-16 08:31:17 +00:00
Aryan Salmanpour
48880a017e
[hip][tests] add a unit test for testing hipLaunchCooperativeKernel ( #1361 )
...
* [hip][tests] add a unit test for testing hipLaunchCooperativeKernel
* use __ockl_grid_sync function
* remove already defined __ockl_grid_sync function
* use sync function for grid synchronization
2019-09-16 08:31:09 +00:00
Evgeny Mankov
05e689aa73
Merge pull request #1421 from emankov/hipify-clang
...
[HIPIFY] Scripting related refactoring
2019-09-15 21:17:49 +03:00
Evgeny Mankov
b7df0627b1
[HIPIFY] Scripting related refactoring
...
+ Move generating hipify-perl and hipify-python into separate namespaces and files
+ Move getAbsoluteFilePath and getAbsoluteDirectoryPath to StringUtils
2019-09-15 21:13:56 +03:00
Evgeny Mankov
8e901f028f
Merge pull request #1420 from emankov/hipify-clang
...
[HIPIFY][perl] Syn with hipify-clang
2019-09-15 15:26:54 +03:00
Evgeny Mankov
287ceac229
[HIPIFY][perl] Syn with hipify-clang
2019-09-15 15:25:40 +03:00
Evgeny Mankov
78f2386a8e
Merge pull request #1419 from emankov/hipify-clang
...
[HIPIFY][perl] Generate sub warnUnsupportedDeviceFunctions
2019-09-15 15:20:13 +03:00
Evgeny Mankov
fd8a7675f8
[HIPIFY][perl] Generate sub warnUnsupportedDeviceFunctions
2019-09-15 15:18:58 +03:00
Evgeny Mankov
6770e993a4
Merge pull request #1417 from emankov/master
...
[HIPIFY][doc] Update README.md
2019-09-13 18:58:22 +03:00
Evgeny Mankov
e51880cbca
Merge pull request #1416 from gargrahul/HACK_hipFreeForManangedMalloc
...
[HACK] Temporary fix in hipFree for hipManagedMalloc use
2019-09-13 18:57:21 +03:00
Evgeny Mankov
182058fa0b
[HIPIFY][doc] Update README.md
...
+ Adding -DLLVM_TARGETS_TO_BUILD="NVPTX" to LLVM's cmake command line for minimizing LLVM's build (tested)
+ Test logs update
+ Versions update
2019-09-13 18:56:48 +03:00
foreman
184ee3f5a0
P4 to Git Change 1997981 by cpaquot@cpaquot-ocl-lc-lnx on 2019/09/13 11:17:32
...
SWDEV-203438 - [HIP] AllGather RCCL test issue
The test tries to launch a kernel on two devices at once and they need to communicate with each other.
For that, it uses a custom stream for each devices.
Problem is in getNullStream we used to call syncStreams all the time
and it was syncing all the streams even the ones on different devices.
So that made the second kernel launch (on 2n dev) to wait for the first kernel to finish which
would never occur since the first one was waiting for the second one.
The fix is to not call syncStreams from getNullStream because we sync already anyway prior in general.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_context.cpp#21 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_event.cpp#16 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_internal.hpp#40 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#70 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#41 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_stream.cpp#24 edit
2019-09-13 11:28:33 -04:00
Jenkins
828e18eff7
Merge 'master' into 'amd-master'
...
Change-Id: I449c7c1d18e70c4f7f25b264d7871a49b75fc77c
2019-09-13 05:10:52 -04:00
Rahul Garg
f0a1f95864
Fix typo and colon in comment
2019-09-13 03:09:01 +05:30
Rahul Garg
fe47b2185c
[HACK] Temporary fix for hipFree for hipManagedMalloc
2019-09-13 02:10:21 +05:30
foreman
a508c497ec
P4 to Git Change 1997621 by cpaquot@cpaquot-ocl-lc-lnx on 2019/09/12 14:16:09
...
SWDEV-145570 - Fixed some logging issues regarding hip(Ext)ModuleLaunchXXX.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#40 edit
2019-09-12 14:35:36 -04:00
Evgeny Mankov
6739a6d656
Merge pull request #1414 from emankov/master
...
[HIPIFY] Add unsupported fp16 functions
2019-09-12 18:53:08 +03:00
Evgeny Mankov
34540002dd
[HIPIFY] Add unsupported fp16 functions
2019-09-12 18:51:25 +03:00
Evgeny Mankov
63198e0d66
Merge pull request #1413 from emankov/master
...
[HIPIFY] Add more unsupported device functions
2019-09-12 17:55:43 +03:00
Evgeny Mankov
58c1a144d1
[HIPIFY] Add more unsupported device functions
2019-09-12 17:53:25 +03:00
Jenkins
121206adf9
Merge 'master' into 'amd-master'
...
Change-Id: I99fcb0d69a58ce09065991a580662660d2941f9f
2019-09-12 06:06:14 -04:00
Evgeny Mankov
2cf9922754
Merge pull request #1410 from emankov/master
...
[HIPIFY][#1404 ] Fix: add cooperative_groups.h -> hip/hip_cooperative_groups.h
2019-09-11 18:13:05 +03:00
Evgeny Mankov
0426035468
[HIPIFY][ #1404 ] Fix: add cooperative_groups.h -> hip/hip_cooperative_groups.h
2019-09-11 18:07:09 +03:00
Evgeny Mankov
ccf8ffc0c9
Merge pull request #1406 from emankov/master
...
[HIPIFY][#1400 ] Fix Template Instantiation kernel launch (clang & perl)
2019-09-11 15:49:46 +03:00
Jenkins
c0f79c5411
Merge 'master' into 'amd-master'
...
Change-Id: I836a4d36b2f43db1e1850c55e8b174755abd414f
2019-09-10 10:07:07 -04:00
Evgeny Mankov
56ab105e9d
[HIPIFY][ #1400 ] Fix Template Instantiation kernel launch (clang & perl)
...
+ Enclose template instantiation kernel calls into round brackets, leave regular kernel names unchanged (hipify-perl doesn't handle cases with macros).
+ Fix corresponding tests.
PS. hipify-perl couldn't handle correctly the following cases due to macros expansion disability, thus hipify-clang should be used instead:
#define KERNEL_NAME_MACRO axpy<float>
#define KERNEL_CALL_MACRO axpy<float><<<1, 2>>>
#define KERNEL_ARG_LIST_MACRO a, x, y
// CUDA:
KERNEL_NAME_MACRO<<<1, 2>>>(KERNEL_ARG_LIST_MACRO);
KERNEL_CALL_MACRO(KERNEL_ARG_LIST_MACRO);
// hipify-perl:
hipLaunchKernelGGL(KERNEL_NAME_MACRO, dim3(1), dim3(2), 0, 0, KERNEL_ARG_LIST_MACRO);
KERNEL_CALL_MACRO(KERNEL_ARG_LIST_MACRO);
// hipify-clang:
hipLaunchKernelGGL((KERNEL_NAME_MACRO), dim3(1), dim3(2), 0, 0, KERNEL_ARG_LIST_MACRO);
hipLaunchKernelGGL((axpy<float>), dim3(1), dim3(2), 0, 0, KERNEL_ARG_LIST_MACRO);
2019-09-10 15:59:06 +03:00
Aditya Lad
90acfb809c
Replacment of RENAME command in CMakeLists.txt ( #1402 )
2019-09-10 08:35:16 +00:00
foreman
66a6e703f2
P4 to Git Change 1995672 by skudchad@skudchad_rocm on 2019/09/09 14:45:42
...
SWDEV-199293 - Fix build
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_rtc.cpp#4 edit
2019-09-09 14:50:37 -04:00
foreman
f3e2d275c5
P4 to Git Change 1995665 by skudchad@skudchad_test2_win_opencl on 2019/09/09 14:36:57
...
SWDEV-199293 - Implement hiprtcGetLoweredName
ReviewBoardURL = http://ocltc.amd.com/reviews/r/17960/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_rtc.cpp#3 edit
2019-09-09 14:42:47 -04:00
Jenkins
cbe6b65e2c
Merge 'master' into 'amd-master'
...
Change-Id: Ibb1dbe78aa52c8aeb77851ee1471e1edaec4c654
2019-09-09 05:10:28 -04:00
Evgeny Mankov
29d09eb269
Merge pull request #1398 from emankov/master
...
[HIPIFY] Add device functions support
2019-09-07 19:00:00 +03:00
foreman
8167dee7d5
P4 to Git Change 1994974 by skudchad@skudchad_test2_win_opencl on 2019/09/06 19:09:36
...
SWDEV-199293 - Implement some hipRTC apis.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/17953/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/build/Makefile.hip#18 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_rtc.cpp#2 edit
2019-09-06 19:11:43 -04:00
Evgeny Mankov
6602fadc16
[HIPIFY] Add device functions support
...
+ Add a corresponding matcher cudaDeviceFuncCall to match only (__device__ or __global__) and not __host__ functions.
+ Add a corresponding device functions mapping:
only unsupported are listed, cause supported are exactly the same as of CUDA and do not need transformation;
make FindAndReplace for device functions separated from host API calls.
+ Add a test to distinguish device functions and user-defined.
2019-09-06 18:34:12 +03:00
foreman
5db35686db
P4 to Git Change 1994498 by skudchad@skudchad_test2_win_opencl on 2019/09/06 04:25:22
...
SWDEV-202136 - Fix windows build
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#26 edit
2019-09-06 04:33:24 -04:00
foreman
ee1b139bd9
P4 to Git Change 1994199 by michliao@hliao-dev-00-hip.rocm-workspace on 2019/09/05 16:35:36
...
SWDEV-202136 - Fix typos in `hip_hcc.def.in`.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#25 edit
2019-09-05 16:45:19 -04:00
foreman
3b5927d7fe
P4 to Git Change 1994120 by axie@axie-hip-pal on 2019/09/05 14:22:40
...
SWDEV-201667 - Linker errors when using hipArrayCreate and hipMemcpyParam2D
Tests:
In Windows 10, the linker errors has been fixed with this code change.
http://ocltc.amd.com:8111/viewModification.html?modId=125600&personal=true&tab=vcsModificationBuilds
Review:
http://ocltc.amd.com/reviews/r/17943/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#24 edit
2019-09-05 14:34:47 -04:00