Commit Graph

1062 Commits

Author SHA1 Message Date
Maneesh Gupta 4a4745e466 Merge pull request #1069 from mhbliao/hliao/master/test_cleanup
[test] Remove unused common routines.

[ROCm/clr commit: 0fffbbe67a]
2019-05-07 16:02:57 +05:30
Michael LIAO f644d4daaa [test] Add device variant of std::declval.
- Current clang disallows any invocation of wrong-side functions even
  under context with type-inspection only. Work around that by adding a
  variant of `std::decl` with `__device__` attribute.


[ROCm/clr commit: 32f69c8bc4]
2019-05-03 15:58:31 -04:00
Michael LIAO fa74e75fc1 [test] Use explicit cast for address space cast.
[ROCm/clr commit: a27877794f]
2019-05-03 14:56:00 -04:00
Maneesh Gupta c73c864fdf Merge pull request #1058 from mhbliao/hliao/master/devfunc
[Device Function] Fix implementation

[ROCm/clr commit: ad070d4da5]
2019-05-03 17:47:51 +05:30
Konstantin Pyzhov 1be1dd207a hipMultiThreadDevice test: Reduced maximum number of created HIP streams on Windows.
[ROCm/clr commit: e04e408a37]
2019-05-03 05:43:30 -04:00
Konstantin Pyzhov b2e3880696 Refined hipSetDevice test.
[ROCm/clr commit: 29fd3812e5]
2019-05-03 05:39:45 -04:00
Michael LIAO 9823184397 [test] Remove unused common routines.
- Remove used common routines, which variadic functions, not supported
  feature, are used.


[ROCm/clr commit: aad19f6171]
2019-05-02 12:34:07 -04:00
emankov c27f0b952e [HIPIFY][tests] Add cuSPARSE CSR-BCSR-SPMV-conversions example
[ROCm/clr commit: d5c3e5ea71]
2019-04-30 17:37:34 +03:00
Michael LIAO eb43303d0b [Device Function] Fix implementation of __bitinsert_u64
- It's a common mistake by assuming 1 << shamt would be promoted to
  64-bit, if shamt is a 64-bit integer. That's not the case. Replace
  that left shift to a 64-bit one to ensure it won't fall into undefined
  behavior.
- Fix the host-side implementation as well for device function testing.


[ROCm/clr commit: 2380eb8ecc]
2019-04-30 08:59:13 -04:00
Evgeny Mankov 3c3255fbe5 [HIPIFY][DNN] cudnnSetFilter4dDescriptor support
[ROCm/clr commit: 72a809caf6]
2019-04-25 12:18:51 +03:00
Rahul Garg d69edbbb7f Add hipMallocManaged default functional support (#1036)
* Add hipMallocManaged default functional support

* Fix build error

* Add dtest


[ROCm/clr commit: 94769fc8dd]
2019-04-24 16:50:03 +05:30
Rahul Garg c6dd7c9678 Fix D2DAsync test
[ROCm/clr commit: f1dc017167]
2019-04-18 07:35:06 +05:30
Evgeny Mankov 17ac0d3e19 [HIPIFY][tests] Fix typo in test for CUDA 10.x
[ROCm/clr commit: 5b59f87305]
2019-04-01 19:52:08 +03:00
Evgeny Mankov 36f569cba6 [HIPIFY][tests] Update lit testing infrastructure
+ Set -D__LP64__ in case of 64-bit hipify-clang binary
  [partial workaround for clang's bug https://bugs.llvm.org/show_bug.cgi?id=38811]

  C:/GIT/LLVM/trunk/llvm-64-release-vs2017/dist/lib/clang/9.0.0\include\__clang_cuda_device_functions.h(1609,45): error GEF7559A7: no matching function for call to 'roundf'
  __DEVICE__ long lroundf(float __a) { return roundf(__a); }

  #if defined(__LP64__)
  __DEVICE__ long lround(double __a) { return llround(__a); }
  __DEVICE__ long lroundf(float __a) { return llroundf(__a); } // ok: llroundf should be used when 64-bit
  #else
  __DEVICE__ long lround(double __a) { return round(__a); }
  __DEVICE__ long lroundf(float __a) { return roundf(__a); } // error
  #endif

+ Print more system info while testing in the following form:
  ========================================
  CUDA 9.0 - will be used for testing
  LLVM 9.0.0svn - will be used for testing
  AMD64 - Platform architecture
  Windows 10 - Platform OS
  64 - hipify-clang binary bitness
  32 - python 3.7.2 binary bitness
  ========================================


[ROCm/clr commit: f138f89bc8]
2019-03-06 19:26:05 +03:00
Maneesh Gupta b1fdc46049 Merge pull request #932 from ROCm-Developer-Tools/feature_maybe_dlopen_test
Add one test case for complex dynamic loading behavior

[ROCm/clr commit: b9809cb2b6]
2019-03-06 17:32:23 +05:30
Maneesh Gupta 5c939d7f27 Update hipMemset.cpp
Address build issues on nvcc path.

[ROCm/clr commit: 8af4e2b5e4]
2019-03-05 12:11:11 +05:30
Wen-Heng (Jack) Chung 21cf5b0ae4 Add direct test for hipMemsetD32 and hipMemsetD32Async
[ROCm/clr commit: 365d08535b]
2019-03-04 17:20:32 +00:00
Rahul Garg b3ba23ba81 Fix review comments
[ROCm/clr commit: 5900416629]
2019-03-02 23:38:37 +05:30
Rahul Garg e6f5850ebb Fix hipMemset test for HIP/NVCC
[ROCm/clr commit: 41afe4d947]
2019-03-01 03:46:57 +05:30
Rahul Garg 1a97c0fafc Update hipMemset test
[ROCm/clr commit: 0156388a6b]
2019-02-28 06:54:49 +05:30
Wen-Heng (Jack) Chung 3faa98b08d Add one test case for complex dynamic loading behavior
Existing HIT syntax doesn't seem to support the expected build and run steps
for this test.


[ROCm/clr commit: 8c5a92a789]
2019-02-25 17:03:31 +00:00
Evgeny Mankov 7d96cc8b63 [HIPIFY][tests] caffe2 test fix
[ROCm/clr commit: 9c000f4b57]
2019-02-25 17:12:32 +03:00
Evgeny Mankov e42eda2744 [HIPIFY][Caffe2] Initial Caffe2 support
[ROCm/clr commit: 59533a7309]
2019-02-23 20:46:22 +03:00
Evgeny Mankov 656baeebe3 [HIPIFY][DNN] Revise and sync with CUDNN 10 and hipDNN
+ CUDNN_API_supported_by_HIP.md update
+ CUDA_Runtime_API_functions_supported_by_HIP.md update
+ libraryPropertyType_t is moved to Runtime
+ test update


[ROCm/clr commit: b22c0de428]
2019-02-13 19:40:59 +03:00
Evgeny Mankov 29f6a802ff [HIPIFY][RAND] revise
+ add missing functions
+ add minimum rocRAND support
+ updated CURAND_API_supported_by_HIP.md


[ROCm/clr commit: d1ea9d9d97]
2019-02-08 17:54:28 +03:00
Evgeny Mankov b0b83332df [HIPIFY][BLAS][tests] Add tests on hipifying to 'roc'
[ROCm/clr commit: 72c41d4ebf]
2019-02-07 19:25:23 +03:00
Maneesh Gupta 53c5bc3a86 Merge pull request #863 from aaronenyeshi/add-dot-instructions
Add Dot functions as amd_mixed_dot function

[ROCm/clr commit: 95fe4b770f]
2019-01-29 06:01:01 +05:30
Aaron Enye Shi 859c3bc699 Fix mixed dot for Jenkins rocm_2_0
[ROCm/clr commit: 2ab6fd374c]
2019-01-24 22:48:35 +00:00
Aaron Enye Shi d54135c38d Exclude mixed dot functions from nvcc path
[ROCm/clr commit: 77de87d0cd]
2019-01-24 21:00:45 +00:00
Evgeny Mankov 866ba6db18 [HIPIFY] CUDA Runtime API functions total revise (up to CUDA 10.0)
+ for all CUDA versions starting from 7.0
+ add missing functions
+ sync with HIP
+ update CUDA_Runtime_API_functions_supported_by_HIP.md
+ update affected tests
+ fix typos
+ formatting, annotating


[ROCm/clr commit: 229b34fc79]
2019-01-24 14:05:36 +03:00
Aaron Enye Shi 5024e71039 Fix mixed dot for Jenkins rocm_head
[ROCm/clr commit: a014668d3a]
2019-01-23 16:39:25 +00:00
Aaron Enye Shi 0e1860dd53 Add tests for dot functions
[ROCm/clr commit: bbe5a0381f]
2019-01-22 22:32:28 +00:00
Evgeny Mankov 9ee1b594e9 [HIPIFY][tests] Exclude one more test if CUDA == 7.0, which is of CUDA 7.5
[ROCm/clr commit: 994ee4bab0]
2019-01-22 22:26:43 +03:00
Yaxun Sam Liu 34f021d374 Fix get_time in tests for nvcc
[ROCm/clr commit: f49aed237b]
2019-01-21 14:07:50 -05:00
Yaxun Sam Liu 25b47bfed7 Use chrono instead of sys/time.h in test
sys/time.h is not available on Windows. Use C++11 <chrono> instead so that the tests
compile on Windows.


[ROCm/clr commit: 75a67c216f]
2019-01-20 00:05:27 -05:00
Evgeny Mankov 175c294553 [HIPIFY] option cuda-path might be used with '--' as well as with a single dash: '-'
+ Update README.md


[ROCm/clr commit: 9b6af0a81d]
2019-01-10 18:01:33 +03:00
Evgeny Mankov 7d81558657 [HIPIFY] Introduce CUDA installation path option '-cuda-path'
Repeats clang's '--cuda-path' option.

[Reason]
In case of absence of any other clang's options setting '-cuda-path' allows not to specify separator '--' before clang's '--cuda-path'.

+ Tests and scripts are updated accordingly.


[ROCm/clr commit: c4268855e3]
2019-01-09 20:18:36 +03:00
Maneesh Gupta e7195f7dc2 Update hipAsynchronousStreams.cpp
This test needs to be excluded from nvcc platform

[ROCm/clr commit: 79251fc4e2]
2019-01-07 09:26:21 +05:30
Sameer Sahasrabuddhe 0a99432a01 introduce test for asynchronous streams
This test is identical to the HCC tests under
tests/Unit/OCKL/AsyncStreams, but only for synchronous operation.


[ROCm/clr commit: 39574dba4c]
2019-01-04 05:51:50 -05:00
emankov 0207ebf990 [HIPIFY][tests] CUDA 7.0 is also supported
+ Exclude 1 test for CUDA 7.0
+ Update README.md


[ROCm/clr commit: 9808b22864]
2019-01-04 01:32:30 +03:00
emankov e8d2dab718 [HIPIFY][tests] Exclude CUDA 8.0 (9.0) tests if CUDA < 8.0 (9.0)
[ROCm/clr commit: 852bc8d545]
2019-01-03 18:06:49 +03:00
emankov 18ce5086c5 [HIPIFY][tests] Reduce cudaRegister.cu test
[ROCm/clr commit: 410f8f0b92]
2019-01-03 17:17:21 +03:00
Maneesh Gupta 5b8b5b842b Merge pull request #818 from yxsamliu/fix-hip-config
Add default arguments for hipConfigureCall

[ROCm/clr commit: aa932d542a]
2019-01-01 10:25:25 +05:30
Maneesh Gupta 389ef21eca Merge pull request #821 from gargrahul/remove_redundancy_memset3d_test
[dtest] Remove redundant arguments from hipMemset3D test

[ROCm/clr commit: 7647bdb664]
2018-12-31 15:32:19 +05:30
emankov 170add36d8 [HIPIFY] Get rid of setting '--cuda-gpu-arch='
[Reasons]
+ We don't compile kernel code at least for now as HIP kernel syntax is almost equal CUDA's;
+ clang always includes PTX in its binaries, so e.g. a binary compiled with --cuda-gpu-arch= would be forwards-compatible with e.g. sm_35 GPUs.


[ROCm/clr commit: 6f6228cf36]
2018-12-29 19:48:28 +03:00
emankov 843f936503 [HIPIFY] Starts implicitly setting '-x cuda' by hipify-clang itself
+ No need in setting '-x cuda' for the user anymore;
+ Testing and Readme updated accordingly.


[ROCm/clr commit: cea1e5a1a6]
2018-12-29 17:04:59 +03:00
emankov 1e3c49e8f2 [HIPIFY][win] Set -std=c++14 if MSVC
+ Due to latest Windows SDK implementation based on c++14 features, '-std+c++14' is a must;
+ Doesn't affect older versions of MSVC 2017, 2015 (the latest supported is MSVC 2015);
+ On linux still '-std=c++11' is a must;
+ Exclude '-std=' from lit config.


[ROCm/clr commit: b24e89b40f]
2018-12-29 15:51:06 +03:00
Rahul Garg 489f3ef7ba Remove redundant arguments from hipMemset3D test
[ROCm/clr commit: 10d2a9a269]
2018-12-20 06:51:22 +05:30
Yaxun Sam Liu 51702bed29 Add default arguments for hipConfigureCall
[ROCm/clr commit: 02deeadaa4]
2018-12-18 15:48:21 -05:00
Aaron Enye Shi 56262f0970 Add *_rn functions back into HIP intrinsics
Add back the round-to-nearest-even intrinsics back to HIP math
intrinsics as it caused regression.


[ROCm/clr commit: 2d50392c17]
2018-12-18 19:31:54 +00:00