f68bee02f5
[HIPIFY][tests] Rename the ambiguous call as well
2019-10-25 16:07:31 +03:00
9529e1d91d
[HIPIFY][tests] Fix ambiguous call to cusparseGetErrorString declared in cusparse.h
2019-10-25 16:04:20 +03:00
40522e2b6a
Add missing operators, fix GCC compilation. ( #1589 )
2019-10-25 15:44:24 +05:30
f909a393ff
Fix deadlock, remove old __sync_* use. ( #1584 )
...
This fixes a deadlock introduced by the switch to TTAS loops, and is therefore mildly urgent (to prevent the CI from hoovering in the broken code).
2019-10-25 15:44:17 +05:30
66a3c874c8
[dtest] Fix hipMemset2D test ( #1579 )
...
Reverts changes made in #1399 . This is a RT api test. For testing hipMemAllocPitch , a new test should be written and that should use correct memset API.
2019-10-25 15:44:05 +05:30
14b870d1ce
Add hipMemcpy2DfromArray ( #1510 )
...
Adds hipMemcpy2DFromArray and hipMemcpy2DFromArrayAsync equivalent to cudaMemcpy2DFromArray and cudaMemcpy2DFromArrayAsync.
2019-10-25 15:43:33 +05:30
ff8d3fa446
Update profiling doc ( #1576 )
2019-10-24 17:51:55 +05:30
f53b1a1755
Adding New Analyze Target Merging with cppcheck ( #1583 )
2019-10-24 17:46:06 +05:30
170c4f0270
Add HIP checks in texture driver sample ( #1581 )
2019-10-24 17:45:51 +05:30
f25692b399
Hip vdi profiling header ( #1577 )
...
Add HIP-VDI profiling interface for GPU timing collection.
2019-10-24 17:45:42 +05:30
26914ec76e
Make CAS loops use the TTAS idiom. ( #1573 )
...
* Make CAS loops use the TTAS idiom.
* More efficient re-formulation of TTAS.
* Fix typo.
* The typo was not quite a typo
2019-10-24 17:45:20 +05:30
6c5fbf9b4a
Fix occupany APIs ( #1560 )
...
Addresses SWDEV-205006
2019-10-24 17:44:47 +05:30
15a699688e
Improve performance of v2 arg handling ( #1539 )
...
* Improve performance of v2 arg handling
* Missing change to `std::string`
2019-10-24 17:44:05 +05:30
84d5b399f6
Improve scalar access into vector types. ( #1531 )
...
The improvement is based on the ideas here: https://t0rakka.silvrback.com/simd-scalar-accessor . It yields significantly better ISA when the base's .xyzw members are used.
2019-10-24 17:43:49 +05:30
93c688a0c9
[hip] add support for implicit kernel argument for multi-grid sync ( #1456 )
...
* [hip] add support for implicit kernel argument for multi-grid sync
* modified code for calculating the prev_sum
* change the impCoopArg type to size_t
* add memory clean up
* launch init_gws and main kernels into two separate loops
2019-10-24 17:43:30 +05:30
465581612e
Merge pull request #1559 from vsytch/win10_aligned_alloc
...
Fixes for hipMemcpy_simple on Windows
2019-10-23 13:10:59 -07:00
2435567e70
[HIPIFY][cmake][ #1571 ] Take into account building hipify-clang as a part of building HIP while installing
...
[Algorithm]
[Release]
If CMAKE_INSTALL_PREFIX is set by the user:
If BIN_INSTALL_DIR is set by HIP, use it as CMAKE_INSTALL_PREFIX, otherwise CMAKE_INSTALL_PREFIX is used unchanged.
If the user does not set CMAKE_INSTALL_PREFIX (CMAKE_INSTALL_PREFIX_INITIALIZED_TO_DEFAULT):
If BIN_INSTALL_DIR is set by HIP, use it as CMAKE_INSTALL_PREFIX, otherwise use PROJECT_BINARY_DIR/bin for installation.
[Debug]
If CMAKE_INSTALL_PREFIX is set by the user:
CMAKE_INSTALL_PREFIX is used unchanged.
If the user does not set CMAKE_INSTALL_PREFIX (CMAKE_INSTALL_PREFIX_INITIALIZED_TO_DEFAULT):
use CMAKE_CURRENT_SOURCE_DIR/bin for installation.
Standalone build left unchanged: CMAKE_INSTALL_PREFIX is used if set.
2019-10-23 18:54:45 +03:00
7ab06b3892
[HIPIFY] Disable delayed template parsing
...
By implicit unconditional passing -fno-delayed-template-parsing option (which appeared in LLVM 3.8.0, thus doesn't need compatibility wrapping) to hipify-clang.
[Reason] To parse uncalled template functions otherwise they are not parsed without calling, thus not hipified.
Affects cub_03.cu test, which has uncalled global template function.
2019-10-22 19:07:37 +03:00
e2191e23e6
[HIPIFY][ #1569 ] Fix
2019-10-22 11:08:37 +03:00
3233a845f6
[HIPIFY][tests] Set max clang's CudaArch for corresponding CUDA major.minor version
...
[Reason] To support maximum CUDA features in offline tests
+ Add defined(__CUDA_ARCH__) && __CUDA_ARCH__ >= 600 restriction for atomicAdd on doubles in atomics.cu.
So if LLVM < 7 and --cuda-gpu-arch doesn't work, __CUDA_ARCH__ is unset too (350 by default in clang);
if LLVM >= 7 --cuda-gpu-arch is used and __CUDA_ARCH__ is set based on it.
2019-10-21 17:50:00 +03:00
9633cdbd8a
[HIPIFY][perl] Support of 'using namespace cub'
2019-10-21 17:15:05 +03:00
9fc7afa738
[HIPIFY][tests] Set max clang's CudaArch for corresponding CUDA version
...
[Reason] To support maximum CUDA features in offline tests
+ Add CUDA_VERSION >= 800 restriction for atomics.cu
[TODO] Find a way to use or exclude atomicAdd for doubles if LLVM < 7, because
LLVM 6.0.1 and older do not use --cuda-gpu-arch in clang's Driver code at all (option is only declared)
2019-10-21 15:51:25 +03:00
ff6057d1ff
[HIPIFY][tests] Set -I for CUDA path instead of --cuda-path for LLVM < 4
2019-10-20 20:08:56 +03:00
5bf1ff19ff
[HIPIFY][tests] Exclude all CUB tests if CUDA_CUB_ROOT_DIR is not set
2019-10-20 20:03:18 +03:00
432380aa5d
Remove extra #endif.
2019-10-18 16:40:29 -04:00
44a897a146
[HIPIFY][tests] Test clean-up
2019-10-18 18:55:52 +03:00
86f6756b02
[HIPIFY][CUB][ #1460 ] Add "using namespace cub" translation support
...
+ Add cub_03.cu
2019-10-18 18:51:40 +03:00
eb6690bbba
Merge pull request #1558 from aaronenyeshi/fix-hipify-cmake-version
...
[HIPIFY][cmake] Make CMakeLists use default 3.5.1 for Ubuntu 16.04
2019-10-18 06:39:35 +03:00
07eed1e5bf
Merge pull request #1550 from yxsamliu/new-launch
...
Add -fhip-new-launch-api to hipcc for HIP/VDI
2019-10-17 19:07:32 -07:00
f4440817cb
_aligned_malloc() on Windows first takes size, then alignment, which is the opposite of how the similar function behaves on Linux. Memory allocated by it also has to be freed using _aligned_free(), unlike Linux where we can use regular free().
...
Edit aligned_alloc() macro and add a aligned_free() one to align with the above behaviour.
2019-10-17 18:58:32 -04:00
b3ea58abe7
[HIPIFY][cmake] Make CMakeLists use default 3.5.1 for Ubuntu 16.04
2019-10-17 21:21:24 +00:00
1165e6bd71
[HIPIFY][doc] Update README.md
...
+ Versions, testing
2019-10-17 22:26:48 +03:00
446718f990
Revert "hipcc defaults to code object v3 ( #1298 )"
...
This reverts commit c8f38a139a
2019-10-17 13:27:28 -04:00
8c3dff7ab9
[HIPIFY][cmake] Add install rule for clang-resource-headers
...
+ Fix: set destination for all installing files to ${CMAKE_INSTALL_PREFIX}
2019-10-17 15:05:55 +03:00
a21fe1443b
Merge pull request #1544 from vsytch/master
...
QoL changes to the hipMemset family
2019-10-16 18:54:20 -07:00
e557563947
[HIPIFY][CUB][ #1460 ] Add cub:: namespace support in TemplateInstantiation of cudaLaunchKernel
...
+ Update cub_02.cu test accordingly
2019-10-16 19:02:13 +03:00
86d0c5fa5a
hipMemset2D and hipMemset3D tests should be passing by default.
2019-10-16 11:02:38 -04:00
0a20048759
[HIPIFY] Refactor a couple of matcher functions
...
+ Separate out GetSubstrLocation function for finding substr SourceLocation in a given SourceRange
2019-10-16 13:43:56 +03:00
5555d46e66
[HIPIFY][CUB][ #1460 ] Implement cubFunctionTemplateDecl matcher
...
+ Add cub_02.cu test
+ Partial fixes #1460
2019-10-16 13:08:11 +03:00
9d571e3c9e
Use the correct return type in runTest in 11_texture_driver sample. ( #1546 )
...
Fixes SWDEV-203394.
Currently in runTest() returns true, even if the texture reference copy does not happen. Using the existing testResult Flag to return from runTest().
2019-10-16 10:52:15 +05:30
c2aadd4d12
Update hipMathFunctions, hipTestHalf and hipTestNativeHalf tests to support Navi10 and Navi14. ( #1545 )
2019-10-16 10:51:48 +05:30
9773f94c71
[hipcc] Temporary add -D_OPENMP to clang options to workaround cmake issue ( #1540 )
...
* Temporary add -D_OPENMP to clang options in hipcc to allow using CMake OpenMP detection with hip-clang (until updated CMake version is available).
2019-10-16 10:51:28 +05:30
d16963c9d5
Guard against division by zero for no VGPR usage (e.g., in an empty kernel) ( #1528 )
...
* guard against division by zero for no VGPR usage (e.g., in an empty kernel)
* fix bracket format
* clean up parenthesis
2019-10-16 10:49:56 +05:30
b3351561c5
Adding code object manager to rtc ( #1526 )
...
Adding Code Object Manager file to rtc to resolve address of Bundled_code_object in libhiprtc.so
2019-10-16 10:49:16 +05:30
f9b8a01c77
Fix struct declaration for C ( #1524 )
...
This change is necessary for HPCToolkit to use Roctracer to produce code centric profiling view.
2019-10-16 10:48:55 +05:30
739530d53b
Add -fhip-new-launch-api to hipcc for HIP/VDI
2019-10-15 21:47:33 -04:00
cc5abec092
In the hipMemset2D and hipMemset3D tests synchronize with the default stream after performing an async memset.
2019-10-15 17:15:49 -04:00
f402b6d01a
Update indentation in the hipMemset3D test. Replace all tabs with four spaces.
2019-10-15 15:29:14 -04:00
c83b6adb33
Add async subtest to hipMemSet3D
2019-10-15 14:24:04 -04:00
39e42d4056
hipMemset2D test should pass only if both async and sync subtests pass.
2019-10-15 14:20:14 -04:00
Evgeny Mankov
Alex Voicu
Rahul Garg
Jatin Chaudhary
gandryey
satyanveshd
searlmc1
Aryan Salmanpour
Vladislav Sytchenko
Aaron Enye Shi
kjayapra-amd
vsytch
kpyzhov
Nick Curtis
Xiaozhu Meng
Yaxun (Sam) Liu