3dae7945ff
Update HIP-Clang build instruction ( #1764 )
...
[ROCm/hip commit: af697eaefe
2020-01-06 12:32:47 +05:30
2a70e452a5
Document FMA settings ( #1717 )
...
[ROCm/hip commit: 47525e72db
2020-01-06 12:32:38 +05:30
6ac4d18753
Merge pull request #1759 from emankov/master
...
[HIP] Unify hipError_t (Step 2)
[ROCm/hip commit: 0dadb23327
2019-12-30 19:21:09 +03:00
6d2ec032fe
Change to generate hipVersion.h ( #1726 )
...
HIP_VERSION_MAJOR, HIP_VERSION_MINOR, HIP_VERSION_PATCH and HIP_VERSION pre-processor macros are now defined in hipVersion.h instead of being set by hipcc.
[ROCm/hip commit: aa4aea0754
2019-12-30 12:44:24 +05:30
ffea90f865
[hip] refactoring cooperative kernel launch APIs ( #1737 )
...
This PR is a follow-up on PR# #1698 and it makes two more APIs (hipLaunchCooperativeKernel/hipLaunchCooperativeKernelMultiDevice) inline so that they can work correctly with lazy binding.
[ROCm/hip commit: 6968aeb841
2019-12-30 12:42:17 +05:30
8f0daf4019
Fix texture 3D & 2D layered with N components ( #1746 )
...
SWDEV-151670: Issue with 3D texture with 4 components
SWDEV-151671: Issue with 2D layered texture with 4 components
Fixed memcpy when memory is allocated with driver API's.
Github issues: #1755
Fixed 3D default case when array type is not set during memory allocation.
[ROCm/hip commit: 391e99cd12
2019-12-30 12:41:42 +05:30
6ebd0592da
[HIPIFY][doc] LLVM 9.0.1 is the latest stable release now
...
+ Tested on Windows and Linux: no changes are needed
[ROCm/hip commit: 3254c3e2c1
2019-12-23 20:00:33 +03:00
abef353b5b
[HIP] Clean-up deprecated HIP error codes
...
hipErrorMemoryAllocation -> hipErrorOutOfMemory
hipErrorInitializationError -> hipErrorNotInitialized
hipErrorMapBufferObjectFailed -> hipErrorMapFailed
hipErrorInvalidResourceHandle -> hipErrorInvalidHandle
[ROCm/hip commit: 4921678b6c
2019-12-23 17:01:35 +03:00
d10837db9d
Add macro __HIP_ENABLE_CUDA_WRAPPER_FOR_OPENMP__ ( #1761 )
...
This is to allow force enable cuda wrapper for OpenMP for flexibility
[ROCm/hip commit: 3c90d57072
2019-12-23 19:24:54 +05:30
bb7fa73dcc
Fix the return type of demangle function so that its compatible across ABIs ( #1744 )
...
[ROCm/hip commit: 080b0b9a68
2019-12-23 19:11:40 +05:30
1f5ecc0f6a
Fix late-coming issues. ( #1724 )
...
Implementation for hipMemcpyWithStream.
[ROCm/hip commit: 75a11330aa
2019-12-23 19:11:24 +05:30
91f82210fb
replace array designator C99 ( #1694 )
...
* replace array designator C99
* Update texture_functions.h
Highlight valid and invalid values in texFormatToSize
Co-authored-by: Maneesh Gupta <maneesh.gupta@amd.com >
[ROCm/hip commit: 7d6634ce9d
2019-12-23 19:10:24 +05:30
b7803eb0c6
static inline in a header, just like excess sugar in a diet, causes bloat (#1692 )
...
[ROCm/hip commit: be70b9f7e7
2019-12-23 19:09:38 +05:30
97535f3e97
[HIP] Fix typo
...
[ROCm/hip commit: 9544682e2c
2019-12-23 12:06:44 +03:00
30ae3942dc
[HIP] Unify hipError_t (Step 2)
...
Step 2. Make a few hipError codes deprecated
Update hipify-clang, hipify-perl, docs and samples accordingly
[ROCm/hip commit: dbad4d9b7f
2019-12-22 02:05:31 +03:00
bf2f77846e
Merge pull request #1734 from vsytch/win10-hipMemcpy-fixes
...
Fix OOM issues with hipMemcpy test on Windows
[ROCm/hip commit: 9c91632194
2019-12-21 22:02:34 +03:00
ddc169cf4e
Update texture_functions.h
...
Highlight valid and invalid values in texFormatToSize
[ROCm/hip commit: d92169c05a
2019-12-21 12:25:36 +05:30
78dc660cb6
bump version to 3.1 ( #1756 )
...
* Bump version to 3.1
* [ci] Enable tests on ROCm 3.0
[ROCm/hip commit: 18c1003469
2019-12-20 12:10:25 -08:00
19fd269632
[HIPIFY][doc] Update Readme.md: versioning
...
[ROCm/hip commit: dd0bf8c7ff
2019-12-20 20:55:19 +03:00
ef3ee5908f
Fix for windows dtest build failure ( #1742 )
...
[ROCm/hip commit: e2fc00da65
2019-12-19 13:10:43 -08:00
ded857e2b9
[hip] Add macro guarding the enum conversion for scalar accessor. ( #1748 )
...
- That's a high overhead part, which needs enabling ONLY if necessary.
[ROCm/hip commit: 99a3b66110
2019-12-19 10:08:37 -08:00
4b6e343269
Merge pull request #1739 from emankov/hipify-clang
...
[HIP] Unify hipError_t (Step 1)
[ROCm/hip commit: 191808fe72
2019-12-19 07:57:02 +03:00
baf7e660a2
[HIPIFY] Revise HIP runtime
...
+ hipMemAllocHost is deprecated, use hipHostMalloc instead
+ hipMemAllocManaged -> hipMallocManaged
+ hipFreeHost -> hipHostFree
+ cudaMemset3D(Async) -> hipMemset3D(Async)
+ update docs and hipify-perl accordingly
[ROCm/hip commit: 3b00057922
2019-12-16 20:09:55 +03:00
6015efe6f5
Add comments explaining choice on the cap for system memory on Windows.
...
[ROCm/hip commit: bf3df9d7c0
2019-12-13 21:38:27 -05:00
29ca9f7d53
Add explicit cast when computing the available amount of system memory.
...
[ROCm/hip commit: 12634879e2
2019-12-13 21:37:30 -05:00
ca99107c7b
[HIP] Unify hipError_t (Step 1)
...
Step 1. Set the same values for RT error codes as for analogous Driver's
[Reason] RT's and Driver's error codes unification in CUDA 10.2
[ROCm/hip commit: d8737ba50c
2019-12-13 19:40:16 +03:00
1eeb648716
[HIPIFY][build] Fix building with LLVM trunk
...
[ROCm/hip commit: 96839c82ae
2019-12-13 15:42:27 +03:00
9827755cb5
[HIPIFY][BLAS] Sync with cuBlas/rocblas/hipBlas
...
[ROCm/hip commit: 396a8a7a00
2019-12-12 19:52:18 +03:00
9dade23425
Reduce the amount of free host memory to 40% of what is reported on Windows, otherwise we can run into OOM situations.
...
[ROCm/hip commit: b12c53cceb
2019-12-11 20:21:12 -05:00
e1f55b0cee
Only add hipExtLaunchMultiKernelMultiDevice for non-HCC compilers. ( #1729 )
...
[ROCm/hip commit: 444c931641
2019-12-10 10:32:25 -08:00
65f80f80a1
fix linking of vector types with gcc ( #1690 )
...
* fix linking of vector types when linking hipcc objects with gcc
* use __atribute__((vector_size)) with both clang and gcc
and reinstate nonaligned n=3 vector type
* use implicit conversion to value and ext_vector_type when available
* Alternate formulation for GCC compatibility
* Built-in arrays don't mix well with placement new
* Fix typo
* Add conversions to enum
* Fix Scalar_accessor assignment.
* Update hip_vector_types.h
* stir up the underlying_type hideous mess
This fixes the HIP build issue "error: only enumeration types have underlying types".
[ROCm/hip commit: 00d735cdc9
2019-12-10 09:40:15 +05:30
69ac658878
[HIPIFY] Add cleanupHipifyOptions functionality
...
Needed for sorting input files to avoid clang errors on unsupported options
[ROCm/hip commit: 40bac6b144
2019-12-08 15:13:28 +03:00
92401a01c2
[HIPIFY] main source file refactoring
...
[ROCm/hip commit: b472f41896
2019-12-07 13:50:13 +03:00
8ba707b016
[HIPIFY][fix][ #1246 ][ #1655 ] Sort input files based on their dependency graph
...
+ Implemented by using clang Driver infrastructure
[ToDo] Add tests for in-place hipification of multiple files
[ROCm/hip commit: 6868b38d1b
2019-12-07 11:20:17 +03:00
7703031d33
Fix hipExtLaunchMultiKernelMultiDevice refactoring. ( #1714 )
...
- Use the correct condition for HIP VDI runtime.
[ROCm/hip commit: e9da934ac6
2019-12-06 09:49:17 -08:00
ebdd811632
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.
[ROCm/hip commit: e53fc316f1
2019-12-06 09:51:53 +05:30
90d5c25f69
Merge pull request #1712 from emankov/hipify
...
[HIPIFY] CUDA 10.2 (functions only)
[ROCm/hip commit: 496f0c53b3
2019-12-04 18:39:10 +03:00
970724661d
[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
[ROCm/hip commit: ec9b6a95a3
2019-12-04 18:37:31 +03:00
8e620bc59b
Merge pull request #1710 from emankov/hipify
...
[HIPIFY] CUDA 10.2 fp_16 support
[ROCm/hip commit: c518929fbc
2019-12-04 13:28:43 +03:00
c9d68c898f
[HIPIFY] CUDA 10.2 fp_16 support
...
[ROCm/hip commit: e3c3c6d86e
2019-12-04 13:27:46 +03:00
261b0a6f50
[packaging] Rename hip packages to conform to debian package renaming rules ( #1708 )
...
[ROCm/hip commit: 1f6b3e2aa5
2019-12-04 15:55:37 +05:30
6a6aa3226d
[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
[ROCm/hip commit: 5a03e8ccb9
2019-12-04 12:25:20 +03:00
fa87abd58a
Fix hipMemcpyWithStream ( #1707 )
...
* Fix hipMemcpyWithStream
* Add a dtest for the same.
[ROCm/hip commit: 892d2a562e
2019-12-04 11:56:23 +05:30
abe7531676
[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.
[ROCm/hip commit: 68cc787781
2019-12-04 11:50:51 +05:30
ee44fffd35
Revert "[packaging] Conform to debian package naming rules ( #1647 )" ( #1688 )
...
This reverts commit 8016fa32df4585b7f6f3
2019-12-01 14:47:43 +05:30
389263471e
[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
[ROCm/hip commit: d33dd71ed3
2019-11-29 19:29:11 +03:00
4e564f783e
Revert changes for atomic FADD support when address is in LDS ( #1701 )
...
This reverts PR #1591 and follow-on PR #1695
[ROCm/hip commit: 32442c6506
2019-11-29 11:58:12 +05:30
dc9bc8ee86
Uniform is_shared query. ( #1695 )
...
[ROCm/hip commit: b6514fffb9
2019-11-28 13:39:05 +05:30
f1544b621d
fix hipMempcy precondition, test for zero size first ( #1697 )
...
[ROCm/hip commit: 299bbe2325
2019-11-25 19:03:03 -08:00
023ee64ecf
replace array designator C99
...
[ROCm/hip commit: edf29b8673
2019-11-25 16:51:49 +05:30
Yaxun (Sam) Liu
Evgeny Mankov
Sarbojit2019
Aryan Salmanpour
ansurya
saleelk
Alex Voicu
Maneesh Gupta
mhbliao
Vladislav Sytchenko
jglaser
Rahul Garg
Jeff Daily
Anusha Godavarthy Surya