Ben Sander
e417eb5d35
Refactor hipHostRegister test.
...
Run all tests in one command.
Run 128 offsets.
2017-05-05 17:29:04 -05:00
Ben Sander
444e4a20ba
Update streamEventTEst.
...
- add checks for events across devices.
- refactor test to make sure it runs long enough to sensitive sync
techniques.
- add tests for DeviceSync, streamWaitEvent.
2017-05-05 17:29:04 -05:00
wsttiger
a10d37e5e6
Improve hipStreamWaitEvent test.
...
- use addOne kernel, use local initializer rather than init_array.
- use addOneReverse test to add from back of array. Test alternate fwd and backward to stress dependency logic.
- check device-side dependencies.
2017-05-05 17:29:04 -05:00
Evgeny Mankov
054bf0859d
[HIPIFY] LLVM 3.9 support
...
3.8 and 3.9 are both supported. 3.8 is stable, 3.9 needs more testing.
2017-05-05 21:28:02 +03:00
Rahul Garg
f77059218c
hipMemcpy2DAsync for HIP/NVCC
...
Change-Id: I46f0057fef49bdaaac41c1df80c3e27432b8f376
2017-05-04 06:47:55 +05:30
Evgeny Mankov
2636d91ba4
[HIPIFY] CUDA RT memcpy functions update.
...
cudaMemcpyFromSymbol -> hipMemcpyFromSymbol
cudaMemcpyFromSymbolAsync -> hipMemcpyFromSymbolAsync
cudaMemcpy2DAsync -> hipMemcpy2DAsync
2017-05-03 23:05:44 +03:00
Evgeny Mankov
72df25a0c7
[HIPIFY] Blas update: add a few functions, supported by HIP.
...
cublasDaxpy -> hipblasDaxpy
cublasDgemv -> hipblasDgemv
cublasDger -> hipblasDger
cublasDgemm -> hipblasDgemm
cublasDgemmBatched -> hipblasDgemmBatched
cublasGetStream -> hipblasGetStream
cublasSetStream -> hipblasSetStream
cublasDaxpy -> hipblasDaxpy
2017-05-03 22:45:56 +03:00
Evgeny Mankov
3b40776244
[HIPIFY] HIPIFY and HIP sync with CUDA Driver API data types.
...
+ Update CUDA_Driver_API_functions_supported_by_HIP.md.
+ Final update of HIPIFY with CUDA driver data types.
[TODO] Syncing HIPIFY and HIP by CUDA Driver API functions.
2017-05-03 21:59:48 +03:00
Rahul Garg
b136e80a45
Added support for hipMemcpy2DAsync in HIP/HCC
...
Change-Id: Ia4a8306f2dc1e33a81a7195ec29aef652fcccc4b
2017-05-03 22:29:12 +05:30
Evgeny Mankov
3e89277a9b
[HIPIFY] Rename enumConstantDecl -> enumDecl
...
Reason: not to mix up with clang's enumConstantDecl, used for enum DeclRefExpr (enum constant).
2017-04-28 21:59:33 +03:00
Evgeny Mankov
85a32c3987
* [HIPIFY] [FIX] Replacement error: enum cudaMemcpyKind kind -> hipMemcpyKindyKind kind
...
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/72
[Solution] [Workaround]
Offset calculation for enum VarDecl as param decl, declared with enum type specifier.
[Result]
enum cudaMemcpyKind kind -> enum hipMemcpyKind kind
[ToDo]
Test on terminal qualifiers (const, etc).
2017-04-28 21:45:36 +03:00
Evgeny Mankov
5f76cf3098
[HIPIFY] [FIX] replacement error: cudaError_t -> hipError_t_t
...
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/71
[Solution]
getUnqualifiedType for enumConstantDecl's type is added, except ordinary enum declarations (w/o typedef).
[ToDo]
Find more appropriate way of distinguishing redefined enum declarations and ordinary ones.
2017-04-28 21:03:03 +03:00
Aditya Atluri
208bdbbcbb
fixed hipFuncSetCacheConfig on rocm path
...
Change-Id: I937a3afbf115edc94a753a0beb2230ed60a6f021
2017-04-28 11:53:11 -05:00
Aditya Atluri
f368271872
fixed fast math expf and exp10f
...
Change-Id: I73963220f902efebb0a7404c5f8966dffb4c35ca
2017-04-26 19:01:10 -05:00
Aditya Atluri
e91c35fde0
added hipFuncSetCacheConfig API for nvcc path
...
Change-Id: I87fae35bc0e10a0dca5ae1c5015fe5d9e52a1d0d
2017-04-26 18:56:57 -05:00
Maneesh Gupta
a5fb13bab3
Merge branch 'amd-master' into amd-develop
2017-04-26 23:55:13 +05:30
Sun, Peng
50daa408aa
fix hip_complex.h header on NV path
...
Change-Id: Ia95d003ca1b284bab1c76723050e6b3b89178f65
2017-04-25 00:15:20 -05:00
Ben Sander
fbf28a8728
Add host offset checking
2017-04-24 22:30:54 -05:00
Ben Sander
e7af4ef641
Refactor hipHostRegister test.
...
- Add more testing for offsets.
- Parse cmdline options and use --tests.
2017-04-24 22:30:54 -05:00
Ben Sander
5ba167b82b
Refactor hipHostRegister to cover misaligned cases.
2017-04-24 22:30:54 -05:00
Ben Sander
3da8e94cbf
Tailor pointer info for src/dst before calling HCC copy routines.
...
HCC sometimes uses the srcPtrInfo or dstPtrInfo to determine the
pointer. Make sure these use the actual pointer and not the base
of the allocation.
2017-04-24 22:30:54 -05:00
Ben Sander
d120b2dd12
Add test for non-page-aligned mem copies.
2017-04-24 22:30:54 -05:00
Ben Sander
b44a3eefd1
Add corrected test for offsets
2017-04-24 22:30:54 -05:00
Ben Sander
6b16f56f38
Refactor hipMemcpy test to share mem alloc for multiple copies.
2017-04-24 22:30:54 -05:00
Ben Sander
693e5abc1c
Add negative testing for memory full condition.
2017-04-24 22:30:54 -05:00
Ben Sander
dfacfbb641
Fix hip debug for case where copyAgent is null (host-to-host)
2017-04-24 22:30:54 -05:00
Ben Sander
fb7eee01ff
Fix hipMalloc to return error code if allocation fails.
2017-04-24 22:30:54 -05:00
Aditya Atluri
1f532b06f6
fixed build issues with hipPointerGetAttributes
...
Change-Id: I3f5fbc05bdaef720884ba949075928752a070377
2017-04-24 15:31:07 -05:00
Aditya Atluri
85c189c846
changed arguments for hipPointerGetAttributes
...
Change-Id: Ia7a7c4722c1f7d0a23f0e5cc3dd6dea6c01c1fd8
2017-04-24 15:24:16 -05:00
Maneesh Gupta
1186af91d7
Merge branch 'amd-develop' into amd-master
...
Change-Id: I312fb9d1181733ef5160d1e993e2ae57ced0f6b3
2017-04-24 08:51:03 +05:30
Maneesh Gupta
c16e0fb088
Updated release notes
...
Change-Id: Ia98aff420ea9d488924dce8fe9168cec9da301ab
2017-04-24 08:48:35 +05:30
James Edwards
197a29f142
Specify full path of hip libraries in link file.
...
Change-Id: I49b788f3489e7abff6b11006ff97fdfca4e5942c
2017-04-22 15:54:14 -05:00
James Edwards
c33c84a596
Properly link hip cmake file into top level lib directory.
...
Change-Id: I2113a86ca6985f34fd0cfb091abdbce0f632cfc2
2017-04-21 22:34:26 -05:00
Evgeny Mankov
f66780f39b
[HIPIFY] [DOC] Readme.md update: Ubuntu 16.04 support
2017-04-22 01:01:31 +03:00
Evgeny Mankov
342c63d398
[HIPIFY] Initial sync HIPIFY with HIP by CUDA Driver API data types.
...
+ Introduce CUDA_Driver_API_functions_supported_by_HIP.md.
+ Initial update of HIPIFY with CUDA driver data types.
+ Initial sync HIP types against CUDA Driver and Runtime API types.
+ Typo fixes.
2017-04-21 21:46:30 +03:00
Ben Sander
f1f907cb51
Fix compilation error with nvcc (c++ nullptr)
2017-04-21 09:01:34 -05:00
Sun, Peng
c224209200
add support of hipLaunchKernelGGL on NV path
...
Change-Id: I0aeafd80c2181873be385d985f1d8ed86a98d136
2017-04-19 11:00:25 -05:00
Sun, Peng
ee299695a8
fix broken header in NV path
...
Change-Id: Ia3aff2a89d9ba49547f51ce03a3304dfab58ba25
2017-04-19 10:47:40 -05:00
James Edwards
edfefc9aee
Fix RPM HIP packages from specifying /opt
...
Change-Id: Iec3c3b81eef4c8888d425eefc80b12488a8d20a1
2017-04-19 12:35:42 +05:30
James Edwards
c5e1235f64
Fix RPM HIP packages from specifying /opt
...
Change-Id: Iec3c3b81eef4c8888d425eefc80b12488a8d20a1
2017-04-17 23:58:34 -05:00
Maneesh Gupta
137c04c536
FindHIP: Apply remove_item on non-empty lists only
...
Change-Id: Ib7fcb992d7e1bb679d4d86676fe3d980ba204815
2017-04-18 10:21:20 +05:30
Ben Sander
710c83b73b
Update bugs - Add CreateKernel, new signature for static kerns.
2017-04-16 14:22:48 -05:00
Maneesh Gupta
e45ee8c37a
FindHIP: Handle remove_item from empty lists
...
Change-Id: I6adf31b32edeae9e8454b1a2528064cf3985fca1
2017-04-13 14:41:11 +05:30
Maneesh Gupta
fb1689346d
Merge branch 'amd-develop' into amd-master
...
Change-Id: I05572d2b32f1df70b54e2efeb32c8a4d8055912d
2017-04-13 12:39:36 +05:30
Maneesh Gupta
b3b1ba1e5d
Do not rebuild cmake cache by default
...
Change-Id: Ie21e99beaa3465b54b5a6a77439c455f34de98b3
2017-04-13 12:38:38 +05:30
Maneesh Gupta
6db8c7c69e
dtests should ignore HIP_PATH env var
...
Change-Id: I27b1cdab6e6b799987dad3ce97b56c764b1b8867
2017-04-13 12:12:04 +05:30
Maneesh Gupta
0d4f1c2d0c
Add hip-config.cmake to hip_hcc package
...
Due to the way hip packages are generated, for the purpose of
packaging hip-targets*.cmake are not generated at build time.
However hip-config*.cmake are generated at build time.
This will be fixed in future.
Change-Id: I5d79bc58a4f7a324ae06457130d8372ffe403830
2017-04-13 10:34:33 +05:30
Ben Sander
8bd34535b4
Add integer abs (initial implementation, can be optimized with OCML)
...
Change-Id: I1f568c8c0e2333af1fda4c313dc48ea0c5b6ab00
2017-04-11 01:16:28 +00:00
Ben Sander
29af40fa76
Doc update for Serialization.
...
Describe workaround for partial specialization
2017-04-10 19:45:18 -05:00
Sun, Peng
3c04722d10
Fix ifndef guard in hip_fp16.h
...
Change-Id: I0215556e7aa98a74e8a984e4de3fb6e8cafdfb24
2017-04-10 11:17:05 -05:00