97fa7aeef6
added half data type and vector destructors
...
1. Added half data types to hip_fp16.h
2. Added destructor to vector data types
Change-Id: Id5ae76a663bb90a4bde2839ec79c58fbaee5072f
[ROCm/clr commit: fdcc223842
2017-06-02 11:19:33 -05:00
afbf55a9dc
[HIP] [HIPIFY] CUDA Driver API 8.0.44 JIT options support.
...
[ROCm/clr commit: 463c026976
2017-05-31 18:55:29 +03:00
0ca3262f0a
Add event controls for release fences.
...
Env var : HIP_EVENT_SYS_RELEASE
Event allocation flags : hipEventReleaseToDevice, hipEventReleaseToSystem
(remove hipEventDisableSystemRelease)
Update test for new functionality.
[ROCm/clr commit: 942ec0eff8
2017-05-27 16:02:34 -05:00
9442c6dd2d
Updates so hip compiles on CUDA.
...
[ROCm/clr commit: 8dc968f036
2017-05-27 15:55:07 -05:00
d302498787
Add hipHostMallocCoherent, hipHostMallocNonCoherent
...
Provide per-allocation control over coherent/non-coherent mem.
These overrid the default HIP_COHERENT_HOST_ALLOC setting.
[ROCm/clr commit: dda70ae514
2017-05-24 00:48:10 -05:00
f848e9f117
fix GGL helper header file, reorder for C++17
...
Change-Id: I3d9ddfe670bf7e3e8e7bd85e52cc61f48c19c213
[ROCm/clr commit: 2523f8a492
2017-05-22 08:52:43 -05:00
052f630bd3
[HIP] [HIPIFY] [FIX] cuModuleLoadDataEx -> hipModuleLoadDataEx
...
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/81
1. Do not use JIT options on HCC path, call hipModuleLoadData instead.
2. NVCC path is unchanged, to call cuModuleLoadDataEx with all options.
3. Get rid of manual hipification, based on #ifdef #else for NVCC/HIP.
4. Update documentation accordingly.
[ROCm/clr commit: ae9f14ef9c
2017-05-19 17:22:14 +03:00
905ab9a689
fixed vector type issues by reverting to old code, changed __half2 to map to vector types in llvm
...
Change-Id: I7317408c25e8c1a0c02a346042c9137e160c8bbd
[ROCm/clr commit: 0433a2e608
2017-05-18 10:51:07 -05:00
0318e91450
changed vector types to make sure it generate proper llvm vector types
...
Change-Id: I6c4616dae137dc4eac35e5827dc5b7f3251e0247
[ROCm/clr commit: 93fa174900
2017-05-16 21:35:40 -05:00
4ac6ac9d1d
Add initial HIP_SYNC_NULL_STREAM=0 mode.
...
This eliminates host-synchronization for null stream. Instead, the
null-stream uses GPU-side events to wait for other streams.
Default is OFF pending additional testing.
Add enhanced null-stream test.
Also refine HIP_TRACE_API.
[ROCm/clr commit: 8bc6ee5932
2017-05-16 19:04:25 -05:00
293524f1dc
Added hipMallocPitch on HIP/NVCC path
...
Change-Id: Ie3ba7d3f95acac23805efa919531043b350a3f21
[ROCm/clr commit: d22b731f95
2017-05-16 07:15:13 +05:30
9b222c0ae4
Add hipEventDisableSystemRelease flag.
...
[ROCm/clr commit: 2c2625cb9e
2017-05-12 16:06:03 -05:00
b5f0a9471d
added guard against hip_runtime.h so that non-hcc compilers can use it
...
Change-Id: I3d68deda9ce8a5956e21e15a69e549d6c21e3e39
[ROCm/clr commit: 14930dc594
2017-05-09 10:14:16 -05:00
ba1a3360ab
hipMemcpy2DAsync for HIP/NVCC
...
Change-Id: I46f0057fef49bdaaac41c1df80c3e27432b8f376
[ROCm/clr commit: f77059218c
2017-05-04 06:47:55 +05:30
9739d5d60f
Added support for hipMemcpy2DAsync in HIP/HCC
...
Change-Id: Ia4a8306f2dc1e33a81a7195ec29aef652fcccc4b
[ROCm/clr commit: b136e80a45
2017-05-03 22:29:12 +05:30
6dccc18fa0
fixed hipFuncSetCacheConfig on rocm path
...
Change-Id: I937a3afbf115edc94a753a0beb2230ed60a6f021
[ROCm/clr commit: 208bdbbcbb
2017-04-28 11:53:11 -05:00
afec5e075b
added hipFuncSetCacheConfig API for nvcc path
...
Change-Id: I87fae35bc0e10a0dca5ae1c5015fe5d9e52a1d0d
[ROCm/clr commit: e91c35fde0
2017-04-26 18:56:57 -05:00
a82ecbcc13
fix hip_complex.h header on NV path
...
Change-Id: Ia95d003ca1b284bab1c76723050e6b3b89178f65
[ROCm/clr commit: 50daa408aa
2017-04-25 00:15:20 -05:00
05c136e7b8
Fix hipMalloc to return error code if allocation fails.
...
[ROCm/clr commit: fb7eee01ff
2017-04-24 22:30:54 -05:00
119f5a1e67
changed arguments for hipPointerGetAttributes
...
Change-Id: Ia7a7c4722c1f7d0a23f0e5cc3dd6dea6c01c1fd8
[ROCm/clr commit: 85c189c846
2017-04-24 15:24:16 -05:00
4ac3084503
add support of hipLaunchKernelGGL on NV path
...
Change-Id: I0aeafd80c2181873be385d985f1d8ed86a98d136
[ROCm/clr commit: c224209200
2017-04-19 11:00:25 -05:00
6a3bee59f1
fix broken header in NV path
...
Change-Id: Ia3aff2a89d9ba49547f51ce03a3304dfab58ba25
[ROCm/clr commit: ee299695a8
2017-04-19 10:47:40 -05:00
ab55162481
Add integer abs (initial implementation, can be optimized with OCML)
...
Change-Id: I1f568c8c0e2333af1fda4c313dc48ea0c5b6ab00
[ROCm/clr commit: 8bd34535b4
2017-04-11 01:16:28 +00:00
a5e8222af8
Fix ifndef guard in hip_fp16.h
...
Change-Id: I0215556e7aa98a74e8a984e4de3fb6e8cafdfb24
[ROCm/clr commit: 3c04722d10
2017-04-10 11:17:05 -05:00
588dba559c
add math.h to cover sqrtf function
...
Change-Id: Ia37752710cea4ca77e0a4e61f8e69a0355d9488d
[ROCm/clr commit: bfa61eac2c
2017-04-10 08:53:12 -05:00
4cd6fb07d5
Update the define of __global__ for GGL
...
Change-Id: I563bb2a132403bcbe9e9f279b55406cf0255af7d
[ROCm/clr commit: 9884f2e83f
2017-04-07 16:39:33 -05:00
7cedda4996
Add more operator overloading for float2 type, contributed by Aditya
...
Change-Id: If1ab7fb24d64bb5304142aed0951c9bd5ad47d20
[ROCm/clr commit: 19987ede22
2017-04-07 14:17:41 -05:00
9fd6c7179d
update GGL to log launched kernel information
...
Change-Id: Ied0aa6055673c687071b4a579aecd17f0f3f09ce
[ROCm/clr commit: f699e02715
2017-04-07 14:06:31 -05:00
dc9e957a36
Finish adding start/stop event support to hipHccModuleLaunchKernel.
...
Change interface to use hipEvent_t rather than hipEvent_t*
Change-Id: I259062dc087a13d51dc27f84e1e8861f332a104d
[ROCm/clr commit: e9eaadd135
2017-04-06 21:02:50 -05:00
c2acfe24cc
Improve documentation for hipModuleLaunch functions.
...
Change-Id: I0e22621e499775740c3301347b7416d5f98c2414
[ROCm/clr commit: fb4e2307f1
2017-04-06 21:02:50 -05:00
950dfd28d6
fixed header structure for complex data types
...
Change-Id: I16bf19005d933f42e8c8603c5d0b2df8ea3ad04f
[ROCm/clr commit: 1358bd8f07
2017-04-06 16:43:26 -05:00
9d6a04d334
GGL update, add while 0 guard for hipLaunchKernel API
...
Change-Id: Ie48ef8ca2ab5e26a51febfcd92417902c33fbf66
[ROCm/clr commit: 49a38da6e3
2017-04-06 10:48:11 -05:00
d1c334df66
added __host__ for complex functions and corrected memset and memcpy test
...
Change-Id: I9ffefb7a0025aa111a54d20d2766982df15532e7
[ROCm/clr commit: 42739c37ef
2017-04-06 09:29:44 -05:00
388f35c67a
add extra guard to grid_launch_GGL header
...
Change-Id: I120619c08ea2d084804fcb1639efbe6c4648dde9
[ROCm/clr commit: 335e107f5f
2017-04-05 22:25:41 -05:00
e342bbc7fb
Add more include header file for GGL, to make it self-compilable
...
Change-Id: I833cb194784450fb86e7961a7f9fe196ce3c7da5
[ROCm/clr commit: 2ca9eec9fc
2017-04-03 11:16:35 -05:00
5f99f3811b
Enable promote free HCC by default
...
Change-Id: I5f82a8d958dd675a8a46d9d17458c71321daab7c
[ROCm/clr commit: cea13c34c3
2017-04-03 11:18:30 +05:30
0e23f09b47
Fix warpSize, for related issue in hipeigen and torch
...
Change-Id: Ic66b24923a363304dca189011869ba7a0a6f8895
[ROCm/clr commit: a7d7797781
2017-04-01 15:42:53 -05:00
5a4ac0929f
GGL update, fix for thread-safe access to streams (accelerator_views).
...
Change-Id: I6dd329a85b3ba7de23d52823febee0c53857a981
[ROCm/clr commit: df4cee1a57
2017-04-01 14:50:39 -05:00
e3d3cacebc
Fixed copyright and header names
...
Change-Id: Id595c65ea3b7289e87be4c42db5d8a31905a4fdd
[ROCm/clr commit: 3859f59189
2017-03-31 12:40:29 -05:00
e54234667c
fixed header names
...
Change-Id: I21650d6398187d3767b28e8ac81b2642d3b89a0e
[ROCm/clr commit: 614e7db513
2017-03-31 12:18:55 -05:00
3651cb672a
added new api hipHccModuleLaunchKernel
...
1. hipHccModuleLaunchKernel is same as hipModuleLaunchKernel with OpenCL workitem model
2. Added copy right
3. Fixed header naming
Change-Id: I6a7c35a3566e2f8d3f5056613e34193775d4b236
[ROCm/clr commit: 1cead6a4cd
2017-03-31 12:11:34 -05:00
ace18c6a57
remove extra GGL header info
...
Change-Id: I09f0d1b64a7a31eb2e926f19b69b7bafbacc7787
[ROCm/clr commit: e3a4547703
2017-03-30 18:16:56 -05:00
56b2c7d82e
Rename hipLaunchKernelV3 to hipLaunchKernelGGL
...
Change-Id: I303daae006db41e9b04eb591e0b09b2717a7cf66
[ROCm/clr commit: 9983fdb9e2
2017-03-30 18:10:17 -05:00
8b91454077
Enable GGL as the default kernel launch method
...
Change-Id: I8022d126ee28ff7e4d9a96209e399d4243d39d8b
[ROCm/clr commit: c081099de1
2017-03-30 13:36:46 -05:00
bf84371e0d
Update GGL to fix one Torch build issue
...
Change-Id: I95a2a335902e3c368ed29f075ac72eabbb64c97e
[ROCm/clr commit: a32f50b5c7
2017-03-29 09:03:21 -05:00
d64768262f
resolve GGL merge conflicts
...
Change-Id: I7a5ec3696cf2dd1a77b1686536a1cb84cbfed66e
[ROCm/clr commit: 6f38d0f55e
2017-03-27 22:46:41 -05:00
30eb97392a
Fix for MemcpyFromSymbol on HIP/NVCC path
...
Change-Id: Ice38307f72870ae468cbf0861e104f0fa46dfd56
[ROCm/clr commit: 41f0ebebf0
2017-03-27 00:35:10 +05:30
8a5d4a6ab8
Added support for Primary Context Management APIs
...
Change-Id: I70f91b4492e112dd8e12ecf511fdc18a27944a06
[ROCm/clr commit: 33a8cdc6d3
2017-03-26 23:45:54 +05:30
feae5cda6b
update GGL implementation to use hipLaunchKernel
...
Change-Id: Ibc08185c814bb07d54f3e68016b10eb7b9f2bf4b
[ROCm/clr commit: 4ca09e977d
2017-03-21 15:46:56 -05:00
962ac22fe0
Update GGL implementation to extended overload set for make_lambda_wrapper
...
Change-Id: I949f113671ddf155db8689e8a7f23d415839a7b5
[ROCm/clr commit: 6b9be30b70
2017-03-21 15:46:44 -05:00
Aditya Atluri
Evgeny Mankov
Ben Sander
pensun
Rahul Garg
Maneesh Gupta