301a7a199b
Missing __device__.
...
[ROCm/clr commit: b72d82f982
2018-06-01 19:48:36 +01:00
4306856ec9
Fix typos / address review comments.
...
[ROCm/clr commit: 91d9ec75d7
2018-06-01 16:20:21 +01:00
130e9d409e
Re-sync with upstream.
...
[ROCm/clr commit: f2d7f112ab
2018-06-01 15:49:05 +01:00
11e33a1203
Map hipArray alloc flags on NVCC
...
[ROCm/clr commit: b1b9a477a2
2018-06-01 17:28:43 +05:30
2a9a7b9aa3
Merge pull request #447 from ROCm-Developer-Tools/feature_native_vector_types
...
Switch over to using native vector types, for better codegen. Remove noise.
[ROCm/clr commit: 3fec282097
2018-06-01 13:58:07 +05:30
125f6201c8
Merge pull request #466 from ROCm-Developer-Tools/feature_use_Float16
...
Feature use _Float16 and match CUDA __half behaviour.
[ROCm/clr commit: 8ecb3eeb55
2018-06-01 13:50:12 +05:30
a9cabfc8df
Re-sync with upstream. Add integer abs.
...
[ROCm/clr commit: e03ca1a72e
2018-05-31 16:38:00 +01:00
37c73225e3
Merge branch 'feature_use_Float16' of https://github.com/ROCm-Developer-Tools/HIP into feature_use_Float16
...
[ROCm/clr commit: e20380319a
2018-05-31 15:27:31 +01:00
840d5c5fe0
Add missing interop with volatile. Fix unit tests.
...
[ROCm/clr commit: 208f5a41c6
2018-05-31 15:27:12 +01:00
2924783ee4
Switch to using ROCDL directly, as opposed to via HC. Add missing bits.
...
[ROCm/clr commit: 14e6a04387
2018-05-31 03:17:26 +01:00
bd44b8c22a
Fix __syncthreads for hip-clang
...
[ROCm/clr commit: 0e0b028846
2018-05-30 16:33:18 -04:00
999df03229
Add more function declarations for hip-clang
...
[ROCm/clr commit: 27fe2283d6
2018-05-30 06:51:03 -04:00
92757a7033
Merge pull request #469 from yxsamliu/fun
...
Add math function declaration for hip-clang
[ROCm/clr commit: e5b8358b05
2018-05-30 15:01:27 +05:30
b04a576c12
Add signature for hipFuncGetAttributes to nvcc_details
...
Change-Id: I06d5b50028fcfe3ea9d81f749b02be6b094c0977
[ROCm/clr commit: a923cac5d5
2018-05-30 11:24:48 +05:30
a4aa29f24b
Fix hipBindTexture on NVCC path
...
[ROCm/clr commit: d74353b112
2018-05-30 10:59:07 +05:30
c510903a32
Merge pull request #471 from Jorghi12/patch-4
...
Adding double/long int signatures for abs
[ROCm/clr commit: fde0613554
2018-05-30 08:31:32 +05:30
377b028628
GCC only supports power-of-2 component counts, so round up to next square.
...
[ROCm/clr commit: c059619e2c
2018-05-29 01:22:26 +01:00
f40b06757c
Add 1d texture types for NVCC path
...
[ROCm/clr commit: 3699e74857
2018-05-28 15:02:06 +05:30
8a2d7a25b1
Update math_functions.h
...
CUDA also has a function named labs.
[ROCm/clr commit: b33876ee4d
2018-05-26 16:22:10 -04:00
0a2d514cf8
Add missing alias half / half2 aliases
...
[ROCm/clr commit: 0f4a135e5f
2018-05-26 12:10:50 +01:00
43d55a79e4
Adding double/long int signatures for abs
...
Adding overloads for abs that are found in cuda's math_functions.
[ROCm/clr commit: b127488ea9
2018-05-26 00:41:24 -04:00
c7d481a812
Add more function declarations to hip-clang
...
[ROCm/clr commit: 9c43970a27
2018-05-25 16:18:46 -04:00
0dfdaf1808
Missing bits.
...
[ROCm/clr commit: 25ac508214
2018-05-25 20:12:21 +01:00
494f7f6a38
Add math function declaration for hip-clang
...
[ROCm/clr commit: f53e86e1b8
2018-05-25 12:22:16 -04:00
2b16b4e31b
Missing commit.
...
[ROCm/clr commit: ecefdd6541
2018-05-23 17:57:47 +01:00
e72516bd2f
Merge pull request #445 from ROCm-Developer-Tools/feature_func_attributes
...
Add support for the hipFuncGetAttributes interface.
[ROCm/clr commit: 85342d73b5
2018-05-22 09:37:41 +05:30
96e4122845
Add hipMalloc3D to nvcc detail
...
Change-Id: I8a5654066ed1504e3b05eddbbdebf05fd52aa149
[ROCm/clr commit: 5133299804
2018-05-21 11:33:09 +05:30
cc87c11be1
defined hipPitchedPtr
...
Added a define for hipPitchedPtr to resolve a compiler error
[ROCm/clr commit: 5c5d87b0a3
2018-05-18 09:11:50 -04:00
106f3f591e
Merge pull request #433 from gargrahul/add_hipmemset3d
...
Added hipMemset3D
[ROCm/clr commit: 1c93e11cdf
2018-05-18 14:54:15 +05:30
e3823fd9e8
Merge pull request #440 from yxsamliu/assert2
...
Add __assert_fail, __device_trap and hipErrorAssert for clang
[ROCm/clr commit: 43f7705278
2018-05-18 14:13:27 +05:30
3dd565a6d7
Switch over to using native vector types, for better codegen. Remove noise.
...
[ROCm/clr commit: fcce9f5337
2018-05-12 00:14:17 +01:00
4e6996282d
Add support for the hipFuncGetAttributes interface.
...
[ROCm/clr commit: bf9529aaa8
2018-05-11 03:35:10 +01:00
7d7125195c
Fix warning about inlined function is not defined
...
[ROCm/clr commit: 4c2d2b9e69
2018-05-08 16:38:50 -04:00
cd2b1f1baf
Add __assert_fail, __device_trap and hipErrorAssert for clang
...
[ROCm/clr commit: eeedfc5567
2018-05-08 15:42:27 -04:00
16d8a56909
add intrinsics mbcnt_lo, mbcnt_hi, lane_id
...
[ROCm/clr commit: 4f12ddea8f
2018-05-08 13:43:53 -04:00
a7be5be448
Added hipMemset3D
...
[ROCm/clr commit: 78568435da
2018-05-07 10:24:30 +05:30
db7e579529
Cleaned up
...
[ROCm/clr commit: 976b9ed200
2018-05-02 12:54:17 +05:30
6f05c37cde
Fix texture 3D for HIP/NVCC
...
[ROCm/clr commit: 1d76c48e3d
2018-05-02 11:56:37 +05:30
d6f360e65d
Merge pull request #419 from gargrahul/fix_hip_gcc_host_build
...
Fix gcc build issues for host code
[ROCm/clr commit: 2e6b46f308
2018-05-01 12:42:50 +05:30
9ac65ca55c
Merge pull request #415 from deven-amd/master
...
Checkin to fix bugs in math functions.
[ROCm/clr commit: af0c227df4
2018-05-01 12:29:03 +05:30
1bc33c9356
Merge pull request #410 from gargrahul/hipCtx_deprecation_warning
...
Mark hipCtx APIs as deprecated
[ROCm/clr commit: f8431772a8
2018-05-01 12:23:22 +05:30
60a66aa10d
Merge pull request #406 from gargrahul/fix_tex1D_coordinate_type
...
Resolve tex1D float to int conversion issue
[ROCm/clr commit: 885e51e1a5
2018-05-01 09:18:11 +05:30
f932a51c8e
Fix gcc build issues for host code
...
[ROCm/clr commit: 720e209366
2018-04-25 12:30:11 +05:30
13ab5368d0
Checkin to fix bugs in math functions.
...
This change fixes the following bugs that were discovered while debuggnig TF unit test failures (cwise_ops_test)
1. __hisinf and __hisnan routines
Both had incorrect implementations.
2. abs
A "long long" (64bit int) version was missing, resulting in the 32bit version being used for 64bit ints (which resulted in incorrect results, when the value passed in was outside the 32bit int range)
3. lgamma
We seemed to have a custom version for the 'double' datatype (which was giving incorrect results). Replaced it with a call to the 'double' version of the underlying 'hc::precision_math::lgamma'
[ROCm/clr commit: 65a90c55e7
2018-04-24 18:10:07 +00:00
3111ff5e40
Moved deprecated macro to the end
...
[ROCm/clr commit: bc61260168
2018-04-20 21:57:23 +05:30
bc59051858
Mark hipCtx APIs as deprecated
...
[ROCm/clr commit: 89674c54a7
2018-04-20 17:34:48 +05:30
65b792cb9b
Resolve tex1D float to int conversion issue
...
[ROCm/clr commit: a898ccfb3f
2018-04-18 14:15:18 +05:30
39a19cc9e8
Added hipMemset2DAsync support
...
[ROCm/clr commit: fcc0866681
2018-04-17 18:27:27 +05:30
d8256b0652
Merge pull request #395 from lmoriche/master
...
Add missing __device__ function attributes.
[ROCm/clr commit: 2ab05a2069
2018-04-12 14:51:32 +05:30
957651a021
Add missing __device__ function attributes.
...
[ROCm/clr commit: 8dcfbf5bee
2018-04-11 09:29:37 -07:00
Alex Voicu
Rahul Garg
Maneesh Gupta
Yaxun (Sam) Liu
Jorghi12
founta
Siu Chi Chan
Deven Desai
Laurent Morichetti