Mark Searles
ff4df50cd5
Add keyword 'explicit' to avoid ambiguity
...
Older gcc, e.g., 5.x, see an ambiguity in some calls. Example error
output as seen with gcc 5.5:
In file included from benchmark_wino.cpp:1:
In file included from ./miopen.hpp:13:
In file included from /usr/lib/gcc/x86_64-linux-gnu/5.5.0/../../../../include/c++/5.5.0/regex:38:
In file included from /usr/lib/gcc/x86_64-linux-gnu/5.5.0/../../../../include/c++/5.5.0/algorithm:62:
In file included from /usr/lib/gcc/x86_64-linux-gnu/5.5.0/../../../../include/c++/5.5.0/bits/stl_algo.h:66:
In file included from /usr/lib/gcc/x86_64-linux-gnu/5.5.0/../../../../include/c++/5.5.0/random:51:
/usr/lib/gcc/x86_64-linux-gnu/5.5.0/../../../../include/c++/5.5.0/bits/random.tcc:1324:27: error: call to 'abs' is ambiguous
const double __y = -std::abs(__n) * __param._M_sm - 1;
^~~~~~~~
/opt/rocm/hip/include/hip/hcc_detail/hip_complex.h:345:31: note: candidate function
__DEFINE_HIP_COMPLEX_REAL_FUN(abs, hipCabs)
[ROCm/hip commit: f0183dfaac ]
2018-08-31 10:36:05 -07:00
Maneesh Gupta
236c7002b4
Merge pull request #623 from ROCm-Developer-Tools/tf18
...
Fix declaration conflict when hip/math_functions.h is included first
[ROCm/hip commit: d8d45a906c ]
2018-08-10 09:36:27 +05:30
Maneesh Gupta
28739aa597
Merge pull request #613 from ROCm-Developer-Tools/hcc-compat
...
Add HCC compatibility mode for hip-clang
[ROCm/hip commit: 11cccaa5d0 ]
2018-08-10 09:32:31 +05:30
Yaxun Sam Liu
9c2a33b5c9
Fix __HIP_ARCH_* not defined after including math_functions.h
...
hcc_detail/math_functions.h used to include hcc_detail/hip_runtime.h.
Removing it has caused regression in TensorFlow 1.8.
Put it back for backward compatibiliity.
[ROCm/hip commit: b8a0c14171 ]
2018-08-08 08:55:28 -04:00
Yaxun Sam Liu
735e7ac097
Fix __HIP_DEVICE_COMPILE__ not defined when hip/math_functions.h is included
...
This fixes build failure in TensorFlow 1.8 for HCC
[ROCm/hip commit: 450ea49df0 ]
2018-08-07 17:14:39 -04:00
Yaxun Sam Liu
fd1b704358
Fix declaration conflict when hip/math_functions.h is included first
...
This fixes build failure in TensorFlow 1.8 for HCC
[ROCm/hip commit: dacb18414e ]
2018-08-07 15:44:59 -04:00
Wen-Heng (Jack) Chung
38b6402860
Revert "HIP program state re-initialization logic"
...
This reverts commit e5ca9eb081 .
[ROCm/hip commit: 2604f33930 ]
2018-08-03 17:03:04 -05:00
Maneesh Gupta
9ddcc6e6c6
Merge pull request #603 from b-sumner/master
...
Drop use of NVCC_COMPAT since it is always set
[ROCm/hip commit: 78ca12d583 ]
2018-08-03 08:45:18 +05:30
Maneesh Gupta
f0dfb51e39
Merge pull request #610 from aaronenyeshi/enable-hcc-module-launch
...
Enable hipHccModuleLaunchKernel on HIP-Clang
[ROCm/hip commit: d0cdc69676 ]
2018-08-02 08:08:46 +05:30
Yaxun Sam Liu
9e3a94dbf0
Add hc_get_workitem_absolute_id for hip-clang HCC compatibility mode
...
[ROCm/hip commit: ef1d1d57f7 ]
2018-08-01 15:14:38 -04:00
Yaxun Sam Liu
43437a3689
Add HCC compatibility mode
...
[ROCm/hip commit: 43bb490947 ]
2018-08-01 15:14:38 -04:00
Aaron Enye Shi
12414015fa
Enable hipHccModuleLaunchKernel on HIP-Clang
...
[ROCm/hip commit: d1149a9b64 ]
2018-07-31 20:38:20 +00:00
Aaron Enye Shi
415e5e0c7f
Protect macro TYPEDEF_VECTOR_VALUE_TYPE and prefix __hip_ for vector_value_types
...
[ROCm/hip commit: 99e6496898 ]
2018-07-30 20:41:24 +00:00
Aaron Enye Shi
390fedfb35
Ensure correct vector value type used in texture
...
Implement the vector value types in terms of SCALAR_TYPE##SIZE_vector_value_type for float, int, uint and sizes 2, 3, 4, 8, and 16.
[ROCm/hip commit: e0e785a45d ]
2018-07-30 19:43:12 +00:00
Brian Sumner
94afa0fc08
Drop use of NVCC_COMPAT since it is always set
...
[ROCm/hip commit: 71f96388ed ]
2018-07-27 09:29:29 -07:00
Aaron Enye Shi
c53a494be9
Enable HIP texture functions with HIP vector types
...
Include texture_functions.h on HIP-Clang path. Use HIP vector types instead of hc::short_vectors for texture_functions, and remove HCC headers.
[ROCm/hip commit: 8c50bcfdab ]
2018-07-25 19:52:19 +00:00
Maneesh Gupta
f87491f61c
Merge pull request #584 from ROCm-Developer-Tools/fix-vector-explicit-cast
...
Fix explicit cast required
[ROCm/hip commit: 06db076b10 ]
2018-07-23 10:13:05 +05:30
Aaron Enye Shi
89a00bc280
Fix explicit cast required
...
This is required for PyTorch which runs into an issue with narrowing types. Requires an explicit cast.
[ROCm/hip commit: 4ddb81424f ]
2018-07-20 16:34:15 -04:00
Aaron Enye Shi
3408e6440d
[HIPClang] Switch ordering on math functions
...
Header math_functions.h should be included after including __clang_cuda_math_forward_declares.h to avoid warning: attribute declaration must precede definition.
[ROCm/hip commit: ffefae77d7 ]
2018-07-20 13:25:04 -04:00
Maneesh Gupta
9648844caf
Merge pull request #543 from ROCm-Developer-Tools/add-std-complex
...
Support std::complex for hip-clang
[ROCm/hip commit: 12202dae90 ]
2018-07-19 07:14:45 +05:30
Yaxun (Sam) Liu
30e14f699b
Support std::complex for hip-clang
...
[ROCm/hip commit: 80438035c2 ]
2018-07-18 00:08:04 -04:00
Alex Voicu
4f0111994e
Update hip_vector_types.h
...
[ROCm/hip commit: 487bb6f2ef ]
2018-07-17 17:28:48 +01:00
Alex Voicu
45018c46cb
Not yet.
...
[ROCm/hip commit: 73d0091328 ]
2018-07-17 17:21:10 +01:00
Alex Voicu
6005a9d8d6
Constrain variadic constructor to ranks > 1 and add missing __device__s
...
[ROCm/hip commit: 3e2755adab ]
2018-07-17 17:17:48 +01:00
Maneesh Gupta
c86d32350a
Merge branch 'master' into move-memcpy
...
[ROCm/hip commit: b5cfa773ef ]
2018-07-17 10:51:42 +05:30
Maneesh Gupta
d24617e118
Merge branch 'master' into feature_native_vector_types
...
Change-Id: I943b94ee230871a960d8f9c321ad5e096183b415
[ROCm/hip commit: fb5199dee6 ]
2018-07-17 09:51:22 +05:30
Maneesh Gupta
d655537bc6
Merge pull request #515 from ROCm-Developer-Tools/hipclang-add-amdgcn-funcs
...
Add hipclang amdgcn functions
[ROCm/hip commit: fbbe2599dd ]
2018-07-17 09:25:09 +05:30
Alex Voicu
b47b4c771d
Merge branch 'master' of https://github.com/ROCm-Developer-Tools/HIP into feature_native_vector_types
...
[ROCm/hip commit: fe289dbebe ]
2018-07-15 11:44:48 +01:00
Evgeny Mankov
b063b3eb26
[HIPIFY] Support of cudaComputeMode / CUcomputemode
...
+ update docs
+ fix typo in hip_runtime_api.h
[ROCm/hip commit: 55e21055e3 ]
2018-07-14 16:08:19 +03:00
Maneesh Gupta
0d92d4d9e3
Merge pull request #563 from gargrahul/add_gettexrefoffset_nvcc
...
Add hipGetTextureAlignmentOffset and hipGetChanDesc on NVCC path
[ROCm/hip commit: 4aef1351fc ]
2018-07-11 15:01:20 +05:30
Maneesh Gupta
4a8751b6e7
Merge branch 'master' into add_bindtextoarrayref_overload_api
...
[ROCm/hip commit: 65e82f785d ]
2018-07-11 12:55:49 +05:30
Maneesh Gupta
12569a1aa9
Merge pull request #549 from gargrahul/correct_bindtex_syntax_nvcc
...
Corrected enum type for TextureReadMode
[ROCm/hip commit: 583b6b270a ]
2018-07-11 12:52:50 +05:30
Rahul Garg
00f7f9488b
Added hipGetChanDesc for NVCC path
...
[ROCm/hip commit: 6bf979d9e6 ]
2018-07-11 12:37:07 +05:30
Rahul Garg
54881b7a12
Add hipGetTextureAlignmentOffset on NVCC path
...
[ROCm/hip commit: c93f216fe8 ]
2018-07-11 12:17:33 +05:30
Yaxun (Sam) Liu
9de153ce91
Include stddef.h in device_functions.h and remove use of uint8_t
...
[ROCm/hip commit: e22c35e124 ]
2018-07-10 18:12:41 -04:00
Yaxun (Sam) Liu
244c9875e2
Move __hip_hc_memcpy and __hip_hc_memset from device_utils.cpp to device_functions.h as inline functions
...
[ROCm/hip commit: 8136a348ab ]
2018-07-10 18:12:41 -04:00
Aaron Enye Shi
2b89bb572a
Fix min/max, icmp asm and add comment for conversion functions
...
[ROCm/hip commit: 4a7ad93655 ]
2018-07-10 18:56:48 +00:00
Aaron Enye Shi
a13caad541
Add func attributes to match ocml and ockl
...
[ROCm/hip commit: 1f871736e5 ]
2018-07-10 18:27:32 +00:00
Aaron Enye Shi
e0f0c2c312
Remove activelanemask asm using ockl and llvm instrinsics
...
Replace implementation of __any and __all functions using OCKL functions and replaced __ballot implementation to use llvm intrinsic llvm.amdgcn.icmp.i32 instead of calls to __activelanemask_v4_b64_b1 which is not convergent.
[ROCm/hip commit: d1323f4f9b ]
2018-07-10 17:56:57 +00:00
Yaxun (Sam) Liu
c2350905f1
Add workaround __local_to_generic
...
[ROCm/hip commit: 5ec7973a6f ]
2018-07-09 17:56:33 +00:00
Yaxun (Sam) Liu
e97e1fd176
Add declare of __get_dynamicgroupbaseptr for host compilation
...
[ROCm/hip commit: e0390ddc9f ]
2018-07-09 17:56:25 +00:00
Aaron Enye Shi
0dde5be16b
Enable surface functions on HIP clang path
...
Fix surface test on HIP clang path.
[ROCm/hip commit: 22b60afa0d ]
2018-07-05 20:49:47 +00:00
Aaron Enye Shi
17453a6278
Implement hip_ldg Functions into HIP header
...
Move all the function definitions for hip_ldg.cpp into hip_ldg.h header and enable for HIP clang path.
[ROCm/hip commit: 76f86ef097 ]
2018-07-05 20:38:46 +00:00
Aaron Enye Shi
8e1a7fdd09
Implement min/max functions in HIP header
...
Remove using hc::precise_math min and max. Instead we can use ocml directly for device and std:: for host.
[ROCm/hip commit: 47d78e372e ]
2018-07-05 20:15:41 +00:00
Rahul Garg
9735598cd1
Added another variant of bindtextoarray for direct porting
...
[ROCm/hip commit: bd4816fc85 ]
2018-07-05 23:41:31 +05:30
Rahul Garg
399ec56015
Corrected enum type
...
[ROCm/hip commit: 4acf489bd5 ]
2018-07-05 23:11:39 +05:30
Rahul Garg
0d40be6866
Added tex2dlayered mapping for HIP/NVCC
...
[ROCm/hip commit: 19bae58e48 ]
2018-07-05 23:00:41 +05:30
Aaron Enye Shi
0ea959ba9d
Implement Memory Fence Functions in header
...
Enabled __llvm_fence_* functions for seq_cst.
[ROCm/hip commit: 930a16bccd ]
2018-07-04 23:35:24 +00:00
Aaron Enye Shi
6bcb3cca86
Workaround cast warning of smaller integer type for __to_local
...
For now, guard the __to_local function for device compile only since a local pointer should be same size as unsigned int on GPU compile. Also change to void* instead of char*.
[ROCm/hip commit: 07de5cb334 ]
2018-07-04 20:52:55 +00:00
Aaron Enye Shi
e8fac0449e
Merge branch 'master' into hipclang-add-amdgcn-funcs
...
[ROCm/hip commit: 2975f2a10a ]
2018-07-04 17:36:08 +00:00