Commit Graph

780 Commits

Author SHA1 Message Date
Aaron Enye Shi 1282ef2e0c Fix explicit cast required
This is required for PyTorch which runs into an issue with narrowing types. Requires an explicit cast.

[ROCm/clr commit: 032ebd41ea]
2018-07-20 16:34:15 -04:00
Maneesh Gupta 746e56a61b Merge pull request #543 from ROCm-Developer-Tools/add-std-complex
Support std::complex for hip-clang

[ROCm/clr commit: 5632faf928]
2018-07-19 07:14:45 +05:30
Yaxun (Sam) Liu dc1257a16c Support std::complex for hip-clang
[ROCm/clr commit: a8dc1257df]
2018-07-18 00:08:04 -04:00
Alex Voicu 8b6a48d0d1 Update hip_vector_types.h
[ROCm/clr commit: 252776d4aa]
2018-07-17 17:28:48 +01:00
Alex Voicu 766211b400 Not yet.
[ROCm/clr commit: 14281c0a40]
2018-07-17 17:21:10 +01:00
Alex Voicu 2c568a5541 Constrain variadic constructor to ranks > 1 and add missing __device__s
[ROCm/clr commit: 1734eaead8]
2018-07-17 17:17:48 +01:00
Maneesh Gupta c96a617b78 Merge branch 'master' into move-memcpy
[ROCm/clr commit: 1fc3ef3cd4]
2018-07-17 10:51:42 +05:30
Maneesh Gupta 112c945016 Merge branch 'master' into feature_native_vector_types
Change-Id: I943b94ee230871a960d8f9c321ad5e096183b415


[ROCm/clr commit: 9e142ea66a]
2018-07-17 09:51:22 +05:30
Maneesh Gupta 8eeb225c04 Merge pull request #515 from ROCm-Developer-Tools/hipclang-add-amdgcn-funcs
Add hipclang amdgcn functions

[ROCm/clr commit: 7d8c1ae761]
2018-07-17 09:25:09 +05:30
Alex Voicu 438a2c1fb9 Merge branch 'master' of https://github.com/ROCm-Developer-Tools/HIP into feature_native_vector_types
[ROCm/clr commit: a2b6508912]
2018-07-15 11:44:48 +01:00
Evgeny Mankov 3a45d7016c [HIPIFY] Support of cudaComputeMode / CUcomputemode
+ update docs
+ fix typo in hip_runtime_api.h


[ROCm/clr commit: 4e0d07f07a]
2018-07-14 16:08:19 +03:00
Maneesh Gupta 6e7579dbd7 Merge pull request #563 from gargrahul/add_gettexrefoffset_nvcc
Add hipGetTextureAlignmentOffset and hipGetChanDesc on NVCC path

[ROCm/clr commit: 56984d04b0]
2018-07-11 15:01:20 +05:30
Maneesh Gupta c25b0f61e9 Merge branch 'master' into add_bindtextoarrayref_overload_api
[ROCm/clr commit: afe1396b35]
2018-07-11 12:55:49 +05:30
Maneesh Gupta 8001ecea35 Merge pull request #549 from gargrahul/correct_bindtex_syntax_nvcc
Corrected enum type for TextureReadMode 

[ROCm/clr commit: 8820eee3a2]
2018-07-11 12:52:50 +05:30
Rahul Garg f25e75ee6b Added hipGetChanDesc for NVCC path
[ROCm/clr commit: bab48b86a6]
2018-07-11 12:37:07 +05:30
Rahul Garg 4fdb1a8ee5 Add hipGetTextureAlignmentOffset on NVCC path
[ROCm/clr commit: 5b53e278a5]
2018-07-11 12:17:33 +05:30
Yaxun (Sam) Liu 6666bf623d Include stddef.h in device_functions.h and remove use of uint8_t
[ROCm/clr commit: 2531815dfe]
2018-07-10 18:12:41 -04:00
Yaxun (Sam) Liu 007a005cf3 Move __hip_hc_memcpy and __hip_hc_memset from device_utils.cpp to device_functions.h as inline functions
[ROCm/clr commit: 903dfa94eb]
2018-07-10 18:12:41 -04:00
Aaron Enye Shi 0e111de9f4 Fix min/max, icmp asm and add comment for conversion functions
[ROCm/clr commit: 9f3404cfea]
2018-07-10 18:56:48 +00:00
Aaron Enye Shi 8d0936bbe4 Add func attributes to match ocml and ockl
[ROCm/clr commit: bc8bc8cac4]
2018-07-10 18:27:32 +00:00
Aaron Enye Shi 327b0fb0e4 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/clr commit: f8746ecc64]
2018-07-10 17:56:57 +00:00
Yaxun (Sam) Liu d1dbc3dbc5 Add workaround __local_to_generic
[ROCm/clr commit: 9aaa792286]
2018-07-09 17:56:33 +00:00
Yaxun (Sam) Liu 2d5e831c7f Add declare of __get_dynamicgroupbaseptr for host compilation
[ROCm/clr commit: 1f34993ac1]
2018-07-09 17:56:25 +00:00
Aaron Enye Shi 0e5ef747e6 Enable surface functions on HIP clang path
Fix surface test on HIP clang path.


[ROCm/clr commit: 5862ae4b6a]
2018-07-05 20:49:47 +00:00
Aaron Enye Shi eab9cacef2 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/clr commit: d39508b4fd]
2018-07-05 20:38:46 +00:00
Aaron Enye Shi 94cc27d4d6 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/clr commit: b362c53475]
2018-07-05 20:15:41 +00:00
Rahul Garg 1751bc7c26 Added another variant of bindtextoarray for direct porting
[ROCm/clr commit: 017f668a72]
2018-07-05 23:41:31 +05:30
Rahul Garg 61b632944b Corrected enum type
[ROCm/clr commit: b15b08d0db]
2018-07-05 23:11:39 +05:30
Rahul Garg c548c3d833 Added tex2dlayered mapping for HIP/NVCC
[ROCm/clr commit: f080abe5a5]
2018-07-05 23:00:41 +05:30
Aaron Enye Shi f78a8d7fd6 Implement Memory Fence Functions in header
Enabled __llvm_fence_* functions for seq_cst.


[ROCm/clr commit: 96ab7c7b25]
2018-07-04 23:35:24 +00:00
Aaron Enye Shi 303a8be7d3 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/clr commit: ce9ca744dc]
2018-07-04 20:52:55 +00:00
Aaron Enye Shi 4b941a0db8 Merge branch 'master' into hipclang-add-amdgcn-funcs
[ROCm/clr commit: f8109c5b06]
2018-07-04 17:36:08 +00:00
Maneesh Gupta b1b4302bc4 Merge pull request #530 from ROCm-Developer-Tools/hipclang-compute-mode
Add HIP Compute Mode

[ROCm/clr commit: e54ffe52a0]
2018-07-04 10:55:48 +05:30
Alex Voicu 7f4b5810bf Add scalar operands and fix C implementation.
[ROCm/clr commit: f931980a6a]
2018-06-29 05:23:49 +01:00
Alex Voicu 32b5fc5b83 Re-sync with upstream.
[ROCm/clr commit: c061d885f6]
2018-06-29 00:16:34 +01:00
Maneesh Gupta aaee068ac0 Merge pull request #503 from ROCm-Developer-Tools/fix-complex
Fix hip_complex.h

[ROCm/clr commit: 1340adce1a]
2018-06-28 12:15:22 +05:30
Maneesh Gupta 3ed562a12b Merge pull request #495 from ROCm-Developer-Tools/hip-clang-upstream
add more device functions for hip-clang

[ROCm/clr commit: b6f5c61a2c]
2018-06-28 11:42:01 +05:30
Maneesh Gupta 90a6b96c5f Fix typo
[ROCm/clr commit: f426491f17]
2018-06-28 11:19:22 +05:30
Maneesh Gupta e33d1d0dda Merge pull request #536 from ROCm-Developer-Tools/hotfix_remove_unimplemented_math
Removes use of unimplemented OCML functionality.

[ROCm/clr commit: 06c1c93945]
2018-06-28 11:15:57 +05:30
Aaron Enye Shi c60682797a Implement __shfl_* funcs into HIP headers
[ROCm/clr commit: e44eaa1a1e]
2018-06-26 18:32:11 +00:00
Alex Voicu 18ce8e6251 Existence is a complex affair.
[ROCm/clr commit: 2acf7f033c]
2018-06-26 00:41:35 +01:00
Alex Voicu e0c1000e9c Be nice to GCC, it is old and worthy of respect.
[ROCm/clr commit: b9ead38478]
2018-06-25 22:59:07 +01:00
Yaxun (Sam) Liu 9d35209f8f Include host_defines.h in hip_fp16.h since it uses __host__ __device__ attributes
[ROCm/clr commit: c74682ef1b]
2018-06-25 15:34:34 -04:00
Alex Voicu 4fcfb64a88 Removes use of unimplemented OCML functionality.
[ROCm/clr commit: 9452298fbe]
2018-06-25 19:16:27 +01:00
Alex Voicu 89d6db12fe Let's try this again...
[ROCm/clr commit: 3ff1d4081e]
2018-06-25 17:49:50 +01:00
Aaron Enye Shi 4ad4f6a489 Add HIP Compute Mode
[ROCm/clr commit: c73cd64f5b]
2018-06-22 14:37:19 +00:00
Alex Voicu 8b9ac66103 Merge branch 'master' of https://github.com/ROCm-Developer-Tools/HIP into feature_native_vector_types
[ROCm/clr commit: 0e0d43bfe1]
2018-06-22 12:19:32 +01:00
Aaron Enye Shi 3f19f0979b Replace __hip_hc_ir_ inline asm with __ockl_* functions
[ROCm/clr commit: 5d4986d470]
2018-06-20 20:40:14 +00:00
Aaron Enye Shi 7cb3cec513 Implement __ballot, __any, __all into HIP headers
[ROCm/clr commit: 9a295a732d]
2018-06-20 17:39:39 +00:00
Aaron Enye Shi d09a2f9237 Implement hip_hc.ll into HIP headers
Move all __hip_hc_ir_* functions from hip_hc.ll into HIP header as inline asm. Remove hip_hc.ll and build dependencies from HIP.


[ROCm/clr commit: 175c87f2bf]
2018-06-20 17:39:31 +00:00