17e35e382c
Added support for pinned 2D SDMA copy
...
[ROCm/clr commit: 5eb11b58f3
2018-07-31 14:05:35 +05:30
9512246b5c
Protect macro TYPEDEF_VECTOR_VALUE_TYPE and prefix __hip_ for vector_value_types
...
[ROCm/clr commit: e42af991d1
2018-07-30 20:41:24 +00:00
ad6066eebd
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/clr commit: ff1e49d191
2018-07-30 19:43:12 +00:00
57c15b815d
Remove device mapping from shareWithAll memory
...
When shareWithAll memory (e.g., host memory) is allocated, set appId
in hc::AmPointerInfo to -1 to indicate that this memory is not mapped
to any device. Peer checking in ihipStream_t::canSeeMemory is not
necessary if memory is shared with all devices. Thus, it is skipped.
Note that earlier host memory is always mapped to device 0 and HIP
always performs peer checking for all kinds of hipMemcpy. Since the
peer checking process requires context locking, hipMemcpy from/to host
memory always grabs device 0's context lock. Therefore, if there is
another thread holding the context lock of device 0 (e.g.,
hipDeviceSynchronize on device 0), hipMemcpy will have to wait for the
lock until it can actually perform memcpy. This can significantly
deteriorate execution performance.
Signed-off-by: Sarunya Pumma <sarunya.pumma@amd.com >
[ROCm/clr commit: 84aadb9274
2018-07-28 23:15:16 -07:00
351f385ea9
[HIPIFY][BLAS] cuBLAS support revise
...
+ doc for CUBLAS API supported by HIP
+ 3 tests: 0 & 1 based indexing, sgemm matmul
[ROCm/clr commit: e00cfefd4f
2018-07-28 21:00:44 +03:00
d539617d9b
Do not use std::complex in test hipDeviceMalloc
...
[ROCm/clr commit: fc1ed91ddb
2018-07-27 17:07:00 -04:00
ea77788a5f
Fix thread index calculation in __hip_malloc
...
[ROCm/clr commit: f08f26942c
2018-07-27 16:24:51 -04:00
9a4744566f
Support malloc/free for hip-clang
...
[ROCm/clr commit: d17dc21c9c
2018-07-27 16:24:51 -04:00
beb7f13068
Drop use of NVCC_COMPAT since it is always set
...
[ROCm/clr commit: 8d80abb30f
2018-07-27 09:29:29 -07:00
56ebef737b
Merge pull request #596 from mangupta/patch5-fixhiplaunchparm
...
[tests] Fixed hipLaunchParm test on nvcc path
[ROCm/clr commit: 17f612def1
2018-07-26 06:17:50 +05:30
951d920990
Merge pull request #595 from mangupta/patch4-fixhipStdComplex
...
[tests] Fixed build & disabled run of hipStdComplex on nvcc path
[ROCm/clr commit: 86188f04c8
2018-07-26 06:17:04 +05:30
e58d3b16b4
Merge pull request #592 from mangupta/patch1-initialize-cudapath
...
[hipcc] Initialize CUDA_PATH before printing
[ROCm/clr commit: ec6995485c
2018-07-26 06:16:31 +05:30
4431a7ad9f
Merge pull request #590 from ROCm-Developer-Tools/fix-ballot-anyall-return
...
Fix return EXIT_FAILURE if ballot or anyall fails
[ROCm/clr commit: ddcb85a77b
2018-07-26 06:16:23 +05:30
65a1df76dc
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/clr commit: bf6fdca1e8
2018-07-25 19:52:19 +00:00
24b33ee4ef
[tests] Fixed hipLaunchParm test on nvcc path
...
- Uses c++11 features. Added it to nvcc options
- Arguments for some kernels exceeded 4096 bytes which is the limit
imposed by nvcc. Reduced BLOCK_DIM_SIZE to 512 to handle this
- Fixed compilation issues on nvcc path
Change-Id: I14f6b28afcb7c6b24a085fd707b2104e2ed64627
[ROCm/clr commit: 51307ac705
2018-07-25 15:41:04 +05:30
7033561423
[tests] Fixed build & disabled run of hipStdComplex on nvcc path
...
Change-Id: I885e96902faa065f6559183f33b6a7aa1f37e393
[ROCm/clr commit: 843020bfe6
2018-07-25 15:39:45 +05:30
34bd297a65
[FindHIP] Remove rdynamic from CMAKE_SHARED_LIBRARY_LINK_HIP_FLAGS
...
CMAKE_SHARED_LIBRARY_LINK_CXX_FLAGS contains -rdynamic which does not go
well with nvcc
Change-Id: I10e6d21553de0e43c6a7ae5fb3ec2418087959ff
[ROCm/clr commit: 7a5987e289
2018-07-25 15:32:14 +05:30
4403d1dc04
[hipcc] Initialize CUDA_PATH before printing
...
Change-Id: Ifd448b628fa4a7055f4a439ed2c817c5f11f9bef
[ROCm/clr commit: 2027675b3f
2018-07-25 15:29:57 +05:30
13338a6873
P4 to Git Change 1584863 by yaxunl@yaxunl-lc10 on 2018/07/24 17:00:42
...
SWDEV-145570 - [HIP] - Update prototype of __hipRegisterVar
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#13 edit
[ROCm/clr commit: 5ac4970a68
2018-07-24 17:14:35 -04:00
6353687511
Fix return EXIT_FAILURE if ballot or anyall fails
...
[ROCm/clr commit: 8ab8a37812
2018-07-24 20:39:48 +00:00
792b9ebe7c
Null check before setting offset
...
[ROCm/clr commit: f17c897751
2018-07-24 12:25:40 +05:30
99758b3064
Set offset in hipGetTextureAlignmentOffset
...
[ROCm/clr commit: 0ec64e8b24
2018-07-24 10:11:26 +05:30
6d9f3c2cf1
Merge pull request #582 from ROCm-Developer-Tools/set-vdi-path
...
Let hipcc use proper include and lib path for HIP/VDI
[ROCm/clr commit: cab5e80b22
2018-07-24 07:32:54 +05:30
e4ceff98e9
Fix typo
...
[ROCm/clr commit: 25ba9f6f94
2018-07-24 07:30:05 +05:30
1f5303c7c3
Fixing a HIP bug that causes DEBUG symbols to be dropped.
...
I've noticed that in the generated makefiles, we'd have
if(NOT build_configuration)
set(build_configuration Debug)
endif()
However, only definitions with all capitals are provided.
e.g.
set(HIP_HCC_FLAGS_DEBUG -g)
This is because of lines 442 to 448. We capitalize the configuration ${config_upper}.
This PR fixes this bug.
[ROCm/clr commit: 25c780c89e
2018-07-23 18:16:00 -04:00
5f73275382
Fix warnings about unused command line option --hip-link
...
and set search path for libhip_hcc.so
[ROCm/clr commit: 555c2f9c08
2018-07-23 14:55:07 -04:00
b0ff0e4657
Document hip-clang fatbinary format and initialization function
...
This is for documenting the change in hip-clang: https://reviews.llvm.org/D49083
[ROCm/clr commit: 674b45902d
2018-07-23 09:28:40 -04:00
f45c9e9d61
Fix hipcc for -M for hip-clang
...
[ROCm/clr commit: 6053c71323
2018-07-23 09:26:42 -04:00
192601f691
Let hipcc use proper include and lib path for HIP/VDI
...
Add support of environment variable HIP_VDI_HOME.
By setting environment variable HIP_VDI_HOME to the distribution directory of HIP/VDI, hipcc
will choose proper include and lib path for hip-clang.
[ROCm/clr commit: 4627aa0085
2018-07-23 09:26:42 -04:00
57dabd81c3
Merge pull request #583 from emankov/cuDNN
...
[HIPIFY][DNN] cuDNN support revise
[ROCm/clr commit: a5afadec7e
2018-07-23 14:13:54 +03:00
2a2d857b6e
Merge pull request #568 from Srinivasuluch/sprint2_9tests
...
Adding a few more struct scenarions and a ResultValidation()
[ROCm/clr commit: e2615335c0
2018-07-23 14:44:06 +05:30
9c8a5d72b6
Merge pull request #558 from Jorghi12/patch-8
...
Fix the FindHIP.cmake script.
[ROCm/clr commit: 00c1542487
2018-07-23 11:48:42 +05:30
bc49031828
Fix hipHostRegister mem leak and copy kind in hipMemcpyAsync ( #587 )
...
[ROCm/clr commit: e8a5ad629f
2018-07-23 11:04:14 +05:30
e858a79470
Fix typo in FindHIP.cmake
...
[ROCm/clr commit: 4637e11a24
2018-07-23 10:39:45 +05:30
3e92bf10bb
Merge pull request #559 from rpathani/master
...
Adding functor unit tests
[ROCm/clr commit: 137d7f559f
2018-07-23 10:15:16 +05:30
9f23f7fc99
Merge pull request #584 from ROCm-Developer-Tools/fix-vector-explicit-cast
...
Fix explicit cast required
[ROCm/clr commit: 123af1a9c6
2018-07-23 10:13:05 +05:30
48335701ff
Merge pull request #581 from ROCm-Developer-Tools/fix-warning-math-decl
...
[HIPClang] Switch ordering on math functions
[ROCm/clr commit: 4d6a3f3eba
2018-07-23 10:12:01 +05:30
dd1b7d970c
Merge pull request #554 from gargrahul/cleanup_hipmemset3d_test
...
Cleaned up hipMemset3D test
[ROCm/clr commit: c53d0f967e
2018-07-23 10:11:49 +05:30
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
23fb9e51f2
[HIPIFY][DNN] cuDNN support revise
...
+ add doc for CUDDN API supported by HIP
+ cuDNN softMax test
[ROCm/clr commit: 618b1f5171
2018-07-20 21:00:24 +03:00
5e2395deeb
[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/clr commit: 98fca64220
2018-07-20 13:25:04 -04:00
77ab2afbbb
Update hipLaunchParm.cpp
...
Added struct bit field test, test number#21
[ROCm/clr commit: 996cfcd62a
2018-07-20 12:38:06 +05:30
a0bb8820c5
Merge pull request #578 from emankov/testing
...
[HIPIFY][test] Add static and dynamic shared memory tests
[ROCm/clr commit: 5184a69410
2018-07-19 20:48:25 +03:00
b4eefc9e5c
Merge branch 'master' of https://github.com/rpathani/HIP
...
# Conflicts:
# tests/src/kernel/hipLaunchParmFunctor.cpp
[ROCm/clr commit: 6ae4df95b5
2018-07-19 12:11:23 +05:30
e2f3e4d67e
Merge branch 'master' of https://github.com/rpathani/HIP
...
# Conflicts:
# tests/src/kernel/hipLaunchParmFunctor.cpp
[ROCm/clr commit: c2108cb82a
2018-07-19 12:02:42 +05:30
a05aa9d71f
Update FindHIP.cmake
...
Using Bool expression to handle empty case.
[ROCm/clr commit: 4abcf5dbc4
2018-07-18 23:02:11 -04:00
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
235def5d29
P4 to Git Change 1582432 by cpaquot@cpaquot-ocl-lc-lnx on 2018/07/18 17:16:05
...
SWDEV-145570 - [HIP] Fixed __hipRegisterVariable to __hipRegisterVar
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#7 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.map.in#7 edit
[ROCm/clr commit: c07510d622
2018-07-18 17:51:20 -04:00
aae1e3e73f
[HIPIFY][test] Add static and dynamic shared memory tests
...
[ROCm/clr commit: 7e4e821fbb
2018-07-18 18:33:44 +03:00
b425c11139
Fix test hipStdComplex.cpp to emit expected output for passing
...
[ROCm/clr commit: e20e93db65
2018-07-18 07:21:24 -04:00
Rahul Garg
Aaron Enye Shi
Sarunya Pumma
Evgeny Mankov
Yaxun Sam Liu
Brian Sumner
Maneesh Gupta
foreman
Jorghi12
Evgeny Mankov
saleelk
Srinivasuluch
rohit pathania