1ee794c7f3
P4 to Git Change 2031768 by cpaquot@cpaquot-ocl-lc-lnx on 2019/11/16 17:21:58
...
SWDEV-212440 - [HIP] Memory access fault observed on Pytorch while running performance tests with Microbenchmarking script
We need to loop through all the default stream to sync them in case
the app call hipFree on a different current stream and another current stream
is using the memory.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#85 edit
2019-11-16 17:25:59 -05:00
4d0a234183
[ci] Fix failures on nvcc path ( #1654 )
...
Change-Id: I97f6b51314e14965618ab510bf57a2b034183e78
2019-11-15 07:58:40 +05:30
5a03482690
[packaging] Conform to debian package naming rules ( #1647 )
...
* [packaging] Conform to debian package naming rules
* Add hip_hcc metapackage for backward compatibility
2019-11-14 11:40:34 +05:30
8d29a5192d
P4 to Git Change 2029899 by kjayapra@0_HIPWS_LNX1_ROCM on 2019/11/13 12:29:19
...
SWDEV-144570 - Implementing ihipUnbindTextureImpl API.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#35 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.map.in#33 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_texture.cpp#27 edit
2019-11-13 12:34:13 -05:00
574ec4298b
Merge 'master' into 'amd-master'
...
Change-Id: I263f0f3c3e288df178a5313d2c142574a098dfa2
2019-11-13 05:10:40 -05:00
6efa1d4655
P4 to Git Change 2029039 by kjayapra@0_HIPWS_LNX1_ROCM on 2019/11/12 12:52:46
...
SWDEV-210844 - Implementing hipExtGetLinkTypeAndHopCount
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_device_runtime.cpp#22 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#344 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.hpp#47 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#141 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#44 edit
2019-11-12 12:55:17 -05:00
cf79be0f64
Merge pull request #1651 from emankov/hipify
...
[HIPIFY][doc] A few words about LLVM's FileCheck in README.md
2019-11-12 16:06:25 +03:00
bb6f1648ec
[HIPIFY][doc] A few words about LLVM's FileCheck in README.md
2019-11-12 16:04:28 +03:00
316b792b09
Merge 'master' into 'amd-master'
...
Change-Id: I3e1ce565f4faaeb5b53d6f8c7a4023c498fbf621
2019-11-11 05:10:19 -05:00
0a2682625f
Bump version to 3.0 ( #1648 )
...
Change-Id: Id1e6eb222080a339af938212e74a52c39c305eab
2019-11-10 20:23:58 -08:00
4c73950e31
Merge pull request #1643 from mhbliao/hliao/master/remove.redundant.decl
...
Remove redundant declarations.
2019-11-07 11:16:40 -08:00
d28ad401c9
Remove redundant declarations.
...
- The revised `hip/hip_ext.h` have that declarations.
2019-11-07 10:11:22 -05:00
c96534901d
Merge 'master' into 'amd-master'
...
Change-Id: Ia44cb648c8ae16ff424bcf64fb9a9338c7678c0c
2019-11-07 05:22:21 -05:00
806ae1f43e
Merge pull request #1642 from emankov/hipify
...
[HIPIFY] Clang style formatting
2019-11-07 12:49:52 +03:00
7f143585f6
[HIPIFY] Clang style formatting
2019-11-07 12:48:05 +03:00
367261ea0e
Merge pull request #1641 from emankov/hipify
...
[HIPIFY][fix] Abandon canCompileHostAndDeviceInOneJob check for --cuda-compile-host-device option
2019-11-07 11:35:15 +03:00
4d2424ad36
[HIPIFY][fix] Delete canCompileHostAndDeviceInOneJob declaration as well
2019-11-07 11:32:53 +03:00
21c3d25403
[HIPIFY][fix] Abandon canCompileHostAndDeviceInOneJob check for --cuda-compile-host-device option
...
[Reason] It turned out that it is not so: 2 jobs are always
2019-11-07 11:30:40 +03:00
6f4ada4196
Merge pull request #1640 from emankov/hipify
...
[HIPIFY] Reorder options propagation to clang
2019-11-07 11:23:38 +03:00
f18e63b8a9
[HIPIFY] Reorder options propagation to clang
...
[Reason] Leave overriding opportunity for -D, -std=c++XX, etc.
2019-11-07 11:21:06 +03:00
854147e911
General hipMemset improvements ( #1495 )
...
* hipMemset et al can use HSA API directly for synchronous cases
* lock and flush stream in hipMemset, hold lock until complete
* move hipMemset async check to front of conditional
* use hsa_amd_memory_fill for additional sync memset cases
code cleanup/review for all memset calls
* Fix inversion of execution mutating value.
* ihipMemsetSync fall back to kernel if HSA memset fails
* Never fallback, never surrender.
* Allow NULL stream.
* Optimise memset kernel. Remove deadwood.
* Update hip_memory.cpp
* Clean up stream logic in sync memset
* Revert "Clean up stream logic in sync memset"
This reverts commit 6117dedf673367f44cc704192573a117a3d92477.
2019-11-07 13:19:54 +05:30
d38cc8efba
Remove native vector support from the GCC case, since it never worked ( #1637 )
2019-11-07 13:19:14 +05:30
97fca3439d
[HIP] Fixed hipStreamAddCallback [SWDEV#165185] ( #1425 )
...
Fixed hipStreamAddCallback() as requested in SWDEV#165185
Added unit test to test the behavior
2019-11-07 13:18:12 +05:30
dc8f556460
Fixed texture 2D mapping for pitched arrays & 3D Texture read ( #1415 )
...
Texture 2D image mapping for pitched arrays:
github issue: Texture Object's Buffer seems to be Misaligned #886
JIRA ticket: SWDEV-199313
SWDEV-151670 : Fixed issue with 3D texture with 4 components
SWDEV-151671 : Issue with 2D layered texture with 4 components
2019-11-07 13:17:46 +05:30
dfee3ae279
Rename hip/hip_hcc.h to hip/hip_ext.h ( #1341 )
...
* Rename hip/hip_hcc.h to hip/hip_ext.h
* Deprecate hip_hcc.h
2019-11-07 13:17:10 +05:30
39587efda6
P4 to Git Change 2026174 by eshcherb@evgeny-hip on 2019/11/06 19:37:26
...
SWDEV-197289 - tracing callbacks call fix
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_prof_api.h#5 edit
2019-11-06 19:47:10 -05:00
6d0c96993d
P4 to Git Change 2026152 by cpaquot@cpaquot-ocl-lc-lnx on 2019/11/06 17:50:08
...
SWDEV-206239 - [HIP] RCCL: finegrain VRAM does not work
Implemented fine grained VRAM allocation via ATOMICS.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#84 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#138 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#43 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmemory.cpp#45 edit
2019-11-06 18:00:44 -05:00
d8e9a09064
Merge pull request #1630 from ROCm-Developer-Tools/hotfix_volatile_accessors
...
__half2 should walk like CUDA and talk like CUDA
2019-11-06 10:25:03 -08:00
5af7ca4445
Merge pull request #1620 from jeffdaily/hipEventRecord_patch
...
fix race condition in hipEventRecord
2019-11-06 10:23:24 -08:00
c0f9c854a6
Merge pull request #1636 from emankov/hipify-clang
...
[HIPIFY][Win][fix] canCompileHostAndDeviceInOneJob is true only for LLVM >= 10
2019-11-06 19:27:40 +03:00
97a0670d29
[HIPIFY][Win][fix] canCompileHostAndDeviceInOneJob is true only for LLVM >= 10
2019-11-06 19:25:42 +03:00
d9016ab9b7
Merge pull request #1635 from emankov/hipify-clang
...
[HIPIFY][doc] Update README.md
2019-11-06 19:20:02 +03:00
3f737625ee
[HIPIFY][doc] Update README.md
...
+ Supported versions, testing
2019-11-06 19:18:13 +03:00
b2af1820f5
hipEventRecord only needs one lock; remove locked_eventIsReady
2019-11-06 15:56:32 +00:00
444122dfb9
remove ihipEvent_t::refreshEventStatus(), new hipEventElapsedTime
...
This fixes possible races in hipEventElapsedTime.
2019-11-06 15:46:46 +00:00
2777b24abb
Merge pull request #1634 from emankov/hipify-clang
...
[HIPIFY][format] Apply clang style formatting in LLVMCompat
2019-11-06 14:46:45 +03:00
4cad5fab78
[HIPIFY][format] Apply clang style formatting in LLVMCompat
2019-11-06 14:43:22 +03:00
d0c5ab0e00
Merge pull request #1633 from emankov/hipify-clang
...
[HIPIFY] Add a compatibility for CharSourceRange if LLVM < 7
2019-11-06 14:32:47 +03:00
3dcae8760d
[HIPIFY] Add a compatibility for CharSourceRange if LLVM < 7
2019-11-06 14:30:39 +03:00
1df423165b
Remove leftover noise.
2019-11-06 02:46:21 +02:00
55fd1363e2
__half2 should walk like CUDA and talk like CUDA
2019-11-06 02:43:04 +02:00
6833886449
Merge pull request #1626 from mhbliao/hliao/master/fix.windows.build
...
Use portable macro for deprecation message.
2019-11-05 10:08:43 -08:00
7b1ca3dd91
Merge pull request #1595 from emankov/cmake
...
[HIP][cmake] Simplify `UNIX` related code (the beginning)
2019-11-05 20:30:55 +03:00
7ca43b98d1
Use portable macro for deprecation message.
2019-11-05 11:51:00 -05:00
cf5d1caab3
Merge pull request #1624 from emankov/hipify
...
[HIPIFY][#1409 ] Fix for kernel launch macro expansion
2019-11-05 14:04:51 +03:00
a1c380f38d
[HIPIFY][ #1409 ] Fix for kernel launch macro expansion
...
+ Add a corresponding test kernel_launch_01.cu
+ Add isBefore() check to avoid crash on Replacement with negative length
TODO:
+ Compatibility with former LLVM versions
+ More complicated kernel launch tests
2019-11-05 14:00:13 +03:00
8b3fce8069
Deprecate HIP Markers ( #1622 )
...
* Deprecate HIP markers
* Deprecate profiler start/stop
2019-11-05 12:32:59 +05:30
7a3d592ed0
[hip][tests] Add two more workgroup sizes for testing hipLaunchCooperativeKernel ( #1613 )
2019-11-05 12:32:46 +05:30
cbbe0e78cf
P4 to Git Change 2024544 by cpaquot@cpaquot-ocl-lc-lnx on 2019/11/04 16:35:14
...
SWDEV-145570 - [HIP] Fetched properties from current device and not default 0 one.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_device_runtime.cpp#21 edit
2019-11-04 16:42:31 -05:00
bd34138596
P4 to Git Change 2024543 by cpaquot@cpaquot-ocl-lc-lnx on 2019/11/04 16:32:23
...
SWDEV-206239 - [HIP] Return hipErrorMemoryAllocation for fine grained VRAM for now
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#83 edit
2019-11-04 16:39:35 -05:00
foreman
Maneesh Gupta
Jenkins
Evgeny Mankov
Rahul Garg
Michael LIAO
Jeff Daily
Alex Voicu
Sarbojit2019
ansurya
Aryan Salmanpour