AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
5,275 Commits 3 Branches 0 Tags
4e7b1cdbd2d360edd3252204e8b3393ebd35ab36
Commit Graph

5275 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
949f45ac 9210263727 Provide correct __mul64hi and __umul64hi builtins, using code from ROCm-Device-Libs 2018-05-14 08:34:56 +02:00
Alex Voicu fcce9f5337 Switch over to using native vector types, for better codegen. Remove noise. 2018-05-12 00:14:17 +01:00
foreman c6b9701abb P4 to Git Change 1553709 by cpaquot@cpaquot-ocl-lc-lnx on 2018/05/11 11:24:30 
SWDEV-145570 - [HIP] Make streamSet global and protect it
	By default from the spec, streamSet should be global and not per thread.
	There is a flag to make it per thread but we don't handle this yet. We
	would just add another variable that will be thread local and use it instead.

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_context.cpp#12 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_device_runtime.cpp#10 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_internal.hpp#11 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_stream.cpp#8 edit
 2018-05-11 11:34:01 -04:00
Aaron Enye Shi 4488f9f7a7 Fix hipMathFunction for gfx906 2018-05-11 10:53:07 -04:00
Alex Voicu eded014abc Don't use magic constants, they're evil. 
Also clarify that the register count cannot be queried at the moment.
 2018-05-11 11:31:46 +01:00
Alex Voicu bf9529aaa8 Add support for the hipFuncGetAttributes interface. 2018-05-11 03:35:10 +01:00
Siu Chi Chan 368affcea4 initial gfx906 support 2018-05-10 19:28:00 +00:00
Evgeny Mankov e3568a744d Merge pull request #443 from emankov/cuDNN 
[HIPIFY][DNN] support of cuDNN 7.1.3 - continuation
 2018-05-10 19:42:05 +03:00
Evgeny Mankov b31df9cc60 [HIPIFY][DNN] support of cuDNN 7.1.3 - continuation 
- not finished yet.
- based on https://github.com/ROCmSoftwarePlatform/hipDNN.
- testing on https://github.com/baidu-research/DeepBench.
 2018-05-10 17:36:51 +03:00
foreman 1426982a65 P4 to Git Change 1552807 by skudchad@skudchad_test2_win_opencl on 2018/05/09 18:08:07 
SWDEV-145570 - [HIP] - Release a stream first before taking it off from the set.
	- Queue::create() needs to be called before returning a valid queue.

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/14830/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_stream.cpp#7 edit
 2018-05-09 18:15:41 -04:00
Jenkins 3df55f552f Merge 'master' into 'amd-master' 
Change-Id: Ic79e86ea3cd8cd7f85d7f9f2a24e6acfad02d942
 2018-05-09 04:09:39 -05:00
foreman d4bb6a0a11 P4 to Git Change 1552209 by cpaquot@cpaquot-ocl-lc-lnx on 2018/05/08 20:24:32 
SWDEV-145570 - [HIP] Use the as_amd()->asHostQueue with streamSet

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_stream.cpp#6 edit
 2018-05-08 20:30:17 -04:00
foreman 813c770d73 P4 to Git Change 1552198 by skudchad@skudchad_rocm on 2018/05/08 18:57:32 
SWDEV-145570 - [HIP] - Synchronize the legacy default stream with other blocking streams as per the spec.

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/14796/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_context.cpp#11 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_internal.hpp#10 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_stream.cpp#5 edit
 2018-05-08 19:04:35 -04:00
Yaxun (Sam) Liu 4c2d2b9e69 Fix warning about inlined function is not defined 2018-05-08 16:38:50 -04:00
foreman 434912d2ca P4 to Git Change 1552011 by skudchad@skudchad_test2_win_opencl on 2018/05/08 14:48:45 
SWDEV-145570 - [HIP] Fix offset calculation when getting a memory object. Also include case when destination VA may just be a CPU host VA and not nessarily device alloced.
	- Fix hipMemset* to write each byte and now a dword as per the spec

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/14787/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#24 edit
 2018-05-08 15:43:13 -04:00
Yaxun (Sam) Liu eeedfc5567 Add __assert_fail, __device_trap and hipErrorAssert for clang 2018-05-08 15:42:27 -04:00
Evgeny Mankov e8dd9297ae Merge pull request #438 from emankov/hipBLAS 
[HIPIFY][Blas] Sync with CUDA 9.1
 2018-05-08 20:50:39 +03:00
Siu Chi Chan 4f12ddea8f add intrinsics mbcnt_lo, mbcnt_hi, lane_id 2018-05-08 13:43:53 -04:00
Evgeny Mankov 57d9ced72a [HIPIFY][Blas] Sync with CUDA 9.1 2018-05-08 20:42:30 +03:00
Evgeny Mankov 76372e9bfa Merge pull request #435 from emankov/hipBLAS 
[HIPIFY] Sync with hipBLAS
 2018-05-08 19:22:32 +03:00
Evgeny Mankov 089ac77b97 [HIPIFY] Sync with hipBLAS 2018-05-08 19:20:47 +03:00
Jenkins b5fa30cfc6 Merge 'master' into 'amd-master' 
Change-Id: I387d49269f0314b38db5e77eacc1be636280620d
 2018-05-08 04:09:43 -05:00
Maneesh Gupta 52bc057dc1 Merge pull request #432 from moosichu/patch-1 
Add space between `###` and `Notes` in hip_terms
 2018-05-08 12:30:27 +05:30
Rahul Garg 78568435da Added hipMemset3D 2018-05-07 10:24:30 +05:30
Tom Maenan Read Cutting 06538c2b6d Add space between ### and Notes in hip_terms 
Makes `Notes` an H3 heading.
 2018-05-05 13:30:11 +01:00
foreman 6b7ec8f427 P4 to Git Change 1550648 by skudchad@skudchad_test2_win_opencl on 2018/05/04 20:56:00 
SWDEV-145570 - [HIP]
	- Implement hipMemcpyHtoD/DtoH/DtoD/ and their Async APIs
	- Combine logic for hipMemset/Memcpy/Memset2D/Memcpy2D that can be shared across multiple APIs

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/14782/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#23 edit
 2018-05-05 00:34:05 -04:00
Evgeny Mankov 0884499e61 Merge pull request #431 from emankov/master 
[HIPIFY][test] Undo commit "Apply .clangformat to all repo source files"
 2018-05-04 22:23:19 +03:00
emankov 0d756720b2 [HIPIFY][test] Undo commit "Apply .clangformat to all repo source files" 
Commit broke tests due to code and comments formatting changes, thus FileCheck fails on checks, which are in comments.
 2018-05-04 22:23:16 +03:00
foreman 05b48c7450 P4 to Git Change 1550526 by cpaquot@cpaquot-ocl-lc-lnx on 2018/05/04 15:00:44 
SWDEV-145570 - [HIP] Missing HIP_INIT_API macro in hipDeviceSynchronize.

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_device_runtime.cpp#9 edit
 2018-05-04 15:12:48 -04:00
foreman a6c58d73f6 P4 to Git Change 1550486 by cpaquot@cpaquot-ocl-lc-lnx on 2018/05/04 14:00:33 
SWDEV-145570 - [HIP] Fix multithread init
	Make the g_ihipInitialized variable per thread
	And make sure to assign a default g_context

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_context.cpp#10 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_internal.hpp#9 edit
 2018-05-04 14:13:13 -04:00
emankov cbecca665b Merge branch 'master' of https://github.com/ROCm-Developer-Tools/HIP 2018-05-04 21:04:34 +03:00
Evgeny Mankov ec51b980e3 Merge pull request #429 from emankov/master 
[HIPIFY][test] add cuDNN test
 2018-05-04 12:40:48 +03:00
Evgeny Mankov 0bf8e42caa [HIPIFY][test] add cuDNN test 2018-05-04 12:37:15 +03:00
Evgeny Mankov bc77bf5ca0 Merge pull request #428 from emankov/docs 
[HIPIFY][doc] Readme.md update
 2018-05-04 11:06:59 +03:00
Evgeny Mankov b6d4c7cfe6 [HIPIFY][doc] Readme.md update 
+ supported CUDA version to LLVM version correspondence table is added.
+ Test section is rewritten.
+ Windows support is added.
 2018-05-04 10:50:18 +03:00
Evgeny Mankov 6c31248cac Merge pull request #423 from emankov/cuDNN 
[HIPIFY] Initial cuDNN support
 2018-05-04 10:01:47 +03:00
foreman f404808c04 P4 to Git Change 1550010 by cpaquot@cpaquot-ocl-lc-lnx on 2018/05/03 21:16:06 
SWDEV-145570 - [HIP] Fixed region initialization

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#22 edit
 2018-05-03 21:26:33 -04:00
foreman d6ffa28a5f P4 to Git Change 1549971 by skudchad@skudchad_rocm on 2018/05/03 18:39:37 
SWDEV-145570 - [HIP] -  Fix build

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#21 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.hpp#21 edit
 2018-05-03 18:56:01 -04:00
foreman f1f8ea8266 P4 to Git Change 1549965 by skudchad@skudchad_test2_win_opencl on 2018/05/03 18:15:18 
SWDEV-145570 - [HIP] -  Minor fixes. Use coarse grain for hipMallocPitch

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/14768/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#20 edit
 2018-05-03 18:26:45 -04:00
Jenkins a40e597b4e Merge 'master' into 'amd-master' 
Change-Id: I1d6d23c5381138aaf2d4555229183a6810612da2
 2018-05-03 04:09:39 -05:00
Evgeny Mankov 4d19bd1da5 [HIPIFY] Initial cuDNN support 
- based on https://github.com/ROCmSoftwarePlatform/hipDNN.
- lit testing was supplemented with CUDA_DNN_ROOT_DIR for cuDNN testing.
- single cuDNN test was added.
 2018-05-03 11:33:40 +03:00
Evgeny Mankov 0ad4d520ed Merge pull request #417 from emankov/master 
[HIPIFY] Sync with HIP (Execution Control, Surfaces, Memory)
 2018-05-03 10:45:51 +03:00
Maneesh Gupta de0148dcca Merge pull request #422 from luckynikki/NULL-FIXES 
Null checks added for hipmallocpitch and hipmemcpy apis
 2018-05-03 10:14:40 +05:30
Lakhan Singh 12d8a47c0c Null checks added for hipmallocpitch and hipmemcpy apis 2018-05-03 09:27:50 +05:30
foreman 16f6cf55e9 P4 to Git Change 1549067 by skudchad@skudchad_rocm on 2018/05/02 17:40:37 
SWDEV-145570 - [HIP] - Fix build

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#19 edit
 2018-05-02 22:25:53 -04:00
foreman ed35f0693f P4 to Git Change 1549060 by skudchad@skudchad_test2_win_opencl on 2018/05/02 17:36:53 
SWDEV-145570 - [HIP] - Add hipMemcpyDeviceToDevice/hipMemcpyHostToHost/hipMemcpyDefault cases to hipMemcpy/hipMemcpyAsync, hipMemset2DAsync

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/14746/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#18 edit
 2018-05-02 21:08:53 -04:00
foreman 57a05ba4d5 P4 to Git Change 1549012 by lmoriche@lmoriche_opencl_dev2 on 2018/05/02 16:56:02 
SWDEV-145570 - [HIP] - Add hipMemset2DAsync, fix the HCC triple for device binaries (make sure to update to the latest HCC).

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#4 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.map.in#4 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#17 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#10 edit
 2018-05-02 19:14:52 -04:00
emankov 3958e930ab [HIPIFY] Initial cuDNN support 
- based on https://github.com/ROCmSoftwarePlatform/hipDNN.
- lit testing was supplemented with CUDA_DNN_ROOT_DIR for cudnn testing.
- single cuDNN test was added.
 2018-05-02 22:11:05 +03:00
Maneesh Gupta 563bdd813f Merge pull request #421 from gargrahul/fix_tex3d_nvcc 
Fix texture 3D for HIP/NVCC
 2018-05-02 13:49:00 +05:30
Rahul Garg 976b9ed200 Cleaned up 2018-05-02 12:54:17 +05:30