Gráfico de commits

13197 Commits

Autor SHA1 Mensaje Fecha
Maneesh Gupta 5e81056c48 Merge branch 'amd-develop' into amd-master
Change-Id: I3ba6305d6cd8de2381615a7ea54d7954e5514216
2017-06-14 10:06:14 +05:30
Maneesh Gupta 4fa2090b78 Bump HIP base version to 1.2
Change-Id: I8ecc164afed4383f78579ed86a5c8c11a73b0780
2017-06-14 09:45:46 +05:30
Sandeep Kumar 01842faa1c Add peer2peer bandwidth and latency test
Change-Id: I6d88e4aa9f6e64096af16579eebef4740734203e
2017-06-14 09:44:56 +05:30
Evgeny Mankov 82b37fe481 [HIPIFY] Initial sync HIPIFY with HIP by CUDA Driver API functions.
+ CUDA_Driver_API_functions_supported_by_HIP.md update.
+ Initial update of HIPIFY with CUDA driver API functions:
  1.Error Handling, 2.Initialization, 3.Version Management, 5-6.Device Management, 7.Primary Context, 8-9.Context, 10.Module Management, 11.Memory Management.
+ Sync HIP functions against CUDA Driver and Runtime API functions.
+ Typo fixes.

ToDo: 12-30 modules of CUDA Driver API.
2017-06-13 20:25:11 +03:00
Rahul Garg 0efd737767 Input args NULL check in hipChooseDevice
Change-Id: I1a7b8cded2f81d739645bbf3dab2f04bb9c3c796
2017-06-13 13:35:50 +05:30
foreman 7544a92aec P4 to Git Change 1421594 by johtaylo@johtaylo-jtincrementor-increment on 2017/06/13 03:00:04
SWDEV-2 - Change OpenCL version number from 2439 to 2440.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2187 edit
2017-06-13 03:09:39 -04:00
foreman cd61586801 P4 to Git Change 1421454 by yaxunl@yaxunl_stg_win50 on 2017/06/12 17:18:49
SWDEV-123891 - Bump up OpenCL SC_CLIENT_INTERFACE_MAJOR_VERSION from 1 to 4 Part 2/2

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/backends/gpu/scwrapper/EGNI/scCompile89.cpp#3 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/backends/gpu/scwrapper/SI/scCompileSI.cpp#6 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/backends/gpu/scwrapper/scState.cpp#6 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/complibdefs#7 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/complibdefs#50 edit
... //depot/stg/opencl/drivers/opencl/compiler/sclibdefs.opencl#15 edit
2017-06-12 17:31:11 -04:00
foreman 628503e964 P4 to Git Change 1421208 by gandryey@gera-w8 on 2017/06/12 13:15:22
SWDEV-124171 - adding support for p2p OCL in rocm stack
	- Add cl_amd_copy_buffer_p2p extension for P2P transfers. The extension adds a new API entry - clEnqueueCopyBufferP2PAMD() which allows to transfer CL buffers between different CL contexts on different GPUs. If P2P isn't possible, then double copy performed
	- Also the app can query the P2P support capabilities for the device. A list of P2P accessible devices can be returned for the current device

	http://ocltc.amd.com/reviews/r/12913/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_context.cpp#54 edit
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_device.cpp#62 edit
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_p2p_amd.cpp#1 add
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_p2p_amd.h#1 add
... //depot/stg/opencl/drivers/opencl/api/opencl/khronos/headers/opencl2.0/CL/cl_ext.h#29 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/cpu/cpuvirtual.hpp#14 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#287 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.hpp#141 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.hpp#26 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#54 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#22 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmemory.cpp#24 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#39 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#12 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/command.cpp#79 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/command.hpp#84 edit
2017-06-12 13:31:09 -04:00
foreman 9ba3948388 P4 to Git Change 1421208 by gandryey@gera-w8 on 2017/06/12 13:15:22
SWDEV-124171 - adding support for p2p OCL in rocm stack
	- Add cl_amd_copy_buffer_p2p extension for P2P transfers. The extension adds a new API entry - clEnqueueCopyBufferP2PAMD() which allows to transfer CL buffers between different CL contexts on different GPUs. If P2P isn't possible, then double copy performed
	- Also the app can query the P2P support capabilities for the device. A list of P2P accessible devices can be returned for the current device

	http://ocltc.amd.com/reviews/r/12913/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_context.cpp#54 edit
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_device.cpp#62 edit
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_p2p_amd.cpp#1 add
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_p2p_amd.h#1 add
... //depot/stg/opencl/drivers/opencl/api/opencl/khronos/headers/opencl2.0/CL/cl_ext.h#29 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/cpu/cpuvirtual.hpp#14 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#287 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.hpp#141 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.hpp#26 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#54 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#22 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmemory.cpp#24 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#39 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#12 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/command.cpp#79 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/command.hpp#84 edit
2017-06-12 13:31:09 -04:00
Rahul Garg ce6e455677 Update P2P test for using memcpy and kernel tests
Change-Id: Ib0f8fc9425e6e85fd11d7d02395c52bc713dcb37
2017-06-12 11:53:25 +05:30
Maneesh Gupta 3d87f73924 Merge branch 'amd-develop' into amd-master
Change-Id: I7f2fba7875ed1c89dfc768f7415ed6fb0d1c6407
2017-06-12 11:22:53 +05:30
Maneesh Gupta 6174e69f87 Updated RELEASE.md
Change-Id: Ic451612555c66f3ed7131514fc97fcc41091370a
2017-06-12 11:20:28 +05:30
Maneesh Gupta 15a3464630 Update directed tests README.md
Change-Id: I395245454d376508f04e5a4a62c8933895cb3867
2017-06-12 11:19:55 +05:30
Patrick Flick 821c238bad fix typo 2017-06-12 10:15:27 +05:30
Maneesh Gupta d2b90ad93c Merge branch hipify-updates into amd-develop
Change-Id: I13d8750027a2a8787e4eb2e1ed525cf69d14b805
2017-06-12 10:10:19 +05:30
Maneesh Gupta c5366a55f1 Initial implementation of hipify-cmakefile
Change-Id: Id365da9f887b5c3409639f000b430d093fd4f6b3
2017-06-12 09:57:17 +05:30
foreman 209a1a1ec2 P4 to Git Change 1420896 by johtaylo@johtaylo-jtincrementor-increment on 2017/06/10 03:00:04
SWDEV-2 - Change OpenCL version number from 2438 to 2439.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2186 edit
2017-06-10 03:18:05 -04:00
foreman d2a7bdf187 P4 to Git Change 1420426 by johtaylo@johtaylo-jtincrementor-increment on 2017/06/09 03:00:04
SWDEV-2 - Change OpenCL version number from 2437 to 2438.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2185 edit
2017-06-09 03:17:08 -04:00
Sun, Peng 682dda4418 Fix error related to undefined reference of __get_dynamicgroupbaseptr().
Change-Id: I14951e1725e35dd5f5e53805f81cdb58661f59f2
2017-06-08 19:24:32 -05:00
Sun, Peng 5450021f93 Add clang version guard so the hip_fp16.h header won't be picked up by gcc
Change-Id: Ia21335a455bc93210901b44bc8c76a7f4a385b55
2017-06-08 19:24:32 -05:00
foreman 057a019e40 P4 to Git Change 1420248 by skudchad@skudchad_test_win_opencl2 on 2017/06/08 18:13:23
SWDEV-107271 - [OpenCL][GFXIP9 Bring up]
	- Correct the pageMigration flag in runtime

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#48 edit
2017-06-08 18:40:45 -04:00
foreman 3152da6be5 P4 to Git Change 1419977 by gandryey@gera-w8 on 2017/06/08 11:16:05
SWDEV-121409 - [SSG][Vega] new LF change requires integration in OpenCL
	- Add lfInit() and lfTerminate() calls

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/platform/context.cpp#43 edit
2017-06-08 11:22:06 -04:00
foreman c7ab8428c4 P4 to Git Change 1419768 by johtaylo@johtaylo-jtincrementor-increment on 2017/06/08 03:00:05
SWDEV-2 - Change OpenCL version number from 2436 to 2437.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2184 edit
2017-06-08 03:12:54 -04:00
foreman a2b0ebf70c P4 to Git Change 1419586 by skudchad@skudchad_test_win_opencl2 on 2017/06/07 16:09:44
SWDEV-107271 - [OpenCL][GFXIP9 Bring up]
	- Detect HBCC and correctly support Vega10 with XNACKs

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/12911/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#47 edit
2017-06-07 16:14:38 -04:00
foreman 406d9ff7c8 P4 to Git Change 1419517 by lmoriche@lmoriche_palamida on 2017/06/07 14:42:22
SWDEV-94644 - Update runtime after the recent merge
	- Update CMakeLists.txt.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/CMakeLists.txt#5 edit
2017-06-07 14:59:18 -04:00
foreman 04412e2b5a P4 to Git Change 1419430 by lmoriche@lmoriche_palamida on 2017/06/07 12:28:09
SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment
	- Enable building the ICD and clinfo.

Affected files ...

... //depot/stg/opencl/drivers/opencl/CMakeLists.txt#11 edit
... //depot/stg/opencl/drivers/opencl/tools/clinfo/CMakeLists.txt#2 edit
2017-06-07 12:55:38 -04:00
foreman 5adbcd8f12 P4 to Git Change 1419430 by lmoriche@lmoriche_palamida on 2017/06/07 12:28:09
SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment
	- Enable building the ICD and clinfo.

Affected files ...

... //depot/stg/opencl/drivers/opencl/CMakeLists.txt#11 edit
... //depot/stg/opencl/drivers/opencl/tools/clinfo/CMakeLists.txt#2 edit
2017-06-07 12:55:38 -04:00
foreman db1916dcbc P4 to Git Change 1419386 by kzhuravl@kzhuravl-fiji-ocllc on 2017/06/07 11:13:10
SWDEV-94644 - Update runtime after the recent merge
	 - AMDGPUCodeObjectMetadata.h was moved to include/llvm/Support

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/build/Makefile.pal#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.hpp#12 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#40 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.hpp#18 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/build/Makefile.oclrocm#17 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.hpp#25 edit
2017-06-07 11:19:18 -04:00
Ben Sander 9bfc7b0e13 Use amHostCoherentFlag. Requires new HCC version. 2017-06-07 09:06:40 -05:00
Maneesh Gupta ff4fae7d20 hip_hcc package: add libstdc++-static as a rpm dependency
Change-Id: I83a79353492a6be3d788b7c0ce4a8f3aa740d9d9
2017-06-07 15:50:28 +05:30
Maneesh Gupta ff8ade59aa hipMemcpy-size test: reduce max size to make it work correctly on nvcc path
Change-Id: I9ce9f5a9e141ffd8ddf961269010b33358e02771
2017-06-07 15:25:54 +05:30
Maneesh Gupta a50f5ca0ac hipDeviceMemcpy test: make it functional on nvcc path
Change-Id: Id10c79b48747ed701adbd0a233c53cd60cfa743b
2017-06-07 15:24:44 +05:30
Maneesh Gupta a7dc938ec0 p2p_copy_coherency test: gracefully handle single gpu case
Change-Id: I216663f67ef58c673136332635dab8b57079b909
2017-06-07 15:23:37 +05:30
foreman 062360faef P4 to Git Change 1419239 by johtaylo@johtaylo-jtincrementor-increment on 2017/06/07 03:00:03
SWDEV-2 - Change OpenCL version number from 2435 to 2436.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2183 edit
2017-06-07 03:14:42 -04:00
foreman 53f8a48c7c P4 to Git Change 1419209 by lmoriche@lmoriche_palamida on 2017/06/07 01:42:42
SWDEV-101666 - No need to force disable SDMA, the ROCR now enables SDMA for known good configurations.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#53 edit
2017-06-07 01:56:08 -04:00
Ben Sander c2baa4f6e6 Enable HCC_OPT_FLUSH=1.
Requires appropriate HCC with this support   :
commit 38e392b517a46a09a3b1c8f388e6a0db3741c510
2017-06-07 00:15:05 -05:00
Maneesh Gupta 2015eff64e Merge branch 'amd-develop' into amd-master
Change-Id: I3802a78577a3979172c0e5fbf39c3db2a044824d
2017-06-06 15:38:03 +05:30
Sun, Peng 3b6a863eef Improve HIP kernel names, attributes and codegen, contributed by Alex Voicu
Change-Id: I2cafbdc5a98e26c7f4fad84739c915e7dc09993c
2017-06-05 11:39:00 -05:00
Ben Sander 344b6cb0c0 Enable HIP_SYNC_NULL_STREAM=0 optimization. 2017-06-05 08:50:41 -05:00
Ben Sander 823281dcba Fix HIP_SYNC_NULL_STREAM=0 mode.
- Fix null-stream sync
- hipStreamDestroy of null stream returns hipErrorInvalidResourceHandle
- Update documentation.
- Add tests for null stream sync, hipEventElapsedTime.
- Rename internal enum hipEventStatusRecorded to hipEventStatusComplete
- refactor hipStreamWaitEvent to streamline control-flow
2017-06-05 08:50:22 -05:00
Ben Sander 863b7c3f56 Update tests.
Fix some NVCC issues.
Add hipStreamSync2, record_event tests.
2017-06-04 20:18:37 -05:00
Ben Sander 15f54fb943 Update tests, add p2p coherency test. 2017-06-03 17:11:34 -05:00
foreman 1b6f1b949a P4 to Git Change 1417876 by johtaylo@johtaylo-jtincrementor-increment on 2017/06/03 03:00:04
SWDEV-2 - Change OpenCL version number from 2434 to 2435.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2182 edit
2017-06-03 03:18:41 -04:00
Aditya Atluri fdcc223842 added half data type and vector destructors
1. Added half data types to hip_fp16.h
2. Added destructor to vector data types

Change-Id: Id5ae76a663bb90a4bde2839ec79c58fbaee5072f
2017-06-02 11:19:33 -05:00
emankov c5f9758f4b [HIPIFY] annotation 2017-06-02 16:33:48 +03:00
emankov e7779650e9 [HIPIFY] rename legacy hipify perl script and its usage to hipify-perl 2017-06-02 16:30:43 +03:00
foreman 04a9e0aac2 P4 to Git Change 1417347 by johtaylo@johtaylo-jtincrementor-increment on 2017/06/02 03:00:18
SWDEV-2 - Change OpenCL version number from 2433 to 2434.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2181 edit
2017-06-02 03:14:25 -04:00
foreman c9760c9f39 P4 to Git Change 1417118 by gandryey@gera-w8 on 2017/06/01 15:09:16
SWDEV-111607 - [Vega][Redstone] Blender 2.78 Corruption observed on generated image
	- Add submit() call for CU cache invalidation to make sure the HW operation was counted

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.hpp#25 edit
2017-06-01 15:16:52 -04:00
Evgeny Mankov ee85243bcd [HIPIFY] All CUDA 8.0.44 API functions update
(for both Driver and Runtime APIs)

1) P2P
cuDeviceGetP2PAttribute   cudaDeviceGetP2PAttribute

2) Memory Mngmnt
cuMemPrefetchAsync        cudaMemPrefetchAsync
cuMemAdvise               cudaMemAdvise
cuMemRangeGetAttribute    cudaMemRangeGetAttribute
cuMemRangeGetAttributes   cudaMemRangeGetAttributes

3) Streams (Driver API only, no analogues in Runtime API)
cuStreamWaitValue32
cuStreamWaitValue32
cuStreamWriteValue32

4) Texture Reference Mngmnt (Driver API only, no analogues in Runtime API)
cuTexRefSetBorderColor
cuTexRefGetBorderColor
2017-06-01 21:08:33 +03:00
foreman e67ecb4135 P4 to Git Change 1416741 by johtaylo@johtaylo-jtincrementor-increment on 2017/06/01 03:00:04
SWDEV-2 - Change OpenCL version number from 2432 to 2433.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2180 edit
2017-06-01 03:16:21 -04:00