İşleme Grafiği

13197 İşleme

Yazar SHA1 Mesaj Tarih
Aditya Atluri 1673bdefec added sample for module api which works on both cuda and rocm
Change-Id: Id88abb3698a224177fd0531e3e16013dde6dd95e
2016-08-29 15:00:26 -05:00
Ben Sander d76372a049 Doc update. Describe memcpytosymbol, threadfence_system workarounds 2016-08-29 13:45:20 -05:00
foreman 97cbf919ba P4 to Git Change 1307926 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/29 03:00:09
SWDEV-2 - Change OpenCL version number from 2210 to 2211.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1957 edit
2016-08-29 03:12:55 -04:00
Aditya Atluri 645975a05b Fixed nvcc path for module apis
Change-Id: I43c7a6bc226f4f270c37f3c4af86b3b3efd0b175
2016-08-28 22:50:44 -05:00
Aditya Atluri 71991ed8f8 Cleaned up module api
- Moved HIP_INIT_API from internal to all public apis
- renamed hipLaunchModuleKernel to hipModuleLaunchKernel
- Changed tests according to the changed api

Change-Id: I822ff63c7c5b7dad340ece49456baf9d89428e9f
2016-08-28 16:48:57 -05:00
Ben Sander 46c3ff52a6 Remove cutnpaste check on uninitialized hsa status
Change-Id: Icc0256b9b1bd01d45cca4feb1106cfe7427bcd15
2016-08-28 16:35:27 -05:00
Ben Sander f7ee14e637 Remove deprecated message from hipLaunchModuleKernel
Change-Id: I87675453ae4363e3340a9d1491bb00543fa8c6e0
2016-08-28 16:34:34 -05:00
foreman dfd10f4c8c P4 to Git Change 1307882 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/28 03:00:08
SWDEV-2 - Change OpenCL version number from 2209 to 2210.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1956 edit
2016-08-28 10:03:23 -04:00
foreman 0f68300697 P4 to Git Change 1307868 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/27 03:00:08
SWDEV-2 - Change OpenCL version number from 2208 to 2209.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1955 edit
2016-08-27 20:49:42 -04:00
Aditya Atluri 5e7c396bcd added memset and memcpy device functions
- Added memcpy and memset device functions
- Added test for memcpy and memset

Change-Id: Icd21a8dd964953b86d5e92889bf1664bee647219
2016-08-27 11:13:56 -05:00
Aditya Atluri f722a132a4 Added explicit memory copy direction apis
- Fixed stale printf in context api
- Added 4 sync memcpy apis
  1. hipMemcpyHtoD
  2. hipMemcpyDtoH
  3. hipMemcpyDtoD
  4. hipMemcpyHtoH
- Added test for added apis

Change-Id: I4a9c382445b62631f8d0bcbb9a670322288b72b1
2016-08-26 13:11:01 -05:00
Ben Sander 8b8181df41 Add docs on porting driver API 2016-08-26 10:46:39 -05:00
Ben Sander 8d02cf466f Doc update to clarify supported / unsupported features 2016-08-26 10:46:39 -05:00
Aditya Atluri 74a8fb132d Added NVCC support and name changes
- Added NVCC support for module APIs
- Changed hipFunction and hipModule data types to hipFunction_t and hipModule_t
- Created new intenal ihipModuleGetFunction as it is used twice
- Changed test to match with the new data types

Change-Id: I300a1c7fd40ed7065b1b8b9de97e3a06b96ed729
2016-08-26 10:32:01 -05:00
Rahul Garg 452108b0e6 Added logic to update primary ctx when ctx stack is empty, updated hipCtxDestroy and ctxGetCurrent functions
Change-Id: Ia0a8943c121bc1279788a1cfa9be59af614b04a6
2016-08-26 19:03:23 +05:30
Rahul Garg 77991293ba Resolved errors due to hipCtxXXX APIs
Change-Id: Iffac0095c4352864eca622ea318d2291571b5153
2016-08-26 15:32:49 +05:30
Rahul Garg da80bf6326 NVCC path support for hipCtxXXX APIs
Change-Id: Ic7dbfbdaee9d00c0de1363c50758e5e29a96a8b2
2016-08-26 14:10:36 +05:30
Rahul Garg a1d401336b Addition of hipCtxEnablePeerAccess and hipCtxDisablePeerAccess functions
Change-Id: I381c8cbbde17eae7d9bb5d4cb1596cebf4bda039
2016-08-26 13:51:33 +05:30
foreman f2b2b57a65 P4 to Git Change 1307446 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/26 03:00:08
SWDEV-2 - Change OpenCL version number from 2207 to 2208.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1954 edit
2016-08-26 03:08:43 -04:00
foreman bd32c277bb P4 to Git Change 1307417 by lmoriche@lmoriche_opencl_dev on 2016/08/26 01:38:31
SWDEV-94610 - Setup the (hidden) arguments for global_offset [x,y,z].

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#16 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#10 edit
2016-08-26 01:47:40 -04:00
Maneesh Gupta 0c051fede1 hipcc needs to link against supc++
Change-Id: Ica0949099a0bdfe2a493341dc1cd96ec93f34f11
2016-08-26 10:35:30 +05:30
foreman 08f5a25e7b P4 to Git Change 1307381 by lmoriche@lmoriche_opencl_dev on 2016/08/25 20:34:15
SWDEV-94610 - Wait on every kernel dispatch if env.GPU_FLUSH_ON_EXECUTION is set.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#9 edit
2016-08-25 20:40:27 -04:00
Aditya Atluri 25be4fb532 Changed how hipEvent_t is typedefed internall
- Mapped hipEvent_t directly to ihipEvent_t* instead of a handle

Change-Id: I5a8bcca0ef962932e0738c03eb1fc914d23022ae
2016-08-25 14:34:41 -05:00
Aditya Atluri f22a3189a3 Added hipModuleGetGlobal and hipModuleLoadData
Change-Id: Iaec873f7d86b72911b6ad32e067a4dfe3d552fe6
2016-08-25 14:16:53 -05:00
foreman 82f13f6ba1 P4 to Git Change 1307211 by smekhano@stas-nova-hsa on 2016/08/25 13:09:04
SWDEV-101354 - HSA HLC: fix unify metadata pass
	When we link multiple modules we have metadata duplicated, so after we link with our library bitcode is twice bigger than needs to be.
	Besides we did not unify llvm.ident metadata since llvm 3.6 merge.

	Fix that:
	1. Add llvm.ident to the processing;
	2. Do not duplicate strings within unified metadata;
	3. Run unification pass post link, not before the link.

	Now since our library is compiled for OpenCL 2.0 we will always get OCL version 2.0 as a maximum. That is not really correct, and since
	the pass was not really working before that would lead to regression, as we would fail to identify correct kernel's OpenCL version and
	perform simplifications for 1.2. Now the pass will pick the first version, which shall represent the kernel module. That might not be
	100% correct because we may have several kernel modules, but a proper fix would require to correctly identify library as 1.2, which is
	troublesome. In the current state that just keeps the status quo.

	Testing: smoke, precheckin
	Reviewed by Evgeny Mankov

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/linker.cpp#152 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/include/AMDFixupKernelModule.h#2 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/lib/AMDFixupKernelModule.cpp#7 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/tools/opencl-link/opencl-link.cpp#10 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Transforms/Scalar/AMDUnifyMetadata.cpp#2 edit
2016-08-25 13:14:09 -04:00
foreman 8539fc5c2e P4 to Git Change 1307201 by lmoriche@lmoriche_opencl_dev on 2016/08/25 12:44:28
SWDEV-1307193 - Back out changelist 1307198

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#14 edit
2016-08-25 12:00:21 -05:00
foreman ed1144e0c0 P4 to Git Change 1307198 by lmoriche@lmoriche_opencl_dev on 2016/08/25 12:39:46
SWDEV-1306648 - Add OptimizeLLVMBitcode to the ROCm-OpenCL-Driver. Call the optimizer in roc::HSAILProgram::linkImpl, between linking with the bitcode built-in libraries and code generation. Use the default optimization level, with at a minimum -strip -instcombine -always-inline.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#13 edit
2016-08-25 12:49:33 -04:00
Evgeny Mankov 5cca5b3dca clang-hipify: code refactoring - API (Driver/Runtime/Blas) distinguishing is added. 2016-08-25 19:36:37 +03:00
foreman ff561c4f21 P4 to Git Change 1306999 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/25 03:00:07
SWDEV-2 - Change OpenCL version number from 2206 to 2207.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1953 edit
2016-08-25 03:08:36 -04:00
foreman 3d384e6412 P4 to Git Change 1306787 by skudchad@skudchad_test_win_opencl2 on 2016/08/24 17:43:49
SWDEV-94442 - [IQE] [EllesmereGL] [Win10 TH2] GPU2GPU ocltst causes hard hang intermittently
	- Add gpu events for marker wait in runtime

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/11174/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#405 edit
2016-08-24 17:50:16 -04:00
foreman 01ed632c6c P4 to Git Change 1306648 by lmoriche@lmoriche_opencl_dev on 2016/08/24 15:17:25
SWDEV-94610 - [OpenCL/LC] Program Manager: Implement multi-file linking and separate compilation. Implement saving and loading programs, libraries, and executables.

Affected files ...

... //depot/stg/opencl/drivers/opencl/make/clang.git/include/clang/Config/config.h#3 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#277 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/roccompiler.cpp#9 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#7 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#12 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.hpp#7 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/program.cpp#82 edit
2016-08-24 15:25:49 -04:00
foreman f3612fa220 P4 to Git Change 1306607 by lmoriche@lmoriche_opencl_dev on 2016/08/24 13:05:06
SWDEV-94644 - Fix the build after the latest update from git

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#6 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmetadata.hpp#2 edit
2016-08-24 13:11:03 -04:00
Evgeny Mankov 207eb02736 clang-hipify: code refactoring and performance improvement 2016-08-24 18:51:36 +03:00
foreman 1940203fcb P4 to Git Change 1306553 by gandryey@gera-w8 on 2016/08/24 10:48:34
SWDEV-86035 - Use local memory for kernel allocation

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.cpp#9 edit
2016-08-24 11:00:15 -04:00
Aditya Atluri af4dc556eb changed internal structure of hipFunction and hipModule
Change-Id: Ifa343782e29d7e056efc47e56253311013005093
2016-08-24 09:47:11 -05:00
foreman 0a76bb941c P4 to Git Change 1306438 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/24 03:00:08
SWDEV-2 - Change OpenCL version number from 2205 to 2206.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1952 edit
2016-08-24 02:07:07 -05:00
Aditya Atluri 768287f2a2 Module test correction and hipModuleUnload API
- Corrected the hipModule.cpp test to minimal code
- Added hipModuleUnload API
- Added hipModuleUnload API test

Change-Id: I9c40337043d7972a570b795e1bfc104bd2c4d8aa
2016-08-23 14:19:15 -05:00
foreman 840d785d11 P4 to Git Change 1306185 by gandryey@gera-w8 on 2016/08/23 14:55:34
SWDEV-79445 - OCL generic changes and code clean-up
	- Report LogError with the build log if it's not empty

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#201 edit
2016-08-23 15:12:24 -04:00
Aditya Atluri 2f426aaaea Added module api test with gcn binary
Change-Id: I61c3ecc2b34168d10f1a7b15d668630eb2c69c8c
2016-08-23 13:50:19 -05:00
foreman 862e3a1a79 P4 to Git Change 1306133 by gandryey@gera-w8 on 2016/08/23 14:00:09
SWDEV-79445 - OCL generic changes and code clean-up
	- Update staging copy path with a flush so CPU copy and SDMA transfer could run asynchronously.
	- Tune chunk size for transfers

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpublit.cpp#121 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palblit.cpp#6 edit
2016-08-23 14:12:24 -04:00
foreman 53a0c0add6 P4 to Git Change 1306079 by gandryey@gera-w8 on 2016/08/23 11:52:31
SWDEV-95905 - OpenCL on PAL - Device Enqueue
	- Move disapatch funciton outside of the compiler lib. Client must provide 4 new functions to the scheduler: GetCmdTemplateHeaderSize(), GetCmdTemplateDispatchSize(),EmptyCmdTemplateDispatch(), RunCmdTemplateDispatch().

	http://ocltc.amd.com/reviews/r/11142/

Affected files ...

... //depot/stg/opencl/drivers/opencl/library/hsa/hsail/src/devenq/schedule.cl#13 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#34 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palschedcl.cpp#2 edit
2016-08-23 12:05:38 -04:00
foreman ce1ae03c57 P4 to Git Change 1305891 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/23 03:00:09
SWDEV-2 - Change OpenCL version number from 2204 to 2205.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1951 edit
2016-08-23 03:14:15 -04:00
Aditya Atluri c0fbe7891c Added stream synchronisation for hipLaunchModuleKernel
- The module kernel launch is now in sync with commands in its stream
- Moved launch kernel inside ihipStream

Change-Id: Ic00cfcf4882bf81b6203c36881a52575ea68b529
2016-08-22 14:17:55 -05:00
Aditya Atluri 98c740c51f Added nvcc path for hipComplex APIs
- Changed from inline to static inline for hipComplex AMD APIs
- Added NVCC path for hipComplex APIs mapped to cuComplex APIs

Change-Id: I809cf3a11b5b1c8bbc7a57c5fbcc3dc6745ccb95
2016-08-22 10:29:46 -05:00
Rahul Garg b0ba622ed5 Added support for hipCtxSynchronize and hipCtxGetFlags,modified hipDeviceSynchronize
Change-Id: If7bac667a262fa8c0cb3dc93e97f2534855acd07
2016-08-22 16:15:27 +05:30
foreman 8b645f00ac P4 to Git Change 1305449 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/22 03:00:08
SWDEV-2 - Change OpenCL version number from 2203 to 2204.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1950 edit
2016-08-22 03:11:55 -04:00
Rahul Garg 778bb2e78a Added initial draft for performance optimizations, started with unpinned memory transfers
Change-Id: Icbce2aec347d015bc66cc0c08f6193057bf36b4c
2016-08-22 11:37:37 +05:30
foreman 9674e8eb01 P4 to Git Change 1305360 by lmoriche@lmoriche_opencl_dev on 2016/08/21 17:34:39
SWDEV-94640 - Set the O# level and cl-std from the command args or default settings.
	Select the correct header (CL1.2/C2.0) from the given cl-std.
	Reorder the linked libraries in reverse order of dependency (was failing linking).
	Make the built-in library objects depend on the pre-compiled header.
	Disable the timestamp on the pre-compiled header.
	Disable image support reporting.

Affected files ...

... //depot/stg/opencl/drivers/opencl/make/amdgcn.git/headers/build/Makefile.headers#7 edit
... //depot/stg/opencl/drivers/opencl/make/amdgcn.git/irif/build/Makefile.irif#4 edit
... //depot/stg/opencl/drivers/opencl/make/amdgcn.git/ockl/build/Makefile.ockl#5 edit
... //depot/stg/opencl/drivers/opencl/make/amdgcn.git/oclc/build/Makefile.oclc#5 edit
... //depot/stg/opencl/drivers/opencl/make/amdgcn.git/ocml/build/Makefile.ocml#5 edit
... //depot/stg/opencl/drivers/opencl/make/amdgcn.git/opencl/build/Makefile.opencl#6 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/roccompiler.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#10 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#11 edit
2016-08-21 17:46:14 -04:00
foreman e2c43f9651 P4 to Git Change 1305332 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/21 03:33:56
SWDEV-2 - Change OpenCL version number from 2202 to 2203.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1949 edit
2016-08-21 03:46:32 -04:00
foreman 6e831e634a P4 to Git Change 1305300 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/08/20 03:00:10
SWDEV-2 - Change OpenCL version number from 2201 to 2202.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1948 edit
2016-08-20 03:08:27 -04:00