Commit Graph

11723 Commits

Author SHA1 Message Date
Yaxun Sam Liu d3295e61e2 Fix __HIP_ARCH_* not defined after including math_functions.h
hcc_detail/math_functions.h used to include hcc_detail/hip_runtime.h.

Removing it has caused regression in TensorFlow 1.8.

Put it back for backward compatibiliity.


[ROCm/clr commit: 87de95975a]
2018-08-08 08:55:28 -04:00
foreman 44f389b7eb P4 to Git Change 1590683 by chui@ocl-promo-incrementor on 2018/08/08 02:56:36
SWDEV-2 - Change OpenCL version number from 2711 to 2712.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2459 edit


[ROCm/clr commit: 1575bb0507]
2018-08-08 03:21:33 -04:00
foreman eb836ba642 P4 to Git Change 1590524 by gandryey@gera-lnx-rcf on 2018/08/07 17:38:31
SWDEV-160552 - [CQE OCL][DTB][BD][Perf][QR][G] Soft hang is observed with ProRender after few subtests due to base driver.
	- Reset the pointer to  nullptr if PAL platform was destroyed

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#102 edit


[ROCm/clr commit: 1a61f6b9e5]
2018-08-07 17:56:54 -04:00
foreman 20e1c977c9 P4 to Git Change 1590512 by gandryey@gera-lnx-rcf on 2018/08/07 17:06:25
SWDEV-160552 - [CQE OCL][DTB][BD][Perf][QR][G] Soft hang is observed with ProRender after few subtests due to base driver.
	- Delete PAL platform object if runtime didn't find any supported device.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#101 edit


[ROCm/clr commit: 769cc7cddb]
2018-08-07 17:29:56 -04:00
Yaxun Sam Liu 6e153a743f Fix __HIP_DEVICE_COMPILE__ not defined when hip/math_functions.h is included
This fixes build failure in TensorFlow 1.8 for HCC


[ROCm/clr commit: 058eb7535b]
2018-08-07 17:14:39 -04:00
foreman 289f9e9c69 P4 to Git Change 1590448 by gandryey@gera-w8 on 2018/08/07 15:31:50
SWDEV-159036 - HBCC is off by default in OCL on WX9100
	- Disable HBCC, since it causes ocltst failures on TC

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#293 edit


[ROCm/clr commit: 99cb78b186]
2018-08-07 15:53:13 -04:00
Yaxun Sam Liu 6df5aef807 Fix declaration conflict when hip/math_functions.h is included first
This fixes build failure in TensorFlow 1.8 for HCC


[ROCm/clr commit: 69bbf45b44]
2018-08-07 15:44:59 -04:00
foreman da2cfffc52 P4 to Git Change 1590417 by gandryey@gera-w8 on 2018/08/07 14:30:31
SWDEV-159036 - HBCC is off by default in OCL on WX9100
	- Add HBCC size to the global memory. It seems KMD fixed all regressions.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#292 edit


[ROCm/clr commit: abefff3eb2]
2018-08-07 14:51:30 -04:00
foreman d05fc25898 P4 to Git Change 1590078 by chui@ocl-promo-incrementor on 2018/08/07 02:56:00
SWDEV-2 - Change OpenCL version number from 2710 to 2711.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2458 edit


[ROCm/clr commit: 60bea15438]
2018-08-07 03:13:19 -04:00
foreman ee0e1fafe2 P4 to Git Change 1590010 by skudchad@skudchad_test2_win_opencl on 2018/08/06 21:18:48
SWDEV-145570 - [HIP] - Fix some issues in hip runtime
	- Set stream for event
	- Free mem needs to be reported in bytes but runtime backends reports in Kb

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/15586/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#40 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#15 edit


[ROCm/clr commit: b4ccaeffe1]
2018-08-06 21:24:14 -04:00
foreman d4ff3dee52 P4 to Git Change 1589968 by skudchad@skudchad_test2_win_opencl on 2018/08/06 17:25:53
SWDEV-145570 - [HIP] Increase the default size for SVM to 16GB for HIP

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/15578/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#100 edit


[ROCm/clr commit: aa99feb153]
2018-08-06 17:32:00 -04:00
foreman 76e0512872 P4 to Git Change 1589611 by chui@ocl-promo-incrementor on 2018/08/04 02:55:53
SWDEV-2 - Change OpenCL version number from 2709 to 2710.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2457 edit


[ROCm/clr commit: f889d2ebba]
2018-08-04 03:10:36 -04:00
sunway513 e1d72394da Add more check to ensure the startup loader only be enabled with the env var set to 1
[ROCm/clr commit: 7d776ae665]
2018-08-04 01:52:27 +00:00
sunway513 ef368c0f43 Add startup loader under HIP_STARTUP_LOADER env var, disable by default
[ROCm/clr commit: 9af862d364]
2018-08-04 01:48:06 +00:00
Wen-Heng (Jack) Chung d7b2bab183 Revert "HIP program state re-initialization logic"
This reverts commit 865fe01b37.


[ROCm/clr commit: 1fe7f1be91]
2018-08-03 17:03:04 -05:00
Wen-Heng (Jack) Chung fbf97dddb5 Revert "Improve performance of re-initialization logic"
This reverts commit 1d1b402e5b.


[ROCm/clr commit: f6178c2159]
2018-08-03 17:02:58 -05:00
Wen-Heng (Jack) Chung 1494fc442f Revert "Keep the map which tracks GPU kernel symbols to grow monotonically"
This reverts commit ca62af218c.


[ROCm/clr commit: 329f3b7a5e]
2018-08-03 17:02:50 -05:00
foreman 4926800fc4 P4 to Git Change 1589476 by axie@axie-rocm-opencl on 2018/08/03 15:54:24
SWDEV-79445 - OCL generic changes and code clean-up
	- Optimize setup of kernel arguments.
	- Add HW ABI support in the abstraction layer
	- Remove arguments parsing loop from the kernel launch. Memory processing will be responsible for dependency tracking and  patching of arguments.

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/15400/

	Tests:
	1. ./run_conformance.py ./opencl_conformance_tests_reallyquick.csv CL_DEVICE_TYPE_GPU for openCL 1.2: OpenCL-GL sharing failed. This is not a regression.
	2. ./ocltst -m oclruntime.so -A oclruntime.exclude
	3. ./run_conformance.py opencl_conformance_tests_lightning.csv CL_DEVICE_TYPE_GPU : PASS
	4. teamcity test: http://ocltc.amd.com:8111/viewModification.html?modId=104598&personal=true&buildTypeId=&tab=vcsModificationBuilds&show_all_builds=true

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#39 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.hpp#23 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#34 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.hpp#14 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#60 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#17 edit


[ROCm/clr commit: 9cd2db67f6]
2018-08-03 16:05:12 -04:00
foreman c1f7312937 P4 to Git Change 1589162 by chui@ocl-promo-incrementor on 2018/08/03 02:56:36
SWDEV-2 - Change OpenCL version number from 2708 to 2709.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2456 edit


[ROCm/clr commit: 6d82075654]
2018-08-03 03:09:01 -04:00
foreman 4897c229a4 P4 to Git Change 1589138 by cpaquot@cpaquot-ocl-lc-lnx on 2018/08/03 01:04:05
SWDEV-145570 - [HIP] Added __gnu_h2f_ieee and __gnu_f2h_ieee

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#9 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.map.in#10 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#16 edit


[ROCm/clr commit: f4add6058a]
2018-08-03 01:15:15 -04:00
Maneesh Gupta 34e14adae2 Merge pull request #588 from Jorghi12/patch-10
Fixing a HIP bug that causes DEBUG symbols to be dropped.

[ROCm/clr commit: ac18a3b701]
2018-08-03 08:48:56 +05:30
Maneesh Gupta 0f033ca572 Merge pull request #603 from b-sumner/master
Drop use of NVCC_COMPAT since it is always set

[ROCm/clr commit: b485957858]
2018-08-03 08:45:18 +05:30
Maneesh Gupta 044a73015e Merge pull request #601 from sarunyap/fix-shareWithAll-mem-dev-mapping
Remove device mapping from shareWithAll memory

[ROCm/clr commit: ad44a77ed4]
2018-08-03 08:45:04 +05:30
foreman 8819b1de20 P4 to Git Change 1589087 by skudchad@skudchad_rocm on 2018/08/02 20:10:02
SWDEV-145570 - [HIP] - hipHccModuleLaunchKernel needs to be a c++ sym

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.map.in#9 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#14 edit


[ROCm/clr commit: 678764d113]
2018-08-02 20:20:23 -04:00
foreman 934eca953a P4 to Git Change 1589062 by cpaquot@cpaquot-ocl-lc-lnx on 2018/08/02 18:34:20
SWDEV-145570 - [HIP] Fix a segfault exposed by rocBlas.

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#39 edit


[ROCm/clr commit: 08a308dd59]
2018-08-02 18:52:37 -04:00
foreman 650747dfe6 P4 to Git Change 1589030 by cpaquot@cpaquot-ocl-lc-lnx on 2018/08/02 17:40:15
SWDEV-145570 - [HIP] Fixed the underline

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#15 edit


[ROCm/clr commit: 388d597321]
2018-08-02 18:00:05 -04:00
foreman 8ca0447f6e P4 to Git Change 1588779 by cpaquot@cpaquot-ocl-lc-lnx on 2018/08/02 12:25:51
SWDEV-145570 - [HIP] Refactored some g_* stuff
	Refactored g_functions into a platform state.
	Added a _vars for registered variables.
	Added an execution stack similar to Hcc-clang.

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#14 edit


[ROCm/clr commit: 2ff260c727]
2018-08-02 12:33:55 -04:00
Rahul Garg 22d59ba040 Remove adipose extension from genco output
[ROCm/clr commit: e9d7584b02]
2018-08-02 15:16:36 +05:30
foreman d9ab745d5b P4 to Git Change 1588573 by chui@ocl-promo-incrementor on 2018/08/02 02:56:34
SWDEV-2 - Change OpenCL version number from 2707 to 2708.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2455 edit


[ROCm/clr commit: 7d877e1589]
2018-08-02 03:12:11 -04:00
foreman 68f06e9f60 P4 to Git Change 1588556 by skudchad@skudchad_rocm on 2018/08/02 02:04:15
SWDEV-145570 - [HIP] - Release eStop if it not a nullptr.

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#13 edit


[ROCm/clr commit: e638c1377f]
2018-08-02 02:18:57 -04:00
foreman a3fc17658c P4 to Git Change 1588551 by skudchad@skudchad_rocm on 2018/08/02 01:55:45
SWDEV-145570 - [HIP] - Refactor ihipModuleLaunchKernel and add hipModuleLaunchKernelExt

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.def.in#8 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_hcc.map.in#8 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#12 edit


[ROCm/clr commit: 11a91fbe0b]
2018-08-02 02:14:04 -04:00
Maneesh Gupta 1039c7952c Merge pull request #610 from aaronenyeshi/enable-hcc-module-launch
Enable hipHccModuleLaunchKernel on HIP-Clang

[ROCm/clr commit: 0354367a07]
2018-08-02 08:08:46 +05:30
Maneesh Gupta 6de7fd4934 Merge pull request #600 from ROCm-Developer-Tools/enable-texture-functions
Enable HIP texture functions with HIP vector types

[ROCm/clr commit: 4544e681ac]
2018-08-02 08:01:00 +05:30
Maneesh Gupta f9dbc19d51 Merge pull request #593 from mangupta/patch2-removerdynamic
[FindHIP] Remove rdynamic from CMAKE_SHARED_LIBRARY_LINK_HIP_FLAGS

[ROCm/clr commit: 1692af19cc]
2018-08-02 07:58:34 +05:30
Maneesh Gupta c440c4d3e8 Merge pull request #514 from luckynikki/GETandALGN
[GETandALGN]: Get channel and alignment details

[ROCm/clr commit: 2960739c43]
2018-08-02 07:57:50 +05:30
Yaxun Sam Liu b8bbc4f1fa Let hipcc handle static library for hip-clang
[ROCm/clr commit: 2ba45a23b8]
2018-08-01 22:03:01 -04:00
Aaron Enye Shi da60937c5d Enable HIP_COMPILER to choose HIP-Clang lib links
[ROCm/clr commit: 7560956233]
2018-08-01 20:37:46 +00:00
Yaxun Sam Liu b587661200 Add hc_get_workitem_absolute_id for hip-clang HCC compatibility mode
[ROCm/clr commit: 9c28758fa0]
2018-08-01 15:14:38 -04:00
Yaxun Sam Liu 2d74df6f34 Add HCC compatibility mode
[ROCm/clr commit: b91691966f]
2018-08-01 15:14:38 -04:00
foreman aea58b76ad P4 to Git Change 1588199 by gandryey@gera-w8 on 2018/08/01 11:53:29
SWDEV-155434 - Add SQTT instrumentation tokens for OpenCL dispatches for RGP support
	- Fix the build.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.cpp#119 edit


[ROCm/clr commit: 9fb27ccc8b]
2018-08-01 12:30:25 -04:00
foreman 7664c89d3e P4 to Git Change 1588185 by vsytchen@vsytchen-win10 on 2018/08/01 11:46:00
SWDEV-145570 - [HIP] Correctly handle calculating the offset when retrieving the amd::Memory object from a prepinned pointer

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/15539/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#38 edit


[ROCm/clr commit: 85777ea939]
2018-08-01 12:08:35 -04:00
foreman 149e3587ed P4 to Git Change 1588181 by gandryey@gera-w8 on 2018/08/01 11:39:16
SWDEV-155434 - Add SQTT instrumentation tokens for OpenCL dispatches for RGP support
	- Switch to the workgroup size report for the dispatch info.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.cpp#118 edit


[ROCm/clr commit: 19968ab78a]
2018-08-01 11:53:18 -04:00
Evgeny Mankov 9ffec20ff2 [HIPIFY][docs] Formatting
[ROCm/clr commit: 55ee734b22]
2018-08-01 17:29:00 +03:00
foreman d0eaa26e88 P4 to Git Change 1588043 by chui@ocl-promo-incrementor on 2018/08/01 02:56:59
SWDEV-2 - Change OpenCL version number from 2706 to 2707.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2454 edit


[ROCm/clr commit: 37677e98a5]
2018-08-01 03:23:36 -04:00
lthakur 313cdfd494 Added offset value verification.
Added offset value verification.

[ROCm/clr commit: b7a4b2c300]
2018-08-01 10:41:30 +05:30
Aaron Enye Shi 83d293a2a4 Enable hipHccModuleLaunchKernel on HIP-Clang
[ROCm/clr commit: 0ad36dfa58]
2018-07-31 20:38:20 +00:00
Evgeny Mankov 66e24cb0d3 [HIPIFY][FFT] Add cuFFT support
+ Add API_FFT support (data types and functions)
+ Add CUFFT_API_supported_by_HIP.md
+ Add simple_cufft.cu test
+ Update README.md


[ROCm/clr commit: c57bfd50e3]
2018-07-31 22:25:48 +03:00
foreman 50b6d51a72 P4 to Git Change 1587799 by vsytchen@vsytchen-win10 on 2018/07/31 14:47:31
SWDEV-145570 - [HIP] refactor hipMemcpy* functions to correctly handle copies using prepinned memory

	The current implementation of hipMemcpy functions picks the copy type based on a flag that the user passes. However, one can use the hipMemcpyHostToDevice/hipMemcpyDeviceToHost flag in a combination with prepinned memory. By using the WriteMemoryCommand/ReadMemoyCommand in this case, we will pin the same host memory twice. This is fine on PAL/Linux, since pinning the same VA range is a noop, but this will start failing once we switch to using device memory with HIP/VDI/HSA.

	The solution is to ignore the hipMemcpyKind flag and let the runtime decide what kind of copy is best to do. Except for the case when hipMemcpyHostToHost is passed, since both host pointers may be prepinned.

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/15482/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#37 edit


[ROCm/clr commit: b02040e934]
2018-07-31 15:07:56 -04:00
Rahul Garg 17e35e382c Added support for pinned 2D SDMA copy
[ROCm/clr commit: 5eb11b58f3]
2018-07-31 14:05:35 +05:30
foreman 6a05f7fd71 P4 to Git Change 1587463 by chui@ocl-promo-incrementor on 2018/07/31 02:56:47
SWDEV-2 - Change OpenCL version number from 2705 to 2706.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2453 edit


[ROCm/clr commit: ce73c861d6]
2018-07-31 03:15:02 -04:00