Commit Graph

13557 Commits

Author SHA1 Message Date
Ben Sander d666fbaafe Add option to deny peer access.
Also fix test.

Change-Id: I1b247f6c4271442b008e560669bca4daf8eb94c7
2016-11-10 23:12:48 -06:00
Ben Sander 6e54a600b6 Use forceUnpinnedCopy to resolve P2p corner cases.
Change-Id: I2aebb419881246cebb696bec87798635bc71acc2
2016-11-10 23:12:48 -06:00
Ben Sander 0eeaa3bcd5 Enable async copy again.
Also add HIP_FORCE_SYNC_COPY chicken bit.

Change-Id: I76a385410494b99bf27305d3c08f55dd81987565
2016-11-10 23:12:48 -06:00
Ben Sander 85e65b55ff Doc change only - add comments to test.
Change-Id: Ie42087cf3c78e49337b18bb71f3f0e1e7950ee1b
2016-11-10 23:12:48 -06:00
Ben Sander e767e0032e Refactor copy and P2P logic.
Prefer use of source-engine for DMA copies, even if user submits copy
in a stream attached to a different device.
The stream is now used only for synchronization, and HIP
makes the most optimal decision for which engine to perform the
copy - typically the source copy engine.

HIP now makes decision on which engine should perform the copy
and passes this to HCC using new apis.
HIP has additional information about peer
visibility and will make a decision which agent should perform
the copy .

Change-Id: I0cf4cfebeae256e6ca795f08a7ed7130f4857d1f
2016-11-10 23:12:48 -06:00
Ben Sander e9835617f1 Improve memory debug
Change-Id: I0f033139aa4e4b47039eb016e404009127bd0a44
2016-11-10 23:12:48 -06:00
foreman 90d4cf8e78 P4 to Git Change 1339888 by lmoriche@lmoriche_opencl_dev on 2016/11/10 21:21:17
SWDEV-102510 - Need a way to control cl_khr/cl_amd extension macros
	- Use -cl-ext option to enable OpenCL extensions

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palcompiler.cpp#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#23 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#44 edit
2016-11-10 21:30:26 -05:00
foreman 30a6048658 P4 to Git Change 1339730 by lmoriche@lmoriche_opencl_dev on 2016/11/10 16:37:37
SWDEV-103793 - [ROCm CQE][OCLonLC][ISV] QCDGPU test fails
	- The runtime should not report cl_amd_fp64, cl_khr_fp64 should be used instead.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palsettings.cpp#10 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#9 edit
2016-11-10 17:04:27 -05:00
pensun 9aa2269d5c Update depreciated information for threadfence_system()
Change-Id: Id13d2f81edb51eb42b896a5c06913d59ec907c55
2016-11-10 11:55:12 -06:00
foreman a9885c5273 P4 to Git Change 1339431 by gandryey@gera-w8 on 2016/11/10 10:47:43
SWDEV-86035 - Add PAL backend to OpenCL
	- Fix a regression after LC path integration

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.cpp#24 edit
2016-11-10 11:02:24 -05:00
Maneesh Gupta cdcf04d744 CMakeLists.txt: Cascade CMAKE_BUILD_TYPE to tests
Change-Id: I53a3ea951c1fd57e43a02381a457c1dedc1a34f7
2016-11-10 21:26:34 +05:30
foreman 655d43d311 P4 to Git Change 1339294 by johtaylo@johtaylo-jtincrementor-increment on 2016/11/10 03:00:04
SWDEV-2 - Change OpenCL version number from 2270 to 2271.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2017 edit
2016-11-10 03:13:54 -05:00
Rahul Garg f86c7b5b3c hipDeviceGetByPCIBusId support for HIP/NVCC
Change-Id: I8f82890e88d2a15f592bff192179e7d5c5362722
2016-11-10 11:40:59 +05:30
Maneesh Gupta 462ffb8117 hipcc: Default to HIP_LIB_TYPE=1
Change-Id: I83b05accd76f7bc94bd724c66ae060fa0095bc8d
2016-11-10 11:34:00 +05:30
Maneesh Gupta 052a580d5b hcc_dialects/Makefile: use clamp-config
Change-Id: I86df82f75b75125825e22d0545209a19386d9936
2016-11-10 11:31:50 +05:30
pensun bbb619c732 resolve conflicts for git pull
Change-Id: Ie353b831e2241bc28042069b6cc7405257e871e1
2016-11-09 21:38:43 -06:00
pensun 94dfff9db2 Add documentation on threadfence_system workaround guidelines.
Change-Id: I9636a3808798f3dabe992285ce5652187cee6eb8
2016-11-09 21:36:30 -06:00
pensun 23de0e1b50 Add option to alloc fingrained system memory
Change-Id: Ia13c8e058cb988b5857e75a590a4d67411362ae1
2016-11-09 21:36:30 -06:00
Maneesh Gupta e3b5eef7c9 Merge branch 'rocm-rel-1.3' into amd-develop
Conflicts:
	include/hip/nvcc_detail/hip_runtime_api.h

Change-Id: I990a7d008da9e8dcc68250cebbc8ee6e723c7e01
2016-11-10 08:56:38 +05:30
foreman 1f5d19b75b P4 to Git Change 1339111 by lmoriche@lmoriche_opencl_dev on 2016/11/09 19:34:29
SWDEV-95488 - Limiting max cap Context on ROCm, it is limited to 100
	- Re-order the member variables to match the initializer list.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#8 edit
2016-11-09 19:58:26 -05:00
foreman 672d37d9b7 P4 to Git Change 1339076 by scharupa@scharupa_temp on 2016/11/09 18:21:08
SWDEV-95488 -  Replacing gpuDevice_ with roc_device_

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#18 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#5 edit
2016-11-09 18:36:38 -05:00
foreman 18f7cb7b5a P4 to Git Change 1339014 by scharupa@scharupa_temp on 2016/11/09 17:18:01
SWDEV-95488 - Fixing build issues

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#17 edit
2016-11-09 17:50:11 -05:00
foreman d833468ea5 P4 to Git Change 1338994 by scharupa@scharupa_temp on 2016/11/09 16:52:17
SWDEV-95488 - fixing build issues introduced at 1338972

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#7 edit
2016-11-09 17:30:22 -05:00
foreman 2d64dbab44 P4 to Git Change 1338972 by scharupa@scharupa_temp on 2016/11/09 16:33:58
SWDEV-95488 - Limiting max cap Context on ROCm, it is limited to 100

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#204 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#279 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#25 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#7 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#6 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#16 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#4 edit
2016-11-09 17:07:33 -05:00
pensun 4a8a6a4697 fix hipProfiler* apis on NV path
Change-Id: I6adca6151fef3a9b35348163eb6bd13f5c414172
2016-11-09 15:44:01 -06:00
pensun e5277ab4b6 fix for hipcallback function on NV path
Change-Id: If80c0cfe60b1f3b1a71627b5f3f79503cba4d491
2016-11-09 11:33:23 -06:00
foreman 621f70b372 P4 to Git Change 1338735 by gandryey@gera-w8 on 2016/11/09 10:46:19
SWDEV-86035 - Add PAL backend to OpenCL
	- Enable split logic in PAL

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palsettings.cpp#9 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.cpp#38 edit
2016-11-09 10:55:17 -05:00
foreman 0428dca9f2 P4 to Git Change 1338576 by johtaylo@johtaylo-jtincrementor-increment on 2016/11/09 03:00:05
SWDEV-2 - Change OpenCL version number from 2269 to 2270.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2016 edit
2016-11-09 03:12:34 -05:00
foreman 94bf561bd7 P4 to Git Change 1338072 by wchau@wchau_WIN_OCL_HSA on 2016/11/08 10:02:39
SWDEV-95007 - OpenCL on PAL - DVR

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.hpp#14 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevicegl.cpp#3 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palresource.cpp#18 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palresource.hpp#9 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.cpp#37 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.hpp#16 edit
2016-11-08 10:11:21 -05:00
Maneesh Gupta 01a38c82a6 Update release notes for 1.0 release
Change-Id: I74fa2b41afc334a76c309b125c27aa141cd59554
2016-11-08 16:31:56 +05:30
foreman 79459fbd6d P4 to Git Change 1337979 by johtaylo@johtaylo-jtincrementor-increment on 2016/11/08 03:00:04
SWDEV-2 - Change OpenCL version number from 2268 to 2269.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2015 edit
2016-11-08 03:12:47 -05:00
foreman 12242ad8de P4 to Git Change 1337788 by asalmanp@asalmanp-opencl-stg on 2016/11/07 17:39:47
SWDEV-95903 - Fixing a regression in Conf RQ on Vega10 introduced by CL#1333973. The SVM size was changed from 4096 MB to 16384 MB in this CL and causing a memory allocation failure (out of VA space) when running conformance (test_basic). Change the size back to 4096 MB for now to unblock testing Vega10 with OCL ToT.
	ReviewBoardURL = http://ocltc.amd.com/reviews/r/11775/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#260 edit
2016-11-07 17:49:07 -05:00
foreman 7f539761ca P4 to Git Change 1337771 by gandryey@gera-w8 on 2016/11/07 17:16:58
SWDEV-106509 - [CQE OCL][OpenCL on PAL][QR] 3 Conformance tests are failing/crashing
	- Add missing constant argument into the qualifiers lookup
	- Initialize rtCU count to the internal constant

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#33 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.cpp#23 edit
2016-11-07 17:32:56 -05:00
foreman 3459794cb3 P4 to Git Change 1337394 by johtaylo@johtaylo-jtincrementor-increment on 2016/11/07 03:00:04
SWDEV-2 - Change OpenCL version number from 2267 to 2268.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2014 edit
2016-11-07 03:08:11 -05:00
Ben Sander a13ec441bf Fix tests to read warpSize from device props.
Change-Id: I9583577793afad49f9eb1ee9069bd4c6963a6023
2016-11-06 04:26:28 -06:00
Ben Sander 0e5cfed3eb Update gitignore for some common output files
Change-Id: I9cd60f042af4dba07fe0fdbd2ee442936ff8c7bd
2016-11-06 04:26:15 -06:00
Ben Sander f3d38c2615 Improve Peer support and testing.
Change-Id: Icadc65988aaf145a265587ab0357c5bf4d26f3eb
2016-11-06 03:22:36 -06:00
foreman 4d705d6ca0 P4 to Git Change 1337201 by johtaylo@johtaylo-jtincrementor-increment on 2016/11/06 02:00:04
SWDEV-2 - Change OpenCL version number from 2266 to 2267.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2013 edit
2016-11-06 02:06:11 -05:00
foreman 0d9068fa5a P4 to Git Change 1337110 by johtaylo@johtaylo-jtincrementor-increment on 2016/11/05 03:00:03
SWDEV-2 - Change OpenCL version number from 2265 to 2266.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2012 edit
2016-11-05 03:11:03 -04:00
Ben Sander 06ecfa3975 Set forceHostCopyEngine for other copy dirs. Support HIP_FORCE_P2P_HOST
Also: more debug for copy and P2p.

Change-Id: I87030c525410e041b2a00baaf6c68e6c0977ff42
2016-11-04 19:53:23 -05:00
foreman b8040b3d6a P4 to Git Change 1337009 by gandryey@gera-w8 on 2016/11/04 18:04:52
SWDEV-79445 - OCL generic changes and code clean-up
	- Use uint64_t for the scratch buffer size calculation

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#558 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.hpp#162 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#175 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.h#63 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#32 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.hpp#13 edit
2016-11-04 18:18:13 -04:00
Ben Sander 60a8a5405d Expand hipP2PSimple testing.
Cover cases where P2P is used for H2D copies, where host is pinned
but not accessible to the copy agent.

Change-Id: I9464b787228b40f93473708c3fde9726e1986365
2016-11-04 16:13:32 -05:00
Ben Sander 926e63c655 Refactor resolve-mem step1
Change-Id: I7b8b2bbb56d7b31a97b48ebd42002641cd07a460
2016-11-04 09:37:56 -05:00
Ben Sander 00276d141e Add debug for Peer APIs. Enable PeerMemcpy APIs by default.
Change-Id: I46e39a9e7b07686a78484c1f3b5495b08e052fbb
2016-11-04 08:51:16 -05:00
Ben Sander 44aee4b61c Print non-peers too
Change-Id: I2a6905edcdf144aa732ae3120c17780477f232ac
2016-11-04 06:34:07 -05:00
Ben Sander d1db786910 Pre-pend HIP_PATH/lib to linker, so we find developer object code
Previously might pick up libs from /opt/rocm/lib.

Change-Id: Ia7adb345defe433d5952aa61706fe03fd7cbcd35
2016-11-04 06:06:04 -05:00
foreman 0794d6b50b P4 to Git Change 1336502 by johtaylo@johtaylo-jtincrementor-increment on 2016/11/04 03:00:03
SWDEV-2 - Change OpenCL version number from 2264 to 2265.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2011 edit
2016-11-04 02:06:26 -05:00
foreman 065133bbc0 P4 to Git Change 1336353 by lmoriche@lmoriche_opencl_dev on 2016/11/03 18:47:50
SWDEV-105604 - [OCL-LC-PAL] OpenCL program manager for LC on PAL
	- Finish the implementation of the program manager to support separare compilation and binary loading.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#22 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.hpp#13 edit
2016-11-03 18:56:31 -04:00
foreman 30b62e264c P4 to Git Change 1336294 by gandryey@gera-w8 on 2016/11/03 17:50:35
SWDEV-86035 - Add PAL backend to OpenCL
	- Use size() method in the segment freeze instead of vmSize(), because vmSize() is aligned in PAL and different heaps may have different alignments

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#21 edit
2016-11-03 18:14:18 -04:00
foreman 4d82398e1f P4 to Git Change 1336110 by gandryey@gera-w8 on 2016/11/03 15:21:09
SWDEV-86035 - Add PAL backend to OpenCL
	- Update printf initialization to match the GSL backend

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.cpp#22 edit
2016-11-03 15:41:39 -04:00