Grafik Komit

13197 Melakukan

Penulis SHA1 Pesan Tanggal
Ben Sander 8939b4f0e5 Add multi-threading synchonization on staging buffers and signals.
Also pre-allocate a couple signals for copies.
2016-02-13 03:18:01 -06:00
Ben Sander 56b3d2e7c4 Enable -O3, style points on array size 2016-02-13 03:17:42 -06:00
foreman 4ae7b701de P4 to Git Change 1236970 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/02/13 03:00:12
SWDEV-2 - Change OpenCL version number from 2034 to 2035.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1781 edit
2016-02-13 03:09:10 -05:00
Ben Sander a002833a89 D2H multi-buffer 2016-02-13 01:15:23 -06:00
Ben Sander 8e3dd664eb Result formatting 2016-02-13 01:14:01 -06:00
Ben Sander 1cd35f6371 Add D2H test 2016-02-12 22:47:26 -06:00
Ben Sander 81c275da13 Add D2H test 2016-02-12 22:46:34 -06:00
Ben Sander ba7497b409 Add Bus Bandwidth test, leveraged from SHOC. 2016-02-12 21:30:43 -06:00
Ben Sander 2353cbb028 Improve copy testing 2016-02-12 18:24:08 -06:00
Ben Sander 1128610801 Improve copy testing implementation.
- add tests for (unpinned/pinned) x H2H x D2D.
- Free memory at end of test.
2016-02-12 18:24:08 -06:00
Ben Sander 90af462b85 Step1 in staging buffer copy.
- use StagingBuffer class for copies.
- refactor g_device to use array rather than vector.
   (keeps pointers from moving).
2016-02-12 18:24:08 -06:00
Ben Sander f464cedcf4 Query tracked memory sizes.
Support more accurate hipMemGetInfo.  Add test to hipPointerAttrib.
2016-02-12 18:24:08 -06:00
Ben Sander f2c1bf3bc0 Remove ! USE_PINNED_HOST support 2016-02-12 18:24:08 -06:00
Ben Sander c04b5d3afb Use memtracker 'appID' to store deviceID associated with ptr 2016-02-12 18:24:08 -06:00
Ben Sander 7216727fba Tracker improvements
- add API to add / remove user-pointers from the tracker.
- test for thread-safety with MultiThreadtest_2 - rapid
  insertions/removal.
- add mutex to provide thread-safety.
- rename tracker interface to "memtracker_..." for consistency.
- add am_memtracker_reset, connect to hipDeviceReset.
-
2016-02-12 18:24:08 -06:00
Ben Sander 721508cc2f Create address tracker for am_alloc.
Tracks device where memory is allocated, pinned-host or device, and
more.

Uses memory-range-based lookups - so pointers that exist anywhere in

the range of hostPtr + size will find the associated AmPointerInfo.

The insertions and lookups use a self-balancing binary tree and
should support O(logN) lookup speed.
2016-02-12 18:24:08 -06:00
Ben Sander f1bc9af294 Fix bug in device bounds comparison.
Shows up in multi-GPU.
2016-02-12 18:24:08 -06:00
Maneesh Gupta c82511258c Documented supported fastmath functions 2016-02-12 14:21:58 +05:30
Maneesh Gupta 2659e70d48 Updated integer intrinsics documentation 2016-02-12 13:58:35 +05:30
foreman 0b9a71219a P4 to Git Change 1236626 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/02/12 03:00:10
SWDEV-2 - Change OpenCL version number from 2033 to 2034.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1780 edit
2016-02-12 03:09:11 -05:00
foreman fffbe5bb3b P4 to Git Change 1236618 by weizhang@msdnweizhang-opencl on 2016/02/11 22:12:34
SWDEV-78299 - Back out changelist 1236441 since OCLCreateBuffer fails.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#541 edit
2016-02-11 22:20:56 -05:00
Evgeny Mankov 460b501cbb Fix typo: maxThreadsPerMultiProcessor -> MaxSharedMemoryPerMultiprocessor
Device property MaxSharedMemoryPerMultiprocessor set equal to totalGlobalMem (HIP path).
Reason: MaxSharedMemoryPerMultiprocessor should be as the same as group memory size. Group memory will not be paged out, so, the physical memory size = total shared memory size = group region size. NVCC path remains untouched: CUDA's device property MaxSharedMemoryPerMultiprocessor is reported.

hipify is updated as well.
2016-02-12 01:29:20 +03:00
Evgeny Mankov 1025341300 Device property maxThreadsPerMultiProcessor set equal to totalGlobalMem (HIP path).
Reason: maxThreadsPerMultiProcessor should be as the same as group memory size. Group memory will not be paged out, so, the physical memory size = total shared memory size = group region size.

NVCC path remains untouched: CUDA's device property maxThreadsPerMultiProcessor is reported.
2016-02-12 00:04:14 +03:00
Evgeny Mankov 658e9f0484 BDFID (BusID/DeviceID/FunctionID) support.
Except FunctionID (or DomainID in CUDA) support, because cudaDeviceProp::pciDomainID is not reported by CUDA.
2016-02-11 22:26:01 +03:00
foreman 31c030ff18 P4 to Git Change 1236441 by weizhang@weizhang-lnx-opencl on 2016/02/11 12:33:49
SWDEV-78299 - [Brahma] Setting max single allocation size by comparing cardMemAvailableBytes with cardExtMemAvailableBytes on Brahma.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#540 edit
2016-02-11 12:41:22 -05:00
sunway513 fe1000df17 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 2016-02-11 22:22:47 +05:30
sunway513 c7cbcfa2e9 Add reminder to keep ROCR runtime on the system library path 2016-02-11 22:22:00 +05:30
foreman c16578eb71 P4 to Git Change 1236419 by asalmanp@asalmanp-opencl-stg on 2016/02/11 11:36:16
SWDEV-79278 - [OpenCL][Ellesmere BU] Exposing Ellesmere in mainline for 16.20 release

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/9750/

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings.h#7 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings_amdil.h#7 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings_amdil64.h#6 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings_hsail.h#6 edit
... //depot/stg/opencl/drivers/opencl/compiler/legacy-lib/utils/v0_8/target_mappings_hsail64.h#6 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings.h#39 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_amdil.h#34 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_amdil64.h#30 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail.h#29 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail64.h#24 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm32/lib/Target/AMDIL/AMDILSubtarget.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm32/lib/Target/AMDIL/Internal/InternalProcessors.td#8 edit
... //depot/stg/opencl/drivers/opencl/compiler/tools/driver/driver.cpp#59 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudefs.hpp#133 edit
... //depot/stg/opencl/drivers/opencl/tests/ocltst/module/runtime/OCLDeviceQueries.cpp#33 edit
2016-02-11 11:49:08 -05:00
dfukalov bae0f00e69 adding ability to build in llvm source tree, updated README 2016-02-11 15:27:00 +03:00
foreman 837ab19841 P4 to Git Change 1236333 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/02/11 03:00:10
SWDEV-2 - Change OpenCL version number from 2032 to 2033.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1779 edit
2016-02-11 03:08:59 -05:00
Maneesh Gupta ed2d86f3a9 Updated readme for test 2016-02-11 13:06:58 +05:30
dfukalov 9d10ae4325 Initial version of CLANG based HIPIFY tool for CUDA -> HIP sources conversion 2016-02-10 20:20:11 +03:00
Ben Sander f3a2e03a6b Initial commit 2016-02-10 09:29:29 -06:00
Evgeny Mankov 3139c72756 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 2016-02-10 17:21:53 +03:00
Evgeny Mankov d9a94191f2 Formatting, no functional changes 2016-02-10 17:21:18 +03:00
streamhsa 90add185fd Remove test for atomicInc and atomicDec 2016-02-10 21:02:52 +08:00
streamhsa 56f1832e70 Updated readme for test 2016-02-10 20:05:59 +08:00
streamhsa 2f8d56e903 Resolved test issues 2016-02-10 20:01:16 +08:00
foreman 4a8a26801c P4 to Git Change 1235953 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/02/10 03:00:10
SWDEV-2 - Change OpenCL version number from 2031 to 2032.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1778 edit
2016-02-10 03:10:47 -05:00
gargrahul 51f46d9ddf Removed atomicInc and atomicDec support from HIP 2016-02-10 04:29:55 +05:30
foreman 45715f0c48 P4 to Git Change 1235782 by asalmanp@asalmanp-opencl-stg on 2016/02/09 15:30:16
SWDEV-68792 - [OpenCL][HWS/CWSR] Adding support for Hardware Scheduler and Compute Wave Save restore (CWSR) feature on ORCA

	Adding a temporary w/a for a CP uCode bug in HWS mode. Due to this bug, CP uCode loops through a RUNLIST unless there is a submission on all queues in HWS mode. This causes some overhead and performance drop in PCMark8 on CZ in HWS mode. To work around this issue, it was suggested to submit a dummy packet during initialization on all available queues on HWS mode so that CP uCode can break the loop. This w/a should be removed once CP uCode provides a final fix for this issue.

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/9616/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#160 edit
2016-02-09 17:25:20 -05:00
Peng Sun b61f0453c0 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into doc_update 2016-02-09 15:08:39 -06:00
Peng Sun 6731505c5f Fix TODO-Doc in hip_texture.h 2016-02-09 10:58:23 -06:00
Evgeny Mankov 4d4ca3ef3f Device property concurrentKernels is added to hipDeviceProp_t struct.
For HCC path concurrentKernels is set to true since all ROCR hardware supports this feature.
For NVCC path concurrentKernels is obtained from CUDA's device property cudaDeviceProp::concurrentKernels.
2016-02-09 17:10:35 +03:00
foreman 9b8568bedb P4 to Git Change 1235458 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/02/09 03:00:10
SWDEV-2 - Change OpenCL version number from 2030 to 2031.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1777 edit
2016-02-09 03:09:16 -05:00
foreman c5265e08c6 P4 to Git Change 1235456 by kebai@kebai_win1 on 2016/02/09 02:47:34
SWDEV-86836 - Enhance caching library class to prepare one-stage kernel caching by:
	0. Moving cache storage setup into constructor
	1. Controlling cache storage size
	2. Explicit cache cleanup
	    a. -kcache-wipe is off by default; when turned on, the caching directory would be wiped off
	    b. Here it's just an option. The implementation (the call of wipeCacheFolders()) will be added in the compiler library
	3. Enforcing cache miss (actual compilation enforcing and adding a new entry to the cache storage).
	    a. -kcache-enforce-miss is off by default; when turned on, the real compilation will be enforced
	    b. Here it's just an option. The implementation will be added in the compiler library
	ReviewBoardURL = http://ocltc.amd.com/reviews/r/9726/

Affected files ...

... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/OPTIONS.def#134 edit
... //depot/stg/opencl/drivers/opencl/compiler/tools/caching/cache.cpp#12 edit
... //depot/stg/opencl/drivers/opencl/compiler/tools/caching/cache.hpp#7 edit
2016-02-09 02:52:58 -05:00
Maneesh Gupta f8bfc7f54c which_hip -> hipconfig 2016-02-09 11:51:26 +05:30
Maneesh Gupta 0df78ac9bf Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 2016-02-09 10:57:46 +05:30
Maneesh Gupta f6e7abd710 Move HIP_DEVICE_COMPILE defines to hip_common.h 2016-02-09 10:57:20 +05:30
streamhsa 310023e273 Rename test hipInfo as hipGetDeviceAttribute 2016-02-09 13:19:32 +08:00