AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
155 Commits 3 Branches 0 Tags
11ca71bd76b93204b1382509379dd8cf16ac2e2f
Commit Graph

56 Commits

Author SHA1 Message Date
pensun 39b44cb484 Test cases for HIP_VISIBLE_DEVICES/CUDA_VISIBLE_DEVICES. 
hipEnvVar is the base test case, to be called by hipEnvVarDriver
at the run time.
Test case includes tests for normal use case of the environment
variable, invalid value/sequence and use CUDA_VISIBLE_DEVICES as a
alternative.
 2016-02-27 14:48:00 -06:00
pensun 1f606261c1 improve the HIP_VISIBLE_DEVICES implementation 2016-02-27 14:14:08 -06:00
pensun 07e56d4666 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 2016-02-27 04:25:28 -06:00
Aditya Avinash Atluri ecadb1623c Merge pull request #4 from AMDComputeLibraries/memtracker 
hipGetPointerAttrib behavioral changes
 2016-02-27 10:51:23 -06:00
Ben Sander ea09557e1b disable rocrv2, properly 2016-02-27 03:31:30 -06:00
Aditya Avinash Atluri 66aa7f2f8a Corrected hipPointerGetAttribute 
Made hipPointerGetAttribute work same as cudaPointerGetAttribute for HCC
 2016-02-26 18:50:40 -06:00
pensun 57f60b34fb relsove conflicts 2016-02-26 09:57:40 -06:00
pensun ee7ac16396 fix compiling error 2016-02-26 09:50:00 -06:00
Ben Sander ff66ef0779 fixes for titan platform 2016-02-26 05:25:30 -06:00
Ben Sander 6e0ccdfb95 Disable ROCR_V2 2016-02-26 23:34:45 -06:00
Ben Sander 369e0d7b5b Merge branch 'memtracker' into privatestaging 
Conflicts:
	include/nvcc_detail/hip_runtime_api.h
 2016-02-26 06:17:05 -06:00
Ben Sander c300ffe458 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 2016-02-26 06:15:09 -06:00
Ben Sander d319299ddb Merge branch 'memtracker' of https://github.com/AMDComputeLibraries/HIP-privatestaging into memtracker 
Conflicts:
	tests/src/hipMemcpy.cpp
 2016-02-25 23:22:51 -06:00
Ben Sander 4adab7b7ef Merge branch 'memtracker' into privatestaging 
Conflicts:
	src/hip_hcc.cpp
 2016-02-25 19:38:46 -06:00
Ben Sander 8b64c0dc62 Improve memory copy and commands switching 
- Add chicken bits to use host-side dependency management.
- Add optional PinInPlace path for unpinned copies
- Synchronize before pinned memcpy path.
- Add mutex to protect two threads launching to same stream.
 2016-02-25 19:19:49 -06:00
Evgeny Mankov 57e212606d Attribute hipDeviceAttributeIsMultiGpuBoard for obtaining Device property isMultiGpuBoard is added. 
On HIP path property obtaining done through hsa_iterate_agents and counting the devices of HSA_DEVICE_TYPE_GPU type.

P.S.
On multi-boards systems it might be problems with detection what board a GPU plugged into (not tested).
 2016-02-25 23:44:39 +03:00
Ben Sander 7090f5c3f9 Add tests for multi-threaded streams 2016-02-23 12:08:22 -06:00
Ben Sander 3886d494f4 Sync review. 
- add calls to ihipInit missing from some routines.
- sync before draining a stream.
 2016-02-23 04:07:11 -06:00
Ben Sander 549b18ce77 Improve async copy implementation. 
- Add device-side signal waits when transitioning between command classes
(Kernel, H2D copy, D2H copy).
- Support waiting in staged memory copies as well.
- Add several chicken bits to control implementation:
    - HIP_DISABLE_ENQ_BARRIER
    - HIP_DISABLE_BIDIR_MEMCPY
    - HIP_ONESHOT_COPY_DEP
- Refactor signal pool to support efficient deallocation based on
signsequnm.
- Deallocate copy signals on eventSynchronize.
- Improve copy tests, add pingpong.
 2016-02-22 23:15:24 -06:00
Ben Sander 0a98db4b5f Merge branch 'memtracker' of https://github.com/AMDComputeLibraries/HIP-privatestaging into memtracker 2016-02-22 08:33:47 -06:00
gargrahul a2fbf06129 Update for shared atomics support 2016-02-22 16:21:52 +05:30
Ben Sander d33d806a5b Track last command to a stream. 
Passing simple tests.
 2016-02-20 11:02:07 -06:00
Evgeny Mankov 833c9e52ad Guard #ifdef USE_ROCR_20 is added for ROCR_20 device properties (memoryClockRate, memoryBusWidth) 
By default isn't defined.
To add ROCR_20 support HIP have to be compiled as follows: make CXX_DEFINES+=-DUSE_ROCR_20
 2016-02-19 13:27:03 +03:00
Evgeny Mankov fbdeee39cd Formatting, no functional changes. 2016-02-18 18:54:19 +03:00
Evgeny Mankov 1c19dbb807 Device property memoryBusWidth implementation. 
+ Device property memoryBusWidth is added to hipDeviceProp_t struct.
+ Device attribute hipDeviceAttributeMemoryBusWidth is added to hipDeviceAttribute_t struct.
+ Tests update.
 2016-02-18 18:15:01 +03:00
Evgeny Mankov 5ea8543d2e Device property memoryClockRate implementation. 
+ Device property memoryClockRate is added to hipDeviceProp_t struct.
+ Device attribute hipDeviceAttributeMemoryClockRate is added to hipDeviceAttribute_t struct.
+ Tests update.
+ Rename hipDevAttrConcurrentKernels to hipDeviceAttributeConcurrentKernels.
 2016-02-18 17:25:28 +03:00
Evgeny Mankov 2b6fda77ca Attribute hipDevAttrConcurrentKernels for obtaining Device property concurrentKernels is added. 2016-02-18 14:34:18 +03:00
Ben Sander c6f8883b0d Enable Tracker and ROCR by default, verify with HCC 2016-02-17 23:03:37 -06:00
Ben Sander d653782d9d Remove HIP-local AM tracker (now in HCC) 2016-02-17 21:33:32 -06:00
Ben Sander 44f40e171a USE_AM_TRACKER=0 works 2016-02-17 21:23:36 -06:00
pensun 8aa4bfce57 1. Bug fix 
2. passed initial tests on different sets of
HIP_VISIBLE_DEVICES: (0),(1),(0,1),(1,2),(2,3),(1,2,3),(2,3,4),(1,5,2,3)
and achieved expected choice of GPU devices at the runtime.
3. Passed HIP test suite.
 2016-02-17 09:32:50 -06:00
pensun c1e120fb1b Implementation of HIP_VISIBLE_DEVICES in runtime 2016-02-17 06:59:18 -06:00
Ben Sander 59379ffb44 more work on async copies 2016-02-17 00:59:12 -06:00
pensun 060439b6ab modify to add remove invalid devices numbers 2016-02-16 10:00:05 -06:00
pensun d40cbef2af Implement to read HIP_VISIBLE_DEVICES to internal global variable 2016-02-16 07:39:04 -06:00
Ben Sander caef9b5ced Add per-stream pool for hsa_signals. 2016-02-16 01:59:13 -06:00
Ben Sander 38c735fd1d Update before checkin to HCC. 
Add support for USE_AM_TRACKER=2 (HCC version).
Add AM_ALLOC, AM_FREE indirection to ease swapping AM implementations.
 2016-02-15 21:16:00 -06:00
Ben Sander db3a63360b Move warpSize to header, have shuffles use default warpsize. 2016-02-15 05:41:09 -06:00
Ben Sander 4637e19da4 Update docs, cleanup 2016-02-15 05:40:12 -06:00
Ben Sander 6420655dc8 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 b314777bc1 D2H multi-buffer 2016-02-13 01:15:23 -06:00
Ben Sander 1bfd3cdbd0 Improve copy testing 2016-02-12 18:24:08 -06:00
Ben Sander 134d7975ce 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 24c1fdb864 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 d7396b5af3 Query tracked memory sizes. 
Support more accurate hipMemGetInfo.  Add test to hipPointerAttrib.
 2016-02-12 18:24:08 -06:00
Ben Sander 0370cd1cfc Remove ! USE_PINNED_HOST support 2016-02-12 18:24:08 -06:00
Ben Sander 00fd172c64 Use memtracker 'appID' to store deviceID associated with ptr 2016-02-12 18:24:08 -06:00
Ben Sander de45e2291e 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 4ee2a5229b 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 e483eea85b Fix bug in device bounds comparison. 
Shows up in multi-GPU.
 2016-02-12 18:24:08 -06:00