AI/rocm-systems
2
0
Förgrening 0
Kod Ärenden Pull-förfrågningar Actions Packages Projekt Släpp Wiki Aktiviteter
198 Incheckningar 3 Grenar 0 Taggar
bf3d3a9745cb4df07afbb2cd6c25175422200921
Commit-Graf

69 Incheckningar

Upphovsman SHA1 Meddelande Datum
Aditya Atluri bf3d3a9745 Revert "Fixed passing unpinned memory to async memcpy" 
This reverts commit c272c53808.


[ROCm/hip commit: 45965aaa2d]
 2016-03-08 03:17:27 -06:00
Aditya Atluri c272c53808 Fixed passing unpinned memory to async memcpy 
[ROCm/hip commit: 216af71480]
 2016-03-08 03:14:49 -06:00
Aditya Atluri 91dbc3114d added feature for hipHostGetFlags for CUDA and HIP 
[ROCm/hip commit: 75952029d6]
 2016-03-06 12:17:30 -06:00
Aditya Atluri 5346c57411 added flags support to be used later 
[ROCm/hip commit: 1f3695b224]
 2016-03-06 08:40:33 -06:00
Aditya Atluri f1b8758919 corrected hipDeviceGetProperties to hipGetDeviceProperties - not docs 
[ROCm/hip commit: d3ba2b9782]
 2016-03-06 08:31:04 -06:00
Aditya Atluri 3c91a6d0a7 Added hipHostAlloc with hipHostAllocMapped flag 
[ROCm/hip commit: 3aa764d5eb]
 2016-03-05 15:57:56 -06:00
Aditya Atluri 52049a5927 v2: Fixed implementation of hipMemcpyDefault flag 
[ROCm/hip commit: bd23279271]
 2016-03-04 06:00:04 -06:00
Aditya Atluri cc9a9cc5db Added support for hipMemcpyDefault 
[ROCm/hip commit: c154e1f4e4]
 2016-03-03 10:30:06 -06:00
Aditya Avinash Atluri 12e7f550b7 Update hip_hcc.cpp 
[ROCm/hip commit: 40eefc1cde]
 2016-03-03 13:59:43 -06:00
Aditya Avinash Atluri 057bcc73f1 Fix output of hipPointerGetAttributes 
The output of hipPointerGetAttributes is fixed to match CUDA counterpart.

[ROCm/hip commit: b6e34a44b0]
 2016-03-03 13:58:18 -06:00
Aditya Atluri 0ed3d943f1 Initialize hip when single kernel is called 
[ROCm/hip commit: ce7ae41d42]
 2016-03-02 08:08:45 -06:00
Aditya Avinash Atluri 5214b314a9 H2H Async memcpy fix 
In this change, the cpu memcpy will wait until all the commands in the current stream are done.
Note that, it only waits on current stream. But not on other streams.

[ROCm/hip commit: 180bc32db0]
 2016-02-29 12:49:50 -06:00
Ben Sander 5fc32bc1e8 Copy dependency bug fixes and test modes. 
Add dependency for host-to-host copy.

Add debug mode for HIP_DISABLE_HW_COPY_DEP and
HIP_DISABLE_HW_KERNEL_DEP - setting these to -1 now ignores
all dependencies.


[ROCm/hip commit: ba9ad6be80]
 2016-02-28 21:19:49 -06:00
Ben Sander cefe65e24a touchup 
[ROCm/hip commit: af22d056e0]
 2016-02-28 21:08:53 -06:00
pensun b9421830a4 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.


[ROCm/hip commit: 39b44cb484]
 2016-02-27 14:48:00 -06:00
pensun 1154872b0d improve the HIP_VISIBLE_DEVICES implementation 
[ROCm/hip commit: 1f606261c1]
 2016-02-27 14:14:08 -06:00
pensun a71c5a6cf0 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 
[ROCm/hip commit: 07e56d4666]
 2016-02-27 04:25:28 -06:00
Aditya Avinash Atluri f2dfb87abf Merge pull request #4 from AMDComputeLibraries/memtracker 
hipGetPointerAttrib behavioral changes

[ROCm/hip commit: ecadb1623c]
 2016-02-27 10:51:23 -06:00
Ben Sander 66d6dbf686 disable rocrv2, properly 
[ROCm/hip commit: ea09557e1b]
 2016-02-27 03:31:30 -06:00
Aditya Avinash Atluri e7631f28e5 Corrected hipPointerGetAttribute 
Made hipPointerGetAttribute work same as cudaPointerGetAttribute for HCC

[ROCm/hip commit: 66aa7f2f8a]
 2016-02-26 18:50:40 -06:00
pensun c864a3f0fd relsove conflicts 
[ROCm/hip commit: 57f60b34fb]
 2016-02-26 09:57:40 -06:00
pensun dc3710cee4 fix compiling error 
[ROCm/hip commit: ee7ac16396]
 2016-02-26 09:50:00 -06:00
Ben Sander 1ac07d2b87 fixes for titan platform 
[ROCm/hip commit: ff66ef0779]
 2016-02-26 05:25:30 -06:00
Ben Sander b46a90852e Disable ROCR_V2 
[ROCm/hip commit: 6e0ccdfb95]
 2016-02-26 23:34:45 -06:00
Ben Sander 193dbe4632 Merge branch 'memtracker' into privatestaging 
Conflicts:
	include/nvcc_detail/hip_runtime_api.h


[ROCm/hip commit: 369e0d7b5b]
 2016-02-26 06:17:05 -06:00
Ben Sander 8a2bcf2da3 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 
[ROCm/hip commit: c300ffe458]
 2016-02-26 06:15:09 -06:00
Ben Sander a10f2d1a93 Merge branch 'memtracker' of https://github.com/AMDComputeLibraries/HIP-privatestaging into memtracker 
Conflicts:
	tests/src/hipMemcpy.cpp


[ROCm/hip commit: d319299ddb]
 2016-02-25 23:22:51 -06:00
Ben Sander 5ca4914e0e Merge branch 'memtracker' into privatestaging 
Conflicts:
	src/hip_hcc.cpp


[ROCm/hip commit: 4adab7b7ef]
 2016-02-25 19:38:46 -06:00
Ben Sander ae365b4ae2 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.


[ROCm/hip commit: 8b64c0dc62]
 2016-02-25 19:19:49 -06:00
Evgeny Mankov 82900a1888 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).


[ROCm/hip commit: 57e212606d]
 2016-02-25 23:44:39 +03:00
Ben Sander 60552f5133 Add tests for multi-threaded streams 
[ROCm/hip commit: 7090f5c3f9]
 2016-02-23 12:08:22 -06:00
Ben Sander 1888acb5f3 Sync review. 
- add calls to ihipInit missing from some routines.
- sync before draining a stream.


[ROCm/hip commit: 3886d494f4]
 2016-02-23 04:07:11 -06:00
Ben Sander f0c734e256 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.


[ROCm/hip commit: 549b18ce77]
 2016-02-22 23:15:24 -06:00
Ben Sander 23b257bca4 Merge branch 'memtracker' of https://github.com/AMDComputeLibraries/HIP-privatestaging into memtracker 
[ROCm/hip commit: 0a98db4b5f]
 2016-02-22 08:33:47 -06:00
gargrahul ccd1ed0a97 Update for shared atomics support 
[ROCm/hip commit: a2fbf06129]
 2016-02-22 16:21:52 +05:30
Ben Sander ebf2700936 Track last command to a stream. 
Passing simple tests.


[ROCm/hip commit: d33d806a5b]
 2016-02-20 11:02:07 -06:00
Evgeny Mankov c3a600c63b 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


[ROCm/hip commit: 833c9e52ad]
 2016-02-19 13:27:03 +03:00
Evgeny Mankov 68621c6702 Formatting, no functional changes. 
[ROCm/hip commit: fbdeee39cd]
 2016-02-18 18:54:19 +03:00
Evgeny Mankov 4fcd9f2542 Device property memoryBusWidth implementation. 
+ Device property memoryBusWidth is added to hipDeviceProp_t struct.
+ Device attribute hipDeviceAttributeMemoryBusWidth is added to hipDeviceAttribute_t struct.
+ Tests update.


[ROCm/hip commit: 1c19dbb807]
 2016-02-18 18:15:01 +03:00
Evgeny Mankov a0cc7134e3 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.


[ROCm/hip commit: 5ea8543d2e]
 2016-02-18 17:25:28 +03:00
Evgeny Mankov 8c1a0d1924 Attribute hipDevAttrConcurrentKernels for obtaining Device property concurrentKernels is added. 
[ROCm/hip commit: 2b6fda77ca]
 2016-02-18 14:34:18 +03:00
Ben Sander 4e0ba06044 Enable Tracker and ROCR by default, verify with HCC 
[ROCm/hip commit: c6f8883b0d]
 2016-02-17 23:03:37 -06:00
Ben Sander a2d8f9d98e Remove HIP-local AM tracker (now in HCC) 
[ROCm/hip commit: d653782d9d]
 2016-02-17 21:33:32 -06:00
Ben Sander 5ab45e9be7 USE_AM_TRACKER=0 works 
[ROCm/hip commit: 44f40e171a]
 2016-02-17 21:23:36 -06:00
pensun b3d70ca271 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.


[ROCm/hip commit: 8aa4bfce57]
 2016-02-17 09:32:50 -06:00
pensun e5ff38e421 Implementation of HIP_VISIBLE_DEVICES in runtime 
[ROCm/hip commit: c1e120fb1b]
 2016-02-17 06:59:18 -06:00
Ben Sander bb8a1fe72d more work on async copies 
[ROCm/hip commit: 59379ffb44]
 2016-02-17 00:59:12 -06:00
pensun 66df1ff930 modify to add remove invalid devices numbers 
[ROCm/hip commit: 060439b6ab]
 2016-02-16 10:00:05 -06:00
pensun 8c230eab73 Implement to read HIP_VISIBLE_DEVICES to internal global variable 
[ROCm/hip commit: d40cbef2af]
 2016-02-16 07:39:04 -06:00
Ben Sander 512163b889 Add per-stream pool for hsa_signals. 
[ROCm/hip commit: caef9b5ced]
 2016-02-16 01:59:13 -06:00