AI/rocm-systems
2
0
Çatalla 0
Kod Konular Değişiklik İstekleri İşlemler Paketler Projeler Sürüm Viki Aktivite
482 İşleme 3 Dal 0 Etiketler
03014e8a1ddd4aaaf59c19b8bf33fa6aeac8ca73
İşleme Grafiği

482 İşleme

Bu Dal
Bu Dal
Tüm Dallar
Yazar SHA1 Mesaj Tarih
Aditya Atluri 3becf113bd Logging dispatch latency through database util 
[ROCm/clr commit: 867554c423]
 2016-03-23 11:39:57 -05:00
Siu Chi Chan 063e277e25 Merge pull request #18 from AMDComputeLibraries/grid_launch_index 
Grid launch index fix

[ROCm/clr commit: 24496b1c38]
 2016-03-23 11:26:28 -05:00
Ben Sander 7f1f07363d Update docs: 
- Some comments in hip_hcc.cpp
 - document hipHostRegister* flags.
 - expand docs on hipHostRegister/Unregister. Use "register" rather than
   "pin" to describe action these take.
 - change required CUDA version to 6 (require unified memory)
 - remake doxygen.


[ROCm/clr commit: 3ca91cf327]
 2016-03-23 23:11:15 -05:00
scchan e1ee01dd60 add compiler version guard to the grid launch fix 
[ROCm/clr commit: 1313f1fb77]
 2016-03-23 11:24:37 -05:00
scchan 56d4054bc8 fix builtins mapping for IDs and Dims 
[ROCm/clr commit: 671128abb0]
 2016-03-23 11:08:28 -05:00
Aditya Atluri 49946441ab fixed hipHostGetFlags test 
[ROCm/clr commit: e9e84d2db6]
 2016-03-23 02:45:21 -05:00
Aditya Atluri 53bdbdc81b Remove redundant hpiHostGetDevicePointer 
[ROCm/clr commit: b1b6d5e648]
 2016-03-23 09:21:02 -05:00
Ben Sander 4f295f4358 Add USE_AV_COPY (replaces calls to am_copy with new acc_view::copy 
[ROCm/clr commit: 6fb2367b66]
 2016-03-23 10:30:53 -05:00
Ben Sander 72332dccb7 use the CPU_Agent for memory copies to indicate direction. 
[ROCm/clr commit: f3ac666a7a]
 2016-03-23 10:29:44 -05:00
Ben Sander f70dc3c245 Only include activity logger if CodeXL installed. 
Fix hipHostMalloc in hipBusBandwidth.


[ROCm/clr commit: 004b4ada93]
 2016-03-22 09:27:10 -05:00
Ben Sander 18a4fdb1b6 remove unneeded files 
[ROCm/clr commit: 2a54c58cac]
 2016-03-23 03:41:01 -05:00
Ben Sander 236ac8e023 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 
Conflicts:
	src/hip_hcc.cpp


[ROCm/clr commit: 8954e4fb26]
 2016-03-23 03:22:09 -05:00
Ben Sander 8c25bc63b9 Add unique stream_id to devices to improve debug 
[ROCm/clr commit: dc86743b35]
 2016-03-23 03:17:19 -05:00
Ben Sander ab9a256f47 Improve trace API 
- Validate compile-time disables.
- Add README.md section explain how to install/use CodeXL tracing
- Add code docs on trace_helper.h
- fix color on hipLaunchKernel to green.


[ROCm/clr commit: fa8deac1ad]
 2016-03-23 02:57:52 -05:00
Ben Sander 454faa062e HIP_TRACE_API prints function args, and in color 
[ROCm/clr commit: aed1a82ccb]
 2016-03-23 02:19:49 -05:00
Ben Sander f418fe4dae use codexl marker interface to mark HIP function/begin end. 
- Creates markers in HIP group and they show up in CodeXL trace
- Marker text includes HIP functioin arguments
- (Add trace_helper to convert arguments to strings)
- Still need to add HIP_INIT_API for ~30 HIP functions.


[ROCm/clr commit: 54704b59dd]
 2016-03-23 01:17:53 -05:00
Ben Sander acfe7cf1cc Describe how to file an issue 
[ROCm/clr commit: 723327cd0f]
 2016-03-23 01:15:05 -05:00
Aditya Atluri 52213587a7 Update CUDA_Runtime_API_functions_supported_by_HIP.md 
[ROCm/clr commit: 55291cc654]
 2016-03-22 10:42:34 -05:00
Ben Sander 37a02661a6 hipHostRegister and hipHostMalloc refactor. 
Note hipHostMalloc (not hipHostAlloc or hipMallocHost).
 -  the hipHost* is used for all HIP APIs dealing with Host memory.
    (including hipHostMalloc, hipHostFree, hipHostRegister,
hipHostUnregister, hipHostGetFlags, hipHostGetDevicePointer).
  - hipMallocHost is consistent with "hipMalloc" for allocating device
    memory.  Enumerations hipHostMalloc* also used as optional
    flags parm to hipHostMalloc.


[ROCm/clr commit: 2d0fade1f7]
 2016-03-22 02:30:10 -05:00
Ben Sander db8bff1235 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 
[ROCm/clr commit: bf83b949f6]
 2016-03-22 01:17:17 -05:00
Aditya Atluri 726ffa5124 Update CUDA_Runtime_API_functions_supported_by_HIP.md 
[ROCm/clr commit: a8a255b6bc]
 2016-03-21 18:33:50 -05:00
Aditya Atluri cb2b9495f8 Revert "Revert "fixed memory free apis"" 
This reverts commit 70d2a03efd.


[ROCm/clr commit: 8af8ee2476]
 2016-03-21 10:40:42 -05:00
Aditya Atluri 399994788f Revert "Revert "fix nvcc for hipHostMalloc* flags."" 
This reverts commit 0978c92dbc.


[ROCm/clr commit: bde1e6182d]
 2016-03-21 10:39:49 -05:00
Aditya Atluri 0978c92dbc Revert "fix nvcc for hipHostMalloc* flags." 
This reverts commit 849395ec02.


[ROCm/clr commit: 83fee90e83]
 2016-03-21 10:36:14 -05:00
Aditya Atluri 70d2a03efd Revert "fixed memory free apis" 
This reverts commit 805dbd6d90.


[ROCm/clr commit: 1fa4d0d4b9]
 2016-03-21 10:36:11 -05:00
Aditya Atluri 4ca3e4e20e Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 
[ROCm/clr commit: e99179edc8]
 2016-03-21 10:34:08 -05:00
Aditya Atluri 805dbd6d90 fixed memory free apis 
[ROCm/clr commit: 71a6b5cb6c]
 2016-03-21 10:32:30 -05:00
Aditya Atluri 0a91b906f1 Disabling default-stream per-thread tests 
[ROCm/clr commit: 93e6362104]
 2016-03-21 14:42:23 -05:00
Ben Sander 849395ec02 fix nvcc for hipHostMalloc* flags. 
[ROCm/clr commit: d495ffb1d3]
 2016-03-21 09:33:46 -05:00
Ben Sander 7fb3f877b2 Remove kind (aka direction) for copy commands. 
This is always auto-detected from the src/dest location.


[ROCm/clr commit: ac1eca47f7]
 2016-03-21 05:35:36 -05:00
Aditya Atluri f3ae40baf7 suppressed warning in hipFreeHost 
[ROCm/clr commit: ea352aba6b]
 2016-03-20 15:31:59 -05:00
Aditya Atluri cad8b62504 Added feature for --default-streams not working tests and hipcc 
[ROCm/clr commit: f6b38b18b6]
 2016-03-20 08:08:33 -05:00
Ben Sander 98106c384c Implement hipHostFree on HCC path 
[ROCm/clr commit: 80d708846a]
 2016-03-19 23:25:11 -05:00
Ben Sander c2fd536c22 fix nvcc compiler 
- MallocHost and FreeHost deprecation.
- Change tests to call new hipHost* equivs.
- Add missing StreamSynchronize.


[ROCm/clr commit: 6984f24d3d]
 2016-03-19 04:20:15 -05:00
Ben Sander 250a2816fd Refactor copy - place common code in resolveMemoryKind. 
[ROCm/clr commit: 03731020f1]
 2016-03-19 22:56:10 -05:00
Ben Sander b1d3df6484 Deprecate hipMallocHost and hipFreeHost. 
These will print compiler warnings if used, so we can weed them out
before removing.

Also add a default flags args for hipHostAlloc, in the C++ functioin
headers.  So you can replace hipMallocHost(&ptr, size( with hipHostAlloc(&ptr, size)


[ROCm/clr commit: 57365eb7a3]
 2016-03-19 22:53:59 -05:00
Ben Sander c75b63b61c Refactor waitALlDevices and async mem copy. 
- move waitAllStreams to device member function.
- create separate stream member function for copyAsync, like copySync.
  hipMemcpyAsync now calls the copyAsync.


[ROCm/clr commit: a88c2b1ec9]
 2016-03-19 05:42:19 -05:00
Ben Sander d41836a4a1 Fix bug: test was allocating host mem instead of device mem. 
Caused assertion when checking free + allocated should
not exceed total.  Bug introduced in hipHostAlloc conversion.


[ROCm/clr commit: 4c6fd4e7ec]
 2016-03-19 04:11:39 -05:00
Ben Sander cf95ea5468 Swap in corrected hipHostAlloc (bad merge) 
[ROCm/clr commit: aaa2429feb]
 2016-03-19 04:11:08 -05:00
Ben Sander 3305c49949 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging 
Conflicts:
	src/hip_hcc.cpp


[ROCm/clr commit: 90ad8ddc5d]
 2016-03-19 03:22:09 -05:00
Ben Sander 3323d2005c disable mt streams tests (for now) 
[ROCm/clr commit: efc9df8805]
 2016-03-19 03:10:31 -05:00
Ben Sander 57c08b7fce Describe HIP env vars 
[ROCm/clr commit: c39b0f9660]
 2016-03-19 03:09:57 -05:00
Ben Sander 1abdd6602f Fix copy and sync bugs. Remove extra sync in default stream. 
- NULL stream was waiting for itself to be empty before each command.
- Force "blocking" streams to wait for NULL to empty.  This was missing
  before.
- async copy was disabling itself via trueAsync=false for common cases.

Refactor:
- rename _null_stream to _default_stream.
- move some null sync function to defaultSync, move to dev member func.


[ROCm/clr commit: 44522eb607]
 2016-03-19 02:44:26 -05:00
Ben Sander d207f3bc26 Add beastperiteration and onesize for testing. 
onesize allows running tests at one specific size.


[ROCm/clr commit: 5197cf250d]
 2016-03-19 02:43:04 -05:00
Ben Sander d3a3e0ba63 Improve formatting - line up cols 
[ROCm/clr commit: c5d0813f03]
 2016-03-18 23:43:04 -05:00
Ben Sander c55187604c Print Pinned or Unpinned in result summary 
[ROCm/clr commit: 3dc6906855]
 2016-03-18 21:28:29 -05:00
Aditya Atluri 2bc7b1f3ba Update CUDA_Runtime_API_functions_supported_by_HIP.md 
[ROCm/clr commit: a17e390283]
 2016-03-18 11:28:06 -05:00
Aditya Atluri 336001aa29 Create CUDA_Runtime_API_functions_supported_by_HIP.md 
[ROCm/clr commit: 564d13a0b3]
 2016-03-18 11:23:44 -05:00
Ben Sander 203eeda2f7 Supported --aliged mode. Add results check for H2D and D2H. 
[ROCm/clr commit: 690486b9eb]
 2016-03-18 03:09:52 -05:00
Ben Sander b1fe0120ca Refactor copy code. 
-Move staging buffer locks inside the staging buffer code.
-Remove dedicated per-device completion_signal + per-device lock -
instead allocated signal from the per-stream pool.   This elimintes
the lock and allows more concurrency.
-remove switch HIP_DISABLE_BIDIR_MEMCPY


[ROCm/clr commit: e64174f47a]
 2016-03-18 03:02:00 -05:00