Gráfico de commits

1830 Commits

Autor SHA1 Mensaje Fecha
wsttiger db3ebbcbae Add missing CUDA tags
[ROCm/hip commit: 512214e930]
2017-08-15 13:22:20 -05:00
Maneesh Gupta 0a25b3e972 Merge pull request #138 from scchan/threadfence_system
implement __threadfence_system

[ROCm/hip commit: f14e31d9d7]
2017-08-10 12:51:05 +05:30
Maneesh Gupta 7390368693 [dtests] Disable hip_threadfence_system.tst on nvcc
The test needs some more modifications before it is ready for building
on nvcc path. Disabling it for now so that the pull request can be
merged.

Change-Id: I19a894fcda7b2159f86a4e4e95a409c5655d3760


[ROCm/hip commit: 69e431f035]
2017-08-10 12:05:58 +05:30
Ben Sander 5709a38ef3 Merge pull request #146 from gargrahul/docs_update_memcpysymbol
docs update for memcpytosymbol support

[ROCm/hip commit: e7a7574cf3]
2017-08-09 17:48:48 -05:00
Ben Sander d4d9836a96 Merge pull request #145 from gargrahul/context_mgmt_changes
Updated context management logic:

[ROCm/hip commit: 660061c3b4]
2017-08-09 17:47:45 -05:00
Ben Sander d2454ffb9f Merge pull request #139 from gargrahul/disable_p2pcoherency_test
Disable failing test p2p_copy_coherency

[ROCm/hip commit: 4a34bce26f]
2017-08-09 16:54:33 -05:00
Maneesh Gupta 8856953fed Merge pull request #147 from mangupta/bump_workweek
Bump min hcc_workweek required for named kernel dispatch to 17312

[ROCm/hip commit: fa3f9c2474]
2017-08-09 14:10:07 +05:30
Maneesh Gupta c050396510 Bump min hcc_workweek required for named kernel dispatch to 17312
Change-Id: I8c7b58306b279ed113d03260e4bc6086bb8b4e68


[ROCm/hip commit: 8d6acec135]
2017-08-08 11:08:55 +05:30
Rahul Garg d571697190 docs update for memcpytosymbol support
[ROCm/hip commit: eb25bd72db]
2017-08-08 09:13:12 +05:30
Rahul Garg 8d254a7113 Updated context management logic:
1) hipSetDevice sets a flag so that next call to hipCtxGetCurrent returns primary context on current device
2) hipCtxGetCurrent returns primary context on current device if TLS context stack is empty
3) hipCtxPopCurrent falls back to primary context on current device as default
4) hipCtxPushCurrent, hipCtxSetCurrent and hipCtxCreate reset the flag set in hipSetDevice


[ROCm/hip commit: 3e84cf4aba]
2017-08-08 07:02:22 +05:30
Rahul Garg 63456bfea0 Disable failing test p2p_copy_coherency
[ROCm/hip commit: 2f617bac6d]
2017-08-02 14:47:13 +05:30
Siu Chi Chan 21c1d69e77 implement __threadfence_system
[ROCm/hip commit: 96e79f97e2]
2017-08-02 08:50:18 +00:00
Ben Sander f88f99276a Merge pull request #136 from mangupta/update_install_markdown
[doc] Add details of HIP build time dependencies to INSTALL.md

[ROCm/hip commit: 6d6c325626]
2017-07-31 11:05:30 -05:00
Ben Sander aba975dd29 Merge pull request #128 from mangupta/cleanup_amdgpu_target_logic
[hipcc] Cleanup amdgpu target logic

[ROCm/hip commit: 0fcd27586b]
2017-07-31 11:05:06 -05:00
Maneesh Gupta eba208b1fd [doc] Add details of HIP build time dependencies to INSTALL.md
Change-Id: Iee87d1345aba716d0ce49f7afb45ba073a16be88


[ROCm/hip commit: e9c995959b]
2017-07-31 14:57:32 +05:30
Maneesh Gupta e7c823b4ed Merge pull request #134 from gargrahul/fix_hipMemcpy2DAsync
fix hipMemcpy2DAsync

[ROCm/hip commit: fce1d0d7f0]
2017-07-31 10:27:18 +05:30
Maneesh Gupta 1be0719063 Merge pull request #135 from bensander/fix_tracing
Some fixes to tracing.

[ROCm/hip commit: 48573a037e]
2017-07-31 10:24:41 +05:30
Maneesh Gupta 5c9d90007f [hipcc] Fix amdgpu target selection logic
- Refactoring introduced a bug when user does not specify any target via
--amdgpu-target, but has an invalid target specified in
HCC_AMDGPU_TARGET. In this case the selection logic was defaulting to
gf803.
- Removed defaulting to any specific target if rocm_agent_enumerator
  fails. hipcc will report this and die if linking was required.

Change-Id: I76131867049fef92331807dd19a926406dcc1d02


[ROCm/hip commit: 85ff6e3ef4]
2017-07-31 10:11:19 +05:30
Ben Sander 4a234edea9 Some fixes to tracing.
[ROCm/hip commit: ecbb494a58]
2017-07-28 22:13:43 -05:00
Rahul Garg 7e9c13f8de fix hipMemcpy2DAsync
[ROCm/hip commit: 0053e4db03]
2017-07-29 06:50:56 +05:30
Ben Sander 8c5df22733 Merge pull request #119 from emankov/master
[HIPIFY] CUDA RT Textures support

[ROCm/hip commit: 98f09debbb]
2017-07-28 12:12:41 -05:00
Maneesh Gupta 514f5cb8fb [hipcc] Cleanup amdgpu target logic
Existing logic has a bug. If user specifies targetA via commandline
options, while enumerator returns targetB, hipcc will create a fatbin
containing targets targetA and targetB. enumerator should only be used
when no target is specified by user (commandline or env var).

Change-Id: I6da857f86860c0e671b5988cd858644a08f723b9


[ROCm/hip commit: 830e85060b]
2017-07-28 16:18:15 +05:30
Maneesh Gupta b763034c65 Merge pull request #127 from bensander/pass_module_kname
Pass kernel name to HCC dispatch_hsa_kernel, for debug/profile

[ROCm/hip commit: 375ec84bc6]
2017-07-28 10:19:15 +05:30
Ben Sander 1ff23db17b Add workweek check to make sure we have a new enough compiler
[ROCm/hip commit: 4980a6d3ab]
2017-07-27 23:00:58 -05:00
Maneesh Gupta 90fa6b18eb Merge pull request #122 from bensander/enable_async_null_stream
Set HIP_SYNC_NULL_STREAM=0.

[ROCm/hip commit: fcc294ed41]
2017-07-28 09:15:56 +05:30
Maneesh Gupta 5ab79f8182 Merge pull request #121 from syifan/master
Allow hip cmake to link libraries

[ROCm/hip commit: 821f8b9e02]
2017-07-28 09:12:15 +05:30
Ben Sander 3ab70ad042 Pass kernel name to HCC dispatch_hsa_kernel, for debug/profile
[ROCm/hip commit: 3a4dfc0f85]
2017-07-27 22:00:15 -05:00
Evgeny Mankov 24fc06496b Update Cuda2Hip.cpp
[ROCm/hip commit: d614f53ad7]
2017-07-28 01:45:24 +03:00
Evgeny Mankov 28f7d83e7a Update CUDA_Runtime_API_functions_supported_by_HIP.md
[ROCm/hip commit: bcca169631]
2017-07-28 01:41:14 +03:00
Evgeny Mankov daf7df6296 Update CUDA_Runtime_API_functions_supported_by_HIP.md
[ROCm/hip commit: 36a9ccacb6]
2017-07-28 01:38:50 +03:00
Ben Sander a9dce0cfe8 Merge pull request #110 from sunway513/fix_hipEnvVar_test
Fix hipEnvVar test

[ROCm/hip commit: 9cede50942]
2017-07-27 13:03:43 -05:00
sunway513 014f1797d9 Change to use hipDeviceGetPCIBusId api for hipEnvVarDriver test
[ROCm/hip commit: ed0d6cf745]
2017-07-27 17:41:49 +00:00
Ben Sander cba82d9e24 Set HIP_SYNC_NULL_STREAM=0.
Optimizes null stream synchronization so it uses GPU-side dependency
resolution. Requires HCC __hcc_workweek__ > 17300.


[ROCm/hip commit: 77fb9893b4]
2017-07-27 11:11:54 -05:00
Yifan Sun 8db7ef071d Allow hip cmake to link libraries
[ROCm/hip commit: 0aaa800109]
2017-07-27 11:46:18 -04:00
Evgeny Mankov 21c012ffca [HIPIFY] CUDA RT Textures support
[ROCm/hip commit: ffcf5d32a0]
2017-07-27 18:39:40 +03:00
Maneesh Gupta da4042554f Merge pull request #116 from bensander/enable_coh_host
Make host memory allocations coherent by default.

[ROCm/hip commit: 526fcb1223]
2017-07-27 10:45:32 +05:30
Ben Sander 9ec4b0b66f Merge pull request #100 from weixingzhang/texture
HIP Texture Support

[ROCm/hip commit: 4f93b99f23]
2017-07-26 19:28:15 -05:00
Ben Sander adeb4b4ff9 Merge pull request #114 from emankov/master
[HIPIFY] CUDA Stream Callback support

[ROCm/hip commit: bcdc6681fa]
2017-07-26 19:27:05 -05:00
Ben Sander d1dbf4f45d Merge pull request #113 from mangupta/avoid_rebuild_cache
[cmake] No longer rebuild cmake cache on each build

[ROCm/hip commit: 50546c7866]
2017-07-26 19:24:43 -05:00
Ben Sander c828572b94 Merge pull request #115 from bensander/README
Update release notes.

[ROCm/hip commit: 5589c59605]
2017-07-26 19:23:26 -05:00
Ben Sander 6834270f15 Make host memory allocations coherent by default.
Associated change is to optimize event recording so it uses
agent-scope releaes (since it was only using system-scope release
to support non-coherent host mem).

Flags and environment variables exist to obtain previous behavior
if desired.  Options are documented in new performance guide.


[ROCm/hip commit: 8b8e97ff28]
2017-07-26 19:20:34 -05:00
Ben Sander fbffeb7350 Update release notes.
[ROCm/hip commit: d9d9a8620f]
2017-07-26 19:03:32 -05:00
Evgeny Mankov 9ba397e8fe [HIPIFY] CUDA Stream Callback support
[ROCm/hip commit: 368aebc4bc]
2017-07-26 19:29:50 +03:00
Maneesh Gupta 0c211125c4 [cmake] No longer rebuild cmake cache on each build
Cmake cache was being rebuilt on each build. This was being done
to update HIP_VERSION, HCC_VERSION, .hipInfo and .hipVersion.
However, rebuilding cache also re-runs HIT parser which is slow.
Removing the cache rebuild should speed up the build. But user
needs to explicitly rebuild the cache in case HIP_VERSION or
HCC_VERSION changes by calling "make rebuild_cache"

Change-Id: Ia5476eb7105aa614239c4dc7968c37f5e6cb0b29


[ROCm/hip commit: e63c8494fd]
2017-07-26 13:51:32 +05:30
Evgeny Mankov e17a4fffa9 [HIPIFY] Finally finished syncing with CUDA 8.0.61 Driver and Runtime… (#105)
[hipify-clang] Finally finished syncing with CUDA 8.0.61 Driver and Runtime API (including missing data types, D3D, OpenGL, VDPAU and EGL interop).

    + All the Modules are supported now:
	1) 4.1 – 4.31 from CUDA 8.0.61 Driver API
	2) CUDA_Driver_API_functions_supported_by_HIP.md updated accordingly
	3) 4.1 – 4.31 from CUDA 8.0.61 Runtime API
	4) CUDA_Runtime_API_functions_supported_by_HIP.md updated accordingly
    + Typos fixing
    + Annotating

[ROCm/hip commit: 85e6593a1b]
2017-07-26 09:38:03 +05:30
Maneesh Gupta edcb83cc2f Merge pull request #111 from bensander/master
Enable HCC_OPT_FLUSH=1 (if HCC compiler new enough)

[ROCm/hip commit: f420af64d0]
2017-07-25 12:02:33 +05:30
Ben Sander 149ec02e84 Enable HCC_OPT_FLUSH=1 (if HCC compiler new enough)
[ROCm/hip commit: 67f5d2de3d]
2017-07-24 18:57:19 -05:00
sunway513 d8d3016e88 Update Jenkinsfile to run all the built tests
[ROCm/hip commit: 6f359dfbe6]
2017-07-24 19:02:00 +00:00
sunway513 64a3220246 Re-enable hipEnvVarDriver in Jenkinsfile
[ROCm/hip commit: 1275ceed67]
2017-07-24 18:53:46 +00:00
sunway513 286ab981a9 Fix indentation, binary path and add wait on popen calls to hipEnvVar test
[ROCm/hip commit: 81bc82d166]
2017-07-24 17:16:13 +00:00