AI/rocm-systems
2
0
Garpu 0
Kode Masalah Tarik Permintaan Actions Packages Projects Rilis Wiki Kegiatan
4.389 Melakukan 3 Cabang 0 Tag
a5ad7bc67b3deb92bbc98772c1fdbe69aac34fe7
Grafik Komit

4389 Melakukan

This Branch
This Branch
All Branches
Penulis SHA1 Pesan Tanggal
Alex Voicu a5ad7bc67b Address post-staging issues in #1809 (#1894) 
Fixes SWDEV-223910 and SWDEV-223663

[ROCm/hip commit: d830dad3be]
 2020-02-27 16:21:12 +05:30
Maneesh Gupta b499822842 bump version to 3.2 (#1898) 
- Bump version to 3.2
- [ci] Enable tests on ROCm 3.1

[ROCm/hip commit: 71e1f87f7e]
 2020-02-27 16:18:31 +05:30
Nick Curtis 2715d1b036 fix long shuffle implementations for windows (#1895) 
Fixes for SWDEV-223694

[ROCm/hip commit: b7dd073d93]
 2020-02-26 15:53:56 +05:30
Yaxun (Sam) Liu 3067349909 Fix hipcc for extra -mllvm option (#1885) 
[ROCm/hip commit: 69404d8e78]
 2020-02-26 15:53:43 +05:30
Sarbojit2019 9a5658146d [HIPIFY] Add back missing execute permission to hipify-perl (#1881) 
hipify-perl script lost its executable permission hence "samples/0_Intro/square" was failing. Fixes SWDEV 223433.

[ROCm/hip commit: c1a70707e0]
 2020-02-19 13:48:20 +05:30
eshcherb 7103e8de5e adding hipExtModuleLaunchKernel to tracing layer (#1880) 
[ROCm/hip commit: 82ec3c1c5b]
 2020-02-19 13:47:49 +05:30
Alex Voicu 745a66c60e Tweak synchronous memcpy implementation (#1809) 
The existing one can have issues on certain systems, therefore this limits use of direct memcpy via largeBAR to sizes where it is unequivocally better.

Also addresses SWDEV-220030 and SWDEV-222237.

[ROCm/hip commit: 9b4f39e1d8]
 2020-02-18 20:50:27 +05:30
Yaxun (Sam) Liu 32297229da Let HIP-Clang inline all functions by default (#1875) 
This is a quick workaround to match HCC behavior for performance since inlining usually
results in more optimization opportunities therefore better performance.

We will fine tuning inline threashold later.

[ROCm/hip commit: 92cc29ae2b]
 2020-02-17 22:49:26 +05:30
Rahul Garg f1746197c7 Fix hipMemcpy3D (#1798) 
Fixes #1790 and #1791. hipMemcpy3D still requires further refactoring for different input and output combinations.

[ROCm/hip commit: 8c5e5e435b]
 2020-02-17 19:35:35 +05:30
Maneesh Gupta 5d7f0b1798 [dtests] Fix random timeout failures in hipModuleLoadDataMultThreaded (#1877) 
Limit the max threads that are launched to 16.

[ROCm/hip commit: 854afef281]
 2020-02-17 11:16:20 +05:30
vsytch cf57b45f3c Add missing __hip_pinned_shadow__ attributes to the texture global vars. (#1866) 
[ROCm/hip commit: 56b8b0d80e]
 2020-02-15 09:52:25 +05:30
Maneesh Gupta e72b5a0187 Use deque instead of vector for code readers so that the iterators and references will be stable (#1851) 
* Use deque instead of vector for code readers so that the iterators and references will be stable

* Fix compile error

* Assign the iterator

* Add multithreaded test

* Make threads a multiple of hardware concurrency

* Output on failure

* Add setDevice to try and initialize the context on cuda

* Create context for cuda

* Set context on each thread

* Reduce threads on cuda

* Skip test on cuda

* Try to initialize the primary context on cuda

* Push ctx to the stack as current

* Revert "Push ctx to the stack as current"

This reverts commit 0fb72d9e7f.

* Revert "Try to initialize the primary context on cuda"

This reverts commit 44e3908680.

* updated test for nvidia path

* Add c++11 option for nvcc

Co-authored-by: satyanveshd <53337087+satyanveshd@users.noreply.github.com>


[ROCm/hip commit: e7120dd876]
 2020-02-15 09:51:24 +05:30
Nick Curtis 6d661ba6ca Implement long / long long shuffles (#1829) 
Implement additional data-types for shuffles (long and long long).
Based upon the double implementation.


[ROCm/hip commit: 797a929a65]
 2020-02-15 09:51:09 +05:30
Siu Chi Chan cc1b402fe3 Disabling HCC code object v3 generation by default. 
Some PyTorch unit tests have regression.  Disabling cov3 to allow more
time to debug and unblock PyTorch

Change-Id: Iba7f425ef3499c20c42ec45d9152b5d27ce97d03


[ROCm/hip commit: f2ab87d872]
 2020-02-14 19:39:27 -05:00
Evgeny Mankov e67aa6d48e Merge pull request #1870 from emankov/HIP 
[HIPIFY][doc] Update README.md: LLVM 10.0.0-rc2 - the latest supported LLVM Release

[ROCm/hip commit: 9a9319c8e7]
 2020-02-14 13:25:43 +03:00
Evgeny Mankov 4ba1c5d87b [HIPIFY][doc] Update README.md: LLVM 10.0.0-rc2 - the latest supported LLVM Release 
[ROCm/hip commit: 115f45d116]
 2020-02-14 13:09:31 +03:00
Rahul Garg 27526f47fb [sample] Add hipDispatchEnqueueRateMT (#1869) 
* [sample] Add hipDispatchEnqueueRateMT



[ROCm/hip commit: 9d97f91fbb]
 2020-02-13 23:21:40 -08:00
Evgeny Mankov ee7fa043e2 Merge pull request #1867 from emankov/HIP 
[HIPIFY][doc] Update README.md: Windows tested configurations

[ROCm/hip commit: 5f2438a6c6]
 2020-02-13 18:48:38 +03:00
Evgeny Mankov 37b3d0d8de [HIPIFY][doc] Update README.md: Windows tested configurations 
[ROCm/hip commit: 084b2fa0f6]
 2020-02-13 18:34:10 +03:00
Satyanvesh Dittakavi 8cfced9bfd Add c++11 option for nvcc 
[ROCm/hip commit: 3fb4135946]
 2020-02-13 19:48:26 +05:30
Satyanvesh Dittakavi 802aa5f44e updated test for nvidia path 
[ROCm/hip commit: ead254cdd5]
 2020-02-13 16:34:05 +05:30
Jeff Daily a352a87a15 missing break statement in hipDeviceGetAttribute (#1865) 
The break is missing for hipDeviceAttributeMaxTexture3DDepth.

[ROCm/hip commit: 03bb658721]
 2020-02-13 14:22:56 +05:30
Sarbojit2019 624cba30a6 [hip] Fix for bug introduced in #1770 when blockSize is non-power of 2 (#1864) 
Fixes SWDEV-222161

[ROCm/hip commit: 1109cbff83]
 2020-02-13 14:22:46 +05:30
Sarbojit2019 603a254cc2 ihipEnablePeerAccess return error if peer is not accessible (#1858) 
hipDeviceEnablePeerAccess returns success and adds peer into the list even if it is not accessible which creates problem in hipMalloc when it tries to share the ptr to peer device.
Proposed change is to check the access status before updating the peer list and update only when it can access the peer.

[ROCm/hip commit: fc5256fd28]
 2020-02-13 14:22:11 +05:30
ansurya fb53186682 Reduce GPU copying based on arch it runs on (#1751) 
Implements SWDEV-213230.

[ROCm/hip commit: 8c6934223b]
 2020-02-13 14:21:51 +05:30
Evgeny Mankov b58d79dfa3 Merge pull request #1830 from asalmanp/coop_flag_define 
[HIP][HIPIFY] Add some missing flags for cooperative launch and occup…

[ROCm/hip commit: 2536a3093d]
 2020-02-12 14:07:39 +03:00
Paul ca0eb390a9 Revert "Try to initialize the primary context on cuda" 
This reverts commit 44e3908680.


[ROCm/hip commit: 26bb6a97a7]
 2020-02-11 12:34:11 -06:00
Paul f86740ff94 Revert "Push ctx to the stack as current" 
This reverts commit 0fb72d9e7f.


[ROCm/hip commit: e82e3c2339]
 2020-02-11 12:34:10 -06:00
Paul 0fb72d9e7f Push ctx to the stack as current 
[ROCm/hip commit: bff8cbe950]
 2020-02-11 11:46:29 -06:00
Paul 44e3908680 Try to initialize the primary context on cuda 
[ROCm/hip commit: fd98514113]
 2020-02-11 11:26:24 -06:00
Aryan Salmanpour 07c3e4f898 fix build error in nvcc path 
[ROCm/hip commit: 959f1b0f0e]
 2020-02-11 12:16:51 -05:00
Jatin Chaudhary 2868c785d8 Revert "Sync hip-targets*.cmake in package with install changes (#1831)" (#1860) 
Fixes SWDEV-222155 & SWDEV-222158
This reverts commit c7562072a3.

[ROCm/hip commit: ab7526f64c]
 2020-02-11 11:56:57 +05:30
Paul 746c36ed8e Skip test on cuda 
[ROCm/hip commit: b9f97ec3fe]
 2020-02-10 17:23:58 -06:00
Paul 92ff502b92 Reduce threads on cuda 
[ROCm/hip commit: 30e8dfdd86]
 2020-02-10 16:37:34 -06:00
Paul 039171f6fd Set context on each thread 
[ROCm/hip commit: 2d9a2d866c]
 2020-02-10 16:01:53 -06:00
Paul a3e2660669 Create context for cuda 
[ROCm/hip commit: e5d077f70e]
 2020-02-10 15:52:34 -06:00
Paul b6e01adc72 Add setDevice to try and initialize the context on cuda 
[ROCm/hip commit: 29a257d79b]
 2020-02-10 13:37:45 -06:00
Aryan Salmanpour 5db60dab96 Fix a typo causing a build error 
[ROCm/hip commit: 5a29f27455]
 2020-02-10 11:44:40 -05:00
Aryan Salmanpour 964a272004 resolve merge conflict 
[ROCm/hip commit: 874b201ee2]
 2020-02-10 10:30:55 -05:00
Maneesh Gupta 837875336f gedit/hip.lang does not need a seperate license 
[ROCm/hip commit: 6614ae33e0]
 2020-02-10 16:27:20 +05:30
Maneesh Gupta 51546808da Update copyright section in gedit/hip.lang 
[ROCm/hip commit: 9acdcf27c5]
 2020-02-10 16:25:38 +05:30
Maneesh Gupta 7753b3e827 Revert "Match Occupancy APIs syntax with CUDA (#1625)" (#1857) 
Reverting this for now till we figure out how to avoid the build
breakage.

This reverts commit e38db9fb6f.

[ROCm/hip commit: f8e1c01900]
 2020-02-10 10:45:28 +05:30
Alex Voicu 5d5344eea1 (Maybe) Match alignment between Clang and GCC. (#1789) 
Should fix #1740 and the related internal bug.

[ROCm/hip commit: dd34ea95d6]
 2020-02-10 10:44:49 +05:30
mhbliao 63ccc49116 [hip] Cleanup compiler wrapper for HIP-Clang. (#1847) 
[ROCm/hip commit: a01b262660]
 2020-02-07 13:28:26 -08:00
Paul fcadd1107d Output on failure 
[ROCm/hip commit: d77ede7015]
 2020-02-07 10:13:28 -06:00
Paul 3fa3e99a9a Make threads a multiple of hardware concurrency 
[ROCm/hip commit: 8e494cfce8]
 2020-02-06 16:23:29 -06:00
Paul 5cfa3c626a Add multithreaded test 
[ROCm/hip commit: 5361424702]
 2020-02-06 16:21:40 -06:00
Michael LIAO b851e564f4 [hipcc] Skip warning on gfx000. 
- The known target checking should skip `gfx000` as well as it won't be
  used in real compilation command formation. The avoid generating
  annoying warning on `gfx000`.


[ROCm/hip commit: 66678b0170]
 2020-02-06 17:09:14 -05:00
Paul 70fa1572a2 Assign the iterator 
[ROCm/hip commit: 85560d38b1]
 2020-02-06 12:15:29 -06:00
Paul 7ee1ccfe24 Fix compile error 
[ROCm/hip commit: f34f96350c]
 2020-02-06 12:04:50 -06:00