AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
4,389 Commits 3 Branches 0 Tags
6a37ef97c46ecb9c48a8024ba4148ec8e2cf2f79
Commit Graph

3903 Commits

Author SHA1 Message Date
Alex Voicu 6a37ef97c4 Address post-staging issues in #1809 (#1894) 
Fixes SWDEV-223910 and SWDEV-223663

[ROCm/clr commit: cea5489f00]
 2020-02-27 16:21:12 +05:30
Maneesh Gupta d805279060 bump version to 3.2 (#1898) 
- Bump version to 3.2
- [ci] Enable tests on ROCm 3.1

[ROCm/clr commit: 3b4b58bda6]
 2020-02-27 16:18:31 +05:30
Nick Curtis e5d029b8a4 fix long shuffle implementations for windows (#1895) 
Fixes for SWDEV-223694

[ROCm/clr commit: a094c87038]
 2020-02-26 15:53:56 +05:30
Yaxun (Sam) Liu d79e247a70 Fix hipcc for extra -mllvm option (#1885) 
[ROCm/clr commit: ff4479fcc7]
 2020-02-26 15:53:43 +05:30
Sarbojit2019 51087a9a46 [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/clr commit: d81a9a712a]
 2020-02-19 13:48:20 +05:30
eshcherb 5c9e5225d4 adding hipExtModuleLaunchKernel to tracing layer (#1880) 
[ROCm/clr commit: 82478dca65]
 2020-02-19 13:47:49 +05:30
Alex Voicu 3057a52be5 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/clr commit: 730f23829e]
 2020-02-18 20:50:27 +05:30
Yaxun (Sam) Liu 5748b7550b 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/clr commit: f4120043ed]
 2020-02-17 22:49:26 +05:30
Rahul Garg 6f1d14c96d Fix hipMemcpy3D (#1798) 
Fixes #1790 and #1791. hipMemcpy3D still requires further refactoring for different input and output combinations.

[ROCm/clr commit: ec84c16d75]
 2020-02-17 19:35:35 +05:30
Maneesh Gupta 734b90c592 [dtests] Fix random timeout failures in hipModuleLoadDataMultThreaded (#1877) 
Limit the max threads that are launched to 16.

[ROCm/clr commit: 2ec55fde47]
 2020-02-17 11:16:20 +05:30
vsytch fd7dfa19d1 Add missing __hip_pinned_shadow__ attributes to the texture global vars. (#1866) 
[ROCm/clr commit: 5afe43b1aa]
 2020-02-15 09:52:25 +05:30
Maneesh Gupta 2b1a99ff12 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 c1b707ff63.

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

This reverts commit d9cbd5a143.

* updated test for nvidia path

* Add c++11 option for nvcc

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


[ROCm/clr commit: e085417370]
 2020-02-15 09:51:24 +05:30
Nick Curtis e9ee5a3719 Implement long / long long shuffles (#1829) 
Implement additional data-types for shuffles (long and long long).
Based upon the double implementation.


[ROCm/clr commit: 5b0f34c59b]
 2020-02-15 09:51:09 +05:30
Siu Chi Chan c74c8d7a1c 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/clr commit: 13052f27c2]
 2020-02-14 19:39:27 -05:00
Evgeny Mankov fcf5546ec1 [HIPIFY][doc] Update README.md: LLVM 10.0.0-rc2 - the latest supported LLVM Release 
[ROCm/clr commit: 0315aefe78]
 2020-02-14 13:09:31 +03:00
Rahul Garg 50395502b6 [sample] Add hipDispatchEnqueueRateMT (#1869) 
* [sample] Add hipDispatchEnqueueRateMT



[ROCm/clr commit: 893ccc4ddf]
 2020-02-13 23:21:40 -08:00
Evgeny Mankov faf676764d Merge pull request #1867 from emankov/HIP 
[HIPIFY][doc] Update README.md: Windows tested configurations

[ROCm/clr commit: 60c76da392]
 2020-02-13 18:48:38 +03:00
Evgeny Mankov 0c84df243f [HIPIFY][doc] Update README.md: Windows tested configurations 
[ROCm/clr commit: fb52a3973b]
 2020-02-13 18:34:10 +03:00
Satyanvesh Dittakavi 373da28853 Add c++11 option for nvcc 
[ROCm/clr commit: 2034ed3561]
 2020-02-13 19:48:26 +05:30
Satyanvesh Dittakavi 360f321e76 updated test for nvidia path 
[ROCm/clr commit: 627d9a1f46]
 2020-02-13 16:34:05 +05:30
Jeff Daily 40688fbfc2 missing break statement in hipDeviceGetAttribute (#1865) 
The break is missing for hipDeviceAttributeMaxTexture3DDepth.

[ROCm/clr commit: fe47fce496]
 2020-02-13 14:22:56 +05:30
Sarbojit2019 194b05efe2 [hip] Fix for bug introduced in #1770 when blockSize is non-power of 2 (#1864) 
Fixes SWDEV-222161

[ROCm/clr commit: a03628335c]
 2020-02-13 14:22:46 +05:30
Sarbojit2019 56398f5269 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/clr commit: e334d3d6ec]
 2020-02-13 14:22:11 +05:30
ansurya 708af7e6ac Reduce GPU copying based on arch it runs on (#1751) 
Implements SWDEV-213230.

[ROCm/clr commit: 888a7f2a90]
 2020-02-13 14:21:51 +05:30
Evgeny Mankov 20b1eb7ffb Merge pull request #1830 from asalmanp/coop_flag_define 
[HIP][HIPIFY] Add some missing flags for cooperative launch and occup…

[ROCm/clr commit: 5930da7f56]
 2020-02-12 14:07:39 +03:00
Paul 069f2672b9 Revert "Try to initialize the primary context on cuda" 
This reverts commit d9cbd5a143.


[ROCm/clr commit: dd6676dadf]
 2020-02-11 12:34:11 -06:00
Paul 509d8096a5 Revert "Push ctx to the stack as current" 
This reverts commit c1b707ff63.


[ROCm/clr commit: cd279cd574]
 2020-02-11 12:34:10 -06:00
Paul c1b707ff63 Push ctx to the stack as current 
[ROCm/clr commit: e8a7cc6957]
 2020-02-11 11:46:29 -06:00
Paul d9cbd5a143 Try to initialize the primary context on cuda 
[ROCm/clr commit: 86cdb30195]
 2020-02-11 11:26:24 -06:00
Aryan Salmanpour 89cb17284b fix build error in nvcc path 
[ROCm/clr commit: 1c013bab3e]
 2020-02-11 12:16:51 -05:00
Jatin Chaudhary 4436f22326 Revert "Sync hip-targets*.cmake in package with install changes (#1831)" (#1860) 
Fixes SWDEV-222155 & SWDEV-222158
This reverts commit e0b4ba96e8.

[ROCm/clr commit: 18a64fe8d4]
 2020-02-11 11:56:57 +05:30
Paul 5e16db1f16 Skip test on cuda 
[ROCm/clr commit: accdd882dc]
 2020-02-10 17:23:58 -06:00
Paul e6663a7c23 Reduce threads on cuda 
[ROCm/clr commit: 5d24a2beef]
 2020-02-10 16:37:34 -06:00
Paul 44c4b72f83 Set context on each thread 
[ROCm/clr commit: 7a6e88480b]
 2020-02-10 16:01:53 -06:00
Paul 8fb4e25cf1 Create context for cuda 
[ROCm/clr commit: bb145e77ae]
 2020-02-10 15:52:34 -06:00
Paul 218994969f Add setDevice to try and initialize the context on cuda 
[ROCm/clr commit: 9494d0e3c9]
 2020-02-10 13:37:45 -06:00
Aryan Salmanpour e7059e518d Fix a typo causing a build error 
[ROCm/clr commit: 504412c1e4]
 2020-02-10 11:44:40 -05:00
Aryan Salmanpour 44ce519211 resolve merge conflict 
[ROCm/clr commit: dfa565d76f]
 2020-02-10 10:30:55 -05:00
Maneesh Gupta 9d451d2d69 gedit/hip.lang does not need a seperate license 
[ROCm/clr commit: 484d67a485]
 2020-02-10 16:27:20 +05:30
Maneesh Gupta 5df85a27e5 Update copyright section in gedit/hip.lang 
[ROCm/clr commit: f22d042bc8]
 2020-02-10 16:25:38 +05:30
Maneesh Gupta a21b1f84cb 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 2c6e880e05.

[ROCm/clr commit: d032637934]
 2020-02-10 10:45:28 +05:30
Alex Voicu 7f9f87b0d9 (Maybe) Match alignment between Clang and GCC. (#1789) 
Should fix #1740 and the related internal bug.

[ROCm/clr commit: 8f96152d3a]
 2020-02-10 10:44:49 +05:30
mhbliao 87056623e9 [hip] Cleanup compiler wrapper for HIP-Clang. (#1847) 
[ROCm/clr commit: 8343d7c9aa]
 2020-02-07 13:28:26 -08:00
Paul 0b46506fcf Output on failure 
[ROCm/clr commit: 2405ab236f]
 2020-02-07 10:13:28 -06:00
Paul daaa1ebb80 Make threads a multiple of hardware concurrency 
[ROCm/clr commit: fb8c7cb3ec]
 2020-02-06 16:23:29 -06:00
Paul b29fe7e893 Add multithreaded test 
[ROCm/clr commit: bf69c7ae32]
 2020-02-06 16:21:40 -06:00
Michael LIAO 0a97bed2bd [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/clr commit: 59bd608bfe]
 2020-02-06 17:09:14 -05:00
Paul 1ee03c9db9 Assign the iterator 
[ROCm/clr commit: 84d6eb985d]
 2020-02-06 12:15:29 -06:00
Paul c9f66e4ae5 Fix compile error 
[ROCm/clr commit: bb26e99c73]
 2020-02-06 12:04:50 -06:00
Paul 0688b554ae Use deque instead of vector for code readers so that the iterators and references will be stable 
[ROCm/clr commit: 1354a447e1]
 2020-02-06 11:56:17 -06:00