AI/rocm-systems
2
0
Форкнуть 0
Код Задачи Запросы на слияние Действия Пакеты Проекты Релизы Вики Активность
3 897 коммитов 3 Ветки 0 Теги
a4cbc99d2cd1a7e33c365dc04a285817ad2d3a6e
Граф коммитов

3897 Коммитов

This Branch
This Branch
Все ветки
Автор SHA1 Сообщение Дата
Evgeny Mankov a4cbc99d2c Merge pull request #1416 from gargrahul/HACK_hipFreeForManangedMalloc 
[HACK] Temporary fix in hipFree for hipManagedMalloc use

[ROCm/hip commit: e51880cbca]
 2019-09-13 18:57:21 +03:00
Rahul Garg 60adfab243 Fix typo and colon in comment 
[ROCm/hip commit: f0a1f95864]
 2019-09-13 03:09:01 +05:30
Rahul Garg e79f2d0ff8 [HACK] Temporary fix for hipFree for hipManagedMalloc 
[ROCm/hip commit: fe47b2185c]
 2019-09-13 02:10:21 +05:30
Evgeny Mankov e626477f72 Merge pull request #1414 from emankov/master 
[HIPIFY] Add unsupported fp16 functions

[ROCm/hip commit: 6739a6d656]
 2019-09-12 18:53:08 +03:00
Evgeny Mankov bc550cc062 [HIPIFY] Add unsupported fp16 functions 
[ROCm/hip commit: 34540002dd]
 2019-09-12 18:51:25 +03:00
Evgeny Mankov 904124fd32 Merge pull request #1413 from emankov/master 
[HIPIFY] Add more unsupported device functions

[ROCm/hip commit: 63198e0d66]
 2019-09-12 17:55:43 +03:00
Evgeny Mankov 2863db60b2 [HIPIFY] Add more unsupported device functions 
[ROCm/hip commit: 58c1a144d1]
 2019-09-12 17:53:25 +03:00
Evgeny Mankov 9528b8277f Merge pull request #1410 from emankov/master 
[HIPIFY][#1404] Fix: add cooperative_groups.h -> hip/hip_cooperative_groups.h

[ROCm/hip commit: 2cf9922754]
 2019-09-11 18:13:05 +03:00
Evgeny Mankov aae3b6dd50 [HIPIFY][#1404] Fix: add cooperative_groups.h -> hip/hip_cooperative_groups.h 
[ROCm/hip commit: 0426035468]
 2019-09-11 18:07:09 +03:00
Evgeny Mankov 023e47adf1 Merge pull request #1406 from emankov/master 
[HIPIFY][#1400] Fix Template Instantiation kernel launch (clang & perl)

[ROCm/hip commit: ccf8ffc0c9]
 2019-09-11 15:49:46 +03:00
Evgeny Mankov ba10db6fc6 [HIPIFY][#1400] Fix Template Instantiation kernel launch (clang & perl) 
+ Enclose template instantiation kernel calls into round brackets, leave regular kernel names unchanged (hipify-perl doesn't handle cases with macros).
+ Fix corresponding tests.

PS. hipify-perl couldn't handle correctly the following cases due to macros expansion disability, thus hipify-clang should be used instead:

#define KERNEL_NAME_MACRO axpy<float>
#define KERNEL_CALL_MACRO axpy<float><<<1, 2>>>
#define KERNEL_ARG_LIST_MACRO a, x, y

// CUDA:
KERNEL_NAME_MACRO<<<1, 2>>>(KERNEL_ARG_LIST_MACRO);
KERNEL_CALL_MACRO(KERNEL_ARG_LIST_MACRO);

// hipify-perl:
hipLaunchKernelGGL(KERNEL_NAME_MACRO, dim3(1), dim3(2), 0, 0, KERNEL_ARG_LIST_MACRO);
KERNEL_CALL_MACRO(KERNEL_ARG_LIST_MACRO);

// hipify-clang:
hipLaunchKernelGGL((KERNEL_NAME_MACRO), dim3(1), dim3(2), 0, 0, KERNEL_ARG_LIST_MACRO);
hipLaunchKernelGGL((axpy<float>), dim3(1), dim3(2), 0, 0, KERNEL_ARG_LIST_MACRO);


[ROCm/hip commit: 56ab105e9d]
 2019-09-10 15:59:06 +03:00
Aditya Lad 0a90bd8c0f Replacment of RENAME command in CMakeLists.txt (#1402) 
[ROCm/hip commit: 90acfb809c]
 2019-09-10 08:35:16 +00:00
Evgeny Mankov cec523f647 Merge pull request #1398 from emankov/master 
[HIPIFY] Add device functions support

[ROCm/hip commit: 29d09eb269]
 2019-09-07 19:00:00 +03:00
Evgeny Mankov 17b734afde [HIPIFY] Add device functions support 
+ Add a corresponding matcher cudaDeviceFuncCall to match only (__device__ or __global__) and not __host__ functions.
+ Add a corresponding device functions mapping:
  only unsupported are listed, cause supported are exactly the same as of CUDA and do not need transformation;
  make FindAndReplace for device functions separated from host API calls.
+ Add a test to distinguish device functions and user-defined.


[ROCm/hip commit: 6602fadc16]
 2019-09-06 18:34:12 +03:00
Jeff Daily 191482e1e4 fix bug where HIP_DB=1 seg faults at startup (#1388) 
[ROCm/hip commit: 8384f487ad]
 2019-09-05 10:04:19 +00:00
mhbliao 4913aec033 [hip] Stop using noduplicate and replace it with convergent. (#1390) 
[ROCm/hip commit: 119ee4b671]
 2019-09-05 10:03:43 +00:00
Paul Fultz II 9082a9990e Use hcc when building with hcc compiler (#1225) 
* Use hcc when building with hcc compiler

* Fix misnamed variable

* Dont set HIP_RUNTIME


[ROCm/hip commit: f6cf7e4e79]
 2019-09-05 10:03:04 +00:00
Evgeny Mankov 1a60014547 Merge pull request #1395 from emankov/master 
[HIPIFY][cmake][#1394] Fix

[ROCm/hip commit: 1954af4777]
 2019-09-05 12:12:08 +03:00
Evgeny Mankov 4d105b10ca [HIPIFY][cmake][#1394] Fix 
[#1394] cmake's VERSION_GREATER_EQUAL introduced in 3.7


[ROCm/hip commit: ab560d8fd2]
 2019-09-05 12:08:56 +03:00
Evgeny Mankov 598065c1ef Merge pull request #1393 from emankov/master 
[HIPIFY][perl] Code cleanup (preparation for generating)

[ROCm/hip commit: b457b31c82]
 2019-09-05 11:54:10 +03:00
Evgeny Mankov 5ba0d63058 [HIPIFY][perl] Code cleanup (preparation for generating) 
[ROCm/hip commit: 3a034a3427]
 2019-09-05 11:52:59 +03:00
Evgeny Mankov 2d00810408 Merge pull request #1391 from emankov/master 
[HIPIFY][perl] Sync hipify-perl with hipify-clang

[ROCm/hip commit: 0ec5b95b69]
 2019-09-05 09:07:32 +03:00
Evgeny Mankov 835b6481ea [HIPIFY][perl] Sync hipify-perl with hipify-clang 
Sync by means of semi-automatic generation: hipify-clang -perl


[ROCm/hip commit: 2a4c008385]
 2019-09-04 17:07:45 +03:00
Evgeny Mankov aa509de0f3 Merge pull request #1389 from emankov/master 
[HIPIFY][perl] Code cleanup and formatting

[ROCm/hip commit: 8feb4e1c9c]
 2019-09-04 16:39:12 +03:00
Evgeny Mankov 2b06d68835 [HIPIFY][perl] Code cleanup and formatting 
[ROCm/hip commit: 5b8843a127]
 2019-09-04 16:37:26 +03:00
Evgeny Mankov 2e6a49de4f Merge pull request #1387 from emankov/master 
[HIPIFY][perl][#259] Fix

[ROCm/hip commit: c9eb8fa1e8]
 2019-09-04 16:11:16 +03:00
Evgeny Mankov 8896d276fe [HIPIFY][perl][#259] Fix 
empty<<<1, 2>>> ( );     >>  hipLaunchKernelGGL(empty, dim3(1), dim3(2), 0, 0);
empty<<<1, 2, 0>>>();    >>
empty<<<1, 2, 0, 0>>>(); >>

instead of erroneous:    >> hipLaunchKernelGGL((empty), dim3(1), dim3(2), 0, 0, );


[ROCm/hip commit: b98330609b]
 2019-09-03 16:44:20 +03:00
Yaxun (Sam) Liu 73b6fe599c Do not include cuda wappers for OMP for hip-clang (#1382) 
[ROCm/hip commit: 8fe8fc18c0]
 2019-09-03 05:13:59 +00:00
Rahul Garg 83d9c2769b Revert "Using HSA API for hipMemsetAsync (#1346)" (#1381) 
This reverts commit 7dca0455e9.

[ROCm/hip commit: 6545521d6c]
 2019-09-03 05:13:46 +00:00
Nicholas Malaya 0915e29c6f Fix Broken Link in hip_porting_guide (#1376) 
The math library equivalents between CUDA-HIP are broken. This is a key feature for converting to AMD hardware. This fix corrects the broken link and moves the library equivalents to sit under the "Porting a New Cuda Project" header.

[ROCm/hip commit: 5a6eafcbf1]
 2019-09-03 05:13:35 +00:00
Sarbojit2019 fed6843ecf Removed hipLaunchKernel macro got missed in Merge (#1374) 
[ROCm/hip commit: e1f9e08ea7]
 2019-09-03 05:13:07 +00:00
Evgeny Mankov 9a4b4292ac Merge pull request #1385 from emankov/master 
[HIPIFY][doc] Update README.md (testing, versions)

[ROCm/hip commit: 7b0480aa69]
 2019-09-02 18:20:44 +03:00
Evgeny Mankov 29c8b80fc4 [HIPIFY][doc] Update README.md (testing, versions) 
[ROCm/hip commit: fbf23ed231]
 2019-09-02 18:18:43 +03:00
Evgeny Mankov 107a8994e3 Merge pull request #1384 from emankov/master 
[HIPIFY][tests] Add occupancy test

[ROCm/hip commit: 0e70229459]
 2019-09-02 17:56:16 +03:00
Evgeny Mankov a4af0f2fc1 [HIPIFY][tests] Add occupancy test 
[ROCm/hip commit: 1bf6deb149]
 2019-09-02 17:54:06 +03:00
Evgeny Mankov 7797c419a7 Merge pull request #1383 from emankov/hipify-clang 
[HIPIFY][fix] Fix for upcoming LLVM 10.0

[ROCm/hip commit: 9f3a90e174]
 2019-09-01 18:16:24 +03:00
Evgeny Mankov 8a0224db1d [HIPIFY][fix] Fix for upcoming LLVM 10.0 
[ROCm/hip commit: ce4140f94d]
 2019-09-01 18:15:03 +03:00
Rahul Garg a7388558d6 Merge pull request #1378 from ssahasra/no-as 
remove obsolete test for OCKL Asynchronous Streams

[ROCm/hip commit: 521a162870]
 2019-08-30 19:05:37 -07:00
Sameer Sahasrabuddhe 8afbecf687 remove obsolete test for OCKL Asynchronous Streams 
The implementation for OCKL AS was recently removed from the device
library since that feature is now superseded by hostcall.


[ROCm/hip commit: 70023c9075]
 2019-08-30 20:41:29 +05:30
Evgeny Mankov 6f13788276 Merge pull request #1375 from emankov/hipify-clang 
[HIPIFY][DNN] cuDNN 7.6.3 support

[ROCm/hip commit: 57760c6a57]
 2019-08-30 17:26:29 +03:00
Evgeny Mankov 4deb26d19f [HIPIFY][DNN] cuDNN 7.6.3 support 
[ROCm/hip commit: b45c26361d]
 2019-08-29 19:55:38 +03:00
Evgeny Mankov ab9f05e6a3 Merge pull request #1372 from emankov/master 
[HIPIFY][#207][fix] Translate all preprocessor's conditional blocks

[ROCm/hip commit: 5d0b628142]
 2019-08-29 10:30:05 +03:00
Yaxun (Sam) Liu a795928420 Fix sample 11_texture_driver for hip-clang (#1370) 
[ROCm/hip commit: c404312ffe]
 2019-08-29 01:03:20 +00:00
Yaxun (Sam) Liu 8aca7f3e25 Do not change opt level for -g for hip-clang (#1369) 
Currently hipcc uses -O3 for hip-clang by default but uses -O0 if -g is used. This
causes surprise for users since -g should not affect default opt level.

[ROCm/hip commit: b046ec698b]
 2019-08-29 01:03:11 +00:00
Sarbojit2019 73e5c52d0d Updated hipErrorString and CUDAErrorTohipError (#1365) 
[ROCm/hip commit: 0722704f35]
 2019-08-29 01:02:59 +00:00
Sarbojit2019 74a3171c6b [HIP] Reclaiming hipLaunchKernel API (#1353) 
* [HIP] Reclaiming hipLaunchKernel API

* Reclaiming hipLaunchKernel : Incorporated review comments

* Incorporated review comments

* Removed hipLaunchKernel Macro from nvcc path


[ROCm/hip commit: 5c4f78bac3]
 2019-08-29 01:02:41 +00:00
satyanveshd f378b67565 [sample] add new cookbook sample - occupancy (#1352) 
* occupancy.cpp with Makefile

* occupancy sample changes according tothe comments

* Changes according to the review comments

* Occupancy Sample Changes

* Changes according to review comments


[ROCm/hip commit: f807cc1a7b]
 2019-08-29 01:01:49 +00:00
mshivama 37779f5bef Device side support for Cooperative Group feature (#1202) 
* first cut of the header implementation of cooperative group feature

* add diclarations for device library functions

* fixed various compile time issues in the CG headers

* enabled copy construction and copy assignment

* fixed a minor bug related to conditional compilation macro

* fixed few more CG constructor issues and added a unit testcase

* fixed typo

* extended unit testcase

* compute size of partitioned CG from mask

* bit of code refactoring

* removed boilerplate code

* fixed few of the review comments by Brian

* Changes to the sigantures of few grid and multi-grid related OCKL functions

* changes to declarations of OCKL functions related to CG feature

* removed all the block level support as it is not planned for 2.9

* Have taken care of review comments by Brian

* Have taken care of review comments by Brian

* removed unused functions which were initially intended to use in block level cg support


[ROCm/hip commit: d75dc4eb29]
 2019-08-29 01:01:25 +00:00
Evgeny Mankov e78eb9bb47 [HIPIFY][#207][fix] Translate all preprocessor's conditional blocks 
+ Start to translate preprocessor's false conditional blocks too:
  based on clang's https://reviews.llvm.org/D66597;
  available only starting from LLVM 10.0 or trunk.
+ Option -skip-excluded-preprocessor-conditional-blocks for skipping excluded conditional blocks:
  the default behavior for hipify-clang built with LLVM < 10.0;
  false by default for hipify-clang built with LLVM 10 or trunk.
+ Add 4 preprocessor unit tests, 2 of which are LLVM 10.0 only
+ Update couple of existing tests by setting -skip-excluded-preprocessor-conditional-blocks option:
  update lit testing accordingly


[ROCm/hip commit: 24be21495d]
 2019-08-28 21:17:35 +03:00
Evgeny Mankov d81b612791 Merge pull request #1371 from emankov/master 
[HIPIFY][cmake] Minor refactoring

[ROCm/hip commit: 1ab81d9ce6]
 2019-08-28 20:42:41 +03:00