AI/rocm-systems
2
0
Форкнуть 0
Код Задачи Запросы на слияние Действия Пакеты Проекты Релизы Вики Активность
4 089 коммитов 3 Ветки 0 Теги
83bbf215c85b786bf2acdff1963a2c53c0d35e88
Граф коммитов

217 Коммитов

Автор SHA1 Сообщение Дата
Aryan Salmanpour 9ab561dd66 [hip] add support for implicit kernel argument for multi-grid sync (#1456) 
* [hip] add support for implicit kernel argument for multi-grid sync

* modified code for calculating the prev_sum

* change the impCoopArg type to size_t

* add memory clean up

* launch init_gws and main kernels into two separate loops


[ROCm/hip commit: 359dc79101]
 2019-10-24 17:43:30 +05:30
Vladislav Sytchenko 370d039d24 Update the declarations of hipMemsetD8, hipMemsetD8Async, hipMemsetD16, hipMemsetD16Async. These functions are type aware and take in as their third argument the number of elements in the buffer, not the buffer size. Change the name of this argument from sizeBytes to count to align with the above description. 
[ROCm/hip commit: 0b52c1d9d8]
 2019-10-15 14:18:42 -04:00
ansurya 66e94b1b1f Added new Memory API's (#1399) 
Added new memory API's hipMemAllocPitch, hipMemAllocHost, hipMemsetD16, hipMemsetD16Async, hipMemsetD8Async
Modified to support all scenarios hipMemcpyParam2DAsync, hipMemcpyParam2D.


[ROCm/hip commit: ba9c6e13e4]
 2019-10-04 13:36:31 +05:30
satyanveshd bfb64c43a4 Reimplement hipMemGetInfo (#1447) 
Addresses SWDEV-136570. hipMemGetInfo changed to compute free memory based on information from kfd instead of relying on hc::am_tracker.


[ROCm/hip commit: 3d661e4706]
 2019-10-01 12:40:36 +05:30
satyanveshd fe1fae3ec2 Fix checks in hipMemcpy[D/H]to[D/H]() (#1397) 
* SWDEV-202043 hipMemcpyDtoD() issue


[ROCm/hip commit: cfbd0e77fe]
 2019-09-16 08:32:38 +00:00
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
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
Rahul Garg 4aa011eec6 Fix memcpy with IPC slowness (#1321) 
* Fix memcpy with IPC slowness

* Make early erroneous returns

* Real Clean up

* Real Clean up++


[ROCm/hip commit: 71559200c0]
 2019-08-23 09:19:18 +00:00
Jatin Chaudhary 7dca0455e9 Using HSA API for hipMemsetAsync (#1346) 
[ROCm/hip commit: ac62d7a5c0]
 2019-08-21 10:00:10 +00:00
Rahul Garg a984acf245 Add hipMemcpy3DAsync (#1320) 
* Add hipMemcpy3DAsync

* Fix CI build error

* Move back stream resolution to internal function

* Remove stream redefinition and check


[ROCm/hip commit: 2405621f62]
 2019-08-16 02:13:16 +00:00
Rahul Garg d42844182c Add hipMemcpyParam2DAsync (#1296) 
* Add hipMemcpyParam2DAsync

* Add NVCC path changes

* Clean up

* Fix build issue

* Fix else use in both sync and async apis


[ROCm/hip commit: 45b73e0961]
 2019-08-09 11:50:37 +00:00
Jeff Daily 9b44993343 consolidate thread local storage (#915) 
* all thread local access now through single struct

* clean up old commented-out code, more use of GET_TLS()

* fewer calls to GET_TLS by passing tls as a funtion argument

* revert unnecessary change to printf

* fix failing tests due to TLS change

* fix merge conflicts in ihipOccupancyMaxActiveBlocksPerMultiprocessor


[ROCm/hip commit: 1eb3dbf065]
 2019-08-05 09:51:02 +00:00
Rahul Garg 0009dc1067 Change hipErrorUnknown to hipErrorInvalidValue 
[ROCm/hip commit: 483aab031f]
 2019-07-31 00:28:30 +05:30
Evgeny Mankov d11063e64c [HIP] Fix segfault on uninitialized struct members in hipArrayCreate and hipArray3DCreate 
[ROCm/hip commit: 09162d9a53]
 2019-07-12 16:38:26 +03:00
Evgeny Mankov 96801f7b3a [HIP][HIPIFY] Split HIP_ARRAY_DESCRIPTOR struct to HIP_ARRAY_DESCRIPTOR and HIP_ARRAY3D_DESCRIPTOR 
[Reason] To be compatible with CUDA [#1133]

Update HIP code, hipify-clang, tests and docs

[TODO] Add support of the corresponding functions on nvcc fallback path


[ROCm/hip commit: c7117df91b]
 2019-07-11 14:58:16 +03:00
Jatin Chaudhary c7f8ffe41e Adding bounds check before hipMemset (#1190) 
* Adding bounds check in ihipMemset

* Adding ihipMemPtrGetInfo to hipMemPtrGetInfo


[ROCm/hip commit: 5ed16432f8]
 2019-07-08 11:00:38 +00:00
Anusha Godavarthy Surya f1d6b56fc4 Added missing NULL checks and corrected API return values as per validation 
[ROCm/hip commit: 3d5f6be1c7]
 2019-06-27 00:19:05 +05:30
Evgeny Mankov cd309b6638 [HIP][HIPIFY] Make hipMemcpyParam2D coherent with cuMemcpy2D 
+ Makes hip_Memcpy2D struct compatible with CUDA_MEMCPY2D struct
+ Add hipMemcpyParam2D support in nvcc fallback path
+ Update hipify-clang, tests and docs accordingly


[ROCm/hip commit: 8f059b0ee9]
 2019-05-22 18:31:39 +03:00
Rahul Garg c4567ad01a Add fine grained host memory lock support (#1095) 
* Add fine grained host memory lock support

* Fix default flag check


[ROCm/hip commit: aeeab1b23f]
 2019-05-13 11:48:26 +05:30
Rahul Garg c01236f679 Add hipMallocManaged default functional support (#1036) 
* Add hipMallocManaged default functional support

* Fix build error

* Add dtest


[ROCm/hip commit: 2bc2c46d4d]
 2019-04-24 16:50:03 +05:30
Jeff Daily a0172ca884 In hipFree, synchronize owner of memory (#1018) 
* In hipFree, if memory is associated with a device, synchronize that device's streams.

This changes the behavior from synchronizing the currently set TLS device.

* All devices sync in hipFree for _appId=-1 case.

* Revert "All devices sync in hipFree for _appId=-1 case."

This reverts commit 1efb34d6a8426661e45bc5f763422a1147aeac10.

* add HIP_SYNC_FREE env var


[ROCm/hip commit: 2b3037a6ea]
 2019-04-16 08:35:55 +05:30
Rahul Garg 0eaa29ad06 Handle D2D in memcpy2D 
[ROCm/hip commit: 0c55db8552]
 2019-03-28 02:21:45 +05:30
Rahul Garg d98d5ca12a Let hipHostMalloc always share/map pinned host ptr 
[ROCm/hip commit: f0af073793]
 2019-03-26 10:19:13 +05:30
Rahul Garg c6ef785464 Avoid double mapping of devices to hostMalloc buffer 
[ROCm/hip commit: 5e917d70f3]
 2019-03-25 23:07:05 +05:30
Maneesh Gupta 82fd86e63f Merge pull request #970 from mangupta/swdev-172995 
hipExtMallocWithFlags implementation

[ROCm/hip commit: 30b5c02ec4]
 2019-03-25 07:46:53 +00:00
Maneesh Gupta 67819c0395 hipExtMallocWithFlags needs hcc workweek 19115 or higher 
[ROCm/hip commit: cab119c8b2]
 2019-03-25 11:41:20 +05:30
Maneesh Gupta 9ac6005d35 hipExtMallocWithFlags implementation 
Change-Id: Iee9e119796472200b2933d5e23be60813f33bc75


[ROCm/hip commit: 73ec5d54b5]
 2019-03-19 11:59:22 +05:30
Rahul Garg a3fb908a0a Add 2D fallback to use copy kernel 
[ROCm/hip commit: 918d7e3a40]
 2019-03-14 13:03:06 +05:30
Alex Voicu 0c16497abd dlopen() fixes (#929) 
* Initial attempt to switch over to internally linked state.

* Add missing CMake update.

* hipLaunchKernelGGLImpl must be inline as well. Ensure internal linkage.

* Ensure global retrieval uses internally linked state.

* Hide HC in the implementation. Minimise ADL woes.

* Strange software exists, and must be catered to.

* Use a less spammy mechanism for ensuring internal linkage / non-export.

* Remove leftover internal detail.


[ROCm/hip commit: ea0fcf3e61]
 2019-03-06 17:31:44 +05:30
Wen-Heng (Jack) Chung da589e38ed Address code review comments to use hipDeviceptr_t 
[ROCm/hip commit: 5cbd28f29b]
 2019-03-05 05:51:05 +00:00
Wen-Heng (Jack) Chung 0b7f38d100 Add hipMemsetD32 and hipMemsetD32Async 
Add 2 extra memset functions which fills memory with integer-typed data

Also change the parameters of ihipMemset to better explain the semantic


[ROCm/hip commit: 7ebbbd3525]
 2019-03-04 17:00:33 +00:00
Wilkin Chau 4a0d68ba3f Fix hipMemset3D test 
Calculate the allocated size based on the width, height and depth.


[ROCm/hip commit: 8d92d1ebd7]
 2019-02-28 22:42:46 +00:00
Evgeny 3f7ff3450e fixing HSA_INIT_API cid args 
[ROCm/hip commit: 0164464bcc]
 2019-01-16 23:45:44 -06:00
Maneesh Gupta 3cf96f31d0 Merge pull request #797 from gargrahul/fixhipPointerGetAttributes 
Fixed hipPointerGetAttributes for hostmalloced ptr

[ROCm/hip commit: 56ce3e37d5]
 2018-12-12 10:16:07 +05:30
Maneesh Gupta 07dcdff9e5 Merge pull request #608 from gargrahul/add_pinned_2d_sdma_copy 
Added support for pinned 2D SDMA copy

[ROCm/hip commit: 0dd26b4f63]
 2018-12-12 07:44:16 +05:30
Rahul Garg b304ff5210 Fixed hipPointerGetAttributes for hostmalloced ptr 
[ROCm/hip commit: 5f12067708]
 2018-12-08 01:42:08 +05:30
Maneesh Gupta 05e09614be Merge pull request #760 from eshcherb/roctracer-hip-frontend-181113 
Roctracer hip frontend 181113

[ROCm/hip commit: 160c509e23]
 2018-11-23 11:08:25 +05:30
Maneesh Gupta e2bc3e49a5 Merge pull request #748 from mkuron/getsymboladdress 
Implement hipGetSymbolAddress and hipGetSymbolSize

[ROCm/hip commit: bcea027bf1]
 2018-11-21 10:32:01 +05:30
Michael Kuron c35dfb71d5 Merge branch 'master' into getsymboladdress 
[ROCm/hip commit: 8610128c3e]
 2018-11-20 12:03:22 +01:00
Rahul Garg 89efed29d7 Fix hipHostRegister 
[ROCm/hip commit: 1a038879a9]
 2018-11-17 05:38:35 +05:30
Evgeny 73e3c4ec42 renaming HIP_INIT_CB_API to HIP_INIT_API 
[ROCm/hip commit: e5ba097afd]
 2018-11-13 15:33:26 +00:00
Evgeny 0a58dc9b7b adding activity prof layer 
[ROCm/hip commit: b8b1637ef7]
 2018-11-13 15:33:26 +00:00
Rahul Garg ecea878072 Fixed hipMemcpyToSymbol doesn't work on GPU other than device 0 SWDEV-166881 
[ROCm/hip commit: 11e7ab8879]
 2018-11-13 00:49:20 +05:30
Michael Kuron f69866eecc Use correct trace macro in hipGetSymbolAddress/hipGetSymbolSize 
[ROCm/hip commit: 6ebcc2922c]
 2018-11-06 20:46:30 +01:00
Michael Kuron cbba8221ee Introduce ihipModuleGetGlobal 
[ROCm/hip commit: 31acf1c268]
 2018-11-06 09:54:34 +01:00
Michael Kuron bc455ccf50 Implement hipGetSymbolAddress and hipGetSymbolSize 
[ROCm/hip commit: 73616582d6]
 2018-11-04 10:39:34 +01:00
Siu Chi Chan 1159b4aa05 Move the global arrays for hip malloc/free 
from a header into a source file such that
there's only an unique copy in an executable
and prevent wasting static memory on the host

Change-Id: Id5b62766f77809c8d7b47892cb7149c490dcbdb9


[ROCm/hip commit: 0ff408a56c]
 2018-11-01 16:20:35 -04:00
Anton Gorenko f2ce51bdf5 Fix allocation size of arrays with multiple and/or non-32-bit channels 
hipMallocArray and hipMalloc3DArray must use sum of bits
of all components.


[ROCm/hip commit: 21f044eac8]
 2018-10-29 18:12:00 +06:00
Rahul Garg 6d53af5a60 Return hipSuccess when sizeBytes=0 in hipMemset 
[ROCm/hip commit: 90f57d452a]
 2018-09-26 12:47:36 +05:30