AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,090 Commits 3 Branches 0 Tags
789d6cc42e352c7170f61ba4029ba198052bc82b
Commit Graph

359 Commits

Author SHA1 Message Date
Maneesh Gupta 789d6cc42e Rename hipComplex.h -> hip_complex.h 
Change-Id: I86af4ddccc6ebb19606156b459e3065d2c979108


[ROCm/clr commit: 4a87f78fa6]
 2016-10-16 11:02:36 +05:30
Maneesh Gupta 96ce4c8e2c include headers: Update copyright header and fix line endings 
Change-Id: If2b0855f4ebf1e966edb54de5667687d154cc574


[ROCm/clr commit: a34090e9c5]
 2016-10-15 22:52:10 +05:30
Ben Sander fc7eefb5c8 Cleanup files from code review. 
- Remove some stale code
- Update docs
- Correct define for __HIP_ARCH_HAS_GLOBAL_INT64_ATOMICS__

Change-Id: Ic5e3cdb8269b1c18f6d2693700b55e08c4d0080e


[ROCm/clr commit: aa3e04cf9a]
 2016-10-15 11:51:20 -05:00
Ben Sander 57adbf2a17 Add code to use new HCC API accelerator_view::dispatch_hsa_kernel. 
Disabed by default, can enable with USE_DISPATCH_HSA_KERNEL=1

Change-Id: I7a6ba76f2bada34952ed47f5335ce695fa2faea5


[ROCm/clr commit: dcadaadb24]
 2016-10-14 23:46:29 -05:00
Maneesh Gupta 426d597cc6 Remove orphaned hip_blas.h from hcc_detail and nvcc_detail 
Change-Id: I7e2dda475b538d30942c52d86fbdb213918c630c


[ROCm/clr commit: 09b3d87767]
 2016-10-14 12:55:50 +05:30
Maneesh Gupta acae18c4ac Remove incorrect executable-bit from non-executable files 
Change-Id: Iacc434374721e01f7d75d0ab54bceabe0b337f54


[ROCm/clr commit: 2e93d2d5d2]
 2016-10-14 12:53:13 +05:30
Aditya Atluri f4bec1719f added limit enum to nvcc 
Change-Id: If9cb6b1205631da36ec18a84f736f2f2f5155885


[ROCm/clr commit: 943878eff4]
 2016-10-13 15:15:02 -05:00
Aditya Atluri 2829a64391 changed hipLimit to hipLimit_t and data type to enum 
Change-Id: I94f408cdcac4b0bb38801d58709b68e9630d44d0


[ROCm/clr commit: 34ab2a55ab]
 2016-10-13 15:13:11 -05:00
Aditya Atluri adca18892a added compiler flag for polaris 
Change-Id: Ib14c14c0618982ac7b48f5bc704c04b54ff40ed9


[ROCm/clr commit: 1970ea07f1]
 2016-10-13 14:16:48 -05:00
Ben Sander 004ca9fdfc Remove hipblas.h header - this is now hosted in the hcBlas repot. 
Change-Id: I5b3350c900741cb1aee75faf4547f2682a9ec385


[ROCm/clr commit: cfccd63b8c]
 2016-10-13 11:51:25 -05:00
Aditya Atluri 7b180aff34 Added HIP_SYMBOL macro to act as a wrapper between HCC and NVCC symbol name parameters 
Change-Id: I008d028b1e29d5a00d0e449af388216396ad2f75


[ROCm/clr commit: 2a55ae10e8]
 2016-10-13 10:31:56 -05:00
Aditya Atluri a208e9be9c added limit api support for nvcc 
Change-Id: Ib6c939e44343158a70e0de7f107d21afc0d0efba


[ROCm/clr commit: 71b98fef5e]
 2016-10-12 21:06:30 -05:00
Aditya Atluri cd32c9418e Added hipDeviceGetLimit api 
1. hipDeviceGetLimit API for HCC path is added
2. Test for hipDeviceGetLimit API is added
3. The feature added only supports querying heap size
4. Corrected indents for malloc and free device functions
5. Removed redundant data structures
6. Added g_heap_malloc_size to store the heap size

Change-Id: If48d1b0ce9270e994f1c542cc283ddbb14746bbb


[ROCm/clr commit: e5325a1ab4]
 2016-10-12 19:58:48 -05:00
Aditya Atluri 9536444ee5 Added copyright for headers 
Change-Id: I35843d8d3b6acd9553ce00150868e15e2baee1da


[ROCm/clr commit: e8029e4b5f]
 2016-10-12 19:12:35 -05:00
Aditya Atluri caa6a298e7 added malloc and free device functions 
1. Added malloc and free device functions
2. Added test which check malloc and free functions
TODO: Need to add support for multiple device. Works only on one device (multi device support id NOT available).

Change-Id: Id11fc36463915d6ad46c264d5a20c8feb2d2c17c


[ROCm/clr commit: 2918ff273d]
 2016-10-12 19:08:34 -05:00
Aditya Atluri 84476c5e33 changed memcpy and memset device functions 
Change-Id: Ia7f450536a75fad4fe13c7fcf5e9e7a9b5450f52


[ROCm/clr commit: ff59ea76f1]
 2016-10-11 17:43:15 -05:00
Aditya Atluri ac89997057 changed hipTestDeviceSymbol test to compile for both nvcc and hcc path 
Change-Id: I041770ad59d4f88d0c8d27d90cdc8a799935ada1


[ROCm/clr commit: 8e8939d775]
 2016-10-11 13:50:31 -05:00
Aditya Atluri 41a18e4a01 added hipMemcpySymbol feature to nvcc path 
Change-Id: I78d45036083fba62d2b2e4e58f9c72cb4e6eb54f


[ROCm/clr commit: 5274244f7e]
 2016-10-11 13:34:54 -05:00
Aditya Atluri 41077d99af added more changes to memcpytosymbol 
1. Refactored code to use HCC internal APIs rather than HCC copy APIs
2. Added hipMemcpyToSymbolAsync
3. Added test for hipMemcpyToSymbolAsync
4. Added new error hipErrorInvalidSymbol

Change-Id: I0e359b2d0ff5d682bbccdf9c2923e16b35e39497


[ROCm/clr commit: 0bf811b875]
 2016-10-11 13:29:46 -05:00
Aditya Atluri 68b3ae7fce Added feature for memcpy to Symbol 
1. Currently works only for __attribute__((addrspace(1))
2. Need to pass in string for name of the variable
3. Added test to check functionality

Change-Id: I4c3cc1bf151cb5423e4aef59fcc4ad5693b31641


[ROCm/clr commit: 89b576da65]
 2016-10-11 12:09:58 -05:00
Aditya Atluri df5370763a added threadfence feature for hcc 
1. Added feature for __threadfence and __threadfence_block
2. Added feature for using LLVM IR files directly while compilation
3. Added test for threadfence and threadfence_block

Change-Id: Ib7e5d89b4cca1a135952b317e5809cd05b56a3c9


[ROCm/clr commit: 09c9953649]
 2016-10-10 15:29:50 -05:00
Maneesh Gupta 4cef1070b2 Add back deprecated hipHostAlloc, hipMallocHost, hipFreeHost 
Change-Id: Ib8494078c852b07e1958c3acc21fa1866542122c


[ROCm/clr commit: b59659f0c7]
 2016-10-09 16:30:46 +05:30
Elias Konstantinidis c7da2fe434 Added support for __mul24 and __umul24 
[ROCm/clr commit: adc763f440]
 2016-10-06 09:29:36 +05:30
Maneesh Gupta 46feb7dcf2 Move include/* to include/hip/* 
Change-Id: I7a7b2839b4df59c7a4c503550f99fdc9e45c0f54


[ROCm/clr commit: bbfc08f419]
 2016-10-04 22:17:18 +05:30
Ben Sander c47374f479 Fix hipMemSetAsync compilation, bounds check on name copy 
[ROCm/clr commit: 48849feba3]
 2016-10-03 19:05:19 -05:00
Aditya Atluri 21ef342e65 Stream fix on nvcc hipMemsetAsync 
Change-Id: Ia0eb81dff0f422af55d93b4635d42e9aa6921377


[ROCm/clr commit: 3edff7dd3b]
 2016-10-03 12:38:41 -05:00
Ben Sander 77014b3165 disable DB_SHOW_TID 
Change-Id: Ia415437e8ba9903de149997380f61f1272488642


[ROCm/clr commit: 0bcb54be35]
 2016-10-02 05:56:38 -05:00
Aditya Atluri 91627f9bae added vector types for C path 
Change-Id: I4ccd4082f1b2ce4f0f4fd9fb498506234783a803


[ROCm/clr commit: c31ee6b25d]
 2016-09-29 19:44:00 -05:00
Aditya Atluri b474890896 Added C guard for math headers 
Change-Id: I0e71819c6436b3e6c99b2deddb2ee2d2a16acedd


[ROCm/clr commit: 2d5bf72b27]
 2016-09-29 13:54:34 -05:00
Aditya Atluri 92429ae17d Fixed issue on git #39 
Change-Id: I5b507fc2b544df0cd3a900a2763c7a3ad6295c13


[ROCm/clr commit: ca033e19cf]
 2016-09-29 10:06:26 -05:00
pensun 3b2e733b19 add hipEvent* macros on NV path 
Change-Id: I28ff0fa24f69560e13366e7cd8d3a485665c67e1


[ROCm/clr commit: 4570877142]
 2016-09-29 01:22:19 -05:00
Ben Sander d5d764a4d4 Refactor asyncCopy and syncCopy to fix deadlock case. 
- Minimize time that locks are held.
- Eliminate copy code that locked stream and ctx at same time.
    - Stream was locked to ensure thread-safe enqueue to the queue.
    - Devices were locked to query peer-lists.

Change-Id: Ibe8880bb7fb995a3da8f90ff911f212d81525018


[ROCm/clr commit: 806e58ca63]
 2016-09-27 15:45:40 -05:00
Ben Sander a0ec18c516 Add debug option to print ThreadID with each message. 
Also print messages with single fprintf to prevents threads from
interleaving.

Change-Id: Ib3999fe6b1e67b4a16cd7dcde82f3dfc99dd48ff


[ROCm/clr commit: 4fd4168796]
 2016-09-27 15:45:40 -05:00
Ben Sander 46a4854bae Fix signal resource issue. 
Remove memory leak with new hc::completion_future.
Implement HIP_LAUNCH_BLOCKING with queue-level wait.

Change-Id: I45975f81c4d239fdeed7776970988d28449865dc


[ROCm/clr commit: f6717e326e]
 2016-09-26 16:47:32 -05:00
Rahul Garg e77925d911 Doxygen documentation changes for hipCtxXXX APIs 
Change-Id: I1e5e3a621b732bd8bdeef1c607c004d6adfdab32


[ROCm/clr commit: 3d1a61f1d9]
 2016-09-26 23:11:07 +05:30
Rahul Garg 1b6c36d22b Doxygen related document updates for texture/ array related functions 
Change-Id: Iedc5b6512b5ddecb98c0e51a353957f08b955fde


[ROCm/clr commit: 25e9308831]
 2016-09-26 16:18:53 +05:30
Rahul Garg 0206224b5e Removed deprecated hipHostAlloc 
Change-Id: Ia592a3545c5d72f37e049ce29f910e404323c01c


[ROCm/clr commit: c313836013]
 2016-09-26 10:58:46 +05:30
Rahul Garg 71d0254bd4 Removed deprecated hipMallocHost 
Change-Id: I3141fe05a34b6a306297a30721509965f49ccb97


[ROCm/clr commit: f5b463702e]
 2016-09-26 05:52:21 +05:30
Rahul Garg eedc7f410e Removed deprecated hipFreeHost 
Change-Id: I9747ea3993090e0da6a2e1f5e9ce318762bc03e1


[ROCm/clr commit: 3a6acd044b]
 2016-09-26 05:35:14 +05:30
Ben Sander 4cf9803703 Peer-to-Peer improvements. 
- Bug fix for peer visibility.  Now contexts correctly detect when they can use SDMA for P2P vs staging buffers.
- Interface to new HCC copy_ext function.
- Improve context and peer print /debug options.
- Add comments and usage to hipPeerToPeer_simple test.


[ROCm/clr commit: 944d4eaa18]
 2016-09-22 14:21:19 -05:00
Ben Sander 976238ad10 Add kind translation for hipMemcpyDeviceToDevice 
[ROCm/clr commit: a397e60ba9]
 2016-09-22 14:21:19 -05:00
Ben Sander 03866f0ee5 Remove unpinned_copy code. Other cleanup. 
Change-Id: Ie3f71439cf1ba729ef223d078917c403d3de879a


[ROCm/clr commit: 926aefa930]
 2016-09-22 14:21:19 -05:00
Ben Sander 8e55e13469 Cleanup. Remove cfs, ihipSignal_t, staging buffer calls. 
Change-Id: I8bb67c484e3a65be06a03665f059217930da2bed


[ROCm/clr commit: 5bc18c8603]
 2016-09-22 14:21:19 -05:00
Ben Sander 8340e57194 Cleanup: Remove HIP signal pool. 
Change-Id: Icebfd0509d12396cc5933d5556d68b53e1be36e0


[ROCm/clr commit: 17e21309f1]
 2016-09-22 14:21:19 -05:00
Ben Sander dbbd5ce037 Remove HIP command dependency tracking. 
Change-Id: I991c13bc5108193959ba70f9f6f9c692c9ad3a5b


[ROCm/clr commit: c365785c5c]
 2016-09-22 14:21:19 -05:00
Ben Sander bafa586cf5 Cleanup, remove preCopyCommand. 
Change-Id: I3768d3789a99be8136b43179d4152fa1875665cb


[ROCm/clr commit: b0509c5a58]
 2016-09-22 14:21:19 -05:00
Ben Sander 2f4c30ff79 Remove USE_AV_COPY, USE_PEER_TO_PEER fallback paths. 
Change-Id: I9c20173e62029c4caebabc98784c6d7697758e4f


[ROCm/clr commit: 60f797cc5a]
 2016-09-22 14:21:19 -05:00
Ben Sander 16f818b60f Remove HIP_STAGING_BUFFER 
Code simplification/cleanup:
Remove stale fallback paths that uses something besides the unpinned engine.
Remove HIP_STAGING_BUFFER env var - now is const 2, 0 no longer has
special meaning.

Change-Id: I7d24cdd1067dd0c244e87b6a83897cb135d307e7


[ROCm/clr commit: 39750da9a5]
 2016-09-22 14:21:18 -05:00
Ben Sander 872c63b03a Move isLargeBar to UnpinnedCopyEngine constructor. 
Change-Id: I7a7d3a40b1d4e0c6ec856658a6a70e5e70d287ce


[ROCm/clr commit: 17d7fc586e]
 2016-09-22 14:21:18 -05:00
Ben Sander 435c528fd4 Refactor Staging Buffer CopyDeviceToHost 
Use copyMode.  Embed algorithm selection inside the unpinned class.

Change-Id: Ic75fd5931717a3160904402794bbed3ccd445112


[ROCm/clr commit: 77c86934c1]
 2016-09-22 14:21:18 -05:00