Commit Graph

183 Commits

Author SHA1 Message Date
Jack Chung 9d07571e5d Fix math functions so they can be used on host codes as well
Add __host__ for some math functions to fix the following tests:

- hipSinglePrecisionMathHost
- hipDobulePrecisionMathHost


[ROCm/hip commit: 453d07daf4]
2016-05-06 08:09:52 +08:00
Ben Sander 0d8b15274c Merge branch 'privatestaging' into grid_launch
[ROCm/hip commit: 89df2f4e2f]
2016-05-02 18:38:20 -05:00
Ben Sander bdcfd53eb1 enable hipHostRegister
[ROCm/hip commit: 4b1d3d2cba]
2016-04-28 17:54:03 -05:00
Ben Sander 8c914ece4f remove kalmar_defines.h
[ROCm/hip commit: db7877320f]
2016-04-28 17:40:17 -05:00
Ben Sander e29b2734a8 remove hacks to disable math or LDG functions.
[ROCm/hip commit: 558975c046]
2016-04-28 14:14:22 -05:00
Ben Sander 8e726af2f7 Properly pair #if/#endif
[ROCm/hip commit: 9026146951]
2016-04-28 13:43:32 -05:00
Ben Sander 022595ca71 merge fix
[ROCm/hip commit: efaa278cbf]
2016-04-28 13:24:40 -05:00
Ben Sander 6e21fa88f4 hipcc uses -stdlib=stdlibc++ by default.
hipcc switches to -stdlib=libc++ if user requests it.
hack to disable LDG.


[ROCm/hip commit: bae6462e3b]
2016-04-28 13:17:49 -05:00
Jack Chung 2b6a21af94 Fix compilation error in hipSinglePrecisionMathHost test
[ROCm/hip commit: 6756b8f0bc]
2016-04-27 17:46:27 -05:00
Aditya Atluri fb0560651d changed to guard from hc.hpp
[ROCm/hip commit: cac8110a4f]
2016-04-27 17:46:27 -05:00
bwicakso 25b6fc89d5 Merge remote-tracking branch 'refs/remotes/origin/privatestaging' into kernel_synchronization
[ROCm/hip commit: 77d4b41118]
2016-04-25 14:43:28 -05:00
bwicakso dcd09cebca Update with original
[ROCm/hip commit: c976e6bcd3]
2016-04-25 14:42:35 -05:00
bwicakso 58e026519b Merge remote-tracking branch 'refs/remotes/origin/privatestaging' into kernel_synchronization
[ROCm/hip commit: f0974e5867]
2016-04-25 13:57:28 -05:00
bwicakso 456ea7c20f Merge from original
[ROCm/hip commit: a080650e18]
2016-04-25 13:55:44 -05:00
Aditya Atluri 9ba9051cec Fixed git spills
[ROCm/hip commit: ec23aba6f9]
2016-04-22 09:19:05 -05:00
Aditya Atluri 6b48b1be17 added workweek to hip_ldg
[ROCm/hip commit: 4dac4789ff]
2016-04-22 09:14:26 -05:00
Maneesh Gupta 90fa90295f Merge branch 'release_0.84.00' into privatestaging
Conflicts:
	include/hcc_detail/hip_runtime.h
	src/hip_hcc.cpp


[ROCm/hip commit: ffdf6ab23b]
2016-04-22 10:55:58 +05:30
Aditya Atluri e39de0f560 reorg make_datatype feature
[ROCm/hip commit: 9ca8584ec0]
2016-04-21 14:51:19 -05:00
Aditya Atluri 552acd5ed5 added full data type support for __ldg
[ROCm/hip commit: ba59ea87ab]
2016-04-21 11:17:26 -05:00
bwicakso d24ba1ecbb Fix for kernel synchronization
The completion future of a particular kernel is lost if there are
multiple kernels in the stream. This can cause a racing condition where
the signal associated with the unreferenced completion_future might get
released by hcc runtime.


[ROCm/hip commit: df98fd8531]
2016-04-20 15:51:39 -05:00
Aditya Atluri 7feb990d97 added copyright for ldg
[ROCm/hip commit: de7952cd06]
2016-04-20 12:28:02 -05:00
Aditya Atluri 15fc041fd6 added support for __ldg
[ROCm/hip commit: 805b268ad4]
2016-04-20 12:25:40 -05:00
pensun 3a25eeca87 update API trace information for hipLaunchKernel
[ROCm/hip commit: 883a3fe831]
2016-04-20 09:57:55 -05:00
pensun b197cb30cb Merge branch 'release_0.84.00' of https://github.com/AMDComputeLibraries/HIP-privatestaging into release_0.84.00
[ROCm/hip commit: 2ddade96d7]
2016-04-20 09:48:52 -05:00
pensun e2327b1b18 update hipLaunchKernel API trace information
[ROCm/hip commit: 7f37fc4ec7]
2016-04-20 09:48:35 -05:00
Ben Sander 5858b85e8d Fix hipDeviceReset synchronization
[ROCm/hip commit: 453615ed57]
2016-04-19 11:56:12 -05:00
Jack Chung 104339f238 Fix compilation error in hipSinglePrecisionMathHost test
[ROCm/hip commit: 07efbcf0d0]
2016-04-19 23:17:09 +08:00
Maneesh Gupta b29742d6f8 Prefix HIP includes with hip/
[ROCm/hip commit: bc4f388b69]
2016-04-19 15:02:12 +05:30
Ben Sander b0c6b89dd3 Add experimental notes around p2p
[ROCm/hip commit: 30233cd2e0]
2016-04-18 23:52:42 -05:00
Ben Sander bd8388bfc3 Add USE_HCC_LOCK_API.
Default off, if set will compile assuming HCC lock APIs exist.


[ROCm/hip commit: 3bcefdaa2d]
2016-04-18 23:50:52 -05:00
Ben Sander 32f9d67741 Fixes for P2P and hipDeviceReset
- devicereset would lose track of default stream and thus subsequent
  synchronization calls might not actually sychronize.
- Also deviceReset now correctly frees streams.
- fix waits in P2P staging copy - first phase (Device0-to-Staging) must
  wait for second phase (Staging to Device1) to finish draining the
  buffer.


[ROCm/hip commit: e020d68309]
2016-04-18 20:49:33 -05:00
Ben Sander 698ea7f30f Move HIP_HCC define to CMake
[ROCm/hip commit: 65abde6626]
2016-04-17 07:40:04 -05:00
Ben Sander 262b69f387 Merge branch 'privatestaging' into p2p
Conflicts:
	include/hcc_detail/hip_hcc.h
	src/hip_hcc.cpp


[ROCm/hip commit: 49cc5aec91]
2016-04-17 06:46:52 -05:00
Aditya Atluri 44db3bb0c1 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
[ROCm/hip commit: 846432a1c1]
2016-04-16 17:10:32 -05:00
Aditya Atluri 6cd86531b3 Corrected Memcpydefault
[ROCm/hip commit: 8dc1bdcbe6]
2016-04-16 17:10:13 -05:00
Ben Sander d4f449a95e P2P Update.
- add P2P staging buffer copy.
- If copy device does not have sufficient access permissions, fall back
  to staging buffer.
- improve docs for which copy device is used.


[ROCm/hip commit: dcabc9dbf7]
2016-04-16 10:18:56 -05:00
streamhsa e33cf35ce2 Add __ldg
[ROCm/hip commit: 76f2670cef]
2016-04-16 21:12:09 +08:00
Maneesh Gupta 44e5430da9 Bump min required hcc to workweek 16155
[ROCm/hip commit: dc4c174a54]
2016-04-16 14:49:10 +05:30
Aditya Atluri 9164588f86 changed to guard from hc.hpp
[ROCm/hip commit: a0aef11da7]
2016-04-15 11:21:45 -05:00
Aditya Atluri c53211e700 improved error returns for hipHostRegister/UnRegister functions
[ROCm/hip commit: 2e80de6dcb]
2016-04-15 10:42:31 -05:00
pensun 42df341237 Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
[ROCm/hip commit: bdd26bd1f1]
2016-04-14 09:02:16 -05:00
Aditya Atluri 8c347cd30a Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
[ROCm/hip commit: facc7a39be]
2016-04-14 09:17:30 -05:00
Aditya Atluri 3b82076adf Enabling make_* device code
[ROCm/hip commit: 573224e55f]
2016-04-14 09:17:13 -05:00
pensun a708c400d3 Add missing cudaMemsetAsync transformation
[ROCm/hip commit: 234727c801]
2016-04-14 09:02:02 -05:00
Maneesh Gupta 3f0256a3b1 Add hip soft-link to include for migrating to new include format
[ROCm/hip commit: 011a0335d4]
2016-04-14 16:37:27 +05:30
Jack Chung 181b44df3f Add missing USE_PEER_TO_PEER macro definition
[ROCm/hip commit: 210ba67b84]
2016-04-13 13:56:26 +08:00
Aditya Atluri 10bc7e683f Removed pragma once
[ROCm/hip commit: 4dfe27e8b5]
2016-04-12 09:22:03 -05:00
Aditya Atluri 817cda91ff added C guard to hip_runtime.h
[ROCm/hip commit: d5feabfa1c]
2016-04-11 13:01:02 -05:00
Ben Sander 5060403b2a Merge branch 'p2p' of https://github.com/AMDComputeLibraries/HIP-privatestaging into p2p
Conflicts:
	RELEASE.md
	include/hcc_detail/hip_hcc.h
	samples/1_Utils/hipInfo/hipInfo.cpp
	src/hip_hcc.cpp
	src/hip_peer.cpp


[ROCm/hip commit: 31dc13d2ec]
2016-04-11 09:17:27 -05:00
Ben Sander 7c92c9cc22 P2p checkpoint.
- set USE_PEER_TO_PEER=3 (requires HCC "am_memtracker_update_peers")
- when enabling peer, turn it on for previously allocated memory.
- hipDeviceCanAccessPeer is no longer self-ware (self does not qualify
  as a peer)
- device peerlist always includes self, so when we call allow_access
  we never remove self access.
- hipDeviceReset() removes old peer mappings.


[ROCm/hip commit: 1f53c55d3e]
2016-04-11 12:52:18 -05:00