a42b0f5909
Add hipSetDeviceFlags implementation for NVCC
...
Change-Id: Id19e965950b728cfe0a514e6cdf65a191d063c10
2016-05-20 12:39:02 +05:30
8c11c333e2
Support for Atomic inc and dec in HIP
...
Change-Id: I783e4917cece5cc379894f0d293382315fbfa8b0
2016-05-12 11:10:48 +05:30
5e86860c05
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
2016-05-06 08:09:52 +08:00
20043d602e
Merge branch 'privatestaging' into grid_launch
2016-05-02 18:38:20 -05:00
9e36ee9b95
enable hipHostRegister
2016-04-28 17:54:03 -05:00
a16c58ed46
remove kalmar_defines.h
2016-04-28 17:40:17 -05:00
9363e495e1
remove hacks to disable math or LDG functions.
2016-04-28 14:14:22 -05:00
c60614eca5
Properly pair #if/#endif
2016-04-28 13:43:32 -05:00
5c098d5852
merge fix
2016-04-28 13:24:40 -05:00
06e54694c5
hipcc uses -stdlib=stdlibc++ by default.
...
hipcc switches to -stdlib=libc++ if user requests it.
hack to disable LDG.
2016-04-28 13:17:49 -05:00
a5508ae185
Fix compilation error in hipSinglePrecisionMathHost test
2016-04-27 17:46:27 -05:00
ec8cedc70e
changed to guard from hc.hpp
2016-04-27 17:46:27 -05:00
5552301de2
Merge remote-tracking branch 'refs/remotes/origin/privatestaging' into kernel_synchronization
2016-04-25 14:43:28 -05:00
a849321bd9
Update with original
2016-04-25 14:42:35 -05:00
4aca1babe8
Merge remote-tracking branch 'refs/remotes/origin/privatestaging' into kernel_synchronization
2016-04-25 13:57:28 -05:00
7175ca8824
Merge from original
2016-04-25 13:55:44 -05:00
c404f60621
Fixed git spills
2016-04-22 09:19:05 -05:00
a2bc6ebc9f
added workweek to hip_ldg
2016-04-22 09:14:26 -05:00
02e6fc27f4
Merge branch 'release_0.84.00' into privatestaging
...
Conflicts:
include/hcc_detail/hip_runtime.h
src/hip_hcc.cpp
2016-04-22 10:55:58 +05:30
48f3762160
reorg make_datatype feature
2016-04-21 14:51:19 -05:00
3aac6d45ad
added full data type support for __ldg
2016-04-21 11:17:26 -05:00
6773a64b22
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.
2016-04-20 15:51:39 -05:00
f46e1bcf0e
added copyright for ldg
2016-04-20 12:28:02 -05:00
b493eac7e0
added support for __ldg
2016-04-20 12:25:40 -05:00
517008f98b
update API trace information for hipLaunchKernel
2016-04-20 09:57:55 -05:00
f0b0bc672a
Merge branch 'release_0.84.00' of https://github.com/AMDComputeLibraries/HIP-privatestaging into release_0.84.00
2016-04-20 09:48:52 -05:00
f8ad557864
update hipLaunchKernel API trace information
2016-04-20 09:48:35 -05:00
ccf2c1c323
Fix hipDeviceReset synchronization
2016-04-19 11:56:12 -05:00
4431a93c2c
Fix compilation error in hipSinglePrecisionMathHost test
2016-04-19 23:17:09 +08:00
c4cc76398c
Prefix HIP includes with hip/
2016-04-19 15:02:12 +05:30
21d733b2e7
Add experimental notes around p2p
2016-04-18 23:52:42 -05:00
9329fb8c92
Add USE_HCC_LOCK_API.
...
Default off, if set will compile assuming HCC lock APIs exist.
2016-04-18 23:50:52 -05:00
6abfa13c34
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.
2016-04-18 20:49:33 -05:00
f9a31e28ad
Move HIP_HCC define to CMake
2016-04-17 07:40:04 -05:00
8d26dfcde3
Merge branch 'privatestaging' into p2p
...
Conflicts:
include/hcc_detail/hip_hcc.h
src/hip_hcc.cpp
2016-04-17 06:46:52 -05:00
95e9de0946
Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
2016-04-16 17:10:32 -05:00
dc61929a3d
Corrected Memcpydefault
2016-04-16 17:10:13 -05:00
c3bd85595d
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.
2016-04-16 10:18:56 -05:00
0426564a7b
Add __ldg
2016-04-16 21:12:09 +08:00
1cd52f5192
Bump min required hcc to workweek 16155
2016-04-16 14:49:10 +05:30
85c04d8a7d
changed to guard from hc.hpp
2016-04-15 11:21:45 -05:00
e8902a6e20
improved error returns for hipHostRegister/UnRegister functions
2016-04-15 10:42:31 -05:00
1ab9b0ac71
Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
2016-04-14 09:02:16 -05:00
c3b854fba4
Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
2016-04-14 09:17:30 -05:00
12e24f4816
Enabling make_* device code
2016-04-14 09:17:13 -05:00
596e8e4e4e
Add missing cudaMemsetAsync transformation
2016-04-14 09:02:02 -05:00
0dfb9cddad
Add hip soft-link to include for migrating to new include format
2016-04-14 16:37:27 +05:30
fe0e5f3541
Add missing USE_PEER_TO_PEER macro definition
2016-04-13 13:56:26 +08:00
3b79b86a96
Removed pragma once
2016-04-12 09:22:03 -05:00
42df534354
added C guard to hip_runtime.h
2016-04-11 13:01:02 -05:00
Maneesh Gupta
Rahul Garg
Jack Chung
Ben Sander
Aditya Atluri
bwicakso
pensun
streamhsa