e2dc1cc27f
Fixes a typo: perforamnce -> performance
...
Change-Id: I85e3b3d22c98c16556227283bfb33530e1bce2cf
2016-12-15 14:01:08 +05:30
a1d1fcfdac
fixed compilation issues
...
Change-Id: I96692538736e2e4f2da9dba9c8c29a164aec4c0d
2016-12-14 16:50:16 -06:00
c20a86d866
added half2 support
...
Change-Id: I0f3b9b7037fed97e80ec99f5369c75a63f001aae
2016-12-14 14:18:48 -06:00
8b0ce6e81b
Fixes in Makefile of couple of samples
...
- modified Makefile for hipblas_saxpy to replaced hcblas.so with hipblas.so as part of HCSWAP-100
- Resolved missing separator issue in peer2peer cookbook Makefile
Change-Id: I678fea267eee1481f02da09379339ed78d3f95f2
2016-12-14 16:27:14 +05:30
01ed8e91e9
added simple half math ops
...
Change-Id: I10b1d1023a9e5f2ba63f28c4a2bbe60ee49a8aee
2016-12-13 20:20:58 -06:00
26934a920c
disabled compiler flag hcc 4.0 for half support
...
Change-Id: I32175113f4c05d43310b3a05c2a14e12f6d48b09
2016-12-13 20:06:56 -06:00
5eed20be12
Add USE_IPC to disable use of IPC APIs. Set to 0.
2016-12-13 15:07:04 -06:00
7a712aa76b
added few type reinterpret cast device functions
...
1. __int_as_float
2. __hiloint2double
Change-Id: Id247c196887b24a12090f0521bf91e13afeec733
2016-12-13 14:41:36 -06:00
02eab122c5
added half math addition ISA support
...
Change-Id: I293b771f695b499b795d7e53f600c9e4fe2a2071
2016-12-13 09:18:34 -06:00
6bcacf83a4
[HIPIFY] Code refactoring and total stat collecting
...
+ Single base class for Preprocessor and MatchFinder classes.
+ Total Stats for multiple files is added.
2016-12-13 18:01:08 +03:00
b73ffa4e7c
[HIPIFY] Rename -n to -examine
2016-12-12 20:03:01 +03:00
d5eb7f494f
[HIPIFY] Multiple source files support.
2016-12-12 19:57:19 +03:00
812cf130b6
Changed threadfences to match target parsing by hcc
...
Change-Id: I28fcabbaacd13495b707f263fd09afaead0665fa
2016-12-12 10:16:58 -06:00
c5f5e662ef
[HIPIFY] -no-backup option is added.
...
Is useful for release hipifying in place.
2016-12-08 23:14:19 +03:00
f5deff4ba3
[HIPIFY] -no-output support.
...
Actually output file is created by clang itself, but isn't updated and is deleted after processing.
In cooperation with -print-stat -no-output (or single -n) is used for examine the source CUDA code.
Conflicting options:
-inplce
-o
2016-12-08 22:45:10 +03:00
a81dfc5a0f
Fix the limitation of supported input files.
...
Actually .cu and .cuda was supported.
+ All the file names are allowed, including file names without extension.
[IMPORTANT]
To hipify CUDA input file, which name is not *.cu, please add option "-x cuda" after tool's options ending marker "--", for instance:
./hipify-clang NeuralNet -- -x cuda
This option will go to clang itself, not the hipify tool.
2016-12-08 20:28:43 +03:00
a6b2f9c3a0
Fixed build error due to GetPCIBusId overloaded function
...
Change-Id: I626446f2c72c8143f08c95367bc1c528abeaf69d
2016-12-08 14:35:58 +05:30
c677041b37
hcc_detail/hip_runtime_api.h: Fix IPC API signature
...
Change-Id: I0be0f09c62f231620341141bd66183c3338be56a
2016-12-08 12:50:25 +05:30
7ac5f2e8c3
HIP IPC implementation on ROCr IPC APIs
...
Change-Id: I1ca9d520f5d0b1b56694211471b81eb7c6c23d16
2016-12-07 15:38:36 -06:00
d8fdd6c6fc
hipDeviceGetPCIBusId int version changes for CUDA runtime API
...
Change-Id: I4d3b995f1d1ac83415ca84808a074e5c8cd72f3c
2016-12-07 12:12:40 +05:30
8f2a1752c1
Fix issue of ihipModule_t double defined
...
Change-Id: I508677e595776fd573a2f224691116d01288dc78
2016-12-06 19:10:17 -06:00
8fcae3acc0
change hipgetPCIID to take int as third parameter
...
Change-Id: I4429b36756a6d868a769abd783bf28a55147c0d0
2016-12-06 14:24:09 -06:00
092924d660
IPC prototyps and part of the implementation included
...
Change-Id: Id88c7f155d23ec63f57a6ef05098fba43f8af336
2016-12-06 14:24:09 -06:00
808e555247
local changes for hipnccl
...
Change-Id: I05a1f0381ce2914a800f573342cc954eb5ff82d9
2016-12-06 14:22:02 -06:00
3f810a203b
Enabled USE_DISPATCH_HSA_KERNEL, with serialization in hipModuleUnload.
2016-12-06 10:14:01 -06:00
afc81c955e
Simple test case for hipDeviceGetPCIBusId
...
Change-Id: I0fb6d1bef4739a5b6280928f7b349d95c1656431
2016-12-06 17:31:54 +05:30
a019f1ce39
Build Error correction in hipDeviceGetPCIBusId
...
Change-Id: I50ff4d95b7a732924c7a991cba60400b1c93c0de
2016-12-06 17:09:21 +05:30
0017419521
Changed hipDeviceGetPCIBusId to return Bus ID as string
...
Change-Id: I6d5aa7362084109d34bc015d948f8723b2a38ee9
2016-12-06 16:55:17 +05:30
b8413a69c8
Remove redundant variable g_malloc_heap_size
...
Change-Id: Idaf47be70488f0deb3eab05a86d9c5a413d3fff7
2016-12-06 10:19:03 +05:30
4cb1579d4a
HIP resource leaks fix from Jack
...
Change-Id: I93f3ad7cb94ff1cba1577bd8acc90e826693d12e
2016-12-05 20:21:33 -06:00
ac93376c26
Revert "Enable USE_DISPATCH_HSA_KERNEL."
...
This reverts commit f8bcbe8680
2016-12-05 16:55:26 +05:30
ac368cc60f
Don't share g_malloc_heap_size between libraries
...
Change-Id: Ic70bf83d4f865bc5c453941fdbc1814c77f0ad9d
2016-12-05 11:03:45 +05:30
f8bcbe8680
Enable USE_DISPATCH_HSA_KERNEL.
...
Optimize hipLaunchModule dispatch latency.
2016-12-04 00:13:19 -06:00
783ac156ce
Add additional controls for forcing serialization and blocking.
...
Move HIP_COHERENT_HOST_ALLOC so it is read once at init time.
Add HIP_LAUNCH_BLOCKING_KERNELS, HIP_API_BLOCKING.
Update docs on debug and chicken bits.
Conflicts:
src/hip_hcc.cpp
2016-12-02 18:03:59 -06:00
fd8b1c7d41
Improve shared and static library support
...
hipcc accepts new parameter -use-staticlib and -use-sharedlib to
control linking behavior. Default is still static library.
Change-Id: I28fb9a939f8177c75abefd8b77d8118a6666d1f4
2016-12-01 15:33:12 +05:30
f14198956f
hipcc: HCC workaround no longer needs env var
...
Change-Id: I81f4eafddbda5e9e2f1082932dd502ab451cfc24
2016-12-01 12:51:58 +05:30
9cab8d4260
Fix some broken directed tests
...
Change-Id: I40f2661a74951f8d91824df8fd9ef0cc2312d183
2016-12-01 12:47:37 +05:30
e6df8cb28c
added half add and fma intrinsic
...
Change-Id: Ifa60c1a7065f524f069291bb00d987b11c836cc4
2016-11-29 19:46:01 -06:00
504fcaf786
Change to use produce device name by default
...
Change-Id: Ie2cee2a2e94a08b5874a2f5abee5d1ab6c9fdf47
2016-11-29 11:34:06 -06:00
bda0704213
Added support for hipMemGetAddressRange
...
Change-Id: I99a796a4eb765152cf15a12d6a86b58684d34f50
2016-11-29 22:04:09 +05:30
8e2980c7ef
Change the parameter type of hipDeviceGetPCIBusID to char*
...
Change-Id: Ia72f403126e95f65da53208fc246f45d1417381f
2016-11-28 10:47:18 -06:00
a504df955e
Add more debug info
2016-11-26 08:56:02 -06:00
a990806b6c
Tweak profiler doc
...
Change-Id: I7be7c44467510e38ae850e1e0a14209a3b4380f1
2016-11-25 14:06:59 -06:00
de89b25d52
added support for rcp for float and double
...
Change-Id: Ibeba3a9f64494fc0a176bcb4a854fb2f56567b55
2016-11-23 20:01:18 -06:00
cc1f8a1011
added fma for double and float
...
1. Added fma intrinsic support for double and float
2. Added test for fma
Change-Id: I909fdbec34a3d12c03ba6eff3a39376a7128ee43
2016-11-23 18:22:05 -06:00
69b43ec17c
Add some missing APIs on nv path and hipify
...
Change-Id: Ic0f4740ab06bf70b1de61b39fedc7a6e7605cb61
2016-11-23 14:36:30 -06:00
a4b43a6610
Add several missing APIs in hipify
...
Change-Id: I58912871cb0b10128f221ef26a11b0d69fb7873c
2016-11-23 14:06:18 -06:00
c2f6ecf264
Added fast math flag
...
1. Use -DHIP_FAST_MATH to make precise math functions compiled to fast math
2. Added double fast math functions for sqrt
3. Changed hipcc to parse -use_fast_math (not working)
4. Added passed tag to hipFloatMath test
Change-Id: I72884b2436b4efe61e9a9297346c1358fee38a2d
2016-11-23 11:19:15 -06:00
ddb1845ab5
Add toc to hip_profiling.md
...
Change-Id: I3ae100f12686d0398a0403b78ca571382acce135
2016-11-23 08:36:08 -06:00
9db93a1b96
Improve docs in some places
...
Change-Id: If31e84fbf0c8595ca72edb842dce7ce47783579b
2016-11-23 08:16:18 -06:00
Peter Steinberger
Aditya Atluri
Sandeep Kumar
Ben Sander
Evgeny Mankov
Rahul Garg
Maneesh Gupta
pensun