Граф коммитов

1203 Коммитов

Автор SHA1 Сообщение Дата
Maneesh Gupta ca142c34b7 Fix broken tests due to ffb2f23
Change-Id: I847c80f8462e1c955bdef957e6de2841a3a6ab29


[ROCm/hip commit: 4fc082ff09]
2016-11-18 12:20:47 +05:30
Aditya Atluri 68c2b11ed8 moved runtime macros to runtime_api.h
Change-Id: Ib47e449328e8e6ec55d1b6ee19899de4b591ea8e


[ROCm/hip commit: 1618cb3f85]
2016-11-17 14:19:18 -06:00
Aditya Atluri 7192577e38 added texture header to memory api source
Change-Id: I1af6d60aca5a9a9ef1cadf8c304bea892acbe061


[ROCm/hip commit: c20c524400]
2016-11-17 11:57:53 -06:00
Aditya Atluri ffb2f233b9 make texture as seperate header as of now
Change-Id: I3c65aa75f2f729eedd8c3292fa3cbc37709c1cfe


[ROCm/hip commit: dc64a732d8]
2016-11-17 11:55:29 -06:00
Aditya Atluri da0a8cd032 Added i8 packed math intrinsics
1. Added add, sub, mul packed math i8 intrinsics
2. Removed c++ packed data structures included from HCC

Change-Id: I1d109c5ce10c48b7cd3ea059478b88fc1de78499
TODO: Add better packed data structures support


[ROCm/hip commit: 12dd9df88f]
2016-11-17 01:09:12 -06:00
Maneesh Gupta c7f6ddd612 Enable USE_COPY_EXT_V2 by default
Change-Id: I2c0dc80f85a0ccb5744715b5418a604e38b249ed


[ROCm/hip commit: 888a3528d2]
2016-11-15 10:42:27 +05:30
Ben Sander d35f12558c tweak hcc demangler
[ROCm/hip commit: 0c624c009b]
2016-11-14 15:26:27 -07:00
Sandeep Kumar 857fd273f1 Add p2p for cookbook
Change-Id: Id2e77ab31123ef95885d665efe34bc0d4596733a
(cherry picked from commit 6fbd0352713ca36e399b1ed4f17c486207a53875)


[ROCm/hip commit: 09b157ca8c]
2016-11-14 06:10:36 +05:30
Maneesh Gupta 3b85d1aec1 Revert "hipcc: Turn back linking hip_ir.ll by default"
This reverts commit b602b3d168.


[ROCm/hip commit: fd1483ce35]
2016-11-14 06:05:31 +05:30
Ben Sander dbe9548077 Add draft doc on profiling with hip.
Change-Id: I79727dd2500333b3f16acb381dd5852a15ed408a


[ROCm/hip commit: faf2a1e01a]
2016-11-13 10:01:05 -06:00
Ben Sander 3ce663c5ae Add &nbsp to demangler
Change-Id: I89586c7c17f5152b7a6850d0d6c2aa1d3ebc8190


[ROCm/hip commit: c9401cb95f]
2016-11-11 16:50:56 -06:00
pensun a446102fcc Add direct test case for threadfence_system workaround
Change-Id: I5b21b590e957c901044741ac94e816cd8b1426f9


[ROCm/hip commit: 50867efa10]
2016-11-11 15:09:43 -06:00
Aditya Atluri 96f7a083c9 fixed multi-dim module kernel launch
Change-Id: Id1d81f2375d058979ab526433f905cf0ea3d23d6


[ROCm/hip commit: abf6872b2b]
2016-11-11 12:25:23 -06:00
Ben Sander 1c7e71e1b3 Add option to deny peer access.
Also fix test.

Change-Id: I1b247f6c4271442b008e560669bca4daf8eb94c7


[ROCm/hip commit: 1e5515ee9f]
2016-11-10 23:12:48 -06:00
Ben Sander 99ce555300 Use forceUnpinnedCopy to resolve P2p corner cases.
Change-Id: I2aebb419881246cebb696bec87798635bc71acc2


[ROCm/hip commit: 65584e48de]
2016-11-10 23:12:48 -06:00
Ben Sander 9828c7e51c Enable async copy again.
Also add HIP_FORCE_SYNC_COPY chicken bit.

Change-Id: I76a385410494b99bf27305d3c08f55dd81987565


[ROCm/hip commit: d3d6feb4de]
2016-11-10 23:12:48 -06:00
Ben Sander 8e1fc87b52 Doc change only - add comments to test.
Change-Id: Ie42087cf3c78e49337b18bb71f3f0e1e7950ee1b


[ROCm/hip commit: 8724273f28]
2016-11-10 23:12:48 -06:00
Ben Sander e0ef3bb319 Refactor copy and P2P logic.
Prefer use of source-engine for DMA copies, even if user submits copy
in a stream attached to a different device.
The stream is now used only for synchronization, and HIP
makes the most optimal decision for which engine to perform the
copy - typically the source copy engine.

HIP now makes decision on which engine should perform the copy
and passes this to HCC using new apis.
HIP has additional information about peer
visibility and will make a decision which agent should perform
the copy .

Change-Id: I0cf4cfebeae256e6ca795f08a7ed7130f4857d1f


[ROCm/hip commit: ced9d72d94]
2016-11-10 23:12:48 -06:00
Ben Sander 67a305ec9e Improve memory debug
Change-Id: I0f033139aa4e4b47039eb016e404009127bd0a44


[ROCm/hip commit: 2dea3a0b1a]
2016-11-10 23:12:48 -06:00
pensun d88f30f63b Update depreciated information for threadfence_system()
Change-Id: Id13d2f81edb51eb42b896a5c06913d59ec907c55


[ROCm/hip commit: 1ec5761a11]
2016-11-10 11:55:12 -06:00
Maneesh Gupta fcfbfdf318 CMakeLists.txt: Cascade CMAKE_BUILD_TYPE to tests
Change-Id: I53a3ea951c1fd57e43a02381a457c1dedc1a34f7


[ROCm/hip commit: a12d5a8989]
2016-11-10 21:26:34 +05:30
Rahul Garg cfab2571a0 hipDeviceGetByPCIBusId support for HIP/NVCC
Change-Id: I8f82890e88d2a15f592bff192179e7d5c5362722


[ROCm/hip commit: fcb94863f7]
2016-11-10 11:40:59 +05:30
Maneesh Gupta 8aaa4d6744 hipcc: Default to HIP_LIB_TYPE=1
Change-Id: I83b05accd76f7bc94bd724c66ae060fa0095bc8d


[ROCm/hip commit: 669d734624]
2016-11-10 11:34:00 +05:30
Maneesh Gupta ee5d7ba293 hcc_dialects/Makefile: use clamp-config
Change-Id: I86df82f75b75125825e22d0545209a19386d9936


[ROCm/hip commit: 36024deb3a]
2016-11-10 11:31:50 +05:30
pensun bbd526b7e5 resolve conflicts for git pull
Change-Id: Ie353b831e2241bc28042069b6cc7405257e871e1


[ROCm/hip commit: 4d7ac1e091]
2016-11-09 21:38:43 -06:00
pensun e1d9d27254 Add documentation on threadfence_system workaround guidelines.
Change-Id: I9636a3808798f3dabe992285ce5652187cee6eb8


[ROCm/hip commit: 24c621db5b]
2016-11-09 21:36:30 -06:00
pensun 74b0840cc6 Add option to alloc fingrained system memory
Change-Id: Ia13c8e058cb988b5857e75a590a4d67411362ae1


[ROCm/hip commit: f7e9f12bf1]
2016-11-09 21:36:30 -06:00
Maneesh Gupta bbc3811662 Merge branch 'rocm-rel-1.3' into amd-develop
Conflicts:
	include/hip/nvcc_detail/hip_runtime_api.h

Change-Id: I990a7d008da9e8dcc68250cebbc8ee6e723c7e01


[ROCm/hip commit: 72c722c3d6]
2016-11-10 08:56:38 +05:30
pensun 007513296a fix hipProfiler* apis on NV path
Change-Id: I6adca6151fef3a9b35348163eb6bd13f5c414172


[ROCm/hip commit: 57cd3c8244]
2016-11-09 15:44:01 -06:00
pensun ca0c0b4061 fix for hipcallback function on NV path
Change-Id: If80c0cfe60b1f3b1a71627b5f3f79503cba4d491


[ROCm/hip commit: 76c3c20da6]
2016-11-09 11:33:23 -06:00
Maneesh Gupta 56a686a0b9 Update release notes for 1.0 release
Change-Id: I74fa2b41afc334a76c309b125c27aa141cd59554


[ROCm/hip commit: 7c68768a0d]
2016-11-08 16:31:56 +05:30
Ben Sander 58bd8617e8 Fix tests to read warpSize from device props.
Change-Id: I9583577793afad49f9eb1ee9069bd4c6963a6023


[ROCm/hip commit: 7106dc6796]
2016-11-06 04:26:28 -06:00
Ben Sander b478d8a0b8 Update gitignore for some common output files
Change-Id: I9cd60f042af4dba07fe0fdbd2ee442936ff8c7bd


[ROCm/hip commit: c59fe09a81]
2016-11-06 04:26:15 -06:00
Ben Sander 56cd9ce82f Improve Peer support and testing.
Change-Id: Icadc65988aaf145a265587ab0357c5bf4d26f3eb


[ROCm/hip commit: d728819d17]
2016-11-06 03:22:36 -06:00
Ben Sander 2a9aa89491 Set forceHostCopyEngine for other copy dirs. Support HIP_FORCE_P2P_HOST
Also: more debug for copy and P2p.

Change-Id: I87030c525410e041b2a00baaf6c68e6c0977ff42


[ROCm/hip commit: 092b3dacda]
2016-11-04 19:53:23 -05:00
Ben Sander 5741930acb Expand hipP2PSimple testing.
Cover cases where P2P is used for H2D copies, where host is pinned
but not accessible to the copy agent.

Change-Id: I9464b787228b40f93473708c3fde9726e1986365


[ROCm/hip commit: ff3298f40e]
2016-11-04 16:13:32 -05:00
Ben Sander fe1d895f07 Refactor resolve-mem step1
Change-Id: I7b8b2bbb56d7b31a97b48ebd42002641cd07a460


[ROCm/hip commit: 5d79384832]
2016-11-04 09:37:56 -05:00
Ben Sander 1e5540e07f Add debug for Peer APIs. Enable PeerMemcpy APIs by default.
Change-Id: I46e39a9e7b07686a78484c1f3b5495b08e052fbb


[ROCm/hip commit: 3f0a2b8dc1]
2016-11-04 08:51:16 -05:00
Ben Sander 411cf6c80a Print non-peers too
Change-Id: I2a6905edcdf144aa732ae3120c17780477f232ac


[ROCm/hip commit: 5832349e5e]
2016-11-04 06:34:07 -05:00
Ben Sander 6444dddaa7 Pre-pend HIP_PATH/lib to linker, so we find developer object code
Previously might pick up libs from /opt/rocm/lib.

Change-Id: Ia7adb345defe433d5952aa61706fe03fd7cbcd35


[ROCm/hip commit: f0bb817c11]
2016-11-04 06:06:04 -05:00
pensun 9251040786 Update document for workaround suggestion on threadfence_system()
Change-Id: Icccab8270604a0e578a8614b9afb3f95372f4966


[ROCm/hip commit: 774de273d0]
2016-11-02 16:08:27 -05:00
pensun 959880e318 Update hipStreamNonBlocking to use cuda define on NV path
Change-Id: I74ea09db99d602ba1c5f192b36ff7f2781176e6a


[ROCm/hip commit: 4817131cdc]
2016-11-01 20:30:56 -05:00
Aditya Atluri 42130918bb added inter thread data movement intrinsics
Change-Id: I2a8a8ed49429cb7f96439bd28c4b83b5142737df


[ROCm/hip commit: f48c53534e]
2016-11-01 16:37:33 -05:00
Rahul Garg 0aa4468ac3 Added hipDeviceGetByPCIBusId in hip/hcc path
Change-Id: I3cca0dc533d0281689d8a407c7da16ca1ba6a3a8


[ROCm/hip commit: 2d15d0741c]
2016-11-01 10:57:48 +05:30
Evgeny Mankov 0fd5ec74c0 [HIPIFY] wrap kernel name with HIP_KERNEL_NAME macros...
only in case of commas in it.


[ROCm/hip commit: 1b64c167f6]
2016-10-28 20:05:51 +03:00
Evgeny Mankov 4b907cce25 * [HIPIFY] Initial Profiler support.
CUDA Driver API porting to HIP:

+ cuProfilerStart, cuProfilerStop.
- cuProfilerInitialize & cudaProfilerInitialize - unsupported yet by HIP.


[ROCm/hip commit: b0f37d0d2e]
2016-10-28 18:32:13 +03:00
Ben Sander ae8d55ccc1 add hip_profile.h
Change-Id: Id43a4336db53567020584cb7842baf5c1649fd8e


[ROCm/hip commit: 87a2e8f12b]
2016-10-28 07:08:46 -05:00
Maneesh Gupta d1a803e5f2 hipdemangleatp: Try handling HC kernels as well
Change-Id: Ie438ddd28e5bc6067fcd682df849d3183046b40a


[ROCm/hip commit: c39ddad215]
2016-10-28 15:46:59 +05:30
Maneesh Gupta 4009e17202 CMakeLists.txt: Update include paths needed for Fedora support
Change-Id: Ib84f9dba30d2c64f344d6f8e85ddbe15f30af1a0


[ROCm/hip commit: c5a2ad4964]
2016-10-28 14:12:53 +05:30
Maneesh Gupta 6aa40b6f21 hipcc: Update flags for Fedora support
Change-Id: I90be7768410e491b4f11c3b0f08470246d781d80


[ROCm/hip commit: 6fa9bc3b61]
2016-10-28 14:12:13 +05:30