AI/rocm-systems
2
0
Atdalīts 0
Kods Problēmas Izmaiņu pieprasījumi Darbības Pakotnes Projekti Laidieni Vikivietne Aktivitāte
1 926 Revīzijas 3 Atzari 0 Tagi
fbfeaa346cdeeda7bdc8a29bee83389bcf29ff63
Revīziju grafs

606 Revīzijas

Autors SHA1 Ziņojums Datums
Maneesh Gupta 136d220c10 Revert "Enable USE_DISPATCH_HSA_KERNEL." 
This reverts commit bbd9d98af7.


[ROCm/clr commit: ac93376c26]
 2016-12-05 16:55:26 +05:30
Maneesh Gupta b01694f524 Don't share g_malloc_heap_size between libraries 
Change-Id: Ic70bf83d4f865bc5c453941fdbc1814c77f0ad9d


[ROCm/clr commit: ac368cc60f]
 2016-12-05 11:03:45 +05:30
Ben Sander bbd9d98af7 Enable USE_DISPATCH_HSA_KERNEL. 
Optimize hipLaunchModule dispatch latency.


[ROCm/clr commit: f8bcbe8680]
 2016-12-04 00:13:19 -06:00
Ben Sander bccb897f84 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


[ROCm/clr commit: 783ac156ce]
 2016-12-02 18:03:59 -06:00
Aditya Atluri 2c51f72e41 added half add and fma intrinsic 
Change-Id: Ifa60c1a7065f524f069291bb00d987b11c836cc4


[ROCm/clr commit: e6df8cb28c]
 2016-11-29 19:46:01 -06:00
pensun 946e9b4163 Change to use produce device name by default 
Change-Id: Ie2cee2a2e94a08b5874a2f5abee5d1ab6c9fdf47


[ROCm/clr commit: 504fcaf786]
 2016-11-29 11:34:06 -06:00
Rahul Garg 4421ff4fb6 Added support for hipMemGetAddressRange 
Change-Id: I99a796a4eb765152cf15a12d6a86b58684d34f50


[ROCm/clr commit: bda0704213]
 2016-11-29 22:04:09 +05:30
pensun b4b8de3cd7 Change the parameter type of hipDeviceGetPCIBusID to char* 
Change-Id: Ia72f403126e95f65da53208fc246f45d1417381f


[ROCm/clr commit: 8e2980c7ef]
 2016-11-28 10:47:18 -06:00
Ben Sander 68f1a627e0 Add more debug info 
[ROCm/clr commit: a504df955e]
 2016-11-26 08:56:02 -06:00
Ben Sander e2f024fc55 Improve docs in some places 
Change-Id: If31e84fbf0c8595ca72edb842dce7ce47783579b


[ROCm/clr commit: 9db93a1b96]
 2016-11-23 08:16:18 -06:00
Ben Sander 2e63b780f4 Improve debug capabilities. 
Print TID mapping at init when HIP_TRACE_API=1.
Print base host/dev info from tracker during copy.

Change-Id: I84e26d7b801567e5a91baad36126fb590920ec87


[ROCm/clr commit: 111b57ddd0]
 2016-11-23 08:16:18 -06:00
Rahul Garg cc61f4a720 Removed nested HIP calls from hip_device functions 
Change-Id: I18785b0ee27e32fb8950982fa5c3a64d1ae6a9b8


[ROCm/clr commit: 8a2685e6cd]
 2016-11-23 18:37:06 +05:30
Aditya Atluri 06969472b2 added fast math intrinsics to HIP 
1. Added fast math intrinsics for single precision data types
2. Added test to check the intrinsics
3. Added HIP_PRECISE_MATH macro to enable precise math on fast math

Change-Id: Iadacbb6182c31252c5e3252854372d1b80dfd27b


[ROCm/clr commit: d9a3527769]
 2016-11-22 15:26:00 -06:00
Aditya Atluri b8b1459a64 added fast math APIs 
1. Added fast math apis for sin, cos, tan, sincos
2. Added test for trig math functions
3. Added logarithm fast math
4. Changed how hipGetDevice, hipDeviceGetCacheConfig emit errors

Change-Id: Ie6ab594ddd5853cbe85e39a2f6d3479a807fa323


[ROCm/clr commit: 1a85762f53]
 2016-11-22 10:20:09 -06:00
Rahul Garg 5f24847ec0 Removed hsaKmtReleaseSystemProperties call 
Change-Id: I7cb992cccf587c333f0ca0cb518409f3944bdb06


[ROCm/clr commit: afbd278804]
 2016-11-22 06:15:35 +05:30
Aditya Atluri 66bd811c0d fixed error output for hipDeviceGetAttribute 
Change-Id: I1e343a4e4e20e1a550d419f701cc1e60e9d03af4


[ROCm/clr commit: 912426716b]
 2016-11-21 18:07:01 -06:00
Aditya Atluri 59c423b36a fixed compilation bugs 
1. Texture functions are now compiling fine
2. Fixed hipFuncCache to hipFuncCache_t

Change-Id: I8f815887e4de43ee115bbaff249905b236541c39


[ROCm/clr commit: 2611de2477]
 2016-11-21 08:56:30 -06:00
Aditya Atluri df27127960 Fixed hipDeviceGetCacheConfig on nvcc path 
1. Changed test macro to emit line numbers
2. Added getcacheconfig api test for nvcc path
3. Fixed hipFuncCache_t data type

TODO: With this commit, right now there are 2 func cache datatypes
a. hipFuncCache_t for runtime API
b. hipFuncCache for driver API

Map these to a single data type

Change-Id: Ia47c9f5d7c2633638051bf17b1103048a1ede973


[ROCm/clr commit: b3c16ea7b5]
 2016-11-20 12:18:08 -06:00
Maneesh Gupta b13556367d Refactor for building HIP as dynamic library 
Change-Id: I65a3d9d589c4fdbbdcf1611e5427224253be8260


[ROCm/clr commit: 2195e3c37d]
 2016-11-18 14:33:20 +05:30
Aditya Atluri e2dd339cfd added texture header to memory api source 
Change-Id: I1af6d60aca5a9a9ef1cadf8c304bea892acbe061


[ROCm/clr commit: 84d0d10fad]
 2016-11-17 11:57:53 -06:00
Aditya Atluri a997f0f074 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/clr commit: 603bb321ec]
 2016-11-17 01:09:12 -06:00
Maneesh Gupta 65782bbaeb Enable USE_COPY_EXT_V2 by default 
Change-Id: I2c0dc80f85a0ccb5744715b5418a604e38b249ed


[ROCm/clr commit: 0696d4417f]
 2016-11-15 10:42:27 +05:30
pensun dd1061b874 Add direct test case for threadfence_system workaround 
Change-Id: I5b21b590e957c901044741ac94e816cd8b1426f9


[ROCm/clr commit: 992f94b3a1]
 2016-11-11 15:09:43 -06:00
Aditya Atluri a3286737aa fixed multi-dim module kernel launch 
Change-Id: Id1d81f2375d058979ab526433f905cf0ea3d23d6


[ROCm/clr commit: 6dcdf08e0d]
 2016-11-11 12:25:23 -06:00
Ben Sander 5e354dcd77 Add option to deny peer access. 
Also fix test.

Change-Id: I1b247f6c4271442b008e560669bca4daf8eb94c7


[ROCm/clr commit: d666fbaafe]
 2016-11-10 23:12:48 -06:00
Ben Sander 40f8947cc3 Use forceUnpinnedCopy to resolve P2p corner cases. 
Change-Id: I2aebb419881246cebb696bec87798635bc71acc2


[ROCm/clr commit: 6e54a600b6]
 2016-11-10 23:12:48 -06:00
Ben Sander f634f73fef Enable async copy again. 
Also add HIP_FORCE_SYNC_COPY chicken bit.

Change-Id: I76a385410494b99bf27305d3c08f55dd81987565


[ROCm/clr commit: 0eeaa3bcd5]
 2016-11-10 23:12:48 -06:00
Ben Sander ee41609b48 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/clr commit: e767e0032e]
 2016-11-10 23:12:48 -06:00
Ben Sander ae2992bcb9 Improve memory debug 
Change-Id: I0f033139aa4e4b47039eb016e404009127bd0a44


[ROCm/clr commit: e9835617f1]
 2016-11-10 23:12:48 -06:00
pensun daf19a2dbb resolve conflicts for git pull 
Change-Id: Ie353b831e2241bc28042069b6cc7405257e871e1


[ROCm/clr commit: bbb619c732]
 2016-11-09 21:38:43 -06:00
pensun 61635b585f Add option to alloc fingrained system memory 
Change-Id: Ia13c8e058cb988b5857e75a590a4d67411362ae1


[ROCm/clr commit: 23de0e1b50]
 2016-11-09 21:36:30 -06:00
Ben Sander 03fcf556e9 Improve Peer support and testing. 
Change-Id: Icadc65988aaf145a265587ab0357c5bf4d26f3eb


[ROCm/clr commit: f3d38c2615]
 2016-11-06 03:22:36 -06:00
Ben Sander 0af2722827 Set forceHostCopyEngine for other copy dirs. Support HIP_FORCE_P2P_HOST 
Also: more debug for copy and P2p.

Change-Id: I87030c525410e041b2a00baaf6c68e6c0977ff42


[ROCm/clr commit: 06ecfa3975]
 2016-11-04 19:53:23 -05:00
Ben Sander 06b9391974 Refactor resolve-mem step1 
Change-Id: I7b8b2bbb56d7b31a97b48ebd42002641cd07a460


[ROCm/clr commit: 926e63c655]
 2016-11-04 09:37:56 -05:00
Ben Sander 74c9c6e591 Add debug for Peer APIs. Enable PeerMemcpy APIs by default. 
Change-Id: I46e39a9e7b07686a78484c1f3b5495b08e052fbb


[ROCm/clr commit: 00276d141e]
 2016-11-04 08:51:16 -05:00
Aditya Atluri 2d299543bf added inter thread data movement intrinsics 
Change-Id: I2a8a8ed49429cb7f96439bd28c4b83b5142737df


[ROCm/clr commit: f097b6ef81]
 2016-11-01 16:37:33 -05:00
Rahul Garg 5040e8bcc3 Added hipDeviceGetByPCIBusId in hip/hcc path 
Change-Id: I3cca0dc533d0281689d8a407c7da16ca1ba6a3a8


[ROCm/clr commit: 81c91f5b0b]
 2016-11-01 10:57:48 +05:30
Ben Sander c8aad6ee8e Print short hipLaunchKernel correctly. 
Change-Id: I6ca03d7c707cd03d6982199830213953d5855f17


[ROCm/clr commit: 3d0fa30183]
 2016-10-27 23:09:32 -05:00
Ben Sander 4378a14789 Add initial hipProfileStart/Stop 
And modify sample to show how to use.
Still needs some work to understand interaction with CXL.

Change-Id: I2579824d2dd7863ea23874d34f0dabb3cb305d3e


[ROCm/clr commit: 18dbafe6e8]
 2016-10-27 23:09:32 -05:00
Ben Sander 6a06295e31 Add two levels of HIP_PROFILE_API (1=short,2=long) 
Change-Id: I7ef98589f8731fb879db109fd573c62b489f2b61


[ROCm/clr commit: 1d5034f583]
 2016-10-27 23:09:31 -05:00
Ben Sander 44127ad800 Fix scoped marker so begin/end ATP timestamps correct 
Change-Id: Ic944d3fc00d7bc31b756c0e6c327b99eb489537e


[ROCm/clr commit: 90adf98a7e]
 2016-10-27 23:09:31 -05:00
Ben Sander e0a3b2472c Rename HIP_ATP_MARKER and profiling vars 
HIP_PROFILE_API
HIP_DB_START_API
HIP_DB_STOP_API

Change-Id: I6c4da67212ff8217e6356a2622d4c6278a188c34


[ROCm/clr commit: f7c42efa3f]
 2016-10-27 23:09:31 -05:00
Ben Sander fd2e0db1f8 Allow HIP_DB to be number or string flags (ie HIP_DB=api+mem+sync) 
Add callbacks for processing env vars.

Change-Id: I4ddf50e2da56b1dae43f50657bc693b07b23c03d


[ROCm/clr commit: caf446fe53]
 2016-10-27 23:09:31 -05:00
Ben Sander 22ae54b4cc Add HIP_PROFILE_START_API, HIP_PROFILE_STOP_API 
Refactor HIP_INIT_API to call recordApiTrace.

Change-Id: Ieff4b5018236f59e49e1b9841474440a34f821df


[ROCm/clr commit: 3132c36471]
 2016-10-27 23:09:31 -05:00
Ben Sander 147cd45ebf Add per-thread API seqnum to debug 
Change-Id: Ib13733a3e84cd56bae13a32bae40f936c20b7543


[ROCm/clr commit: 45e2dd4d69]
 2016-10-27 23:09:31 -05:00
Ben Sander 5e20bb6370 Don't call allow-access if allocating device's only peer is self. 
Change-Id: Iac58e6c3e460675833f10b1e8b2e393de223654d


[ROCm/clr commit: f8b356dce8]
 2016-10-27 23:09:31 -05:00
Ben Sander 4029bc31e6 Improve HIP TID printing in debug mode. 
Map long thread-id to a short one that is printed with each message.
Remove clunky stirng creation code for tid_tr.
Print TID on every message.

Change-Id: I780a91d8ce789cb4957789036b478bf5cde8c4e4


[ROCm/clr commit: 7d69e858a5]
 2016-10-27 23:09:31 -05:00
Aditya Atluri c58f5af811 reverted change for cache size query 
Change-Id: I44a1f43818cd287a2a3b6265f43d183f9bd5b71c


[ROCm/clr commit: 8145e1b129]
 2016-10-25 11:03:35 -05:00
Aditya Atluri 71ac51b965 correct cachesize to output correct value 
Change-Id: I5db031591eb718b0c12e78a35e4b19349de9526d


[ROCm/clr commit: e28f65604b]
 2016-10-25 09:33:45 -05:00
pensun b0a039cd5e Add workaround for hipStreamAddCallback function: call stream synchronize on host and then add execute the call back function 
Change-Id: If361f8e053949904b19b9e09245d267f05e29f7b


[ROCm/clr commit: 7a0375854a]
 2016-10-22 23:59:39 -05:00