2
0
Gráfico de cometimentos

236 Cometimentos

Autor(a) SHA1 Mensagem Data
Aditya Atluri 3d27bbd3db Added kernel compilation driver apis
1. Added 2 new driver apis, hipModuleLoad, hipModuleGetFunction

Change-Id: If464a7fad178121e3da791c7ac9e17ebc01a9cd0
Issues: When a sample written with them shows Aborted (core dumped) when exiting
2016-08-16 14:36:25 -05:00
Evgeny Mankov 7015675983 #define HIP_DYNAMIC_SHARED_ATTRIBUTE is added 2016-08-16 17:58:57 +03:00
Rahul Garg eec9edef80 Implementation of hipCtxGetDevice
Change-Id: I067572e486323c3aad6f744a2c0c4997c8696af6
2016-08-13 01:17:46 +05:30
Rahul Garg 62d390da58 First implementation of hipCtxXXX functions
Change-Id: I4609cbe6bd90a1fff8655bff4fdd773864397aba
2016-08-13 00:09:08 +05:30
Jeffrey Poznanovic 48491a3978 Adding hipblas include files
Change-Id: I73064d410acd8f655dc62eaeb6f4bdefc5381e35
2016-08-12 11:59:25 +05:30
Ben Sander 89164259ab Context update.
- Remove tls_deviceID.
- Add first passing test.

Change-Id: If3e2f254abf589028cfe4f9e6369745f04160de0
2016-08-10 08:59:47 -05:00
Rahul Garg 2ac93c340d Changed StagingBuffer class to UnpinnedCopyEngine
Change-Id: I1e212bfc8030dcf225ecf78fd7b23fda9b1de92f
2016-08-09 21:29:42 +05:30
Rahul Garg 023b1ecf33 Moved sync copy decision logic to staging buffer class
Change-Id: I5c398772375fcc1f174a7597eea1215ce7bf80b4
2016-08-09 09:28:18 +05:30
Ben Sander 8f402132ba Add initial context implementation.
APIs: hipInit, hipCtxCreate.
Track TLS default ctx.  Set deviceID now changes the ctx.
Add first context test.

Change-Id: If1cb9989b5a04a36147e25e84904336c7b6f3d88
2016-08-08 17:49:02 -05:00
Ben Sander ed0a2c02fe Code cleanup, use camelCase where appropriate.
Change-Id: I5a7ec50df8bbb3e7a3b313c0b12e2dd55ae4a09c
2016-08-08 14:54:38 -05:00
Ben Sander 2a798152d4 Move copy kernel templates into hip_memory.cpp
Change-Id: I862529f3fa8232372c6bacaa5d36f035bbdd32a1
2016-08-08 12:07:12 -05:00
Ben Sander cfdacab32f Split ihipCtx_t into ihipCtx_t and ihipDevice_t .
Major change to existing code base.
    Ctx holds streams, enables peers, and flags.
    Device holds accelerator, hsa-agent, device props.

Add hipCtx_t.

Add peer APIs that accept hipCtx_t (in addition to deviceId)

Compiles and passes directed tests.

Change-Id: Iddab1eb9edbf90caad2ef5959c6b811d658197f1
2016-08-08 11:55:57 -05:00
Ben Sander 2dc3d3238b Change Device->Ctx
Change ihipDevice_t -> ihipCtx_t (new)
Change ihipGetTlsDefaultDevice->ihipGetTlsDefaultCtx
Some other changes from device->ctx where appropriate.

Change-Id: I5c4ae93b2fd42c6303aa23d748eb166b7431925d
2016-08-07 21:47:12 -05:00
Ben Sander e7d7c5cbe8 Remove ihipStream_r::_device_index
Replace with direct pointer to device.  Cleaner, and prep
for transition to contexts.

Change-Id: I0e550f34412923d46c541c0a14bb7d29c3fd4b11
2016-08-07 20:47:06 -05:00
Rahul Garg fcb2fcce1e Region based apis to pool based api changes
Change-Id: If53019eebafe051ab4e811863995f78315297080
2016-08-05 15:05:57 +05:30
Ben Sander 02dd7a7399 Cleanup sync code.
Remove dead depFutures, enqueueBarrier call.
Rename some parms to reflect usage.
Add comments to better explain tricky parts of sync code.

Change-Id: I763296421d9c2b3b58fc8cef5f010b12ab49553c
2016-07-27 18:31:11 -05:00
Aditya Atluri 1859c6e515 Signal Fix: Added signal limit to allocSignal
1. Did not change the logic in allocSignal
2. Added guard to wait on signal limit

Change-Id: I78f29097e6a584b3c3d78319dac19869067bd1fe
2016-07-27 13:48:49 -05:00
Aditya Atluri 0a31b47e2e Signal Fix: Moved kernel count to critical stream
1. Added environment variable HIP_NUM_KERNELS_INFLIGHT
2. Moved kernelcount variable inside stream critical section

Change-Id: I51d24d0a2a109467209170de117a6d02ba4e308e
2016-07-26 17:09:27 -05:00
Aditya Atluri 53d7629a85 Signal Fix: Changed global signal count to per stream signal count
1. The number of kernels that can use signals are increased to 128
2. The kernel count is now specific to the stream

Change-Id: Ie6d1aa3f437aad8f08c3333fe48bd3f46e551e60
2016-07-26 14:03:51 -05:00
Aditya Atluri 4bdf26a82e Added re-fix for memcpy kernel sync
1. The patch uses HIP signal pools to sync between copy and kernel commands
2. The hsa_signal_create is removed
3. Left the redundant enqueueBarrier method just in case

Change-Id: I3dff3e8ee57fff3cd49bec802ff735ed128e5ca1
2016-07-26 09:22:59 -05:00
Rahul Garg 42a3ed544c D2H and H2D unpinned memory transfer support
Change-Id: If6d6c970f435e5d917d5cc6cddc2ee2918cd1c37

Conflicts:
	src/hip_hcc.cpp
2016-07-25 14:36:07 +05:30
Aditya Atluri c756bb3398 Partial fix async after kernel launch signal issue
Change-Id: Ib48d6564379160035bded9493b93663fba361710
2016-07-23 14:54:20 -05:00
pensun f31668fee4 Add empty stubs for threadfence family routines, changes include:
- stubs and documentation in include/hcc_details/hip_runtime.h
    - stubs with "no-op" in src/hip_memory.cpp
    - document update in hip_kernel_language.md, add suggestions to
    disable L1 and L2 caches when using the threadfence routines.

Change-Id: Ic0753170f802003055bca9d7476d7f48817b98b7
2016-07-22 10:40:58 -05:00
Maneesh Gupta 71d51170ef Replace calls to ihipInit with use of HIP_INIT_API macro
Change-Id: Iabf7df79f0238a8ddffea4607fe945df36642850
2016-07-22 15:46:55 +05:30
Maneesh Gupta b23fad53cc Fix using ATP markers
Change-Id: If2d04f80b580237426c569737551e2001a8cd35a
2016-07-21 16:02:51 +05:30
Maneesh Gupta 7022986ab2 Merge branch 'hiparray' into amd-develop
Change-Id: I63ca7b1db7b593ac5cfb3fd7cd5d08d6e4075a4c
2016-07-21 12:29:56 +05:30
Maneesh Gupta d7b040bdba Merge branch 'amd-master' into amd-develop 2016-07-05 21:40:22 +05:30
Aditya Atluri 36b81c1be6 added more nvcc event functions
Change-Id: I79ee20ef444d4c1ab6ada3c0d56730ce754ab6b6
2016-06-30 21:03:19 -05:00
Maneesh Gupta 3f204b8580 Merge branch 'amd-develop' into amd-master
Change-Id: I04f85b207e15e66c1a546675dc0937726ee08362
2016-06-30 18:36:07 +05:30
Aditya Atluri 5633cc34cc moved half support to a source file
Change-Id: I7c09b41877e22c1b743dea25a585e5307427dafd
2016-06-30 18:23:29 +05:30
Aditya Atluri 83210c8ac3 added fp16 software support
Change-Id: Ic0fdd9f8248a66911169fc00d3af71f50b36e233
2016-06-30 18:23:29 +05:30
7SK 8264d5d6bd NVCC_COMPAT
add support for both cuda compatible implementation and hcc(faster)
implementation with test

Change-Id: I79a22344f458391d7dffac5f147619a542e97e4e
2016-06-28 09:36:06 +05:30
Ben Sander 702b9eff1e Use GRID_LAUNCH_VERSION define provided by HCC grid_launch.h
Change-Id: I5a7b31f13c4055cdca3a8854728d59db7dc5a7bb
2016-06-21 23:52:25 -05:00
Ben Sander 9f29cc6989 Grid-launch updates to 2.0 and cleanup of old.
_ Use fields from GRID_LAUNCH_20 structure
  (See USE_GRID_LAUNCH_20 define, currently set to 0)
  "1" will require HCC support.
- Remove old DISABLE_GRID_LAUNCH support.

Change-Id: I584ce648d217251789a6283cf27feb24cb7dc8d1
2016-06-21 23:24:38 -05:00
Ben Sander 100a744d92 Switch to hc_* coordinate builtins (replace amp_)
Change-Id: I0a8871f0c9f047eb45a7391fd032100af2bbd4e0
2016-06-21 23:22:38 -05:00
Maneesh Gupta 7ca5afc20b Merge branch 'amd-master' into amd-develop 2016-06-21 15:14:04 +05:30
Maneesh Gupta 7393ba139c Merge "Add compiler annotations to make_ functions" into amd-master 2016-06-21 05:34:18 -04:00
Aditya Atluri 90cd67e0b5 able to pass non-dim launch parm to kernel launch
Change-Id: I0411849a27efcba597a1a9aa08be179635e04988
2016-06-18 11:28:20 -05:00
Ben Sander 5c9e286ddc Clean up old work-week and USE_* refs
Change-Id: I929c979fa085f8e5205194cbccca46e9b5516aa9
2016-06-17 15:18:57 -05:00
Aditya Atluri ae12867774 added tests for host math functions
Change-Id: I66a5c574a27190e32054586f07ecf20e1ff71292
2016-06-17 15:05:33 -05:00
Maneesh Gupta 59c483aa16 hipLaunchKernel supports empty kernel with no arguments
Change-Id: Iaa46ced7fe0e3d72734d02250dfb48b98bef615d
2016-06-17 14:11:46 +05:30
Ben Sander b0cf03114e Add NVCC implementation for hipPeeakAtLastError 2016-06-16 09:38:34 -05:00
Ben Sander 29552a9ae7 remove old KERNELBEGIN/KERNELEND macros 2016-06-16 09:38:07 -05:00
Maneesh Gupta d12651521f Add compiler annotations to make_ functions
Change-Id: Ic16cfe402dc5cb387b1b63ad311ce8b57ea07361
2016-06-16 16:18:35 +05:30
Ben Sander 2ab19ca505 NVCC improvements.
- Complete translation tables for cudaError <-> hipError_t.
- Remove some odd errors that were not correctly translated or not used.
- Add HIPCHECK_API to test infrastructure.  Used for negative testing
  an API ; if a mismatch occurs it shows the expected return error
  code.  Can also print a warning rather than error.
- Enable hipMemoryAllocate on NV system, and review error coded.
- Add hipErrorName to nvcc.

Change-Id: I680427dcf32a5796d5913cf9e7f3b4c6f6b91599

Conflicts:
	tests/src/CMakeLists.txt

Bug fixes and improved docs for hipFree and hipHostFree.

    - Passing NULL pointer initialized runtime and return hipSuccess
      (not an error like before).
    - add negative test for this. (hipMemoryAllocate, improved)
    - Match NVCC errors for invalid pointers, add to test.
    - Update hipFree and hipHostFree docs.
    - hipGetDevicePointer always set *devicePointer=NULL, even for
      invalid flags.
    - Gate shared memory usage on specific HCC work-week.

Change-Id: I533b4fd3280a3d6cdbf05eb768976f0c7506c012
2016-06-16 06:13:51 +05:30
Aditya Atluri 7cb45e2d87 added more host functions and tests
Change-Id: I9904e65e14c5479ba33d836c5c0b763cb5af71e3
2016-06-15 11:45:19 -05:00
Aditya Atluri d7df02e310 added erfinv software implementation
Change-Id: Ib1a5584f6c81ab3afa70f7bcbfd7780e156454e3
2016-06-14 00:09:41 -04:00
Aditya Atluri 9ac6e422f0 added more device functions
Change-Id: I191919060b393772ee442cc19d83479217c5a4ce
2016-06-13 11:55:12 -05:00
Aditya Atluri 25a273e4df Added more device functions
1. Added copyright for device float test
2. Added device double functions support
3. Added device double functions test
4. Corrected device function signatures in headers

Change-Id: I13c8829682c925992f5cad84062bc9f702fe4048
2016-06-10 09:46:31 -05:00
7SK 1463075608 fix_ldg
Change-Id: I53de5fa91b4f57d496ffe46787d197ae84dde4a4
2016-06-09 16:56:05 -04:00