b1d3df6484
Deprecate hipMallocHost and hipFreeHost.
...
These will print compiler warnings if used, so we can weed them out
before removing.
Also add a default flags args for hipHostAlloc, in the C++ functioin
headers. So you can replace hipMallocHost(&ptr, size( with hipHostAlloc(&ptr, size)
[ROCm/clr commit: 57365eb7a3
2016-03-19 22:53:59 -05:00
c75b63b61c
Refactor waitALlDevices and async mem copy.
...
- move waitAllStreams to device member function.
- create separate stream member function for copyAsync, like copySync.
hipMemcpyAsync now calls the copyAsync.
[ROCm/clr commit: a88c2b1ec9
2016-03-19 05:42:19 -05:00
d41836a4a1
Fix bug: test was allocating host mem instead of device mem.
...
Caused assertion when checking free + allocated should
not exceed total. Bug introduced in hipHostAlloc conversion.
[ROCm/clr commit: 4c6fd4e7ec
2016-03-19 04:11:39 -05:00
cf95ea5468
Swap in corrected hipHostAlloc (bad merge)
...
[ROCm/clr commit: aaa2429feb
2016-03-19 04:11:08 -05:00
3305c49949
Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
...
Conflicts:
src/hip_hcc.cpp
[ROCm/clr commit: 90ad8ddc5d
2016-03-19 03:22:09 -05:00
3323d2005c
disable mt streams tests (for now)
...
[ROCm/clr commit: efc9df8805
2016-03-19 03:10:31 -05:00
57c08b7fce
Describe HIP env vars
...
[ROCm/clr commit: c39b0f9660
2016-03-19 03:09:57 -05:00
1abdd6602f
Fix copy and sync bugs. Remove extra sync in default stream.
...
- NULL stream was waiting for itself to be empty before each command.
- Force "blocking" streams to wait for NULL to empty. This was missing
before.
- async copy was disabling itself via trueAsync=false for common cases.
Refactor:
- rename _null_stream to _default_stream.
- move some null sync function to defaultSync, move to dev member func.
[ROCm/clr commit: 44522eb607
2016-03-19 02:44:26 -05:00
d207f3bc26
Add beastperiteration and onesize for testing.
...
onesize allows running tests at one specific size.
[ROCm/clr commit: 5197cf250d
2016-03-19 02:43:04 -05:00
d3a3e0ba63
Improve formatting - line up cols
...
[ROCm/clr commit: c5d0813f03
2016-03-18 23:43:04 -05:00
c55187604c
Print Pinned or Unpinned in result summary
...
[ROCm/clr commit: 3dc6906855
2016-03-18 21:28:29 -05:00
2bc7b1f3ba
Update CUDA_Runtime_API_functions_supported_by_HIP.md
...
[ROCm/clr commit: a17e390283
2016-03-18 11:28:06 -05:00
336001aa29
Create CUDA_Runtime_API_functions_supported_by_HIP.md
...
[ROCm/clr commit: 564d13a0b3
2016-03-18 11:23:44 -05:00
203eeda2f7
Supported --aliged mode. Add results check for H2D and D2H.
...
[ROCm/clr commit: 690486b9eb
2016-03-18 03:09:52 -05:00
b1fe0120ca
Refactor copy code.
...
-Move staging buffer locks inside the staging buffer code.
-Remove dedicated per-device completion_signal + per-device lock -
instead allocated signal from the per-stream pool. This elimintes
the lock and allows more concurrency.
-remove switch HIP_DISABLE_BIDIR_MEMCPY
[ROCm/clr commit: e64174f47a
2016-03-18 03:02:00 -05:00
a0d3c018c0
Refactor staging buffer and sync copies.
...
- refactor staging buffer to operate on hsa* data structures not
hc::accelerator.
- use hsa_memory_allocate to allocate staging buffers rather than
am_alloc.
- Refactor device reset with single member function. Don't reallocate
staging buffers on reset.
- Properly track dependencies based on command type. Add new deps for
H2D and D2D rather than overloading H2D.
[ROCm/clr commit: 3b45e064f9
2016-03-17 20:09:10 -05:00
dd90a8ff34
Refactor to isolate staging buffer code.
...
[ROCm/clr commit: 1b7cc7d921
2016-03-17 00:20:56 -05:00
622902d408
Start separaration of staging_buffer.cpp code.
...
Still #include staging_buffer.cpp into hip_hcc.cpp.
Directed tests compile hip_hcc to static library and use the library.
[ROCm/clr commit: a1879ba59b
2016-03-16 22:26:49 -05:00
c903229e9a
Add aligned alloc
...
[ROCm/clr commit: c9d46bdcde
2016-03-16 21:55:57 -05:00
931b58f438
Checkpoint code cleanup.
...
- Refactor ihipStream in prep for thread-safe implementation.
- Do some work on PinInPlace implementation.
[ROCm/clr commit: 8acb53e160
2016-03-16 21:16:29 -05:00
b3612a7356
changed flag in hipHostRegister
...
[ROCm/clr commit: 4b96e8c789
2016-03-16 08:01:53 -05:00
420797aed8
src/ fixed hipHostAllocDefault flags
...
[ROCm/clr commit: 6beed19460
2016-03-16 07:32:54 -05:00
488512466a
Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
...
[ROCm/clr commit: 9b78f0a454
2016-03-16 07:17:22 -05:00
d62227266b
Added performance test for memcpy
...
[ROCm/clr commit: 7567a6c0ac
2016-03-16 07:16:51 -05:00
63523fbeb7
tests/src [v4] Added feature for partial writes on CPU
...
[ROCm/clr commit: 6baeb08e1c
2016-03-16 07:04:40 -05:00
9f9a56597c
test/src [v3] clean up
...
[ROCm/clr commit: 9b772c8553
2016-03-16 05:30:29 -05:00
6253fbbfe5
increased iteration size
...
[ROCm/clr commit: 8f527d2288
2016-03-16 05:24:08 -05:00
2a0d9ffb9c
added cudaHostRegister test
...
[ROCm/clr commit: a60649868d
2016-03-16 05:15:03 -05:00
7493d6074a
corrected first and second kernel dispatch
...
[ROCm/clr commit: 369d01086f
2016-03-15 14:22:00 -05:00
343cf7e864
Added single kernel launch to sample
...
[ROCm/clr commit: a6441c3334
2016-03-15 21:05:15 -05:00
649c943833
added performance metrics for kernel dispatch
...
[ROCm/clr commit: b7bf5830c1
2016-03-15 12:37:24 -05:00
dbc1196236
v2 deprecating hipMallocHost with hipHostAlloc
...
[ROCm/clr commit: 2743a5082f
2016-03-15 13:39:15 -05:00
5327f3cbac
Deprecating hipMallocHost to hipHostAlloc
...
[ROCm/clr commit: 4d4d592e0c
2016-03-15 06:30:16 -05:00
0573890c39
print device config info
...
[ROCm/clr commit: bdefb70862
2016-03-14 23:02:49 -05:00
19a0fb0e0e
Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
...
Conflicts:
src/hip_hcc.cpp
tests/src/CMakeLists.txt
[ROCm/clr commit: 15a8e8f8a0
2016-03-14 15:01:26 -05:00
a24528abcb
enable DB, comments
...
[ROCm/clr commit: 0d05517d0a
2016-03-14 14:40:41 -05:00
649f9834e8
Add Bidir copy test and help.
...
[ROCm/clr commit: 87c15a3629
2016-03-14 14:39:23 -05:00
0aa6e9df26
refactor, add support for speccing xfers in bytes
...
[ROCm/clr commit: 0aab91123b
2016-03-13 09:41:06 -05:00
5efb582db5
Remove ROCR_V2, assume=1
...
[ROCm/clr commit: 59f0ef2801
2016-03-12 04:08:13 -06:00
a236942ca5
Remove USE_AM_TRACKER (assume 1)
...
[ROCm/clr commit: de4bb0ca0e
2016-03-12 04:04:57 -06:00
d6ad50c2e0
Improve error reporting.
...
use throw with error class.
fix bug when memcpyDefault resolved to D2D copy.
[ROCm/clr commit: ac272932f6
2016-03-12 04:02:04 -06:00
6162990f44
Added hipHostRegister for hip with tests and added copyright
...
[ROCm/clr commit: 3127969d97
2016-03-08 12:57:22 -06:00
0667928d29
Added a random test for memcpyAsync
...
[ROCm/clr commit: 4503ebcb21
2016-03-08 03:40:56 -06:00
5a501d065f
v2 added support for unpinned async memcpy
...
[ROCm/clr commit: dfa9051574
2016-03-08 03:19:50 -06:00
85f961cc6b
Revert "Fixed passing unpinned memory to async memcpy"
...
This reverts commit e98925075b1a61733699
2016-03-08 03:17:27 -06:00
849d5e76a3
Merge branch 'privatestaging' of https://github.com/AMDComputeLibraries/HIP-privatestaging into privatestaging
...
[ROCm/clr commit: 218cbf1608
2016-03-08 03:15:33 -06:00
e98925075b
Fixed passing unpinned memory to async memcpy
...
[ROCm/clr commit: bf53992605
2016-03-08 03:14:49 -06:00
98d6ab118f
Refactor memccpy tests.
...
- Add "hipMemcpy_simple" - some simple smoke tests.
- Run memcpy 3 times w/ different switches for kinds, sizes,
multi-thread.
[ROCm/clr commit: 1d87879008
2016-03-07 17:15:48 -06:00
cf92bfb8c7
Added hipHostRegister flags
...
[ROCm/clr commit: ffeba62a74
2016-03-07 10:52:40 -06:00
8a8836088b
Added hipHostRegister feature for CUDA backend and its tests
...
[ROCm/clr commit: 496c549141
2016-03-07 03:42:50 -06:00
Ben Sander
Aditya Atluri