Граф коммитов

5458 Коммитов

Автор SHA1 Сообщение Дата
Vlad Sytchenko ba2d3fbbb3 Support creating textures from user ptr with offset
Change-Id: Ied0af9a842fc6ef067e068a9d89dde63ff5dfce1
2020-06-17 14:10:14 -04:00
Vlad Sytchenko 8db45328d9 Fix -Wsign-compare warnings
Change-Id: I874dc007ac657c25a72c6752c1a2da74c028a822
2020-06-17 11:20:45 -04:00
Jason Tang 8933c078f6 Add asicRevision
Change-Id: I59f3ad20b9bdadf77bd1e0725f7a401d7ad423a3
2020-06-16 17:54:20 -04:00
Tao Sang 63ce69224a Make hipHostMalloc() respect hipSetDevice()
Change-Id: I2410240f91b108c24597ee0fa7cf31e7b1f3ac5d
2020-06-15 23:20:09 -04:00
kjayapra-amd 2467e37f3f SWDEV-240589 - Remove guarantee @ __hipRegisterFunction flow for now.
Change-Id: Ic51e2a1f951ac7745c2bbd11cfd2b92521c6966a
2020-06-15 22:32:24 -04:00
Vlad Sytchenko 509a38f50c Enable the use of some warnings when building HIP-ROCclr
Similar to http://gerrit-git.amd.com/c/compute/ec/vdi/+/375385, except no -Wno-strict-aliasing.

With this change only two warnings show up during the build - -Wsign-compare and -Wmisleading-indentation.

Change-Id: Iffa436c65ab6312aeaa6def71ee6af38ed1b4a4e
2020-06-15 20:33:30 -04:00
German Andryeyev b182ad9574 Initial support for HIP managed memory
- Call the new ROCclr interfaces for HMM

Change-Id: I2cd1bf438f712a9e9e328340e7d0c025257ca6c1
2020-06-15 18:10:41 -04:00
Aryan Salmanpour 02c62165b8 expose five missing memcpy hip APIs
exposing the following hip APIs which have been previously implemented:

hipMemcpy2DFromArray
hipMemcpy2DFromArrayAsync
hipMemcpyAtoH
hipMemcpyHtoA
hipMemcpyParam2DAsync

Change-Id: I3f8635bd4265d2bf5aa2084f4eabb51f74fb18ba
2020-06-15 12:30:31 -04:00
kjayapra-amd 0b788c4c67 SWDEV-236178 - Store texture reference metadata for dynamically loaded modules.
Change-Id: I99ecc80da7e29c691341a01a09e4532972f1e3e5
2020-06-11 22:34:50 -04:00
kjayapra-amd 3fab381493 SWDEV-236178 - Remove __hip_pinned_shadow reference from dtests and add hipModuleUnload for all corresponding hipModuleLoad calls.
Change-Id: I405789b430ffbb8fccda1cebdb5d18e83a85c926
2020-06-11 22:34:36 -04:00
Rahul Garg 2d4b865d63 Addback __mbcnt_lo and __mbcnt_hi
Change-Id: Ic3facba2e2245461515799f6a17842da0f5d9933
2020-06-11 21:21:36 -04:00
Vlad Sytchenko e82e9355f2 Revert "Let hipcc not pass -mllvm option to HIP-Clang on Windows"
This reverts commit e796a1ed78.

Change-Id: If29906b9c63f03d6e51144f510d3a956d90935e8
2020-06-11 16:36:48 -04:00
kjayapra-amd 840347f0d0 SWDEV-236178 - Reorganizing Platform/Modules code for easy access.
Change-Id: Ie8920260ffc4ff01e44b48af8cec9ea5aed1aa9b
2020-06-11 10:11:20 -04:00
Siu Chi Chan 32f3374af9 fix uninitialized value in hipcc
Change-Id: I90b070c491f0efc328fcf97de0e111658ec772de
2020-06-10 16:07:51 -04:00
jujiang bb32492438 Fix for SWDEV-239415 to handle hipGetDevice properly while no GPU present
Change-Id: I252cbbf9a89fc76fe1be1fbb8f45778e96c70fb2
2020-06-10 14:18:56 -04:00
kjayapra-amd 9ff22151d2 SWDEV-231701 - Remove amd::memory->svm_ptr from MemObj, instead of the ptr to the object.
Change-Id: I5aab450a2320cfa5417c284e2a8454102df6f99d
2020-06-10 11:49:02 -04:00
Dittakavi Satyanvesh bb785840b9 SWDEV-236670 Address Eigen unit test failure by adding __host__ attribute to half2 functions
Change-Id: Ifdc852c30a1b3704871e0ee58cb7a55d3d37fc6e
2020-06-10 03:01:42 -04:00
Christophe Paquot 20ae4d709f Do not deferred stream creation now that we multiplex HW queues
SWDEV-239856

Change-Id: I156650faf832f86891f00ee167269509edd844ec
2020-06-09 19:16:25 -04:00
Yaxun (Sam) Liu 0a513d8a02 Fix include path and wrapper header
Currently std::complex and some other std functions require uses to
include hip_runtime.h before any other headers to work, which is not
reliable.

changes are made in clang to fix this issue:
https://reviews.llvm.org/D81176

which requires hipcc and HIP headers to make corresponding changes.

This patch will make sure the clang change will not break
HIP/ROCclr during this transition.

After the transition is done, we can remove explicitly setting
include path for HIP-Clang and HIP header in hipcc and hip config
cmake files and rely on clang driver to set it automatically.

Change-Id: I5d226861c2560ffa6c5ab17343a43cc378048061
2020-06-09 17:37:20 -04:00
jujiang 06c6951205 To fix a format in hip_porting_guide.md
Change-Id: I5faa4ec9b3d17625b7cb5cea86b9f44766b1cfa9
2020-06-09 13:14:52 -04:00
Rahul Garg 6aab5fa993 Bump version to 3.6
Change-Id: I739a7bd03a4ed102bbc7c2f60d108e20132f5423
2020-06-09 11:22:20 -04:00
Saleel Kudchadker fbba37070c Modify HIP_RETURN to print useful details
Change-Id: I23892c2d9a738b0298cdf24106d688a792937c73
2020-06-06 02:05:21 -04:00
kjayapra-amd ee2ff4bc5e SWDEV-239327 - Remove amd_mem_obj during unregistervar
Change-Id: I2130eaa21369b9634a9459680061138c61eaaaa4
2020-06-05 23:24:38 -04:00
kjayapra-amd 1dc24194a3 SWDEV-235295 - Move addDeviceProgram() to lazy loading
Change-Id: I8fe07e370e58844496e18c858bb528393556854f
2020-06-05 18:03:32 -04:00
Jason Tang 14c699e9de SWDEV-227909 - Add gcnArchName
Change-Id: Iea6d16b5d693dd0d900fa424d7a321c39315430e
2020-06-05 15:33:55 -04:00
kjayapra-amd 9261a35be9 SWDEV-234295 - Pass flag to ROCclr to not clear device programs during program::build()
Change-Id: I50b9fa1a96da6895f73fdf4a7c0d3f096b1188da
2020-06-05 09:53:11 -04:00
rohit pathania 0920bac577 [ dtest ] hipModuleLaunchKernel multiThreaded n multiGPU scenarios
1.Added hipModuleLaunchKernel multithreaded multi GPU scenario.
2.removed hipCtxCreate API from earlier test as it is deprecated.

SWDEV-238517 for enhancing hip unit tests

Change-Id: Id102d80887b6ff61a59938dbeb9fa2a26a3275b2
2020-06-05 09:40:58 -04:00
Lakhan Singh Thakur 6f87616103 [dtest] merge 'Adding the two test cases to cover scenarios observed in SWDEV-181598.'
SWDEV-238517 for enhancing hip unit tests

Change-Id: Ie61145b46c89b2e970af0ab11e22b6f6286ec90f
2020-06-05 09:10:23 -04:00
Aryan Salmanpour 83b8e1fbac Add support for setting hip stream priority
this change follows CUDA convention where lower number is greater priority

Change-Id: I72596a36449e818cbd8c175bf8519c51f46b1610
2020-06-04 22:50:30 -04:00
Payam f3ee29cdb2 Observed softhang while running hipStreamAddCallbackCatch SWDEV-236746
Workaround hipStream deadlock issue as the same lock was used twice SWDEV-236746

Change-Id: Icc60104ce6edf4cfd2a3a889bab78a6caadd50b7
2020-06-04 14:11:22 -04:00
Aaron En Ye Shi 29c7c9b1c2 Add gfx908 to hip-config.cmake
Support gfx908 as part of the default AMDGPU_TARGETS. MIGraphX requires this change.

Change-Id: I692f87f27829778e04f59c9ca655c6e8cbc00abc
2020-06-04 11:00:09 -04:00
Siu Chi Chan 4b56aaefd6 add constexpr constructor for vector types
Change-Id: I45bb0537d6a24ee50b548c2fd8b4f20518764813
2020-06-04 01:57:03 -04:00
Evgeny 0c0a8fc108 adding hipGetStreamDeviceId() profiling API
Change-Id: I5ccf88ddac123260d7c17defefcf20ff3b2504e2
2020-06-03 18:57:49 -04:00
Aaron En Ye Shi d93134e727 Add compiler-rt library for __fp16 and _Float16
Similar to HCC, link with compiler-rt to support __fp16 and _Float16 type conversions in ONNX models. This should resolve SWDEV-238491.

Change-Id: Iad8dcff568831719f501f562a04023326ae8036c
2020-06-03 18:53:14 +00:00
Siu Chi Chan c414c70e8f update device library path
fix device lib directory

add missing --hip-link switch for link phase

Change-Id: I4b2eeb32648ca3cec72ec1f4e3381ce1fc0a90a5
2020-06-03 14:44:23 -04:00
Jatin 126573df4c Adding changes for hipExtLaunchKernel for rocCLR
Change-Id: Iba52bc3bde7c37f3fb375a55ba0947e87b3cdc9b
2020-06-02 14:16:41 -04:00
Rahul Garg db471e5ed9 Use right perl-which rpm package
Change-Id: I22106a7d1b4b50c99f945bc6416ff3bd8486d15c
2020-06-02 02:17:54 -04:00
Aaron En Ye Shi a436ab865a HIP-Clang Temporarily include HIP_HCC_FLAGS for PyT
Change-Id: I0f5895fba69f5bf0c6cc13e9402bdc44726dcc5a
2020-06-01 15:28:04 -04:00
Aryan Salmanpour bf11ffd175 Add support for missig hipStreamGetPriority API
Change-Id: I2be4b055e5f977eb6ecad0b1f5f9535e72345fe7
2020-06-01 13:33:14 -04:00
agodavar 3375920638 hipMemset dont enqueue command if size bytes is 0
Change-Id: I63bf896f9f23edf254acdf7a8c11c92f8b5ac039
2020-05-30 10:33:44 -04:00
Rahul Garg 27e306686c Add libfile-which-perl dependency SWDEV-237642
Change-Id: I0799fdcbc58a35c957a3bc69a8a1c6a013a3f57c
2020-05-29 20:04:23 -04:00
kjayapra-amd ab17b43d45 SWDEV-229840 - fixing return HIP_RETURN instances in hip.
Change-Id: I48763d7268bf5649bf2242c962c185f5f4af159c
2020-05-29 09:43:58 -04:00
Aryan Salmanpour 200ab30084 [dtest] add a multi stream test for (SWDEV-237846)
Change-Id: I4a1d764df75af7019d0f38313e5e0a6a224818f8
2020-05-28 23:36:10 -04:00
jujiang 017e3b87b3 Update document for hip_faq.md, hip_porting_guide.md and hip_terms2.md
Change-Id: I2c019f802ad70ed43f1608cfd3c9067f1573741e
2020-05-28 17:51:58 -04:00
Christophe Paquot dfec136725 Revert "Call notifyCmdQueue when building the event wait list"
This reverts commit 1cfc9d1860.

Reason for revert: better fix in ROCclr

Change-Id: I9707e69adf42a662c08fe9b3ec7458655d838bdd
2020-05-28 17:01:10 -04:00
kjayapra-amd f2899243f3 SWDEV-233927 - Crash if binary for current device is not found.
Change-Id: I57281ae6c09110f39155664fca5a83ea57bb62b4
2020-05-28 16:18:27 -04:00
Joseph Greathouse 766e708535 Fix a build error on signed/unsigned comparison
Change-Id: Ic79eb4c3ec5c6fd36cea7c4810d990619f08b9e1
2020-05-28 14:27:16 -05:00
kjayapra-amd 55cdef8e45 SWDEV-236465 - Return error code as soon as global creation fails.
Change-Id: I790b8b4fdd6ab8818bc5b6b9a79e6900b840372d
2020-05-28 13:28:23 -04:00
Joseph Greathouse 90453b68d3 Fix occupancy calculation functions in ROCclr path
The hipOccupancyMaxPotentialBlockSize API is meant to return the
number of threads for the highest-occupancy workgroup, and the number
of those workgroups. It was previously calculating the number of
maximum-sized workgroups that would fit on a single CU. This is
a mixture of the API we wanted (to calculate max potential block size)
and the MaxBlocksPerMultiprocessor function.

This patch fixes it up so that the internal occupancy calculation
function works for two uses: the traditional function that calculates
the maximum blocks per multiprocessor when a user passes in a fixed
block size (used for hipMaxBlocksPerMultiprocessor style functions)
and a function that calculates the size of a block that would lead
to maximum occupancy, and how many blocks of that size would be
needed to fill the whole GPU (for hipOccupancyMaxPotentialBlockSize
style functions).

This also updates the occupancy calculation function to prepare for
gfx10, which does not have SGPR-based occupancy limits.

Change-Id: Ie007b3f9d5ebc4e166b50a3a051498af35650f35
2020-05-28 10:22:10 -05:00
Evgeny d863edb8ba adding hipKernelNameRefByPtr function
Change-Id: Iefc18967b10394b85a207ffdb5bbfe5e3601474d
2020-05-28 10:59:48 -04:00