Sarbojit Sarkar
dc2f1c2e7d
host pointer check [SWDEV-243104, SWDEV-243206]
...
Change-Id: Id58dabfe5c35e5f8e42d224e700a431a4a97d003
[ROCm/hip commit: b2b5a821df ]
2020-07-14 07:16:12 -04:00
Tao Sang
05300c8fe3
Implement HIP_ENABLE_DEFERRED_LOADING in right way
...
Change-Id: Ia28dbdea5cde93b3ca69f44c6a3c18ec8a730d42
[ROCm/hip commit: 66fca8be23 ]
2020-07-11 18:14:03 -04:00
kjayapra-amd
314b48f10a
SWDEV-240165 - Move all amd::MemObjMap_ reference to ROCclr and only allow base ptr to get ipc handle.
...
Change-Id: I4f11e56ba455a7ce5a066c0e9e49657c5220a211
[ROCm/hip commit: 7233bb75da ]
2020-07-09 21:19:50 -04:00
German Andryeyev
6b71f99cca
Host device context should return a valid queue
...
Host memory allocations use shared device context, which allows
to have access from any device. That breaks default queue detection.
Change-Id: I27ad1234d49ab0ea8f5edd6f2e3ffa93c47d4b1f
[ROCm/hip commit: b9b982247e ]
2020-07-09 10:56:33 -04:00
Aryan Salmanpour
7a2961fc57
add an error message before aborting if a wrong GPU target was used for compilation
...
Change-Id: I7b41e93d420dbc47978238a681f2956a0def01cb
[ROCm/hip commit: 9121a74ac1 ]
2020-07-07 14:50:44 -04:00
Jatin Chaudhary
ee2361731a
Fix error: cannot pass object of non-trivial type string through variadic function when building in debug mode
...
Change-Id: Idd55ed422addc815fb56c57b05bd4afa9563f2ac
[ROCm/hip commit: 24554e6755 ]
2020-07-06 00:48:40 -04:00
kjayapra-amd
b823a939b0
SWDEV-236178 - Fixing func_attributes that was over written on redesign.
...
Change-Id: I157913ae4fa4a3a149817e901f10475b7750188b
[ROCm/hip commit: 6f15d8ca6b ]
2020-07-01 11:52:45 -04:00
German Andryeyev
31abf7a6c3
Remove hipMemcpyDeviceToDevice check.
...
There is hipMemcpyDefault that requires validation also.
Change-Id: I5ae0ffa1c023b7774547acbdc049b84a0c2cb4c5
[ROCm/hip commit: f2102adc99 ]
2020-06-30 12:59:54 -04:00
Evgeny
bab86e0e28
HIP-PROF: fixing pointers tracing
...
Change-Id: Id2c35972e1d134012980c95a3649ddfdf9303494
[ROCm/hip commit: a19960e79a ]
2020-06-23 14:02:35 -04:00
agodavar
a6c83e443c
SWDEV-239297:hipMemset fix, getMemoryObject with unaligned ptr
...
Change-Id: I37b8139c5099d37b42f34d728a3d694fdf9cb55d
[ROCm/hip commit: 93da1b1f11 ]
2020-06-23 08:56:09 -04:00
German Andryeyev
f490ef4c87
Disable flush on marker
...
Change-Id: I04d0fd995fb4b29a6bc372e1543902a053017d6a
[ROCm/hip commit: 44c1df802e ]
2020-06-22 13:27:39 -04:00
Tao Sang
9c6dbda74b
Fix static lib crashing issue
...
Initialize HIP_ENABLE_DEFERRED_LOADING = 1
Change-Id: I432ccf1926b953c4391fdd086cde7da2380962bb
[ROCm/hip commit: 0eebbb4f87 ]
2020-06-22 13:24:40 -04:00
Tao Sang
c2a9bed3d7
Support numa policy set by user
...
Add hipHostMallocNumaUser flag to hipHostMalloc() in order to support
numa policy set by user.
Change-Id: I6d70ed539a5f97f27187f2242b68849c0e27e4d6
[ROCm/hip commit: 63051ca2e1 ]
2020-06-19 21:23:58 -04:00
Vlad Sytchenko
2780b86df2
Correct texture element size calculations
...
hipChannelFormatDesc contains size of each channel in bits. There's 8 bits per byte, not 4, so we should divide by 8.
Change-Id: I38ae2d28ad54740823c98908ca2afecad0ba0183
[ROCm/hip commit: e9033edc10 ]
2020-06-19 11:39:03 -04:00
Vlad Sytchenko
9a8ba0d8e4
Support creating textures from user ptr with offset
...
Change-Id: Ied0af9a842fc6ef067e068a9d89dde63ff5dfce1
[ROCm/hip commit: 205e4c1c5f ]
2020-06-17 14:10:14 -04:00
Vlad Sytchenko
3af20b1bbb
Fix -Wsign-compare warnings
...
Change-Id: I874dc007ac657c25a72c6752c1a2da74c028a822
[ROCm/hip commit: a6aef2a36a ]
2020-06-17 11:20:45 -04:00
Jason Tang
671b994ff8
Add asicRevision
...
Change-Id: I59f3ad20b9bdadf77bd1e0725f7a401d7ad423a3
[ROCm/hip commit: 38cd2b96c7 ]
2020-06-16 17:54:20 -04:00
Tao Sang
a41347a1db
Make hipHostMalloc() respect hipSetDevice()
...
Change-Id: I2410240f91b108c24597ee0fa7cf31e7b1f3ac5d
[ROCm/hip commit: 7682ec6a7b ]
2020-06-15 23:20:09 -04:00
kjayapra-amd
78474d9ef8
SWDEV-240589 - Remove guarantee @ __hipRegisterFunction flow for now.
...
Change-Id: Ic51e2a1f951ac7745c2bbd11cfd2b92521c6966a
[ROCm/hip commit: 45e0847319 ]
2020-06-15 22:32:24 -04:00
Vlad Sytchenko
59be70d914
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
[ROCm/hip commit: f3480e019e ]
2020-06-15 20:33:30 -04:00
German Andryeyev
61c65cf0b7
Initial support for HIP managed memory
...
- Call the new ROCclr interfaces for HMM
Change-Id: I2cd1bf438f712a9e9e328340e7d0c025257ca6c1
[ROCm/hip commit: f4211c3905 ]
2020-06-15 18:10:41 -04:00
Aryan Salmanpour
969f1651fb
expose five missing memcpy hip APIs
...
exposing the following hip APIs which have been previously implemented:
hipMemcpy2DFromArray
hipMemcpy2DFromArrayAsync
hipMemcpyAtoH
hipMemcpyHtoA
hipMemcpyParam2DAsync
Change-Id: I3f8635bd4265d2bf5aa2084f4eabb51f74fb18ba
[ROCm/hip commit: 17102ff3a1 ]
2020-06-15 12:30:31 -04:00
kjayapra-amd
10919eb6a9
SWDEV-236178 - Store texture reference metadata for dynamically loaded modules.
...
Change-Id: I99ecc80da7e29c691341a01a09e4532972f1e3e5
[ROCm/hip commit: 173bb2af6e ]
2020-06-11 22:34:50 -04:00
kjayapra-amd
37508980de
SWDEV-236178 - Reorganizing Platform/Modules code for easy access.
...
Change-Id: Ie8920260ffc4ff01e44b48af8cec9ea5aed1aa9b
[ROCm/hip commit: 20f05c4228 ]
2020-06-11 10:11:20 -04:00
jujiang
572efac7a5
Fix for SWDEV-239415 to handle hipGetDevice properly while no GPU present
...
Change-Id: I252cbbf9a89fc76fe1be1fbb8f45778e96c70fb2
[ROCm/hip commit: c54590ebbd ]
2020-06-10 14:18:56 -04:00
kjayapra-amd
707ec427bc
SWDEV-231701 - Remove amd::memory->svm_ptr from MemObj, instead of the ptr to the object.
...
Change-Id: I5aab450a2320cfa5417c284e2a8454102df6f99d
[ROCm/hip commit: 2f8e180f01 ]
2020-06-10 11:49:02 -04:00
Christophe Paquot
5a8857b3e9
Do not deferred stream creation now that we multiplex HW queues
...
SWDEV-239856
Change-Id: I156650faf832f86891f00ee167269509edd844ec
[ROCm/hip commit: c8f9afa9da ]
2020-06-09 19:16:25 -04:00
Saleel Kudchadker
75f05a8b01
Modify HIP_RETURN to print useful details
...
Change-Id: I23892c2d9a738b0298cdf24106d688a792937c73
[ROCm/hip commit: 2e8c1e9f24 ]
2020-06-06 02:05:21 -04:00
kjayapra-amd
81e6261ff8
SWDEV-239327 - Remove amd_mem_obj during unregistervar
...
Change-Id: I2130eaa21369b9634a9459680061138c61eaaaa4
[ROCm/hip commit: 421bc7dfcb ]
2020-06-05 23:24:38 -04:00
kjayapra-amd
a3f6bf1161
SWDEV-235295 - Move addDeviceProgram() to lazy loading
...
Change-Id: I8fe07e370e58844496e18c858bb528393556854f
[ROCm/hip commit: 348066d21f ]
2020-06-05 18:03:32 -04:00
Jason Tang
67baf4f506
SWDEV-227909 - Add gcnArchName
...
Change-Id: Iea6d16b5d693dd0d900fa424d7a321c39315430e
[ROCm/hip commit: 1c0d737e1f ]
2020-06-05 15:33:55 -04:00
kjayapra-amd
0bdff1ff07
SWDEV-234295 - Pass flag to ROCclr to not clear device programs during program::build()
...
Change-Id: I50b9fa1a96da6895f73fdf4a7c0d3f096b1188da
[ROCm/hip commit: 8941d19fe8 ]
2020-06-05 09:53:11 -04:00
Aryan Salmanpour
1cdccedef9
Add support for setting hip stream priority
...
this change follows CUDA convention where lower number is greater priority
Change-Id: I72596a36449e818cbd8c175bf8519c51f46b1610
[ROCm/hip commit: 1bb86658cc ]
2020-06-04 22:50:30 -04:00
Payam
dac652e8df
Observed softhang while running hipStreamAddCallbackCatch SWDEV-236746
...
Workaround hipStream deadlock issue as the same lock was used twice SWDEV-236746
Change-Id: Icc60104ce6edf4cfd2a3a889bab78a6caadd50b7
[ROCm/hip commit: a524f13c97 ]
2020-06-04 14:11:22 -04:00
Evgeny
037da0c744
adding hipGetStreamDeviceId() profiling API
...
Change-Id: I5ccf88ddac123260d7c17defefcf20ff3b2504e2
[ROCm/hip commit: cad3f805c0 ]
2020-06-03 18:57:49 -04:00
Jatin
7b52f0a1ea
Adding changes for hipExtLaunchKernel for rocCLR
...
Change-Id: Iba52bc3bde7c37f3fb375a55ba0947e87b3cdc9b
[ROCm/hip commit: 2d517fdcc6 ]
2020-06-02 14:16:41 -04:00
Aryan Salmanpour
dee385497d
Add support for missig hipStreamGetPriority API
...
Change-Id: I2be4b055e5f977eb6ecad0b1f5f9535e72345fe7
[ROCm/hip commit: 4a1536701f ]
2020-06-01 13:33:14 -04:00
agodavar
76cb1b1644
hipMemset dont enqueue command if size bytes is 0
...
Change-Id: I63bf896f9f23edf254acdf7a8c11c92f8b5ac039
[ROCm/hip commit: 39ee03f64c ]
2020-05-30 10:33:44 -04:00
kjayapra-amd
6d2ec4c700
SWDEV-229840 - fixing return HIP_RETURN instances in hip.
...
Change-Id: I48763d7268bf5649bf2242c962c185f5f4af159c
[ROCm/hip commit: aae3e13296 ]
2020-05-29 09:43:58 -04:00
Christophe Paquot
e97106dda4
Revert "Call notifyCmdQueue when building the event wait list"
...
This reverts commit 9263f193d0 .
Reason for revert: better fix in ROCclr
Change-Id: I9707e69adf42a662c08fe9b3ec7458655d838bdd
[ROCm/hip commit: 7e57136aba ]
2020-05-28 17:01:10 -04:00
kjayapra-amd
cdd32d1fd0
SWDEV-233927 - Crash if binary for current device is not found.
...
Change-Id: I57281ae6c09110f39155664fca5a83ea57bb62b4
[ROCm/hip commit: 0599855f0d ]
2020-05-28 16:18:27 -04:00
Joseph Greathouse
6690e7a9e0
Fix a build error on signed/unsigned comparison
...
Change-Id: Ic79eb4c3ec5c6fd36cea7c4810d990619f08b9e1
[ROCm/hip commit: 1f25e4956c ]
2020-05-28 14:27:16 -05:00
kjayapra-amd
7002be9c95
SWDEV-236465 - Return error code as soon as global creation fails.
...
Change-Id: I790b8b4fdd6ab8818bc5b6b9a79e6900b840372d
[ROCm/hip commit: db972c4f2a ]
2020-05-28 13:28:23 -04:00
Joseph Greathouse
72de06128e
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
[ROCm/hip commit: ebe5054e04 ]
2020-05-28 10:22:10 -05:00
Evgeny
14b9967374
adding hipKernelNameRefByPtr function
...
Change-Id: Iefc18967b10394b85a207ffdb5bbfe5e3601474d
[ROCm/hip commit: ef7ff69ff0 ]
2020-05-28 10:59:48 -04:00
Saleel Kudchadker
3c9945976e
Fix elapsed time calculation for null stream
...
SWDEV-237377 - This fixes time calculation where the event may
be recorded on Null stream and work submitted on other streams
Change-Id: Ie36310dea5cee2fed4a514ed01f04db4b47e571c
[ROCm/hip commit: fb2d7bcd2b ]
2020-05-27 18:42:07 -04:00
Christophe Paquot
9263f193d0
Call notifyCmdQueue when building the event wait list
...
SWDEV-237846
Change-Id: I8bf70e7ad19903767a080d8c6e516c83b0dc2545
[ROCm/hip commit: 3ba07f5e3c ]
2020-05-27 12:53:46 -04:00
Christophe Paquot
3f6aa846a4
hipDeviceSynchronize needs to sync NonBlocking streams as well
...
SWDEV-237167
Change-Id: Ie916d8f03ce91e8ef05a2b4edc580a7021520f6f
[ROCm/hip commit: f7ed87b02b ]
2020-05-26 17:59:22 -04:00
Evgeny
d55d242eb7
hip_prof: fixing printing pointer args
...
Change-Id: I93969723650f7c29d5c00a3809d3701c6a3dca44
[ROCm/hip commit: f9e086ae8a ]
2020-05-25 13:17:16 -04:00
kjayapra-amd
542c7874a7
SWDEV - 237467 - Return proper hip error codes incase of ROCclr IPC API failures.
...
Change-Id: I2cc8da543f70bb3d8b82520fa9b2f509d20ce3c0
[ROCm/hip commit: b44ae437bc ]
2020-05-23 10:51:37 -04:00