5436d362b1
SWDEV-301667 - Add a flag for gpuvm kernargs
...
HIP_FORCE_DEV_KERNARG=1 will create a device allocation for kernel arg
segment. Flag is 0 by default.
Change-Id: Iaaf5a149f3be8596568878d5d272268baf067c60
2023-05-22 11:23:48 -04:00
60d9a4ebab
SWDEV-384557 - Do not fall back to compute
...
- Use regular copy API if we exhaust free SDMA engines and not fall back
to compute copy. Falling to compute is affecting performance for
numerous apps that are GPU bound
Change-Id: I75c767eff0b9f5ada324301c5c327fe2c23a9806
2023-05-22 11:23:23 -04:00
06df9e2efd
SWDEV-301667 - Kernelarg gpuvm
...
Add aligned, nontemporal `memcpy` for kernarg.
Change-Id: I5d8ac76904feaf793b45ec2ea5fbd1069be20068
2023-05-22 11:21:14 -04:00
a4c5c44008
SWDEV-371628 - Shift device lib linking into clang driver
...
Previously, we used the following approach and Comgr actions
for device lib linking:
AMD_COMGR_COMPILE_SOURCE_TO_BC (compile with clang driver)
AMD_COMGR_ADD_DEVICE_LIBRARIES (link in device libs with
llvm-link API)
However, the clang driver can link in device libraries as part
of compilation, assuming a --rocm-path is set. In this context,
this is accomplished by using the following Comgr action instead:
AMD_COMGR_COMPILE_SOURCE_WITH_DEVICE_LIBS_TO_BC (compile and
link in device libs with clang driver)
Change-Id: I661465865365afecc44aa15d4df91bfab361af8d
2023-05-17 17:40:56 -04:00
3f4b70cafd
SWDEV-336531 - Fix for relative path issue in debuginfo
...
Change-Id: Ie36d1a93f40ddc3a42823eb83e041af8c55bcdee
2023-05-17 13:06:26 -04:00
a07e38be2f
SWDEV-400119 - Return error from globalFreeMemory() if HSA available memory query fails
...
Change-Id: I33c069fa1e0d31891f64dc8167a24ef9d7ed1426
2023-05-15 11:54:25 +01:00
0b475284e9
SWDEV-398151 - Partly relax static engine allocation
...
Change-Id: I4903b51a34b597a2e84d771b52cf629f877dba05
2023-05-11 00:52:18 -04:00
ffa1deef27
SWDEV-398047 - Disable arena for XNACK.
...
Change-Id: If6c98490bd5b8146a357f45ed132f687def0877f
2023-05-10 23:28:05 -04:00
7624a48de9
SWDEV-366528 – Fix image memory format updating issue
...
Add dstMemory format updating.
Separate format updating for srcMemory and dstMemory.
Change-Id: I1692b92d417bbd742d562679f218ebf8ca532e92
2023-05-08 21:43:42 -04:00
0eb96cbc59
SWDEV-397168 - Enable dynamic call stack size for PAL.
...
Change-Id: I8be51ffb48e6a742117491a4bf6f12f152e4a0b3
2023-05-07 23:26:28 -04:00
04b696abee
SWDEV-353281 - VM support in mempool for graphs
...
The change enables VM support in graphs on Windows. That allows
to avoid caching of all allocations at the cost of map/unmap
overhead during memory create/destroy.
Change-Id: I792be00fba099e5e5d3cd44a963e1dfd6976a86d
2023-05-05 15:31:26 -04:00
f14e8a2dba
SWDEV-393910 - Port gfx94x changes to mainline.
...
Change-Id: Ibf727223bbe5230b132b47c39e0fc1d87cbd3b9c
2023-05-05 15:16:23 -04:00
afa28cdf44
SWDEV-393362 - Move setData() from profilingEnd() to profilingBegin() to eliminate a possible data race
...
- Introduce a state variable to indicate if HwProfiling is enabled to
eliminate a possible data race of vector<> signals_.
Change-Id: Id504cc76d7fa9f7e6455587dd232b60ccbbb735b
2023-05-04 17:19:25 -04:00
0a7708f982
SWDEV-301667 - Better log
...
Change-Id: I7399de7fe7a8840568d02362d2d936173583030c
2023-05-04 16:29:02 -04:00
82e4859e03
SWDEV-392870 - Enable p2p support for image using PAL.
...
Change-Id: I3acbbc01ee11e488e93204814361b8c3431bca4f
2023-05-03 00:34:08 -04:00
c585aefdb7
SWDEV-376453 - Changes to fix crash seen with HIP and recent compiler change with -O0
...
Change-Id: I37de83b7449461ce2378741e97ff8fb8da736523
2023-05-02 17:46:38 -04:00
538d453b86
SWDEV-301667 - Cleanup log
...
Change-Id: I9719ec8b19406a062727937471e7a845a522a0be
2023-05-01 11:07:55 -07:00
9775c38725
SWDEV-392427 - Invalidate Barrier Value AQL header
...
Change-Id: Id8e04ffe44da58641361468957d397af128443bb
2023-05-01 14:06:05 -04:00
d9c00767dc
SWDEV-345751 - fix misinitialize arraySlice in partialMemCopyTo()
...
Change-Id: I622c368957b1bd39e973f9afa4ee01593dd9e8ac
2023-05-01 13:07:44 -04:00
7ef2da5aba
SWDEV-353281 - Move VirtualMem map update to memobj
...
- The implementation in mempool graphs requires refcounting VA object.
That requires release() to update the map only on the actual destruction.
- Add GPU event tracking for paging operation. Otherwise, runtime
may not always flush IB.
Change-Id: Idf99ffb894321a38e04b490116a7ca435635918d
2023-04-28 17:22:11 -04:00
4a73fbec00
SWDEV-355540 - Do not reset core affinities
...
- Do not reset core affinities when spawning new threads
Change-Id: I26a12075a1ff197bc217aafb9ba18ffbb019ef43
2023-04-26 16:52:37 -04:00
3e5803c4c0
SWDEV-381627 - Move cl_vk_amd.hpp into rocclr
...
Rename VK interop to ExternalMemory object, since it should handle
DX interops also
Change-Id: I536ec46d3e53ece35234a2e29030393ad411b96d
2023-04-26 15:25:01 -04:00
2a9d8b5862
SWDEV-1 - Add missing newlines to debug log output
...
Change-Id: I6c98d482079c34931cdc0615075eba34f87241f3
2023-04-25 13:10:57 -04:00
afa15ddb89
SWDEV-367145 - adding support for timeline semaphore
...
Change-Id: I654359b48abe8bbaac2def5118d3ad0bb05d9492
2023-04-25 10:02:43 -04:00
5865c642d4
SWDEV-384557 - Fix engine status query
...
- Maintain a map of SDMA engine# to stream allocated following a greedy
approach
- Anything past that will query SDMA engine status always and go with a
SDMA or Blit copy path
Change-Id: Ibfaed7f951ab84d80cb0430596a4d11b5aec9202
2023-04-21 00:57:26 -04:00
636983de49
SWDEV-387117 - Support negative miplevel for clCreateFromGLTexture()
...
Change-Id: Ieaa340196e6844b2d18cdcc2278f913fc7c7bbb8
2023-04-19 12:46:45 -04:00
78de2ae692
SWDEV-366992 - adding handle type for vk interop
...
Change-Id: I79ee0d89b948c21b96709e9e607abe7901621a97
2023-04-18 16:59:25 -04:00
e82eecf33e
SWDEV-231145 - Add a validation for uniform launch
...
Change-Id: I68d61690c50060d9dcca117b248dddf69204a23f
2023-04-18 12:56:09 -04:00
c95944fd7e
SWDEV-1 - Make gl.h mandatory for build
...
Change-Id: I6f68d1bdb455f219c7b05a05256eb67d00e81a3c
2023-04-17 18:22:18 -04:00
1cd99b5056
SWDEV-1 - Enable PAL interface 792
...
Change-Id: I4b77ecc8e4c129c5e1e62ec84ac4a5381f97a823
2023-04-17 09:44:29 -04:00
667f048d18
SWDEV-383199 - Use SC for luxmark for gfx10+
...
Change-Id: I693b00fd9fe8e48b7602c8c5a87b59cfae69d891
2023-04-14 13:20:00 -04:00
3d87e3848b
SWDEV-2 - Change OpenCL version number from 3580 to 3581
2023-04-05 07:51:04 +00:00
20ca8b8116
SWDEV-384557 - Leverage SDMA engine status query
...
Change-Id: I5f386f2965de24a229ea43b6c4da82099692f91f
2023-04-05 07:50:53 +00:00
471ff60a90
SWDEV-387508 - construct uuid from pal props
...
This reverts commit e83c7a0510195fda6640c22d7ad98f056cf95dcd.
Reason for revert: it is a test issue to consider uuid a string it is array of values
Change-Id: Ifb2ae05a99cdaa80e689b83693fc553dd974a249
2023-04-05 07:50:45 +00:00
d897082a6c
SWDEV-383606 - Find the correct LocalWorkSize
...
when uniformWorkGroupSize_ is true
Change-Id: Ic7fa25d5fa503e59509d481a492f6519c6e52889
2023-04-05 07:50:34 +00:00
391bbd0455
SWDEV-2 - Change OpenCL version number from 3579 to 3580
2023-04-05 07:50:25 +00:00
82d2213cfa
SWDEV-380903 - Add program lock
...
Create seperate locks for program and HSAIL
Change-Id: I7c826ed9c2f2d839d359e5158f578830655616fd
2023-04-05 07:50:16 +00:00
7b419f42eb
SWDEV-389676 - Update stack size limit
...
Change-Id: Id0cf66820e76e1bbd7f6c17ceb110782cdb5f978
2023-04-05 07:50:06 +00:00
899c0e54e7
SWDEV-2 - Change OpenCL version number from 3578 to 3579
2023-04-05 07:49:57 +00:00
91e2cd5d67
SWDEV-339709 - Device Atomic test failed
...
Code change for Linux
Change-Id: I002881c188b9de351618a06ee8b5ce2aa5d9fd95
2023-04-05 07:49:47 +00:00
c2b64f8bf5
SWDEV-339709 - Device Atomic test failed
...
Code change for Windows.
Change-Id: I5ad520b67d1d9ec029cd2c2df7009915caaa8583
2023-04-05 07:49:34 +00:00
f670c97ad8
SWDEV-2 - Change OpenCL version number from 3577 to 3578
2023-04-05 07:49:23 +00:00
88736010fb
SWDEV-381627 - adding cl interop files to vdi
...
Change-Id: Ic40363587a2bc56f977a148eba386dfb73d6286e
2023-04-05 07:48:49 +00:00
53c071199e
SWDEV-2 - Change OpenCL version number from 3576 to 3577
2023-04-05 07:47:41 +00:00
25fe45bb2a
SWDEV-383856 - Remove mmap function for SVM buffer
...
For large bar platform, it's not necessary to mmap
memory allocated on VRAM to cpu again
Change-Id: I0701680476829d4058b3e7b643e8df657d0c6168
Signed-off-by: Ruili Ji <ruiliji2@amd.com >
2023-04-05 07:47:30 +00:00
37849a0726
SWDEV-286150 - [NFC] Refactor setLanguage method into a static function called getCOMGRLanguage
...
* Return the result by value and not through a pointer passed by
parameter
Change-Id: I8f872c95c4a330bebe299d486fb73f36660b469c
2023-04-05 07:47:02 +00:00
88b4b6428c
SWDEV-361630 - Workaround for rocfft build failure
...
Change-Id: Ie15e17361f2e6cd5f11e6fc1e6755c98cef7f4ce
2023-04-03 14:52:44 +00:00
52b78d8e79
SWDEV-361630 - Fix loading of versioned comgr for hiprtc
...
Change-Id: Ib0f604008e9197029eeacd22bc853538cda24cb5
2023-03-31 11:24:50 +00:00
d9ceb6a3a3
SWDEV-361630 - Enable opencl build
...
Change-Id: I19f8aa2e8157b8bfa053c502dcd63c52f0b6f104
2023-03-26 15:42:33 +00:00
1bc186323f
SWDEV-361630 - Enable hip build
...
Change-Id: Icdeaeff555d46e495c3de954b614168995a82a9f
2023-03-26 15:42:33 +00:00
Saleel Kudchadker
Alex Voicu
Jacob Lambert
AravindanC
Ioannis Assiouras
Jaydeep Patel
taosang2
German
kjayapra-amd
Sourabh Betigeri
Jatin Chaudhary
victzhan
pghafari
Jason Tang
Konstantin Zhuravlyov
Chauncey Hui
Shadi Dashmiz
Rahul Garg
Xie.AlexBin
Ajay
Ruili Ji
Juan Manuel MARTINEZ CAAMAÑO
Maneesh Gupta