Commit Graph

4470 Commits

Author SHA1 Message Date
Chauncey Hui 7630e701f6 SWDEV-2 - Change OpenCL version number from 3442 to 3443 2022-04-02 03:00:06 -04:00
kjayapra-amd 378a427d8c SWDEV-325776 - Adding device release scope for kernel dispatch packet
Change-Id: I8ea763f4c0239c410143b748c05822e9f6694412
(cherry picked from commit ec4894f8a27a3330b895a0ded385ab96f5ef242d)
2022-04-01 08:17:29 -04:00
Chauncey Hui 258e981094 SWDEV-2 - Change OpenCL version number from 3441 to 3442 2022-04-01 03:00:06 -04:00
Christophe Paquot 867346520f SWDEV-322620 - Virtual Memory Management Part 1
Adding opaque data handle to memory. This is used to look back the HIP object associated with it.

Change-Id: I1bbb14a915bed79c6c3593a29a627778c7aaf13a
2022-03-31 21:12:26 -04:00
kjayapra-amd 7fb80a027a SWDEV-305527 - Changes to handle memset blit kernel that takes width, height and depth. This also fixes SWDEV-317261.
Change-Id: Ic85f63a95d9d8f48884fc8c7fd95cbb496dfbbca
2022-03-31 09:02:33 -04:00
Chauncey Hui 7a0e83c061 SWDEV-2 - Change OpenCL version number from 3440 to 3441 2022-03-31 03:00:06 -04:00
German Andryeyev 28597ec5b5 SWDEV-328670 - Enable arena for ROCr interops
Add ROCR memory detection and enable arena mem object for possible
access in HIP

Change-Id: Icf86ac789176bfee4ea8d36b0970a817d4c6a2f7
2022-03-30 16:46:36 -04:00
Chauncey Hui 3a8b5cb9c3 SWDEV-2 - Change OpenCL version number from 3439 to 3440 2022-03-30 03:00:04 -04:00
Saleel Kudchadker 61d0b999be SWDEV-301947 - Report regular CU count for OpenCL
Change-Id: I3ea058bba98f3c6554cbde37173bbd772f489cf5
2022-03-29 16:19:14 -04:00
Satyanvesh Dittakavi c1b95b09bf SWDEV-326397 - P2P copies to take SDMA path if there is no pending dispatch
Change-Id: I50cfb8d77f7882151a20a1de7aaf5219b1695b7d
2022-03-29 14:59:11 +00:00
Chauncey Hui 37e6cbd983 SWDEV-2 - Change OpenCL version number from 3438 to 3439 2022-03-29 03:00:03 -04:00
Saleel Kudchadker 802c2c8a9f SWDEV-322225 - Use numa_allocate_bitmask
- Fix a crash with AMD_CPU_AFFINITY=1 as numa_bitmask_alloc isnt the
right api to allocate bitmask
- Do not set affinity for ROCr thread. It worsens performance rather
than any improvement.
- Fix regression from my previous change for event handler.

Change-Id: I3ea75adc2a6333f29752283eddd5b555e9b58cc5
2022-03-26 13:24:51 -04:00
Chauncey Hui f8a6099344 SWDEV-2 - Change OpenCL version number from 3437 to 3438 2022-03-26 03:00:06 -04:00
German Andryeyev a02ae1b851 SWDEV-311271 - Add a key to control memory pool feature
Change-Id: Ibd929592b802e65d0e1a4fd9689050bce5059e98
2022-03-25 19:07:14 -04:00
Ajay 35877b1b13 SWDEV-301667 - return void in getTime(). Avoid warning treated as error
Change-Id: I9445eec554e6f705fb8f248e6be7ff995f163f25
2022-03-25 13:39:55 -04:00
Chauncey Hui a43de338e3 SWDEV-2 - Change OpenCL version number from 3436 to 3437 2022-03-25 03:00:07 -04:00
Saleel Kudchadker 3c3c0ca4c5 SWDEV-301667 - Selectively queue handler
- Queue handler for hipEventRecord(aka marker_ts_) only if there is a
callback associated with it.

Change-Id: I8a9877ae0e342556053abbaacc9510744a8e772a
2022-03-24 19:46:28 -04:00
Saleel Kudchadker 9292abb2d8 SWDEV-328349 - Rate limit hostcall listener
Update timeout for hostcall wait for signal. If the timeout is small it
checks frequent enough to affect performance for certain applications
which may be CPU bound.

Change-Id: I0a879559e4ad111b09a994a5b82a6faf6e4fea3f
2022-03-23 12:18:00 -04:00
Chauncey Hui 9085b482ad SWDEV-2 - Change OpenCL version number from 3435 to 3436 2022-03-23 03:06:16 -04:00
German Andryeyev 3af3fe10de SWDEV-307185 - Move memory allocation under device layer
It can be too early to allocate memory at the begining of
Device::create() under PAL

Change-Id: I4bd76db7be3f6fb246243ea68022d8b0f860471d
2022-03-21 16:17:22 -04:00
Chauncey Hui 54771e2ed7 SWDEV-2 - Change OpenCL version number from 3434 to 3435 2022-03-18 03:00:08 -04:00
sdashmiz 6e506da36d SWDEV-307107 - HipPrintf failure for XGMI on windows
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Iab2740364bb673029f0a80b419d80d5b726b9082
2022-03-17 10:03:56 -04:00
Chauncey Hui c36b593657 SWDEV-2 - Change OpenCL version number from 3433 to 3434 2022-03-17 03:00:10 -04:00
Sarbojit Sarkar 3c2dc1f646 SWDEV-325708 - Query for FineGrained support
Change-Id: Idd20a71467595ab6577bf47c081c437a4b166988
2022-03-16 05:09:42 -04:00
Chauncey Hui 3b730facae SWDEV-2 - Change OpenCL version number from 3432 to 3433 2022-03-16 03:00:06 -04:00
Jason Tang 7356d2ecb1 SWDEV-316665 - Use CMAKE_SIZEOF_VOID_P to detect 32bit build
Change-Id: If959b0d012caa55afbe54f0fe7da8ad3cca6d452
2022-03-15 10:56:26 -04:00
Chauncey Hui c0ada20664 SWDEV-2 - Change OpenCL version number from 3431 to 3432 2022-03-15 03:00:06 -04:00
German Andryeyev 43ea9a8e31 SWDEV-286150 - Switch PAL version to 711
Change-Id: Iaf07946d6327f92ed734223a6b38c433ce368c4a
2022-03-14 13:00:40 -04:00
Chauncey Hui e8501ecda9 SWDEV-2 - Change OpenCL version number from 3430 to 3431 2022-03-12 03:00:10 -05:00
Jeremy Newton f2e5ef5617 SWDEV-323669 - Fix linux arch detection
CMake assumes we're bundling on x86, but for GNU compatible compilers,
we should rely on the compiler target to set the build arch.

For non-gnu compilers, just fall back to assuming x86 (no change).

Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
Change-Id: Iee9794e6f7c3973c781ddaf740ded77f34712c4f
2022-03-11 09:53:07 -05:00
Chauncey Hui c1858dc39e SWDEV-2 - Change OpenCL version number from 3429 to 3430 2022-03-10 03:00:14 -05:00
Saleel Kudchadker 3bf1d5ac97 SWDEV-301667 - Remove guarantee
Remove guarantee from AddMemObj as one can call it multiple times for
different devices

Change-Id: I49dd76068b3c4c709f17541159052302dcdb374d
2022-03-08 16:20:10 -08:00
Chauncey Hui ed42eda11d SWDEV-2 - Change OpenCL version number from 3428 to 3429 2022-03-08 03:00:12 -05:00
German Andryeyev 7b114a2b8b SWDEV-307185 - Create heap for device memory allocator
Pass the allocated heap with the kernel arguments

Change-Id: Icdec09b7f937845c39e21cbca7071dc3ba791af9
2022-03-04 00:44:41 -05:00
Chauncey Hui a6bcb4435a SWDEV-2 - Change OpenCL version number from 3427 to 3428 2022-03-02 03:00:06 -05:00
haoyuan2 58e1867079 SWDEV-298331 - Fix FreeLibrary throw exception issue on Windows
release HIP resource at DLL_PROCESS_DETACH point

Change-Id: I884ab028eeec1aaa2f655746f7eeae130fd009e4
2022-03-01 12:06:05 -05:00
Chauncey Hui ba8172b4ad SWDEV-2 - Change OpenCL version number from 3426 to 3427 2022-03-01 03:00:07 -05:00
Saleel Kudchadker e888c9e491 SWDEV-301947 - Rename device Info element
Rename maxBoostComputeUnits to maxPhysicalComputeUnits_.

Change-Id: I5941515ac4f1f4348b3b10478bf4e01444f0a864
2022-02-25 16:48:15 -08:00
Chauncey Hui af1f395718 SWDEV-2 - Change OpenCL version number from 3425 to 3426 2022-02-25 03:02:14 -05:00
German Andryeyev 2664d8cf9e SWDEV-319375 - Avoid global var size query
Currently COMGR doesn't provide global variable size and runtime
parses ELF binary directly. Avoid parsing for HIP. That can save
5% in hipModuleLoad() time.

Change-Id: I47540d1e957bdb0c2406b6b848222de2920b2504
2022-02-23 21:33:05 -05:00
Chauncey Hui f1d55767a8 SWDEV-2 - Change OpenCL version number from 3424 to 3425 2022-02-22 03:00:06 -05:00
Saleel Kudchadker c02d3fd7d8 SWDEV-301667 - Fix maxNode size for get_mempolicy
Change-Id: Ifd84c94394b86580cf39178ad0e7f85580b24edb
2022-02-19 12:48:22 -05:00
Chauncey Hui f95188ba4c SWDEV-2 - Change OpenCL version number from 3423 to 3424 2022-02-19 03:00:07 -05:00
German Andryeyev 95d55fdfa8 SWDEV-323702 - Use active queue for transfer
Pass active queue for transfers in the cache coherency layer.
That will allow to use device transfer queue only for
cases when active queue isn't available, because using device
transfer queue from another active queue may cause a deadlock

Change-Id: Ifbe7e0303b77dbf6eeda3939ffbc25a3df7472de
2022-02-18 09:10:53 -05:00
Chauncey Hui ffbd63e774 SWDEV-2 - Change OpenCL version number from 3422 to 3423 2022-02-18 03:00:06 -05:00
German Andryeyev fbf531398a SWDEV-323364 - Fix a typo
Change-Id: I2031296ab9451342d5930b8b2d3d2e6277946647
2022-02-17 20:50:29 -05:00
Chauncey Hui 0a2baa91b1 SWDEV-2 - Change OpenCL version number from 3421 to 3422 2022-02-17 03:00:05 -05:00
Saleel Kudchadker 041ddc0c1c SWDEV-322605 - Fix infinite loop condition
If GlobalMemCacheLine reported is 0, runtime may run into an
infinite loop as the KernelSegmentAlignment is chosen as size of the
cache line.

Change-Id: Ide547940cc0407f16fab10ee210b4fd3ae4eaafc
2022-02-16 13:16:18 -05:00
German Andryeyev 3fd4a67670 SWDEV-316824 - Fix P2P compute copy path
Use device memory object for the GPU VA address look-up.

Change-Id: I76bf58b29205f7b3ba1bf68e9fcca69421267203
2022-02-15 13:20:13 -05:00
Chauncey Hui 07d3044160 SWDEV-2 - Change OpenCL version number from 3420 to 3421 2022-02-12 03:00:09 -05:00