344 Commits

Autor SHA1 Mensaje Fecha
kjayapra-amd 765e6f5d2f SWDEV-413997 - Fixing multiple device cases.
Change-Id: I10ad3fbfca887e92cd81f68392fa1acf753cbd2b


[ROCm/clr commit: d52d16c8e6]
2024-04-13 06:14:03 -04:00
kjayapra-amd ed9c629ad6 SWDEV-446298 - Adding error code to the logs on p2p hsa api failure.
Change-Id: Ic41b1ad1b64cca0e31986337a83a5146d52a7328


[ROCm/clr commit: 2b8634bada]
2024-04-10 06:00:00 -04:00
Saleel Kudchadker 4285981222 SWDEV-301667 - Fix SDMA mask reuse
If we are using the mask returned by getLastUsedSdmaEngine() then we
need to apply the SDMA Read/Write mask to it before using with HSA
copy_on_engine API.

Change-Id: I6e5dc6c187eeb3c61ee159e9d2a0fa7b4737c06e


[ROCm/clr commit: 3f0bcf7834]
2024-04-08 15:42:52 -04:00
Sourabh Betigeri 7cc5dd56ab SWDEV-451964 - Limit gpu single allocation percentage for gfx940 only
Change-Id: Iadcdadd734e7aeeb23742e426353defa972d3ad5


[ROCm/clr commit: dbac2976e4]
2024-04-05 09:43:42 -04:00
kjayapra-amd 0f8a9567e6 SWDEV-413997 - Save hsa_handle as ptr in hipMemCreate path.
Change-Id: Ica32017ef7b00326dfb6d1f604e126d40ad5b786


[ROCm/clr commit: 5cbd74b554]
2024-03-26 10:24:29 -04:00
Ioannis Assiouras b46d3c0f8d SWDEV-451166 - Disable kernel args for non-XGMI if HDP flush register is invalid
Change-Id: I227e046e2b9cb25476a50240f5d070adbd558f21


[ROCm/clr commit: 96f5c44851]
2024-03-15 05:27:52 -04:00
kjayapra-amd 8947420e41 SWDEV-423835 - Fixing kernel launch issues on Virtual Memory Management path.
Change-Id: I9f5e8a3d83af3809b2c50b21a10697e26113dd23


[ROCm/clr commit: f5ca620baa]
2024-03-12 17:22:07 -04:00
Saleel Kudchadker 4ff5ec0a02 SWDEV-301667 - Better log
- Print SWq for AQL packets, this helps correlating a stream to the HWq
mapped

Change-Id: I610430c0872a1abc6636027c00163ec46983cd65


[ROCm/clr commit: 984c86f407]
2024-03-01 16:43:06 -05:00
Ioannis Assiouras 80ddb8e4eb SWDEV-446399 - Fixed segfault in hipMemSetAccess
Change-Id: Ia1200d9bee03e8abade211287505f081e635ceec


[ROCm/clr commit: 1f6d416684]
2024-02-20 18:51:05 -05:00
kjayapra-amd bef39a9369 SWDEV-437832 - Changes to update host unified memory and iommuv2 flags.
Change-Id: I88998cf57c21fc446fa28e250f826c607923670b


[ROCm/clr commit: 7d5b4a8f7a]
2024-02-07 06:27:47 -05:00
Saleel Kudchadker 423887c6f6 SWDEV-301667 - Better log
Display queue base pointer in the log. This can be co-related with AQL
packets

Change-Id: I544f9b6db6ae01c85e57e4b3f0b3fffefcd7c2ed


[ROCm/clr commit: 0567c3b720]
2024-02-05 05:08:11 +00:00
kjayapra-amd 9bff044c87 SWDEV-437832 - Adding device property to check if the device is accelerator.
Change-Id: I8349e99c03422c268bbb60a8c143bd492d9cec09


[ROCm/clr commit: b366a7c992]
2024-02-05 05:08:11 +00:00
Satyanvesh Dittakavi d1901f2017 SWDEV-434846 - Limit the gpu single allocation percentage for all MI300 versions
Change-Id: I33dea3eaab249ce3f9a624d38267489f99cd530c


[ROCm/clr commit: 755eb2962c]
2024-01-03 23:47:44 -05:00
German Andryeyev 7130b87d5c SWDEV-436859 - Enable pitch for COPY_HOST_PTR
Original logic didn't use pitch because, abstraction layer had
a sysmem copy without pitch. Since extra sysmem copy was
disabled, the code has to accept pitch values from the app.

Change-Id: Ia9fba7b33ddff4e9109b4e63d0d6afa52f501c8f


[ROCm/clr commit: fb3dfcf889]
2023-12-13 16:50:16 -05:00
Satyanvesh Dittakavi 542044aadc SWDEV-434846 - Correct the vgprs per simd for MI300
Change-Id: Id4862da7611f64392bfc1538fb644801ec0a9e7f


[ROCm/clr commit: b2102fe939]
2023-12-13 03:06:21 -05:00
German Andryeyev e390ec044f SWDEV-432174 - Change the fillBuffer kernel
- Add the new fillBuffer kernel, which allows to launch a limited
number of workgroups for memory fill operation
- Switch fill memory to 16 bytes write by default
- Allow to limit the workgroups with DEBUG_CLR_LIMIT_BLIT_WG

Change-Id: Ibad1822f2d42b2fc71bcfc1917c31409c0623e8e


[ROCm/clr commit: f1dc81f427]
2023-11-16 14:25:55 -04:00
Jatin Chaudhary 2361cc0dab SWDEV-431399 - fix first set of memory leaks in clr, found in rtc tests
change constexpr variable names to match the C++ style we follow.

Change-Id: Ibc59a65d8ff2ca765da7bf5e653c0650fb3714c4


[ROCm/clr commit: ce27581465]
2023-11-14 20:39:45 -05:00
Saleel Kudchadker 85b95f0a45 SWDEV-301667 - Add error logging
Change-Id: I814399dc0e7083bb7fb0ed8bf46dd96bdf664965


[ROCm/clr commit: f06368fd04]
2023-11-10 11:55:54 -05:00
Alex Xie 4c860f41fd SWDEV-430062 - Support GPU_MAX_HEAP_SIZE flag in ROCm
Change-Id: Ibfe82b3524e09c61879b988f23512f394d725024


[ROCm/clr commit: 4fb9f03f9e]
2023-11-07 10:07:24 -05:00
German 005aebbfad SWDEV-429529 - Allocate glb_ctx_ even for one device
Move context allocation into Device::init() method to simplify the logic and handle
HIP_VISIBLE_DEVICES properly

Change-Id: I0fc6f37c7ae39bedbdad0290295d6794c66d6c54


[ROCm/clr commit: a49d633883]
2023-10-27 15:00:15 -04:00
Saleel Kudchadker cba9970220 SWDEV-408180 - Address possible cornercases
- Address corner cases that can arise with the new
hipMemcpyDeviceToDeviceNoCU enum
- Better log

Change-Id: I6035b901f8d616741054b7a5ff4f67956329ac57


[ROCm/clr commit: 5662d4037c]
2023-10-23 16:54:08 -04:00
kjayapra-amd 6d8458ba5a SWDEV-413997 - Initial VMM changes for ROCm path.
Change-Id: I4405fd7b53182eb4c4622835c811c0dc08461537


[ROCm/clr commit: 3ef829939a]
2023-10-16 11:29:16 -04:00
Jatin Chaudhary 15b062b9b4 SWDEV-306642 - [ABI Break] Add texture/surface/device capabilities device struct entries
- alias hipGetDeviceProperties to hipGetDevicePropertiesR0600
- alias hipDeviceProp_t to hipDeviceProp_tR0600
- remove gcnArch from new device property struct
- add new requested struct members

Change-Id: If3f5dbef3d608487d9f6f419285f4bf577ea9bf0


[ROCm/clr commit: 2989840511]
2023-10-12 11:16:18 -04:00
kjayapra-amd 96580585c3 SWDEV-419688 - Do not run GWS init kernel for targets > gfx12 and MI300.
Change-Id: I8e7441268978be71ab8a5a33e7f8bcf69660e500
(cherry picked from commit 36d37ef614909c0f215512aac0c133408d787080)


[ROCm/clr commit: 6a8bc3c718]
2023-10-05 14:57:56 -04:00
Julia Jiang 78ced05253 SWDEV-401314 - Fix failure on MI300 for progvar_prog_scope_misc basic test
Change-Id: I56701b050023199a3e56dc5a78553c0c330b8537


[ROCm/clr commit: 42799c9f21]
2023-09-14 11:32:25 -04:00
Saleel Kudchadker 28fa32a405 SWDEV-384557 - Reset mask at in the constructor
Change-Id: I15dafd281213c03590bcae4bf4544e33c4a7e277


[ROCm/clr commit: 8a538b5639]
2023-08-11 17:22:27 -04:00
Saleel Kudchadker 520b17fb07 SWDEV-384557 - Enable SDMA query
Change-Id: Ibb0a8d131f799985a4d4adbf753261e58c04157f


[ROCm/clr commit: aa6eb555e2]
2023-08-01 18:41:23 -04:00
Alex Xie a7418845f8 SWDEV-409299 - Vega clinfo is not working
Change-Id: Ia48bc6f130bd102dff210b105de6f9c02ebbe012


[ROCm/clr commit: 7912f3af89]
2023-07-10 09:53:50 -04:00
German b55398774b SWDEV-408650 - Report CL_DEVICE_HALF_FP_CONFIG query
If cl_khr_fp16 extension is enabled, then OCL runtime should report CL_DEVICE_HALF_FP_CONFIG.

Change-Id: I7c4ac48387f80bc704a475c57e5b52a462090d1b


[ROCm/clr commit: ad2c1e899a]
2023-07-06 11:58:16 -04:00
kjayapra-amd 03504e2f6c FEAT-47686 - Changes to support new uncached memory segment in ROCr.
Change-Id: I0ba8769d6737cdf1cc8a8644b2e82109f584a430


[ROCm/clr commit: aca7d9e14a]
2023-07-05 20:47:51 -04:00
Rakesh Roy f887f2fc6f SWDEV-405329 - Fix cuMask issue for WGP mode
- Enable CUs adjacent pairwise for WGP mode
- In HostQueue::terminate() do not segfault if virtual device hasn't been created

Change-Id: I94402ff333308af5824878086cc238b3993d534d


[ROCm/clr commit: 8c1232124e]
2023-06-30 01:09:01 -04:00
Saleel Kudchadker 47894de700 SWDEV-384557 - Rename env var
- Rename HIP_USE_SDMA_QUERY to DEBUG_CLR_USE_SDMA_QUERY as this is
supposed to be a temporary env var for debug purposes only.

Change-Id: If6ebd52ab87624375a3df24ceccdcc05c60a65af


[ROCm/clr commit: 770b2a4711]
2023-06-29 13:54:55 -04:00
ajay a8d5879f26 SWDEV-406687 - combining rocblitcl and palblitcl blit kernel defs
Change-Id: Ia312d73584a03491e8d574f424295b64df6de174


[ROCm/clr commit: d6946ffcbc]
2023-06-23 18:38:36 -04:00
German c8e65415ea SWDEV-311270 - Clean-up IPC path in ROCclr
- Make IPC interfaces generic between devices and rely on the IPC buffer
for attach/detach logic

Change-Id: Id3c18d122030329b7ee532bbb6317de9dd6a0bbe


[ROCm/clr commit: 9aa6f25f1c]
2023-06-20 15:18:21 -04:00
Saleel Kudchadker 2e67d925a1 SWDEV-384557 - Include query under env var
Change-Id: Ibe0335168f9ad0cb72abb2ed69a5e40d33ce7c1c


[ROCm/clr commit: 4510caed26]
2023-06-16 15:09:46 -07:00
Saleel Kudchadker a5715294fb 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


[ROCm/clr commit: 5436d362b1]
2023-05-22 11:23:48 -04:00
Ioannis Assiouras ad30c947bf SWDEV-400119 - Return error from globalFreeMemory() if HSA available memory query fails
Change-Id: I33c069fa1e0d31891f64dc8167a24ef9d7ed1426


[ROCm/clr commit: a07e38be2f]
2023-05-15 11:54:25 +01:00
Saleel Kudchadker 0141e6809f SWDEV-398151 - Partly relax static engine allocation
Change-Id: I4903b51a34b597a2e84d771b52cf629f877dba05


[ROCm/clr commit: 0b475284e9]
2023-05-11 00:52:18 -04:00
Jaydeep Patel cb3642e26e SWDEV-398047 - Disable arena for XNACK.
Change-Id: If6c98490bd5b8146a357f45ed132f687def0877f


[ROCm/clr commit: ffa1deef27]
2023-05-10 23:28:05 -04:00
Saleel Kudchadker 8ae30e036d 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


[ROCm/clr commit: 5865c642d4]
2023-04-21 00:57:26 -04:00
Xiaomeng Hou 7729cb184b SWDEV-373400 - Lower the threshold of single alloc percent for APUs w/ small VRAM
Change-Id: Ie8e3504a0c468bc44633991cdfde66bcf0b74f06


[ROCm/clr commit: 35a0855e4e]
2023-03-06 10:12:01 -05:00
Saleel Kudchadker 981dc31e20 SWDEV-380035 - Do not create Arena Memobj for pinned memory
Change-Id: I4da281f0b7139efb4bb46b47f812358a0212445a


[ROCm/clr commit: d429314b21]
2023-02-24 12:31:25 -08:00
Anusha GodavarthySurya 06ffd15060 SWDEV-366636 - Fix performance drop in TF-RCCL models
Change-Id: Idc845bb0dab858b94b9d2720cae8308cac2e7328


[ROCm/clr commit: f98dcf9d7f]
2023-02-17 00:48:13 -05:00
Maneesh Gupta d7fdd9fcb8 SWDEV-368235 - Revert "Remove obsolete env variables"
This reverts commit dfa7790030.

Reason for revert: Deferred to a future release.

Change-Id: Ia66c37f0ab9734dee73c930d10d7469d5fd57254


[ROCm/clr commit: 5dc104b3ea]
2023-02-15 07:25:00 +00:00
Anusha GodavarthySurya 24eec94928 SWDEV-380035 - Check for agent and ptr match for hsa LOCKED ptr
Change-Id: I2503aa05512aebc3535963e188ca8d1dbfab08f0


[ROCm/clr commit: fa35239b4b]
2023-02-15 00:12:42 -05:00
David Yat Sin b8226e4292 SWDEV-365908 - Do not align mem size for IPC
Remove alignment to granularity for IPC handles as ROCr has a patch that
will internally validate pointer sizes against requested size during
allocation instead of size aligned to page size. This patch is needed
together with this patch from ROCr:
f8a42a3a:Use user requested size for memory fragments

Change-Id: I28b25558ea03c836b44fafdb34b7330cf6887424


[ROCm/clr commit: fa1b7925cf]
2023-02-13 17:32:02 -05:00
German Andryeyev b180effdb6 SWDEV-381633 - Use blit manager on app's queue
Heap initialization used device queue, but it shoudl be used for
cooperative launches only. Heap initialization must use the same queue
as the current dispatch.

Change-Id: I856621bf82bbdeb1c2d0fbc4970e90d09af805cb


[ROCm/clr commit: 67af16bdf3]
2023-02-09 09:43:13 -05:00
Saleel Kudchadker 95e07af803 SWDEV-381633 - Better log
Change-Id: Ie151d73aa4b1dae7d0d9acd0457e7dbdf35f21ee


[ROCm/clr commit: 7619b53e9b]
2023-02-08 17:54:11 -05:00
jatang 43e9497462 SWDEV-380792 - Fix floating point exception when maxEngineClockFrequency_ is 0
Change-Id: Ic443ceae586c4c84995ed2abef9bd7f32f8b60f9


[ROCm/clr commit: b798c85272]
2023-02-07 11:43:10 -05:00
German dfa7790030 SWDEV-368235 - Remove obsolete env variables
Change-Id: I7e14d53297e79e2f68b3a6cc40251ad7db9eb5ab


[ROCm/clr commit: 7b50c935f8]
2023-02-03 13:44:24 -05:00