Commit Graph

388 Commits

Author SHA1 Message Date
3792481d33 XeHPC Implicit scaling: put command/ring/semaphore buffer to first memory bank
In direct submission scenario command/ring/semaphore buffer allocations
are placed in the same memory bank to ensure that their memory is updated in
correct order

Related-To: NEO-6698
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2022-03-21 08:10:50 +01:00
754efbdaff Cleanup mocks
Signed-off-by: Daniel Chabrowski daniel.chabrowski@intel.com
Related-To: NEO-6591
2022-03-18 11:54:03 +01:00
c0e2251ceb Skip adding allocations to remote devices if not allocated there
When making graphics allocations resident in multi-GPU scenarios,
we should make them resident only if there's an allocation for that
device. So return appropriate null pointer and skip it.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
2022-03-16 23:58:07 +01:00
a71b88fefb PageFaultHandler: speedup UM allocations lookup
Add a per-instance SVMAllocsManager::nonGpuDomainAllocs container for
all allocations to be removed in
moveAllocationsWithinUMAllocsManagerToGpuDomain. This approach replaces
the current iterative search and performs the task faster.

Add 7 new unit-tests to verify the functionality related to
nonGpuDomainAllocs container, both in expected and unexpected/synthetic
scenarios.

For UTs replace a dummy unifiedMemoryManager pointer with a pointer to
an instace of SVMAllocsManager, otherwise a SegFault error is thrown at
the end of tests.

Perform overall cleanup in related tests implementation, includes but
not limited to removal of:

- givenInitialPlacementGpu\
WhenMovingToGpuDomainThenFirstAccessDoesNotInvokeTransfer

As it is fully covered by:

givenAllocationMovedToGpuDomain\
WhenVerifyingPagefaultThenAllocationIsMovedToCpuDomain

- givenInitialPlacementGpu\
WhenVerifyingPagefaultThenFirstAccessDoesNotInvokeTransfer

As it is fully covered by:

givenTbxAndnitialPlacementGpu\
WhenVerifyingPagefaultThenMemoryIsUnprotectedOnly

Finally, reduce code duplication where possible.

Related-To: NEO-6658
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
2022-03-16 11:18:24 +01:00
c0b7f05897 Add memory prefetch for kmd migrated shared allocations
This feature is disabled by default, controlled with the knob
AppendMemoryPrefetchForKmdMigratedSharedAllocations

Related-To: NEO-6740

Signed-off-by: Milczarek, Slawomir <slawomir.milczarek@intel.com>
2022-03-09 16:02:18 +01:00
a010fb3634 Dont close shared handle on imported allocations
Related-To: LOCI-2272

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
2022-03-06 04:35:47 +01:00
090bfb9642 Reuse kernel allocation
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-02-28 19:26:19 +01:00
e4a93f9870 Set page table cloning on for event device allocs
Related-To: NEO-5968

Signed-off-by: Aravind Gopalakrishnan <aravind.gopalakrishnan@intel.com>
2022-02-24 22:35:03 +01:00
0634aa3f1b Create resource with given address
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-02-24 17:06:19 +01:00
c405fb5c2c Use shared mutex for unified memory manager
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-02-23 19:00:38 +01:00
71f48c89c8 CacheSettingsHelper - timestamp allocations
Related-To: NEO-6664

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2022-02-09 19:32:39 +01:00
61ca84e94b Pass memory placement info to StorageInfo
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2022-02-09 16:55:32 +01:00
a95198521e Initial implementation of CacheSettingsHelper
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2022-02-08 16:18:06 +01:00
c88fce0def Gmm construction cleanup
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2022-02-08 10:20:24 +01:00
4b0d986876 Move AllocationType enum out of GraphicsAllocation class
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2022-02-04 17:49:09 +01:00
45d23868de Remove device enqueue part 15
- remove DEVICE_QUEUE_BUFFER

Related-To: NEO-6559
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
2022-02-03 16:19:07 +01:00
2be98a1e62 Create kmd migrated allocation with initial placement
Implements ZE_HOST_MEM_ALLOC_FLAG_BIAS_INITIAL_PLACEMENT
for zeMemAllocShared with KMD migrated allocation.

Signed-off-by: Milczarek, Slawomir <slawomir.milczarek@intel.com>
2022-02-01 15:42:10 +01:00
ae77bd1bd2 Enable Device Memory to be shared in WSL-2 with L0
- Add getMemoryManagerType to check which memory manager has been init
to determine if Linux + WDDM memory manager is in use.
- Add isNTHandle to test and verify if a handle is an NT handle during
L0 Open IPC Handle.

Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
2022-01-27 23:14:29 +01:00
a5067d6b0b Remove duplicate check in residency container
Remove find in SVMAllocsManager
addInternalAllocationsToResidencyContainer,
not needed, CSR resolves duplicates at makeResident calls

Co-authored-by: Michal Mrozek <michal.mrozek@intel.com>

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2022-01-26 17:09:31 +01:00
592eb4607e [4/n] Improve indirect allocations handling.
Add memory manager interfaces to obtain the status.
Only support when VmBind is supported.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2022-01-26 15:24:32 +01:00
dbf0f90186 Return pageSize in getMemAllocProperties
Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
2022-01-25 18:26:13 +01:00
6df17f5a30 [3/n] Optimize indirect allocations handling.
Add new debug variable to trigger new mode.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2022-01-25 16:40:56 +01:00
52d636394c [1/n] Improve indirect allocations handling.
Add new functions that would treat all indirect allocations as single pack.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2022-01-25 06:27:33 +01:00
151aaf7678 Fix alignment for host allocations.
- it is 4k not 64k.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2022-01-24 13:24:03 +01:00
6082865eb4 Revert "Optimize Level Zero indirect allocations handling."
This reverts commit 3ecbc55ba9.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2022-01-20 11:41:13 +01:00
1b7949432f Add shareable allocation on windows dGPUs
Add default initialization for object members

Related-To: LOCI-2665

Signed-off-by: Kamil Diedrich <kamil.diedrich@intel.com>
2022-01-19 19:03:18 +01:00
3ecbc55ba9 Optimize Level Zero indirect allocations handling.
Make them resident directly instead of populating residency container
Remove finds, not needed, CSR resolves duplicates at makeResident calls
Observed gain is 32x for 10k indirect allocations.


Co-authored-by: Michal Mrozek <michal.mrozek@intel.com>

Co-authored-by: Dominik Dabek <dominik.dabek@intel.com>

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2022-01-19 13:08:35 +01:00
394c0e90e1 Return error when failing on submission
Signed-off-by: Raiyan Latif <raiyan.latif@intel.com>
2022-01-12 16:42:30 +01:00
7fc9b2c3dc Implement UUID using pcie bdf
This patch implements the fallback method for Device UUID
using the BDF of the device for all product families.


Related-To: LOCI-2827

Signed-off-by: Joshua Santosh Ranjan <joshua.santosh.ranjan@intel.com>
2022-01-05 23:31:00 +01:00
3b497270c8 Add support for memory free policies
Add support for ZE_DRIVER_MEMORY_FREE_POLICY_EXT_FLAG_BLOCKING_FREE
added in v1.3.

Related-To: LOCI-2672

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
2021-12-28 08:27:50 +01:00
e040c8c6bb Initial compression support for L0 Device allocations
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-16 15:55:13 +01:00
2fd536104d Extend GfxPartition to support 57b addressing
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-08 16:05:05 +01:00
dfdd3c597a Remove redundant BUFFER_COMPRESSED allocation type
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-07 13:35:49 +01:00
0346a5679f Fix returned local memory size
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-06 15:57:55 +01:00
68aea5bf62 Rename compression flags and helpers
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-03 18:09:02 +01:00
9481d8a88b Compression refactor [2/n]
Dont use allocation type for compression preference

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-03 08:47:46 +01:00
3e1023fa1a Unify memory layout for all multi tile post sync operations
Related-To: NEO-6262

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
2021-12-02 18:00:40 +01:00
55959d4d1d Helper method to check if allocation is compressed
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-02 16:13:53 +01:00
7b55d38e34 Refactor [1/n]: Make compression preference allocation property
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-02 15:24:35 +01:00
56c05a2f08 Update allocateGraphicsMemoryWithAlignment
Allocate Gmm object when the allocation uses compression.

Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
2021-11-09 13:37:32 +01:00
95610188af Fallback to cpu copy when filling work partition allocation
move some command stream receiver tests to shared

Related-To: NEO-6325
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2021-11-03 12:17:02 +01:00
782a287b27 Revert "Fallback to cpu copy when filling work partition allocation"
This reverts commit 93e3d948f5.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2021-10-29 04:39:22 +02:00
515130dbe8 Fail on blitMemoryToAllocation when blitter is not supported
Related-To: NEO-6325
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2021-10-28 16:13:40 +02:00
93e3d948f5 Fallback to cpu copy when filling work partition allocation
move some command stream receiver tests to shared

Related-To: NEO-6325
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2021-10-28 14:30:00 +02:00
5fdcb14d57 WSL fallback to staging buffer on blocking reads
Signed-off-by: Kamil Diedrich <kamil.diedrich@intel.com>
2021-10-28 03:34:58 +02:00
ba767dfbaf Revert "Fail on blitMemoryToAllocation when blitter is not supported"
This reverts commit 7bfeeaab86.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2021-10-27 09:13:57 +02:00
7bfeeaab86 Fail on blitMemoryToAllocation when blitter is not supported
Related-To: NEO-6325
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2021-10-26 11:05:51 +02:00
71ed3eba16 Add mapping based colouring
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2021-10-25 09:51:54 +02:00
d34ef3170a Remove CPP_ATTRIBUTE_FALLTHROUGH macro
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-10-22 17:27:57 +02:00
d2a29ce458 DrmBO: canonize GPU addresses
Related-To: NEO-6364
Signed-off-by: Artur Harasimiuk <artur.harasimiuk@intel.com>
2021-10-21 18:25:48 +02:00