Commit Graph

560 Commits

Author SHA1 Message Date
Mateusz Hoppe 83ac95d293 fix: L0 - remove synchronization with events on appends in tbx mode
Related-To: NEO-9400

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-11-27 10:39:55 +01:00
Lukasz Jobczyk ac8c00048e performance: optimize svm allocation tracking
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2023-11-23 10:54:01 +01:00
Maciej Plewka 560b588cbe fix: Update residency fence value after ring buffer switch
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
2023-11-17 20:46:14 +01:00
Compute-Runtime-Validation 7f61217a44 Revert "performance: optimize svm allocation tracking"
This reverts commit e91ce78ec8.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2023-11-16 11:03:19 +01:00
Lukasz Jobczyk e91ce78ec8 performance: optimize svm allocation tracking
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2023-11-15 13:58:05 +01:00
Dominik Dabek 961a8d91d0 refactor: move gmm constructor flags to struct
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2023-11-15 09:26:13 +01:00
Lukasz Jobczyk 9a8138725a fix: Deferred SVM allocations look up by gpu address
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2023-11-14 15:27:01 +01:00
Dominik Dabek 43841fd2ef fix: clear AllocationsList tail on free all
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2023-11-02 18:06:13 +01:00
Michal Mrozek 80e3632c41 performance: do not free deferred allocations when not needed
Level Zero doesn't use deferred allocations so no point in paying the
price to check for them in cleanAllocationList.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2023-10-27 18:02:43 +02:00
Mateusz Hoppe 5d572b9c8f feature: allow freeing memory in aubstream
Related-To: NEO-2707

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-10-26 17:16:23 +02:00
Compute-Runtime-Validation fca2159430 Revert "fix: if device hierarchy is flat then getSubDevicesCount return 1u"
This reverts commit cb0bb57f49.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2023-10-26 15:40:29 +02:00
Baj, Tomasz cb0bb57f49 fix: if device hierarchy is flat then getSubDevicesCount return 1u
Related-To: NEO-9167

Signed-off-by: Baj, Tomasz <tomasz.baj@intel.com>
2023-10-25 15:51:52 +02:00
Mateusz Hoppe 97faeae16f feature: heapAssigner per root device
- create heapAssigner per root device in memory manager to allow per
device config

Related-To: NEO-7063

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-10-19 19:52:24 +02:00
Spruit, Neil R 58fa968273 fix: Calculate size of buffer surface state given mapped allocations
Related-To: NEO-8350

- given a virtual address part of a mapping to multiple physical
allocations, then the buffer surface state size is increased to
include the allocations which follow the current allocation, which
allows users access to the remainder of the mapped buffer.

Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
2023-10-19 13:38:51 +02:00
Spruit, Neil R 456f07212d fix: Return pageSize/Heap for reservation based on the size and devices
Related-To: NEO-9127,NEO-9142

- Change the heap/pageSize used based on the size of the reservation
required and the device heaps available.
- Return the page size required based on the size requested by the user
for virtual reservation and the devices on the system.
- Check the size passed in by the user in memory map and physical memory
allocation is valid for the heap and page size that is required.

Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
2023-10-19 10:44:13 +02:00
John Falkowski f156a74f54 fix: split chunking prefetch flags
Related-To: NEO-9120

Signed-off-by: John Falkowski <john.falkowski@intel.com>
2023-10-18 19:20:42 +02:00
Dunajski, Bartosz af7bcbf99c Revert "refactor: split CpuInaccessible MemoryPool types to Device and System"
This reverts commit 2e8cf5fdf5.

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
2023-10-17 10:03:14 +02:00
Mateusz Hoppe 8c6b936dac fix: decanonize address passed to AubManager's freeMemory()
- the same address should be used for writeMemory() and freeMemory()

Related-To: NEO-2707

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-10-16 12:58:21 +02:00
Dunajski, Bartosz 2e8cf5fdf5 refactor: split CpuInaccessible MemoryPool types to Device and System
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
2023-10-16 12:47:13 +02:00
Mateusz Jablonski a420e34b10 fix: explicitly remove assign operators when not needed
when class defines copy/move ctor then corresponding assign operator(s)
should be defined or deleted

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-10-12 10:25:15 +02:00
Mateusz Hoppe 5558222337 fix: handle completion in OsAgnosticMemoryManager
- add pollForCompletion() for contexts that used allocation prior to
freeGraphicsMemory()

Related-To: NEO-2707

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-10-11 10:24:30 +02:00
Mateusz Hoppe c2d69e5857 feature: allocate SPECIAL_SSH heap in front window from EXTERNAL heap
- SPECIAL_SSH is used for debug surface SurfaceState which must be
located at bindless offset zero
- limit size of external front window

Related-To: NEO-7063

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-10-09 14:54:39 +02:00
Mateusz Hoppe 8b4242eb32 refactor: move setting GpuBaseAddress to mapGpu call
- set base address of heap after mapping allocation to gpu va
- ensures the same heapIndex is used while choosing heap range and
setting base

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-10-09 12:21:23 +02:00
Mateusz Jablonski ece03e6bbf fix: add checks to avoid invalid behavior
Related-To: NEO-9038
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-10-06 15:45:33 +02:00
Mateusz Jablonski 382fc952f2 refactor: add NonAssignableClass to define classes without assign operator
Related-To: NEO-9038
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-10-05 15:16:58 +02:00
Mateusz Jablonski 95eb5df978 refactor: use reference in for-range loops
Related-To: NEO-9038
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-10-04 15:32:26 +02:00
Mateusz Jablonski fc508212de refactor: pass big parameters as reference instead of by value
Related-To: NEO-9038
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-10-04 14:53:13 +02:00
Mateusz Jablonski 747d44f8fc fix: add unrecoverable in case of allocation type unknown
Related-To: NEO-9038
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-10-04 14:02:36 +02:00
Mateusz Jablonski 895a586083 refactor: use reference in for-range loops
Related-To: NEO-9038
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-10-03 18:56:53 +02:00
John Falkowski 56f05303c9 feature: add support for zeMemGet/SetAtomicAccessAttributeExp
Resolves: NEO-8219

Signed-off-by: John Falkowski <john.falkowski@intel.com>
2023-10-02 15:59:17 +02:00
Dunajski, Bartosz c5f5f60f8d feature: cross CmdList and Event in-order data sharing [1/n]
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
2023-09-29 11:14:18 +02:00
Mateusz Jablonski 7d356795f0 fix: ensure memory placement key does not apply to unknown alloc type
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-09-28 08:43:30 +02:00
Katarzyna Cencelewska e71db368db fix: use proper pointer inside appendWriteGlobalTimestamp
Resolves: NEO-8105
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
2023-09-27 14:00:37 +02:00
Dunajski, Bartosz 202a33ffac feature: track InOrder allocation ownership
CmdList can be released before Event. In this case, GfxAllocation
destruction must be deferred.

Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
2023-09-25 15:03:25 +02:00
Mateusz Hoppe 371412a826 refactor: remove useLocalMem from allocate32BitGraphicsMemoryImpl
- this method allocates System Memory
- argument is not needed - ExternalHeap is selected inside this function
- remove unneeded ults
- allocate memory in Device Pool for external heap allocation in
OsAgnosticMemoryManager

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-09-18 10:08:09 +02:00
Baj, Tomasz e10f39017d fix: Add ImageInfo to createGraphicsAllocation on Linux
Related-To: NEO-6757

Signed-off-by: Baj, Tomasz <tomasz.baj@intel.com>
2023-09-14 12:58:59 +02:00
Slawomir Milczarek 5833d65bc0 feature: Add support for memadvise with system preferred location
Related-To: NEO-7851

Signed-off-by: Slawomir Milczarek <slawomir.milczarek@intel.com>
2023-08-31 13:09:55 +02:00
Mateusz Jablonski a4addad510 refactor: prework for defining preferred allocation method per allocation type
Related-To: HSD-18032442707, HSD-18032440987
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-08-30 12:17:33 +02:00
Mateusz Hoppe 00de13939d feature: bindless addressing allocator - reuse of bindless slots
- introduce 2 reuse pools to bindlessHeapHelper
- one pool stores slots for reuse, second pool stores released slots
- stateCacheDirty flags keep track of state cache - when pools are
switched - flags are set indicating flushing caches is needed after
old slots have been reused for new allocations

Related-To: NEO-7063

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-08-29 09:51:36 +02:00
Mateusz Hoppe 8435160db4 feature: bindless addressing for images
- program surface states for redescribed images correctly. Image copy
to/from memory are using redescribed surface states,
- refactor state base address programming - program address and size
together, set max size at the beginning due to lack of Enable flag
- set GpuBase in WddmAllocation when external heap is used
- return max ssh required size from kernelInfo or based on stateful args

Related-To: NEO-7063

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2023-08-18 15:59:20 +02:00
John Falkowski 2403212dcd fix: chunking prefetch add USER_FENCE
Add USER_FENCE before PREFETCH call and after the BIND

Related-To: NEO-8098

Signed-off by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
Signed-off-by: John Falkowski <john.falkowski@intel.com>
2023-08-17 21:32:47 +02:00
Compute-Runtime-Validation ce7b25d445 Revert "fix: use proper pointer inside appendWriteGlobalTimestamp"
This reverts commit 5a235af209.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2023-08-14 10:54:20 +02:00
Lukasz Jobczyk 3ab72e7d79 fix: Align svm cpu to alignment passed to properties
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2023-08-11 14:57:49 +02:00
Katarzyna Cencelewska 5a235af209 fix: use proper pointer inside appendWriteGlobalTimestamp
Related-To: NEO-8105
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
2023-08-11 08:18:45 +02:00
Compute-Runtime-Validation 820e94e89c Revert "fix: Align svm cpu to alignment passed to properties"
This reverts commit d66da494d4.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2023-08-11 07:14:30 +02:00
Lukasz Jobczyk d66da494d4 fix: Align svm cpu to alignment passed to properties
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2023-08-09 11:52:21 +02:00
Milczarek, Slawomir f3f623900a fix: memory prefetch for kmd migrated shared alloc to check for invalid usm ptr
Fix for SIGSEGV in SYCL_CTSO0Debug-usm_api_prefetch_handler_no_events_l0
./run_l0.sh usm_api_prefetch_handler_no_events test_usm

Related-To: LOCI-4742

Signed-off-by: Milczarek, Slawomir <slawomir.milczarek@intel.com>
2023-08-08 11:29:14 +02:00
Mateusz Jablonski 98fd9f5687 performance: change preferred allocation method on MTL devices
prefer allocating resources by KMD

Related-To: NEO-7194
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-07-31 10:45:43 +02:00
Mateusz Jablonski c68c41c8c4 refactor: prework for setting preferred allocation method per platform
Related-To: NEO-7194
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2023-07-28 19:54:35 +02:00
Milczarek, Slawomir 1195578d96 fix: KW issue with dereference in function call that may return null
Signed-off-by: Milczarek, Slawomir <slawomir.milczarek@intel.com>
2023-07-27 07:39:15 +02:00