Commit Graph

855 Commits

Author SHA1 Message Date
John Falkowski
764fd33a46 refactor: clSetKernelArgSVMPtr ptr error if debug var set
Related-To: NEO-12988

Signed-off-by: John Falkowski <john.falkowski@intel.com>
2025-06-18 20:55:48 +02:00
Compute-Runtime-Validation
763fe6ff86 Revert "feature: Enable defer backing by default"
This reverts commit bf8de245d9.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-06-17 06:31:27 +02:00
Bellekallu Rajkiran
bf8de245d9 feature: Enable defer backing by default
Related-To: NEO-13403

Signed-off-by: Bellekallu Rajkiran <bellekallu.rajkiran@intel.com>
2025-06-16 10:39:08 +02:00
Bartosz Dunajski
9fc244c83a refactor: remove not needed imm cmd list flush task code
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-06-16 08:59:51 +02:00
Oskar Hubert Weber
db020bdd60 feature: add warning macro for readability
Related-To: NEO-14511

Signed-off-by: Oskar Hubert Weber <oskar.hubert.weber@intel.com>
2025-06-13 09:28:18 +02:00
Chodor, Jaroslaw
2c10dc2f53 feature: Adding debug key for custom igc lib name
This commit adds new debug key which can be used
to override loaded igc libary name.

E.g. IgcLibraryName=my_custom_igc.so

Related-To: NEO-14579

Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
2025-06-11 22:07:25 +02:00
John Falkowski
805a716fe3 feature: Implement appendMemoryCopy/Fill for Shared System USM
Related-To: NEO-13697

Signed-off-by: John Falkowski <john.falkowski@intel.com>
2025-06-09 16:18:56 +02:00
Bartosz Dunajski
2399492b74 refactor: debug flag to override max mem alloc size
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-06-04 15:16:34 +02:00
Szymon Morek
48b4a33645 Revert "performance: enable low latency hint on Xe"
This reverts commit b74dc9b7e1.

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
2025-06-02 12:18:43 +02:00
Filip Hazubski
04965470fb fix: Remove redundant debug key
OverrideNumThreadsPerEu debug variable can be used
instead of removed Enable10ThreadsPerEu.

Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
2025-05-29 16:58:04 +02:00
Kamil Kopryk
b7681a2e7b refactor: redirect flush L3 host usm to external by default
Related-To: NEO-13163

Rename ForceL3FlushAfterPostSync to EnableL3FlushAfterPostSync
Rename DisableFlushL3ForHostUsm to RedirectFlushL3HostUsmToExternal

Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
2025-05-28 08:50:01 +02:00
Filip Hazubski
3dc32cec88 feature: Add debug key to override number of threads per EU
Related-To: HSD-22021012755, NEO-14749

Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
2025-05-23 16:20:40 +02:00
Mateusz Jablonski
98bf872fdd fix: return error when cannot obtain debug surface size
remove default max debug surface size
check state save area size only for debug scenarios
reduce state save area size in unit tests - rely on values from mock

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-23 14:48:08 +02:00
Mateusz Jablonski
8353f68307 feature(internal): add debug key to override max debug surface size
reduce the value in unit tests

Related-To: NEO-14097
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-21 14:57:41 +02:00
Bartosz Dunajski
91940d2a79 fix: initial support for single temporary allocations list
Related-To: NEO-14641

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-05-21 13:16:15 +02:00
Szymon Morek
3ba25459e8 performance: set timestamps as non-coherent since xe2
Related-To: NEO-14360

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
2025-05-21 07:49:56 +02:00
Maciej Bielski
0f8ee57f98 feature: add NEO_LOCAL_MEMORY_ALLOCATION_MODE
Allow the application to force storageInfo.localOnly and get the
out-of-memory returned if not possible.

This is a windows-only feature supported on discrete platforms.

Related-To: NEO-13428
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
2025-05-20 14:15:17 +02:00
Bartosz Dunajski
7f7aa36c52 feature: allow enabling copy offload on default cmd list
Related-To: NEO-7067

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-05-19 13:28:46 +02:00
Szymon Morek
b74dc9b7e1 performance: enable low latency hint on Xe
Related-To: NEO-14708

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
2025-05-14 15:47:21 +02:00
Young Jin Yoon
5221b5b00e fix: patch counter values with additional blit properties
Modified BlitCommandsHelper and CommandListHw to patch counter values
when using additional blit properties and in order command list is
enabled.

Related-To: NEO-13003

Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
2025-05-13 04:32:31 +02:00
Lukasz Jobczyk
df2c776aab refactor: Remove unused cmdq round robin engine assign
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2025-05-09 11:03:57 +02:00
Lukasz Jobczyk
e964ccdf57 refactor: remove unused ulls diagnostic mode
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2025-05-09 10:01:51 +02:00
Lukasz Jobczyk
1d1414febc refactor: remove unused dc flush mitigation
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2025-05-09 08:26:26 +02:00
Lukasz Jobczyk
d590172b65 refactor: Remove unused events tracker
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2025-05-08 23:53:47 +02:00
Michal Mrozek
52daf63a6f refactor: remove not needed debug variable
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2025-05-08 14:29:27 +02:00
Filip Hazubski
ba85f7417d fix: Add debug toggle to disable flush L3 for host usm
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
2025-05-07 16:57:31 +02:00
Lukasz Jobczyk
ac7bd757a6 refactor: deallocate via gmm
Related-To: NEO-13051

Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2025-05-06 18:12:46 +02:00
Bartosz Dunajski
f3ad230c45 feature: initial support for different copy offload submission modes
Related-To: NEO-7067

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-04-30 09:18:02 +02:00
Bartosz Dunajski
c2266fc69e refactor: remove not used CB event TS logic
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-04-28 14:31:34 +02:00
Michal Mrozek
c41ac0ad4c refactor: remove not needed code
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2025-04-25 16:44:01 +02:00
Mateusz Hoppe
41efee1e7c feature: add mode to expose single root device
- simplify API to expose and use only single root device. Disallow
returning subdevices. This is experimental mode.

Related-To: NEO-14559

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2025-04-25 13:20:56 +02:00
Compute-Runtime-Validation
30f6ddb6c7 Revert "refactor: deallocate via gmm"
This reverts commit bf861d8d49.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-04-23 03:50:43 +02:00
Lukasz Jobczyk
bf861d8d49 refactor: deallocate via gmm
Related-To: NEO-13051

Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2025-04-22 16:46:36 +02:00
Aravind Gopalakrishnan
b4f95a1a26 feature: Add pidfd option for IPC handle import
Related-To: NEO-10380

Signed-off-by: Aravind Gopalakrishnan <aravind.gopalakrishnan@intel.com>
2025-04-18 13:01:27 +02:00
Szymon Morek
3596522637 refactor: remove unused logic in ULLS controller
Related-To: NEO-13843

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
2025-04-17 18:35:20 +02:00
Chodor, Jaroslaw
eb58e1f964 feature: Allow NEOReadDebugKeys and PrintDebugMessages envs in ocloc
This enables NEO's NEOReadDebugKeys and PrintDebugMessages in ocloc
in the same scenarios as in NEO OCL/L0 RT

Note :
* NEOReadDebugKeys=1 should not be prefixed with NEO_OCLOC_
* PrintDebugMessages=1 should be prefixed with NEO_OCLOC_
(to distinguish from OCL/L0 RT usage), so the full env
is NEO_OCLOC_PrintDebugMessages=1

Related-To: NEO-14473
Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
2025-04-16 18:17:15 +02:00
Chodor, Jaroslaw
b423084861 refactor: Change FinalizerInputType debug var to string
Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
2025-04-16 10:33:19 +02:00
Chodor, Jaroslaw
2ce694c52b feature: Enabling support for redirecting FCL to IGC for OCLOC
This allows for invoking IGC as OCL C translator in OCLOC paths

Accepted debug env :
NEO_OCLOC_UseIgcAsFcl=1 ; forces IGC
NEO_OCLOC_UseIgcAsFcl=2 ; forces FCL
NEO_OCLOC_UseIgcAsFcl=0 ; use defaults

Related-To: NEO-14473
Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
2025-04-15 16:44:35 +02:00
Chodor, Jaroslaw
7bded13638 feature: Enabling support for redirecting FCL to IGC for OCL
This allows for invoking IGC as OCL C translator in OpenCL paths

Accepted debug env :
UseIgcAsFcl=1 ; forces IGC
UseIgcAsFcl=2 ; forces FCL
UseIgcAsFcl=0 ; uses defaults

Related-To: NEO-14473
Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
2025-04-14 18:41:48 +02:00
Jaroslaw Warchulski
3e1aa33924 refactor: cleanup headers
Related-To: NEO-5548
Signed-off-by: Jaroslaw Warchulski <jaroslaw.warchulski@intel.com>
2025-04-14 14:59:40 +02:00
Chodor, Jaroslaw
a079a2654d feature: Enabling support for DisableDeepBind debug env in ocloc
Accepted debug env :
NEO_OCLOC_DisableDeepBind=1

Related-To: NEO-14473
Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
2025-04-12 23:21:19 +02:00
Chodor, Jaroslaw
72ae27d51e feature: Enabling debug variables reader in ocloc
This allows for using debug variables in ocloc with
a few by-design limitations :
* debug var needs to be prefixed with NEO_OCLOC_
(when setting-up at runtime as environment variable
or using .config file)
* debug var needs to be tagged with S_OCLOC scope
(at compile time, in debug_variable .inl files)

Related-To: NEO-14473
Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
2025-04-11 19:58:53 +02:00
Chodor, Jaroslaw
9e14eaaed1 feature: Adding support for scoped debug variables
This allows for limitting debug variables to specific
components.

Related-To: NEO-14473
Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
2025-04-11 17:52:17 +02:00
Fabian Zwoliński
7ef3880793 feature: implement pool allocator for gpuTimestampDeviceBuffer
The patch applies to Level Zero.
Only allocations < 2MB will be fetched from the pool.
Allocations are shared and reused within a given device.

Additionally, I added a new debug flag to control the allocator:
EnableTimestampPoolAllocator

Related-To: NEO-12287
Signed-off-by: Fabian Zwoliński <fabian.zwolinski@intel.com>
2025-04-02 14:28:56 +02:00
Dominik Dabek
915d657420 fix: flag to limit usm reuse based on memory usage
Host usm and device usm for igfx checks system memory usage.
Device usm for dgfx checks local memory usage.

If used memory is above limit threshold:
- no new allocations will be saved for reuse
- cleaner will use shorter hold time of 2 seconds
- cleaner will free all eligible allocations, regardless of async
deleter thread having work

Motivation: in case of gfx memory being full, making resident new
allocations will require evictions which leads to massive slowdown on
enqueue calls.
This change aims to minimize cases where extra memory usage from usm
reuse mechanism leads to above situation.

Related-To: NEO-6893, NEO-14160

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2025-03-27 10:25:19 +01:00
Lukasz Jobczyk
8a85a96ed2 feature: Add 3-level wait scheme with tpause intrinsic
Related-To: NEO-14336

Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2025-03-21 12:12:57 +01:00
Michal Mrozek
6a9716b81c refactor: remove not needed code
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2025-03-19 14:53:57 +01:00
Jitendra Sharma
dda7876d3a feature: Introduce SipClassType::externalLib
Related-To: NEO-13737
Signed-off-by: Jitendra Sharma <jitendra.sharma@intel.com>
2025-03-19 11:06:03 +01:00
Jack Myers
0e25970853 fix: re-add switch case for once writable query
A change related to the tbx fault manager
incorrectly removed a switch case from
`AubHelper::isOneTimeAubWritableAllocationType`.

This fixes that and refactors some APIs to prevent
similar mistakes from happening again by cleaning
up logic.

Addresses show stopper for pre-si pytorch workflows.

Resolves: NEO-14399
Signed-off-by: Jack Myers <jack.myers@intel.com>
2025-03-19 09:54:54 +01:00
Maciej Bielski
a7c46b8213 feature: support L2 cache reservation
Related-To: NEO-12837
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
2025-03-17 19:41:55 +01:00