Commit Graph

5807 Commits

Author SHA1 Message Date
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
Bartosz Dunajski
9e92324171 feature: copy offload support for fill operations
Related-To: NEO-7067

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-05-12 10:34:19 +02:00
Young Jin Yoon
4763d7268f fix: Reset additional properties usage after profiling
Related-To: NEO-13003

Signed-off-by: Aravind Gopalakrishnan <aravind.gopalakrishnan@intel.com>
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
2025-05-10 02:04:11 +02:00
Compute-Runtime-Validation
9dc734ce44 Revert "fix: set correct bindless offsets for L0 bindless images extension wi...
This reverts commit b048d0e557.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-05-09 09:06:13 +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
Young Jin Yoon
195ce3049e fix: avoid updating in-order dependency with useAdditionalBlitProperties
Modified appendMemoryCopyBlitRegion, appendCopyImageBlit,
appendMemoryCopyRegion, appendBlitFill and appendMemoryCopy
in CommmandListCoreFamily to not to update in order dependency
counter if useAdditionaBlitProperties used.

Related-To: NEO-13003

Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
2025-05-09 03:07:18 +02:00
Mateusz Jablonski
e0041df3cc refactor: remove not needed function
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-09 00:45:02 +02:00
Mateusz Jablonski
41ef309ccf performance: cache devices to expose via zeDeviceGet API call
vector with all device handles is initialized only once, during driver init

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-09 00:19:23 +02:00
Young Jin Yoon
415fe245d7 feature: apply or set additional blit properties
Modified appendMemoryCopyBlitRegion, appendCopyImageBlit,
appendMemoryCopyRegion, and appendBlitFill in
CommmandListCoreFamily to change behavior when
useAdditionaBlitProperties is used.
Modified dispatchBlitMemoryFill and
dispatchBlitCommandForImageRegion in BlitCommandHelper to
call applyAdditionalProperties when useAdditionalProperties
is enabled.

Related-To: NEO-13003

Signed-off-by: Aravind Gopalakrishnan <aravind.gopalakrishnan@intel.com>
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
2025-05-09 00:04:57 +02:00
Dominik Dabek
501aebfc45 fix: getMemAddressRange for pooled usm
Use pool methods for getting base ptr and size if passed ptr is in a usm
pool.

Related-To: NEO-6893

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2025-05-08 23:28:26 +02:00
Michal Mrozek
6f4a397cfc refactor: remove not required parameter
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2025-05-08 23:13:06 +02:00
Anvesh Bakwad
00c594831e fix(sysman): Fix Memory read and write counter values calculation for BMG
Related-To: NEO-14848

Signed-off-by: Anvesh Bakwad <anvesh.bakwad@intel.com>
2025-05-08 20:03:34 +02:00
Fabian Zwoliński
915e2ff386 fix: ensure texture cache flush after kernels writing to img on regular cmdlist
- refactored texture cache flush logic for regular command list
- replaced `needsTextureCacheFlushOnBarrier` flag with
`textureCacheFlushPending`
- introduced `consumeTextureCacheFlushPending()` for safer flag
management
- updated synchronization points (appendWaitOnInOrderDependency,
appendWaitOnEvents, appendEventForProfiling ...) to integrate
pending flushes into existing PIPE_CONTROLs using the new
consume method
- if a pending flush cannot be merged into an existing PIPE_CONTROL,
a new PIPE_CONTROL will be created before the next compute walker
- added and updated unit tests
- this patch covers both in-order and out-of-order
executions

Related-To: NEO-14713
Signed-off-by: Fabian Zwoliński <fabian.zwolinski@intel.com>
2025-05-08 19:44:57 +02:00
Naklicki, Mateusz
9008ca7b3f fix: add missing ZE_APIEXPORT and ZE_APICALL keywords
Signed-off-by: Naklicki, Mateusz <mateusz.naklicki@intel.com>
2025-05-08 18:36:17 +02:00
Naklicki, Mateusz
a0e7ef1ce1 fix: explicitly skip programming of unused scratch data ptr
If scratch data is unused, explicitly skip programming.

Related-To: NEO-14130

Signed-off-by: Naklicki, Mateusz <mateusz.naklicki@intel.com>
2025-05-08 16:56:42 +02:00
Kamil Kopryk
b048d0e557 fix: set correct bindless offsets for L0 bindless images extension with heapless
Additionally reorder members in bindless heaps helper.

Related-To: NEO-14710
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
2025-05-08 13:46:46 +02:00
Bartosz Dunajski
2782577f1d feature: allow enabling copy offload for regular cmd lists
Related-To: NEO-7067

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-05-08 12:37:55 +02:00
Compute-Runtime-Validation
7167f45199 Revert "refactor: Gate shared system mem caps with KMD cap"
This reverts commit 8e59ac7576.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-05-08 11:26:52 +02:00
Compute-Runtime-Validation
786bf8a9fa Revert "performance: enable device usm pooling for l0"
This reverts commit 0afc042205.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-05-08 11:20:29 +02:00
Mateusz Jablonski
adc40934ae test: ensure thread safety when capturing input params for wait function
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-08 10:22:14 +02:00
Mateusz Hoppe
f6796c7384 test: use pid as suffix in zello_printf output file
Related-To: NEO-14857

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2025-05-07 19:37:19 +02:00
Mateusz Jablonski
1b42ebf7fb feature: add L0 API to wait for completion of all submissions to given device
Related-To: NEO-14560
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-07 17:26:22 +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
Dominik Dabek
0afc042205 performance: enable device usm pooling for l0
Related-To: NEO-6893

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2025-05-07 15:08:36 +02:00
Anvesh Bakwad
35926a5a18 fix(sysman): Add GDDR6 memory type into Memory Properties
Related-To: NEO-14810

Signed-off-by: Anvesh Bakwad <anvesh.bakwad@intel.com>
2025-05-07 11:36:12 +02:00
Bartosz Dunajski
fdb84aaa3d feature: pass partition count to copy properties
Related-To: NEO-7067

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-05-07 10:54:28 +02:00
John Falkowski
8e59ac7576 refactor: Gate shared system mem caps with KMD cap
Enabled only by setting EnableSharedSystemUsmSupport=1 flag

Related-To: NEO-12988

Signed-off-by: John Falkowski <john.falkowski@intel.com>
2025-05-07 06:38:01 +02:00
Jakub Nowacki
38b178dd9e feature: add ze_image_query_allocation support
Related-To: NEO-9988

Signed-off-by: Jakub Nowacki <jakub.nowacki@intel.com>
2025-05-06 17:08:55 +02:00
Dominik Dabek
6afefa1e31 performance: usm device pool, compression handling
Use device compression defaults when allocating pool.
Ignore compression hints when allocating from pool.

Also remove unused late initialization methods.

Related-To: NEO-6893

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2025-05-06 16:30:58 +02:00
Bartosz Dunajski
201324f804 feature: wait path improvements for dual stream offload
Related-To: NEO-7067

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2025-05-06 14:50:05 +02:00
Mateusz Hoppe
d3b11d1527 fix: correctly patch implicit arg buffer in indirect data
- use correct size alignment of implicit arg buffer, crossthread data
should start after the buffer without extra padding

Related-To: NEO-14449

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2025-05-06 13:25:11 +02:00
Compute-Runtime-Validation
d477935ab9 Revert "refactor: Gate shared system mem caps with KMD cap"
This reverts commit f38fae3b18.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-05-06 12:43:16 +02:00
Mateusz Jablonski
5bab1c3f60 refactor: remove not needed NOLINT comments
Related-To: NEO-12901
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-06 09:11:28 +02:00
John Falkowski
f38fae3b18 refactor: Gate shared system mem caps with KMD cap
Related-To: NEO-12988

Signed-off-by: John Falkowski <john.falkowski@intel.com>
2025-05-06 07:28:59 +02:00
Compute-Runtime-Validation
26d3c7527c Revert "fix: correctly patch implicit arg buffer in indirect data"
This reverts commit e345d55fe5.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-05-06 04:38:50 +02:00
Mateusz Hoppe
e345d55fe5 fix: correctly patch implicit arg buffer in indirect data
- use correct size alignment of implicit arg buffer, crosshtread data
should start after the buffer without extra padding

Related-To: NEO-14449

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2025-05-05 16:03:57 +02:00
Pratik Bari
22ddaea09f fix(sysman): Correct gtId is fetched from the EngineInfo
Related-To: NEO-14626

Signed-off-by: Pratik Bari <pratik.bari@intel.com>
2025-05-05 15:50:50 +02:00
Mateusz Jablonski
07f13fc319 feature: add L0 API to translate between device handle and 32-bit identifier
- zerDeviceTranslateToIdentifier to translate from device handle to
32-bit unsigned integer identifier
- zerIdentifierTranslateToDeviceHandle to translate from identifier to
device handle associated to default driver handle

Related-To: NEO-14560
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-05 15:20:44 +02:00
Mateusz Jablonski
01160c05d0 fix: add missing ZE_APICALL keyword in extension function
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-05-05 13:03:29 +02:00
Jakub Nowacki
373d1f8202 feature: add ze_image_memory_properties support
Related-To: NEO-9988

Signed-off-by: Jakub Nowacki <jakub.nowacki@intel.com>
2025-05-05 12:48:48 +02:00
Aleksandra Nizio
62739986bf fix: print error in sim mode without available aubstream
Resolves: NEO-14693

Signed-off-by: Aleksandra Nizio <aleksandra.nizio@intel.com>
2025-05-05 11:48:35 +02:00
Young Jin Yoon
f844dbc88f feature: Add timestamp node to encode post sync args
Added a timestamp node to EncodePostSyncArgs to use
additional timestamp information if required,
Modified event object to use the timestamp added in
EncodePostSyncArgs.

Related-To: NEO-13003

Signed-off-by: Aravind Gopalakrishnan <aravind.gopalakrishnan@intel.com>
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
2025-05-02 20:55:15 +02:00
Compute-Runtime-Validation
edd230b6cb Revert "performance: Remove global fence allocation from BMG"
This reverts commit c9a21c158e.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-05-02 13:32:47 +02:00
Young Jin Yoon
48b2bb7672 refactor: add apply additional blit properties
Added BlitCommandsHelper::applyAdditionalBlitProperties
for various blitter commands

Related-To: NEO-13003

Signed-off-by: Aravind Gopalakrishnan <aravind.gopalakrishnan@intel.com>
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
2025-05-02 01:42:39 +02:00
Young Jin Yoon
55e7b6a209 feature: support copy function post-sync [3/n]
Refactored EncodePostSync to use more generic terms by renaming
variables.
Added some helper functions in EncodePostSync.

Related-To: NEO-13003
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
2025-05-01 19:20:01 +02:00
Raiyan Latif
ee4be92cc1 feature: Add TimelineWin32 External Semaphore support
Related-To: NEO-14397

Signed-off-by: Raiyan Latif <raiyan.latif@intel.com>
2025-05-01 18:51:58 +02:00
Igor Venevtsev
42efb3d204 feature: add built-in copy kernels for 1D_BUFFER images
Resolves: NEO-14782, HSD-18042093957

Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
2025-05-01 07:51:32 +02:00
Compute-Runtime-Validation
0c3b765942 Revert "refactor: add BlitSyncPropertiesExt to BlitSyncProperties"
This reverts commit b5a259aded.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2025-05-01 03:23:53 +02:00
Young Jin Yoon
b5a259aded refactor: add BlitSyncPropertiesExt to BlitSyncProperties
Added BlitSyncPropertiesExt to provide additional information for
different platform and/or blitter commands.

Signed-off-by: Aravind Gopalakrishnan <aravind.gopalakrishnan@intel.com>
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
2025-04-30 19:11:34 +02:00
Mateusz Jablonski
3438a14fc4 refactor: remove not used function
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2025-04-30 17:10:20 +02:00