intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-29 09:03:14 +08:00
Code Issues Packages Projects Releases Wiki Activity
20,917 Commits 42 Branches 655 Tags
f06bb256c7848f9a8553605fbf8a7d50e5d6ecff
Commit Graph

3655 Commits

Author SHA1 Message Date
Matias Cabral
e53949b8f2 feature: EUStall metrics calculation supports scopes 
Related-To: NEO-15142

Signed-off-by: Matias Cabral <matias.a.cabral@intel.com>
 2025-10-10 10:57:19 +02:00
Brandon Yates
7b27c3668e fix: calculate subSlicesPerSlice correctly on XeKmd 
Populate SliceInfo during topology creation.
Set L0 device props subslicesPerSlice to max
enabled subslice on any slice.

Related-to: NEO-16283

Signed-off-by: Brandon Yates <brandon.yates@intel.com>
 2025-10-10 10:44:23 +02:00
Radoslaw Jablonski
232654ca9b refactor: extend L0 Copy extensions validation 
Signed-off-by: Radoslaw Jablonski <radoslaw.jablonski@intel.com>
 2025-10-10 10:38:24 +02:00
Dominik Dabek
a924c77b5c fix: bindless image from offset ptr 
Handling pitched ptr offset from allocation base gpu address.

Related-To: NEO-16317, GSD-11804

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2025-10-09 14:32:06 +02:00
Bartosz Dunajski
0597f064e5 refactor: bcs split class structure 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-10-09 14:19:51 +02:00
Zbigniew Zdanowicz
e768366eac fix: reset all slm arguments of inactive kernel in mutable group 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-10-09 11:05:55 +02:00
Maciej Bielski
57add55b20 performance: R&R - return early from captureCommand if regular cmdlist 
Skip looping over wait-events when it is known that none of them can be
a potential fork point. E.g. at instantiation stage capturing has
already been disabled but the `captureCommand()` is entered anyway for
any zeAPI call supporting R&R.

Related-To: NEO-16017
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
 2025-10-09 10:52:20 +02:00
Bartosz Dunajski
e9e96fac1d fix: adjust offload detection for appendFill 
Related-To: NEO-16332

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-10-08 18:41:59 +02:00
Lukasz Jobczyk
6d6715bebf performance: Lock svm allocs mutex once in appendLaunchKernelWithArgs 
Resolves: NEO-15774

Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2025-10-08 13:35:52 +02:00
Neil R. Spruit
66a0f0d1bb fix: Corrected flag union reserve size and simplified sharedNT flag 
Related-To: NEO-16317

Signed-off-by: Neil R. Spruit <neil.r.spruit@intel.com>
 2025-10-08 13:35:21 +02:00
Shreyas Kunder
d0e323e410 fix: Add function pointer for Metric Scope api 
Related-To: NEO-16359

Signed-off-by: Shreyas Kunder <shreyas.kunder@intel.com>
 2025-10-08 12:11:02 +02:00
Naklicki, Mateusz
0c619299d0 refactor: cleanup header includes in level_zero 
Related-To: NEO-15383
Signed-off-by: Naklicki, Mateusz <mateusz.naklicki@intel.com>
 2025-10-07 19:58:31 +02:00
Bartosz Dunajski
8a706dcf7f fix: adjust copy offload selection logic 2 
Related-To: NEO-16250

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-10-07 15:08:12 +02:00
Narendra Bagria
90ec875dea feature: use heapless builtins for copyRegion 
Related-To: NEO-15323, NEO-7620

Signed-off-by: Narendra Bagria <narendra.bagria@intel.com>
 2025-10-07 14:52:52 +02:00
Bartosz Dunajski
40dd9221ca refactor: improve bcs split size selection 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-10-07 09:29:12 +02:00
Bartosz Dunajski
d1a2ec4590 fix: add missing lock in bcs split path 2 
Related-To: NEO-14557

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-10-07 08:00:18 +02:00
Compute-Runtime-Validation
fca882ceda Revert "fix: adjust copy offload selection logic" 
This reverts commit 6bee05e894.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2025-10-07 04:54:13 +02:00
Radoslaw Jablonski
54de14a9dc refactor: apply clang-format's InsertBraces rule 
Formats code before InsertBraces rule is enabled.

Signed-off-by: Radoslaw Jablonski <radoslaw.jablonski@intel.com>
 2025-10-06 15:32:46 +02:00
Bartosz Dunajski
6bee05e894 fix: adjust copy offload selection logic 
Related-To: NEO-16250

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-10-06 13:21:22 +02:00
Compute-Runtime-Validation
d3ac2f871b Revert "fix: add missing lock in bcs split path" 
This reverts commit 8ea8e78471.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2025-10-06 11:38:28 +02:00
Dominik Dabek
8575573dbd fix: l0, bindless image from pooled alloc 
Make sure pool offset is applied when growing pools are used.

Related-To: NEO-16317, GSD-11804

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2025-10-06 10:21:05 +02:00
Compute-Runtime-Validation
feff29fd47 Revert "refactor: l0 host usm pool to unique ptr" 
This reverts commit 0cf67302b2.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2025-10-05 06:27:30 +02:00
Young Jin Yoon
5eed1f5b72 fix: return unsupported when using ZE_IMAGE_TYPE_BUFFER 
return ZE_RESULT_ERROR_UNSUPPORTED_FEATURE when ZE_IMAGE_TYPE_BUFFER
is used.

Related-To: NEO-16034
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
 2025-10-04 01:51:45 +02:00
Young Jin Yoon
929ebc7976 fix: return allocation size for zeImageGetMemoryPropertiesExp 
Returned the memory allocation size instead of pixel size
for zeImageGetMemoryPropertiesExp

Related-To: NEO-16033
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
 2025-10-04 01:51:29 +02:00
Kamil Kopryk
f17ac35108 refactor: improve compilation time 
Limit bitset includes,
stl bitset is slow to compile

Related-To: NEO-15390

Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2025-10-03 21:21:25 +02:00
Bartosz Dunajski
8ea8e78471 fix: add missing lock in bcs split path 
Related-To: NEO-14557

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-10-03 21:06:28 +02:00
Katarzyna Cencelewska
0f866e24e6 fix: enable setting timestamp via submission on tbx mode 
Resolves: NEO-16293
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
 2025-10-03 08:43:18 +02:00
Dominik Dabek
0cf67302b2 refactor: l0 host usm pool to unique ptr 
Related-To: NEO-16084

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2025-10-02 15:36:25 +02:00
Mateusz Jablonski
710d70a787 refactor: remove not needed code 
Related-To: NEO-16154

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-10-02 11:41:08 +02:00
Dominik Dabek
7bb2d323d9 performance: enable l0 device usm growing pools 
Related-To: NEO-16084

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2025-10-01 17:06:25 +02:00
Mateusz Jablonski
165305090b refactor: remove always true variable - ftrSupportsInteger64BitAtomics 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-10-01 16:45:16 +02:00
Radoslaw Jablonski
be3427a9e8 fix: L0 Copy extensions validation 
Signed-off-by: Radoslaw Jablonski <radoslaw.jablonski@intel.com>
 2025-10-01 11:47:42 +02:00
Marcel Skierkowski
047f6851f7 refactor: correct programming of SLM size 
Unify reading hw avaialble slm size:
Pre-Xe2 SLMSizeInKb stores total SLM size across all DSS
Xe2+ SLMSizeInKb stores SLM size per SS

apply restrictions for preferred/programmable SLM size

Related-To: NEO-12949
Signed-off-by: Marcel Skierkowski <marcel.skierkowski@intel.com>
 2025-09-30 17:57:18 +02:00
Kamil Kopryk
80dec4fc1f feature: implement host functions api calls 
Related-To: NEO-14577
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2025-09-30 11:42:10 +02:00
Bartosz Dunajski
e88b4d7b4c fix: correct signaling of partitioned aggregated event 2 
Related-To: NEO-14557

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-09-29 16:21:38 +02:00
Maciej Bielski
a440a3e8ea refactor: use Clonable type for pImplicitArgs 
Clonable is a helper type which eases proper management of
dynamically-allocated resources. It allows to reduce the amount of
manual handling.

Related-To: NEO-16017
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
 2025-09-29 15:34:24 +02:00
Bartosz Dunajski
2f82a3a1c4 fix: dont optimize implicit semaphore in copy offload mode 
Related-To: HSD-18043488005

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-09-29 12:30:57 +02:00
Compute-Runtime-Validation
0ef032bb28 Revert "fix: correct signaling of partitioned aggregated event" 
This reverts commit 636498c079.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2025-09-27 16:01:57 +02:00
Bartosz Dunajski
636498c079 fix: correct signaling of partitioned aggregated event 
Related-To: NEO-14557

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-09-25 16:38:53 +02:00
Maciej Plewka
6c5ac160a6 fix: use memory ptr from api when copy from/to image by blt 
Related-To: NEO-15934

Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2025-09-25 15:05:01 +02:00
Bartosz Dunajski
72a5bcbfa1 feature: extend copy offload hint support 
Related-To: NEO-14557

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-09-25 12:56:40 +02:00
Bartosz Dunajski
3717187cbc feature: zexDeviceGetAggregatedCopyOffloadIncrementValue implementation 
Related-To: NEO-14557

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-09-25 11:46:02 +02:00
Bellekallu Rajkiran
9cccbcabe1 fix: Reset kernelWithAssertAppended flag 
On new append calls, reset flag if previous submissions
are completed.

Related-To: NEO-16184

Signed-off-by: Bellekallu Rajkiran <bellekallu.rajkiran@intel.com>
 2025-09-24 18:30:51 +02:00
Kamil Kopryk
16050126e3 feature: add host function dispatch 
- program in place for immediate cmdlist
- add to patch commands for regular cmdlist
- initialize host data allocation for the 1st use in csr

Related-To: NEO-14577
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2025-09-24 17:20:40 +02:00
Dominik Dabek
f54e3fda41 feature: adapt usm pool manager 
Change behavior to continue allocating usm pools as needed.
Intended to replace singular usm pools.

Related-To: NEO-16084

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2025-09-24 15:16:10 +02:00
Bellekallu Rajkiran
91d28f17cc fix: Add RAII wrapper for insert sw tag 
Add wrapper to not miss capturing end tags for early returns.

Signed-off-by: Bellekallu Rajkiran <bellekallu.rajkiran@intel.com>
 2025-09-24 13:50:49 +02:00
Damian Tomczak
68755178a0 feature: isForceStatelessRefactor 
Related-to: NEO-6075

Signed-off-by: Damian Tomczak <damian.tomczak@intel.com>
 2025-09-24 12:10:53 +02:00
Bartosz Dunajski
94d01b4d40 feature: use User event in bcs split path if increment value is the same 
Related-To: NEO-14557

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-09-24 09:55:13 +02:00
Fabian Zwoliński
a1c5fa1a13 feature: add pooling of USM global/constant surface 
Related-To: NEO-12287
Signed-off-by: Fabian Zwoliński <fabian.zwolinski@intel.com>
 2025-09-23 18:24:18 +02:00
shubham kumar
0b6b0e3954 Revert "feature: Use zesInit on windows always" 
This reverts commit 9dadadfa73.

Signed-off-by: shubham kumar <shubham.kumar@intel.com>
 2025-09-23 17:18:12 +02:00