intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-03 14:55:24 +08:00
Code Issues Packages Projects Releases Wiki Activity
20,412 Commits 43 Branches 655 Tags
9527cae63df36ecb7dce08a1e75ceb44a7b9e457
Commit Graph

3522 Commits

Author SHA1 Message Date
Zbigniew Zdanowicz
7bcce38ae4 performance: split mutation data into event and kernel for better locality 
Related-To: NEO-13916

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-24 11:51:13 +02:00
Mateusz Hoppe
575fe7fe33 fix: remove local dispatch support 
Related-To: NEO-14561

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2025-07-24 10:31:18 +02:00
Bartosz Dunajski
e3e1ae0def refactor: helper function to set bcs split transfer direction 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-07-24 10:31:03 +02:00
Matias Cabral
93195e3501 feature: support for Metric Scopes 
Resolves: NEO-15136

Signed-off-by: Matias Cabral <matias.a.cabral@intel.com>
 2025-07-23 21:42:48 +02:00
Szymon Morek
6526a7fb66 Revert "performance: enable staging copy in L0" 
This reverts commit 57e4cc025f.

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2025-07-23 17:32:05 +02:00
Mateusz Jablonski
df7e114d54 fix: correct global device synchronization 
skip uninitialized command stream receivers

Additionally, add L0 black box for kernel with arg slm

Test verifies new simplified L0 API

Functions:
- zerDriverGetDefaultContext
- zeDeviceSynchronize
- zeCommandListAppendLaunchKernelWithArguments
- zerIdentifierTranslateToDeviceHandle
- zerDeviceTranslateToIdentifier
- zerDriverGetLastErrorDescription

Definitions:
- defaultCommandQueueDesc
- defaultHostMemDesc
- defaultDeviceMemDesc

Related-To: NEO-14560
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-23 12:39:29 +02:00
Compute-Runtime-Validation
357fd3108e Revert "refactor: move surface state programming method to base class" 
This reverts commit 5893fb08fb.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2025-07-23 04:37:46 +02:00
Compute-Runtime-Validation
321f285fd9 Revert "fix: correct global device synchronization" 
This reverts commit 8bdc479fe7.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2025-07-22 21:34:43 +02:00
Zbigniew Zdanowicz
5893fb08fb refactor: move surface state programming method to base class 
Related-To: NEO-15374

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-22 19:09:03 +02:00
Oskar Hubert Weber
9055ae8e54 fix: skip teardown clean up when terminating process on Windows 
Per https://learn.microsoft.com/en-us/windows/win32/dlls/dllmain,
it's not safe to clean up resources in DllMain when
fdwReason == DLL_PROCESS_DETACH and lpvReserved != NULL (the process is
terminating), e.g. can result in a hang in some cases.

Related-To: NEO-14121

Signed-off-by: Oskar Hubert Weber <oskar.hubert.weber@intel.com>
 2025-07-22 15:13:11 +02:00
Mateusz Hoppe
b17ed79618 fix: update implicitArgs versions 
- fix layout of implicit args
- add enqueued local size and sync buffer ptr

Related-To: NEO-15160

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2025-07-22 15:12:55 +02:00
Mateusz Jablonski
09ee9bf093 feature: expose DDI Handles extension by default 
Related-To: NEO-13917, NEO-13121
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-22 13:13:57 +02:00
Bartosz Dunajski
1f734d6f6a refactor: add helper to set default split mode 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2025-07-22 11:51:48 +02:00
Tomasz Biernacik
2da77209dc refactor: encapsulate condition 
Related-To: NEO-14844

Signed-off-by: Tomasz Biernacik <tomasz.biernacik@intel.com>
 2025-07-22 10:43:39 +02:00
Szymon Morek
57e4cc025f performance: enable staging copy in L0 
Related-To: NEO-14026

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2025-07-22 07:47:38 +02:00
Mateusz Jablonski
8bdc479fe7 fix: correct global device synchronization 
skip uninitialized command stream receivers

Additionally, add L0 black box for kernel with arg slm

Test verifies new simplified L0 API

Functions:
- zerDriverGetDefaultContext
- zeDeviceSynchronize
- zeCommandListAppendLaunchKernelWithArguments
- zerIdentifierTranslateToDeviceHandle
- zerDeviceTranslateToIdentifier
- zerDriverGetLastErrorDescription

Definitions:
- defaultCommandQueueDesc
- defaultHostMemDesc
- defaultDeviceMemDesc

Related-To: NEO-14560
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-21 19:51:49 +02:00
Zbigniew Zdanowicz
86528a10a3 performance: change mutable structures for better browse and access patterns 
Related-To: NEO-13916

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-21 17:21:02 +02:00
Michal Mrozek
773248dc45 performance: remove redundant vector in mcl 
Related-To: NEO-15498

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
 2025-07-21 14:43:03 +02:00
Michal Mrozek
aa8ab38156 performance: remove redundant vector 
Related-To: NEO-15498
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
 2025-07-21 12:54:22 +02:00
Szymon Morek
86b5660c23 performance: introduce staging copy in L0 
Related-To: NEO-14026

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2025-07-21 08:50:43 +02:00
Zbigniew Zdanowicz
259883c9c7 performance: do not capture and mutate stateful arguments in default mode 
Related-To: NEO-13916

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-18 13:25:47 +02:00
Zbigniew Zdanowicz
0af70b2e18 performance: do not copy host view per-thread data at kernel isa mutation 
- per-thread data will be recreated at mandatory group size mutation

Related-To: NEO-13916

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-18 12:48:46 +02:00
Zbigniew Zdanowicz
95dedbbcd8 performance: limit copying compute walker from host view at kernel mutation 
- at kernel mutation copy post sync and inline data
- at mandatory group count/size mutation copy dispatch data

Related-To: NEO-13916

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-18 12:42:04 +02:00
Eric Mortensen
69f2cda2fa fix: add error checks & return to unMapVirtualMem 
Related-To: NEO-8566

Signed-off-by: Eric Mortensen <eric.mortensen@intel.com>
 2025-07-17 23:53:19 +02:00
Maciej Bielski
35ecde70df refactor: extract parts of KernelImp data for easier capturing 
`KernelImp` is virtual class so capturing the current kernel state
cannot be done by a simple copy-instance of the class. However, this can
be done by extracting the relevant data to a separate class and
capturing a copy of its instance.

Related-To: NEO-15374
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
 2025-07-17 23:00:22 +02:00
Neil R. Spruit
2f6e4d36da fix: Correct all L0 handle definitions to properly handle the DDI ext 
Related-To: NEO-15570

- Fixes for missing handle definitions for DDi Ext compliance and
incorrect definitions for several handle definitions requiring handle
translation.

Signed-off-by: Neil R. Spruit <neil.r.spruit@intel.com>
 2025-07-17 19:13:51 +02:00
Grochowski, Stanislaw
e652679b95 fix: correct pollForCompletion in aub on program termination 
Related-To: NEO-14867

Signed-off-by: Grochowski, Stanislaw <stanislaw.grochowski@intel.com>
 2025-07-17 12:59:00 +02:00
Compute-Runtime-Validation
4c929fcc84 Revert "feature: support for Metric Scopes" 
This reverts commit e92ac1c8ab.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2025-07-17 04:29:44 +02:00
Chodor, Jaroslaw
fcc2038def refactor: modernizing NEO::Range->std::span 
Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
 2025-07-17 01:27:11 +02:00
Matias Cabral
e92ac1c8ab feature: support for Metric Scopes 
Resolves: NEO-14753

Signed-off-by: Matias Cabral <matias.a.cabral@intel.com>
 2025-07-16 19:30:14 +02:00
Chodor, Jaroslaw
7acb9585af feature: Adding fork/join support 
Related-To: NEO-15373

Signed-off-by: Chodor, Jaroslaw <jaroslaw.chodor@intel.com>
 2025-07-16 17:28:26 +02:00
Mateusz Jablonski
4dc4c45bbb fix: correct signature of zeCommandListAppendLaunchKernelWithArguments 
add const to pArguments and pNext types

Related-To: NEO-14560
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-16 16:25:12 +02:00
Mateusz Jablonski
10dc8a52a8 fix: correct shared cross device alloc caps for sub devices 
in case of single exposed device, if there are subdevice, then
subdevices may access between each other

Related-To: HSD-18042639532

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-16 15:47:25 +02:00
Compute-Runtime-Validation
a6149fca1a Revert "fix: Add -emit-lib-compile-errors flag when library compilation is en... 
This reverts commit c9ebf91271.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2025-07-16 14:36:11 +02:00
Zbigniew Zdanowicz
ef5efeac0c refactor: add kernel properties and move implementations to imp class 
Related-To: NEO-15374

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-15 15:08:44 +02:00
Shreyas Kunder
e4158208ae feature: Update Metrics Driver Code to use runtime enable & Disable APIs 
Related-To: NEO-15339

Signed-off-by: Shreyas Kunder <shreyas.kunder@intel.com>
 2025-07-15 14:28:49 +02:00
Tomasz Biernacik
2c5cbec033 feature: control post sync completion check 
Related-To: NEO-14844

Signed-off-by: Tomasz Biernacik <tomasz.biernacik@intel.com>
 2025-07-15 13:51:19 +02:00
Zbigniew Zdanowicz
087d1ecea4 refactor: use encoders instead direct dispatch commands in mcl experimental 
Related-To: NEO-10492

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-15 12:52:13 +02:00
Radoslaw Jablonski
be6fe9294c fix: stop Direct Submission on device teardown 
This prevents a race condition, causing page faults, where device related
resources are released, but are still in use by commands (e.g. cache flushes)
submitted by the Direct Submission Controller.

Related-To: NEO-15017

Signed-off-by: Radoslaw Jablonski <radoslaw.jablonski@intel.com>
 2025-07-15 12:12:59 +02:00
Zbigniew Zdanowicz
b62813297d refactor: move add kernel arg stateful buffer to experimental implementation 
Related-To: NEO-10492

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-15 01:02:26 +02:00
Mateusz Jablonski
8dc24d9e2c feature: add default command queue descriptor to ze_intel_gpu.h 
Related-To: NEO-14560
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-14 16:24:18 +02:00
Zbigniew Zdanowicz
1aefbd709b performance: reserve space for internal vectors 
- reserved space avoids vector memory relocations

Related-To: NEO-13916

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-14 13:13:55 +02:00
Zbigniew Zdanowicz
d31366794a fix: skip and continue kernel isa mutation when same kernel is provided 
Related-To: NEO-10492

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-11 20:28:51 +02:00
Mateusz Jablonski
762b04cf77 feature: add command queue flag to pass copy offload hint 
Related-To: NEO-14560
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-11 19:42:48 +02:00
Mateusz Jablonski
645de5add8 fix: correct shared cross device alloc capabilities 
allow for access in case of multiple exposed devices when
there is only one root device

Related-To: HSD-18042639532
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-11 18:32:06 +02:00
Zbigniew Zdanowicz
1599e5bafe feature: add missing experimental mcl api calls 
Related-To: NEO-10492

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2025-07-11 18:23:37 +02:00
Mateusz Jablonski
8f1903c7dd feature: add zerDriverGetLastErrorDescription API 
this function retrieves last error from default driver

Related-To: NEO-14560
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-07-11 13:37:50 +02:00
Aravind Gopalakrishnan
c1c8eb59c0 refactor: Cache context settings during init 
- cache context related settings once to reuse in subsequent calls

Signed-off-by: Aravind Gopalakrishnan <aravind.gopalakrishnan@intel.com>
 2025-07-10 12:06:20 +02:00
Aleksandra Nizio
c9ebf91271 fix: Add -emit-lib-compile-errors flag when library compilation is enabled 
Related-To: NEO-8608
Signed-off-by: Aleksandra Nizio <aleksandra.nizio@intel.com>
 2025-07-10 10:48:21 +02:00
Mateusz Hoppe
703497b067 fix: read ONEAPI_PVC_SEND_WAR_WA env 
- disable optimization with compiler internal option when env is set to
zero

Related-To: NEO-15378, GSD-10884

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2025-07-09 16:22:18 +02:00