intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-22 01:48:50 +08:00
Code Issues Packages Projects Releases Wiki Activity
8,042 Commits 42 Branches 655 Tags
aa5e1780a2c7123d0fda630aeedea0c897490c38
Commit Graph

210 Commits

Author SHA1 Message Date
Lukasz Jobczyk
49cf6f79f5 Validate host ptr on VM it will be used 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2021-08-16 08:15:53 +02:00
Bartosz Dunajski
fb996a3d5b Add EngineDescriptor helper 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-08-12 15:19:46 +02:00
Dominik Dabek
2a7757651a Change percent of global memory returned 
Related-To: NEO-5796

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2021-08-11 13:33:16 +02:00
Kamil Diedrich
2d5d071e66 Extended import device memory functionality 
Signed-off-by: Kamil Diedrich <kamil.diedrich@intel.com>
 2021-08-10 13:24:56 +02:00
Andrzej Koska
94c97fc14c Allowing the creation of overlapping buffers 
Related-To: NEO-5871

Signed-off-by: Andrzej Koska <andrzej.koska@intel.com>
 2021-08-10 10:59:29 +02:00
Bartosz Dunajski
ab73b61b77 Fix multitile ISA placement 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-08-02 15:16:13 +02:00
Milczarek, Slawomir
eb14d8458b Add helper function to enable stateless compression 
Related-To: NEO-5107

Signed-off-by: Milczarek, Slawomir <slawomir.milczarek@intel.com>
 2021-08-02 14:34:33 +02:00
Lukasz Jobczyk
d99f20d400 Move semaphore buffer to system memory 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2021-08-02 14:24:12 +02:00
Jaime Arteaga
b057935c7c Associate shared-allocations with device in multi-device systems 
When a shared-allocation is created in a multi-device system and
no device is passed during the call, then the allocation needs to be
associated with all the devices. For now, that means we default to
creating the shared-allocation in the host, with no device-backing
memory, so all devices can access it.

However, if a device is passed, then the shared-allocation is
expected to be accessed only by that device, and there is no need
to allocate it solely in the host.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-07-29 20:28:27 +02:00
Mateusz Jablonski
575445dbb5 Minor code cleanup 
prevent nullptr dereference

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-07-22 19:25:23 +02:00
Lukasz Jobczyk
f22251f68a Add remaining external allocation types for local bank selector 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2021-07-22 16:57:37 +02:00
Lukasz Jobczyk
890eec6105 Separate local memory usage tracking for internal and external 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2021-07-22 07:52:43 +02:00
Mateusz Jablonski
6f3c89decb Add mechanism to migrate multi root device memory 
invalidate TLB cache if kernel requires migration


Related-To: NEO-3691

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-07-19 19:49:22 +02:00
Bartosz Dunajski
e939e97384 Dont replicate GPU_TIMESTAMP_DEVICE_BUFFER 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-07-16 15:23:50 +02:00
Bartosz Dunajski
3c88492229 Revert "Extended import device memory" 
This reverts commit ea6555e788c98314160a11898212c2d664999705.

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-07-16 09:56:52 +02:00
Jaroslaw Chodor
fad4ee7e24 Enabling WDDM_LINUX by default 
Based on libdxg headers, we're able
to support WDDM_LINUX (WSL2) in our
default build configuration

Signed-off-by: Jaroslaw Chodor <jaroslaw.chodor@intel.com>
 2021-07-14 12:42:29 +02:00
Kamil Diedrich
d5fdb949eb Extended import device memory 
Signed-off-by: Kamil Diedrich <kamil.diedrich@intel.com>
 2021-07-07 16:12:36 +02:00
Bartosz Dunajski
9166f9223a Device pool allocation in OsAgnosticMemoryManager 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-07-07 13:09:12 +02:00
Lukasz Jobczyk
449c0dc856 Do not register allocation on fail 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2021-07-05 19:23:58 +02:00
Bartosz Dunajski
c7a936d1f4 Add memory banks to Simulated CSR 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-07-05 12:19:58 +02:00
Lukasz Jobczyk
850b38781d Move RING_BUFFER to system memory 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2021-06-29 16:24:43 +02:00
Lukasz Jobczyk
73b0df3211 Disable gem close worker in CSR constructor 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2021-06-29 15:22:51 +02:00
Bartosz Dunajski
96d14967ac Partial support for XE_HP_SDV 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-06-24 18:35:54 +02:00
Zbigniew Zdanowicz
ea214dd26b Select lowest subdevice when USM memory is allocated in not multicontext 
Related-To: NEO-5498

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-06-21 18:45:11 +02:00
Mateusz Hoppe
48feca4f44 Check Isa placement for every root device 
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2021-06-21 11:53:19 +02:00
Jaime Arteaga
5e29dccddc Add IPC events support 
Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-06-08 08:11:15 +02:00
Bartosz Dunajski
8667f3eafa Remove incorrect check for Shared USM creation 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-06-07 14:48:33 +02:00
Maciej Dziuban
13b2c326fe Add AlignmentSelector helper 
Related-To: NEO-5750
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
 2021-06-01 11:44:04 +02:00
Jaroslaw Chodor
9dbffa8d93 Refactoring createMemoryManager 
Signed-off-by: Jaroslaw Chodor <jaroslaw.chodor@intel.com>
 2021-05-25 16:58:21 +02:00
Mateusz Hoppe
0f32231fc8 Use correct INTERNAL heap base address for ISA in system memory 
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2021-05-25 10:22:28 +02:00
Maciej Dziuban
33e8f73775 Enable custom allocation alignments on Linux 
Related-To: NEO-5750
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
 2021-05-24 12:56:51 +02:00
Igor Venevtsev
21690dcea5 Fix debugSurface initialization 
Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
 2021-05-24 12:28:23 +02:00
Mateusz Jablonski
ed31aaedb8 Make BUFFER allocation lockable on DG1 
Related-To: NEO-5733
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-05-21 17:05:51 +02:00
Andrzej Koska
c59f44ecdf Collision avoidance on 36 bit address space 
Signed-off-by: Andrzej Koska <andrzej.koska@intel.com>

Related-To: NEO-5356
 2021-05-21 11:20:35 +02:00
Mateusz Jablonski
b50a6bec82 Correct isBlitCopyRequiredForLocalMemory 
detect not lockable allocation based on gmm flag
Related-To: NEO-5733

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-05-20 13:17:49 +02:00
Jaroslaw Chodor
abf0649a5b Improving cpp compliance 
Signed-off-by: Jaroslaw Chodor <jaroslaw.chodor@intel.com>
 2021-05-20 00:10:06 +02:00
Dominik Dabek
a5d38170ad Move some shared tests to shared 
Tests that are not specific to openCL or level zero

Related-To: NEO-5161

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2021-05-18 13:51:18 +02:00
lgotszal
3bd4bca911 Copyright header update 
Dates corrected in copyright headers to reflect original publication date
(2018 for OpenCL, 2020 for Level Zero).

Signed-off-by: lgotszal <lukasz.gotszald@intel.com>
 2021-05-17 20:38:19 +02:00
Mateusz Jablonski
a304e609fe Respect forceSystemMemory flag at the end of getAllocationData method 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-05-14 10:58:23 +02:00
Mateusz Jablonski
b7c7bc0a24 Correct createMultiGraphicsAllocationInSystemMemoryPool method 
force system memory placement
skip duplicated root device indices

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-05-13 12:11:49 +02:00
Mateusz Jablonski
fcbf1dcf32 Mark SVM_GPU allocation as not lockable 
require blitter usage if allocation is not lockable and in local memory

Related-To: NEO-5733
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-05-12 17:47:15 +02:00
Mateusz Jablonski
296634f5b8 Fallback to CPU copy when blit copy fails 
Related-To: NEO-5733
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-05-12 15:38:24 +02:00
Mateusz Jablonski
a552d1fd65 Revert "Mark SVM_GPU allocation as not lockable" 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-05-12 12:24:02 +02:00
Mateusz Jablonski
a79f7b678b Mark SVM_GPU allocation as not lockable 
require blitter usage if allocation is not lockable and in local memory

Related-To: NEO-5733
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-05-12 09:49:35 +02:00
Bartosz Dunajski
cffac84a3b Simplify creating EventPool allocation 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-05-11 18:26:54 +02:00
Zbigniew Zdanowicz
544890c066 Change default placement of preemption allocation 
Related-To: NEO-5800

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-05-11 15:26:24 +02:00
Lukasz Jobczyk
8e1e213bcf Disable GEM_WAIT when new residency model available 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2021-05-10 14:59:21 +02:00
Konstanty Misiak
c4502ccf66 Fix OCL MOCS index with UNCACHED flag 
Related-To: NEO-5742

Signed-off-by: Konstanty Misiak <konstanty.misiak@intel.com>
 2021-05-04 10:01:44 +02:00
Compute-Runtime-Validation
c453e7fe8b Revert "Mark SVM_GPU allocation as not lockable" 
This reverts commit 5341d0663c.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2021-04-29 18:06:38 +02:00
Mateusz Jablonski
5341d0663c Mark SVM_GPU allocation as not lockable 
use blitter if available and allocation is not lockable

Related-To: NEO-5733

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-04-29 15:55:13 +02:00