Commit Graph

68 Commits

Author SHA1 Message Date
Lukasz Jobczyk 6fdcb83020 Add missing tbx calls to timestamp wait logic
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-04-19 14:24:06 +02:00
Patryk Wrobel e4d56cde21 Detect GPU hangs in blocking enqueue handler calls
This change introduces detection of GPU hangs in blocking
calls to enqueueHandler() function. Moreover, usages of
this function template have been revised and adjusted to
check the exit code. Furthermore, enqueueBlit() and
dispatchBcsOrGpgpuEnqueue() functions returns value now.
ULTs have been added to cover new cases.

Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
Related-To: NEO-6681
2022-03-22 21:06:19 +01:00
Jitendra Sharma f52f3df274 Add platform specific getter of debug surface size
For different platforms based on number of available threads
and debug surface layout, calculate max debug surface size.

Related-To: NEO-6676
Signed-off-by: Jitendra Sharma <jitendra.sharma@intel.com>
2022-03-22 12:18:40 +01:00
Lukasz Jobczyk 9ff1307b4b Fix optimize timestamp packet dependiencies
-program barrier after global fence allocation is programmed
-do not double barrier timestamp in blit enqueue
-flush GPGPU while submitting to BCS when barrier requested

Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-02-03 16:27:09 +01:00
Katarzyna Cencelewska a9ebb49fb5 Remove device enqueue part 1
Remove:
-tests with matcher for device enqueue
-classes: MockDeviceQueueHw, DeviceQueueHw, SchedulerSimulation,
DeviceQueueHwTest, KernelArgDevQueueTest
-functions: forceDispatchScheduler, processDeviceEnqueue, dispatchScheduler

Related-To: NEO-6559
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
2022-01-11 15:29:49 +01:00
Lukasz Jobczyk 1f0c58d0bf Refactor timestamp wait mechanism
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2021-12-16 13:22:19 +01:00
Bartosz Dunajski fe8cb190bc Remove deprecated code
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-12-14 11:36:10 +01:00
Lukasz Jobczyk 63116e4100 Use wait utils while waiting for timestamps
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2021-12-02 17:49:05 +01:00
Lukasz Jobczyk e8cc34d7b0 Add infrastructure to assign engine to command queue with round robin
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2021-11-15 13:13:40 +01:00
Compute-Runtime-Validation 908a7721b2 Revert "Assign engine to command queue using round robin algorithm"
This reverts commit 2fff0f9059.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2021-11-10 07:25:44 +01:00
Lukasz Jobczyk 2fff0f9059 Assign engine to command queue using round robin algorithm
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2021-11-09 11:05:16 +01:00
Compute-Runtime-Validation 9058de77cc Revert "Assign engine to command queue using round robin algorithm"
This reverts commit 1c68ac1cbc.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2021-11-08 16:36:18 +01:00
Lukasz Jobczyk 1c68ac1cbc Assign engine to command queue using round robin algorithm
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2021-11-03 17:16:54 +01:00
Filip Hazubski 28b37aea72 Correct handling unique_ptr in functions
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
2021-10-21 18:17:14 +02:00
Maciej Dziuban cdb7287816 Fix direct submission initialization
Related-To: NEO-6057
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
2021-10-12 11:51:48 +02:00
Maciej Dziuban c910a58aa0 Store bcsEngine in an std::vector in OpenCL CommandQUeue
Related-To: NEO-6057
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
2021-10-07 21:55:21 +02:00
Mateusz Hoppe d4a4c35759 Zero-initialize debug surface
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2021-09-22 14:36:36 +02:00
Maciej Dziuban 858727010f Add selectCsrForBuiltinOperation method to OpenCL CommandQueue
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
Related-To: NEO-6057
2021-09-10 12:26:47 +02:00
Maciej Dziuban 3e6b3cd780 Revert "Create single point for selecting engine for builtin ops"
This reverts commit 6513bd371a69cb821ea45c7c1964ea1f51dd75b2.

Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
2021-09-02 11:58:46 +02:00
Maciej Dziuban 33ac15dabe Pass blitter CSR in blocked queue scenarios
Related-To: NEO-6057
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
2021-08-31 16:29:17 +02:00
Bartosz Dunajski 92000bd75b Improve Device getter to support EngineInstanced Device with single Engine
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-08-31 15:50:23 +02:00
Maciej Dziuban a03b5f8a95 Create single point for selecting engine for builtin ops
- selectCsrForBuiltinOperation selects proper CSR
- selected CSR is passed to dispatchBcsOrGpgpuEnqueue
- Notifications such as notifyEnqueueReadBuffer are now made on correct engine
- Temporary allocs for hostptrs are now created on gpgpuCsr

Related-To: NEO-6057
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
2021-08-31 13:29:17 +02:00
Maciej Dziuban 42524b6bef Limit usage of getBcsCommandStreamReceiver()
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
2021-08-27 12:15:30 +02:00
Bartosz Dunajski cb91474b1a Improve getDevice methods
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-08-26 14:45:00 +02:00
Konstanty Misiak ad19eda689 Fix event profiling for marker commands
Related-To: NEO-5799

Signed-off-by: Konstanty Misiak <konstanty.misiak@intel.com>
2021-08-13 15:02:01 +02:00
Lukasz Jobczyk 9e50138829 Defer ULLS initialization
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2021-07-21 15:55:16 +02:00
Maciej Plewka dfe578754b Fix add pipe controll before marker profiling
Resolves: NEO-6065

Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
2021-07-20 16:23:57 +02:00
Kamil Kopryk ce92b42f63 Code cleanup - avoid copy 2/n
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
2021-07-20 10:44:24 +02:00
Maciej Dziuban 6f6ee90aa5 Extend copy engine selector data
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
2021-07-06 12:28:55 +02:00
Michal Mrozek ad3855ceb3 Revert "Fix event profiling for marker commands"
This reverts commit 5f0167b477993f3ebc804c53a110bc432638615b.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2021-07-01 08:01:33 +02:00
Konstanty Misiak 1df1094465 Fix event profiling for marker commands
Related-To: NEO-5799

Signed-off-by: Konstanty Misiak <konstanty.misiak@intel.com>
2021-06-30 14:56:39 +02:00
Bartosz Dunajski 8523747f91 Clear Event dependencies for Subcapture
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-06-18 12:58:49 +02:00
Kamil Kopryk b2918b5115 Correct enqueueCopyImage params
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
Related-To: NEO-4692
2021-06-15 13:24:57 +02:00
Maciej Plewka 9fe2dddcd3 Estimate command stream size for marker profiling
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
2021-06-14 21:36:07 +02:00
Piotr Zdunowski 7cac94b597 Fix profiling for marker commands.
Resolves: NEO-4867

Signed-off-by: Piotr Zdunowski <piotr.zdunowski@intel.com>
2021-05-27 12:15:07 +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
Matias Cabral 2830073139 Save State Save Area Header in OCL
Signed-off-by: Matias Cabral <matias.a.cabral@intel.com>
2021-05-13 17:58:14 +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
Maciej Dziuban 5318ff1872 Defer OsContext initialization
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
Related-To: NEO-5610
2021-04-19 11:54:34 +02:00
Bartosz Dunajski 5a50ad098c Refactor TagAllocator
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-03-26 15:16:56 +01:00
Maciej Dziuban 3f5b9df122 Pass EngineUsage to getEngine
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
Related-To: NEO-5610
2021-03-16 16:31:44 +01:00
Mateusz Jablonski 056300f01f Pass internal kernel object to enqueueKernel method
Related-To: NEO-5001
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2021-03-08 19:14:08 +01:00
Bartosz Dunajski 9c164a01a6 AuxTranslationMode fallback path
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2021-03-02 12:09:04 +01:00
Kamil Kopryk f84cd9f424 Small refactor getCommandStreamReceiver
Related-To: NEO-4692
Change-Id: Ie8c36982d5bd626735d1f8e6d04084bce67738e5
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
2020-10-27 15:52:18 +01:00
Maciej Dziuban 8fcd51c2c8 Do not obtain command stream if it will not be needed
Change-Id: Id7fa1c6b78e71a085084f8fcb66a7b8e873ad2bc
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
Related-To: NEO-5120
2020-10-08 12:24:03 +02:00
Bartosz Dunajski 28acc2c8f6 Remove lowPriorityEngineType member
Change-Id: I39908addf1c3a12f7807479132d0a6d5a4398d7b
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2020-09-28 17:02:56 +02:00
Lukasz Jobczyk 1185205673 Enable batched direct submission
Related-To: NEO-5112

Change-Id: I191dee8e8466232a1350e0b13974c3417366743f
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2020-09-23 15:27:57 +02:00
Bartosz Dunajski 5b31b31734 Remove low priority and internal engine index variables
Change-Id: I984dbcedd5456cfdee284f4d2eda1371ba77f0d0
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
2020-09-21 16:44:14 +02:00
Konstanty Misiak dcc5316d88 Fix debug surface allocation in multi-queue scenarios
Related-To: NEO-5078

Change-Id: Ife48e703f348340cd8f89dd76f8affce0641adac
Signed-off-by: Konstanty Misiak <konstanty.misiak@intel.com>
2020-09-16 13:12:17 +02:00
Lukasz Jobczyk 9158dec82d Do not override immediate dispatch while direct submission enabled
Related-To: NEO-5007

Change-Id: I32330a275f78e825c847b0e380ce426f0c7412f5
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2020-09-03 11:53:38 +02:00