Add a per-instance SVMAllocsManager::nonGpuDomainAllocs container for
all allocations to be removed in
moveAllocationsWithinUMAllocsManagerToGpuDomain. This approach replaces
the current iterative search and performs the task faster.
Add 7 new unit-tests to verify the functionality related to
nonGpuDomainAllocs container, both in expected and unexpected/synthetic
scenarios.
For UTs replace a dummy unifiedMemoryManager pointer with a pointer to
an instace of SVMAllocsManager, otherwise a SegFault error is thrown at
the end of tests.
Perform overall cleanup in related tests implementation, includes but
not limited to removal of:
- givenInitialPlacementGpu\
WhenMovingToGpuDomainThenFirstAccessDoesNotInvokeTransfer
As it is fully covered by:
givenAllocationMovedToGpuDomain\
WhenVerifyingPagefaultThenAllocationIsMovedToCpuDomain
- givenInitialPlacementGpu\
WhenVerifyingPagefaultThenFirstAccessDoesNotInvokeTransfer
As it is fully covered by:
givenTbxAndnitialPlacementGpu\
WhenVerifyingPagefaultThenMemoryIsUnprotectedOnly
Finally, reduce code duplication where possible.
Related-To: NEO-6658
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
Logic related to programming non coherent and thread arbitration policy for
gens 9 and 11 has been moved to EncodeComputeMode object, where similar
logic for gens gen12lp and newer is located.
Functions PreambleHelper::programThreadArbitration and
PreambleHelper::getThreadArbitrationCommandsSize have been removed.
Redundant setForceNonCoherent call has been removed from XE HPG
Related-To: NEO-6728
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
This change introduces detection of GPU hangs
in CommandMapUnmap::submit() as well as in Event::submitCommand().
ULTs have been added to cover the new code.
Related-To: NEO-6681
Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
When we build L0 right after OpenCL or vice versa we expect shared code
to be reused from ccache (when enable). Proper setting of project() and
add_subdirectory() helps to achieve this.
Signed-off-by: Artur Harasimiuk <artur.harasimiuk@intel.com>
Memory module translates memory regions without any
runtime query on underlying hardware/kernel. Use
ioctl helpers for memory region translation.
Related-To: LOCI-3006
Signed-off-by: Bellekallu Rajkiran <bellekallu.rajkiran@intel.com>
- Added support for Dynamic Module link to allow usage of
symbols that exist thru a circular dependency from the export->import
module & chained dependencies with importModule -> ExportModule ->
ExportModule2.
- Adds a dependency walker which collects all dependent modules whose
export surface state must be made resident for a given kernel
execution to successfully access each function resolved on different
modules.
- Expanded the Dynamic Link Black Box test to allow for testing either
simple linking and circular + chained linkage.
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
When trying to set the same values we don't need to set it again.
The pointer and allocId need to be the same.
Related-To: NEO-6737
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
Buffer's default bytesPerPixel value always equals 1 and as
IMAGE1D_BUFFER is originally an image, X coordinate needs to be
multiplied by bytesPerPixel in both copySize and (src/dst)Size.
Signed-off-by: Rafal Maziejuk <rafal.maziejuk@intel.com>
Related-To: NEO-6134
This Patch adds support for collecting IP Metrics using
StreamerOpen, StreamerClose and StreamerReadData
Related-To: LOCI-2755
Related-To: LOCI-2756
Signed-off-by: Joshua Santosh Ranjan <joshua.santosh.ranjan@intel.com>
Do not try to include empty directories, as cmake will not
explicitely return directiories, which would be searched
by compiler anyway. Such found should still be considered
as successful.
Signed-off-by: Grzegorz Choinski <grzegorz.choinski@intel.com>
This commit removes ZebinTargetMetadata struct, and uses
ZebinTargetFlags for both target validations: via machine type, and
via intel gt notes.
Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
