In case of debuggable context device should be additionally
initialized by early empty submission issue.
Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
- Do not iterate when all devices are parsed
- Early continue if given device not present in context
200ns (+10%) in below scenario from compute-benchmarks
ZE_AFFINITY_MASK=0.0 PrintDebugSettings=1 ./api_overhead_benchmark_l0
--test=ExecuteCommandListImmediate --api=l0 --UseProfiling=0
--CallsCount=1 --MeasureCompletionTime=0 --useBarrierSynchronization=0
--KernelExecutionTime=1 --iterations=1000
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
With this change, module's data sections will be allocated in USM device pool
instead of SVM or USM shared.
Signed-off-by: Sebastian Luzynski <sebastian.jozef.luzynski@intel.com>
When pointer type differs from the type, which was obtained
via new[] operator, then delete[] is undefined behavior.
Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
Discarding RAII lock returned from function almost always
is a bug. This change introduces usage of [[no_discard]]
attribute from C++17 to prevent such misues.
Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
With compiler LSC WAs this gives better performance.
If debugger is active, policy will not be changed ie.
will be WBP.
Related-To: NEO-7003
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
- pass correct deviceIndex when notifying cmdQs
- fix allocating debug surface - single tile allocation when single
bit set in storageInfo
Related-To: NEO-6980
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
When header is included for the first time in translation unit,
then preprocessor simply copy-pastes its content. If we define a
constant in a header file and this constant has internal linkage
then each and every translation unit, which includes this header
will have its own copy of this constant.
C++17 introduces inline variables, which are meant to allow creation
of variables in header files, which do not cause multiple instances.
The inline variable has a single instance when:
- constexpr is used without static (constexpr implicitly implies inline)
- inline is used without static
- inline const is used without static (const does not imply internal linkage
when used with inline)
Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
With compiler LSC WAs this gives better performance.
If debugger is active, policy will not be changed ie.
will be WBP.
Related-To: NEO-7003
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
This change is a baseline for tight control over
when dispatch pipeline state commands and which
pipeline state properties can be changed for a
given hardware platform
Related-To: NEO-5019
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
Certain platforms might not require prefetcher to
be disabled in direct submission. This change
provides a way to control that behaviour.
Signed-off-by: Rafal Maziejuk <rafal.maziejuk@intel.com>
Related-To: NEO-7218
This commit adds support for ZEX_NUMBER_OF_CCS flag which can be used
for limiting number of CCS engines
Format is as follows:
ZEX_NUMBER_OF_CCS=RootDeviceIndex:NumberOfCCS;RootDeviceIndex:NumberOfCCS...
i.e. setting Root Device Index 0 to 4 CCS, and Root Device Index 1 To 1 CCS
ZEX_NUMBER_OF_CCS=0:4,1:1
Related-To: NEO-7195
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
Mateusz Jablonski