Adds a per-kernel and per-device query to determine the
number of GRF registers that a kernel was compiled for.
This is an informal query for now, but may be added to
a formally supported extension in the future.
Related-To: NEO-9807
Signed-off-by: Ben Ashbaugh <ben.ashbaugh@intel.com>
- check releaseHelper support when selecting bindless mode, if not
disabled, prefer bindless mode in L0 API
- bindless mode can be forced with DebugVariable: UseBindlessMode
Related-To: NEO-7063
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
Program barrier to task stream, before next enqueue kernel.
This will reduce the number of batch buffer starts for sequences of
enqueue, barrier, enqueue, ... .
Related-To: NEO-8147
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
Program barrier immediately to task stream.
This will reduce the number of batch buffer starts.
Related-To: NEO-8147
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
denorm support is controlled by IGC, we should just set zero by default
Related-To: NEO-8059
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
denorm support is controlled by IGC, we should just set zero by default
Related-To: NEO-8059
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
- getGlobalBindlessHeapConfiguration() should be used to choose global
alloctor for SSH
- remove not needed and incorrect unit tests
- remove not needed branches
- bindless mode controls bindless compilation only
Related-To: NEO-7063
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
Allows the user to use alignments > 64KB in `createUnifiedMemoryAllocation`
So that the restriction in `piextUSMDeviceAlloc` of the DPC++ runtime
could be lifted
Related-To: LOCI-4168
Signed-off-by: Lu, Wenbin <wenbin.lu@intel.com>
adjust thread group dispatch size on pvc if chosen size does not evenly
divide dimension
this is to avoid leftover thread groups
Related-To: NEO-7927
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
Unit tests should not write output to the console.
Instead, every output should be captured.
Signed-off-by: Fabian Zwolinski <fabian.zwolinski@intel.com>
- For static create() method for Kernel and MultiDeviceKernel force errcodeRet
parameter to be passed via reference (instead of a pointer)
- Move part of kernel's creation logic to initialize() method
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
When state base address tracking is enabled and command list use private heaps
then command list at destroy time must calls all compute CSRs that were using
that heap to invalidate state caches.
This allows new command list to reuse the same heap allocation for different
surface states, so before new use cached states are invalidated.
Related-To: NEO-5055
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
Allows the user to use alignments > 64KB in `createUnifiedMemoryAllocation`
So that the restriction in `piextUSMDeviceAlloc` of the DPC++ runtime
could be lifted
Related-To: LOCI-4168
Signed-off-by: Lu, Wenbin <wenbin.lu@intel.com>
