- Move logic to enqueueHandler to cover all scenarios
- Create BlockedCommandsData not only for Kernel enqueue
- KernelOperation cleanup
Change-Id: Ie4a673cbbc986c685996a38ab296444d38e7bbd5
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
- It could happen that 2 threads try to update the same event within tree
- First thread starts to submit command, it releases cmdToSubmit
- Second thread doesn't see the command and follows
- Second thread thinks that it submitted the command so it follows to
subsequent steps which are child events notification
- We end up with corrupted enqueue sequence as child may submit prior to
parent.
- With this change each submit step is synchronized basing on task count
- When second thread enters submit command without task count being set
it wait for first thread to properly set it.
Change-Id: Ic2ddaea17f9af8cab6781320edae2c268dd0b189
Signed-off-by: Mrozek, Michal <michal.mrozek@intel.com>
- Sequence is as follows :
Enqueue returning event , blocked on user event.
clFinish.
There are 2 additional threads:
- one is calling updateExecutionStatus in a loop on returned
event
- one is calling setUserEventStatus on parent user event
In such case virtual event may be submitted but task count will
not be assigned.
Change-Id: Ia097bd59b276cc9213945c476cf289398b8f5934
Signed-off-by: Mrozek, Michal <michal.mrozek@intel.com>
- is queue blocked to release virtual event when it is submitted
- do not wait on completion after ndr submit.
Change-Id: I63c54e6940f6241b32f0dbdd5880f849c46fd12e
Signed-off-by: Mrozek, Michal <michal.mrozek@intel.com>
- remove hwInfo from the csr functions where it was passed as a parameter,
now csr functions have access to hwInfo by Execution Environment
Change-Id: I756ae63d9728c9c963571147bab97f9e1c15797b
Signed-off-by: Adam Stefanowski <adam.stefanowski@intel.com>
Include files are now grouped and sorted in following order:
1. Header file of the class the current file implements
2. Project files
3. Third party files
4. Standard library
Change-Id: If31af05652184169f7fee1d7ad08f1b2ed602cf0
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
- fill AllocationData in one place
- remove allocateGraphicsMemoryForSVM function
- refactor SVM manager tests
Change-Id: I6f4ecd70503da8031cced50ea98a54162fd8e5d3
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
- remove method allocateGraphicsMemory(size_t size)
- pass allocation type in allocation properties
- set allocation type in allocateGraphicsMemoryInPreferredPool
Change-Id: Ia9296d0ab2f711b7d78aff615cb56b3a246b60ec
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
Move definitions of objectNotUsed and objectNotResident to GraphicsAllocation
Change-Id: I2aec604a865cc6c975e9d1121028cbdd35c0b18a
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
- It allows to easily turn ON the power saving mode for the driver.
- In that mode, whenever GPU is not ready it will put current thread to sleep,
thread is resumed when GPU completes.
- PowerSaving mode reduces CPU utilization on blocking calls, while increasing
completion latencies.
Change-Id: I3de83713687952ac31a1ec8c397f48aa4212781d
Move definition of allocations list method to internal_allocation_storage.cpp
Change-Id: I4c6038df8fd1b9335e8a74edbab33b78f9293d8f
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
