- do not allocate state save area every time attention event
is handled
- keep allocated memory for subsequent events
- remove not needed DBEUG_BREAK
Related-To: NEO-8183
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
- check threads with attention state before reading SR_IDENT
- memory read is costly, for threads already in stopped state
there is no need to read sr_ident and check thread state again
- single stepping one thread performance is substantially improved,
by the factor of 6x
Related-To: NEO-8183
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
- reading state save area for every threads takes too long when all
application threads have completed and there are stale ATT events to
process
- on detach gdb seemed to be frozen waiting for ATT event to be handled
- fix is to read state save area once - and check SIP counter for every
thread in ATT bitmask
Related-To: NEO-7897
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
- before marking interrupt request check exception reason. If there is
exception other than forced exception or forced external halt treat
thread as stopped and generate distinct event for it.
Related-To: NEO-7869
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
Topology map was only being created when ZET_ENABLE_PROGAM_DEBUGGING was
set. This was not correct. Now it is unconditionally created at init,
and debug attach will fail if it is not valid.
Related-to: LOCI-3937
Signed-off-by: Yates, Brandon <brandon.yates@intel.com>
Some L0 debug CTSs intentionnally written to exit w/o proper
resource clenup, f.e do not call zetDebugDetach() etc.
On windows it could be the situation when cleanup of DebugSession is
called in context of DllMain(DLL_PROCESS_DETACH).
At this point all threads other then main already terminated by Windows,
see remarks for DLL_PROCESS_DETACH in
https://learn.microsoft.com/en-us/windows/win32/dlls/dllmain
In this case worker thread object still exists, handle and Id are not
null but corresponding Windows thread does not exist any more and
application waits forever for threadFinished variable. We can safely
omit this waiting since join() will either return immediately in case of
thread was killed by Windows or wait until thread is terminated in normal way.
Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
- enable tile attach mode by default
- both root device and subdevice may be attached to
Related-To: NEO-7347
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
We have to always generate module create/destroy events to give
debugger a chance to insert bp, read/write debuggee memory etc
even there is no debug data was generated. In this case ELF will not be
reported to debugger, just ISA GpuVA
Related-To: NEO-6723
Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
- ThreadId is preferred beacuse it uniquely identifies thread and is
not related to device topology
Related-To: NEO-5784
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
MMIO will store SBA tracking buffer address for current context.
This change helps in extracting this address and use to read SBA virtual
register.
Related-To: NEO-6765
Signed-off-by: Jitendra Sharma <jitendra.sharma@intel.com>
Mateusz Hoppe