Why _ThreadProcessSignalsFromIrq() in new exception processing?

[Sergei Organov]

Valette Eric <eric.valette at free.fr> writes:
> Sergei Organov wrote:
> > Could somebody please explain what is the purpose of call to
> > _ThreadProcessSignalsFromIrq() in the new-exception-processing from the
> > shared_raw_irq_code_entry? Should it be implemented in the old exception
> > processing for proper RTEMS behavior?
> 
> I guess, you will not get fully working posix signaling without doing this
> (look at pthreadkill or signalsend). This block of code is used to trigger a
> processing on the newly switched thread that end up calling
> _API_extensions_Run_postswitch.

"Guess" is better than nothing :-) Does it mean I'm OK even without the call
if I don't use pthreads?

> 
> NB : The code was already there on ix86 when I implemented the new exception
> model I just keeped it. When going to PPC, the call was missing so I added
> it.

So maybe those one who wrote it initially for ix86 will comment more...

> 
> In most advanced RTOS, there are things that should be done on the current
> thread that have been trigerred as the result of a system call from isr and
> that must be deffered until the thread runs again...

Sorry, don't quite understand. Care to give an example scenario when it is
required? I feel myself rather bad without understanding of how things are
supposed to work.

> In that case having a context is very usefull. I wanted to use it to
> implement a call to debugger from keyboard in particular....
> 
> > BTW, is the new exception processing model documented somewhere as this
> > call is not the only violation of the old documentation describing how
> > exceptions/interrupts should be handled by BSPs?
> 
> If documentation is not accurate, trust the code.

The trick is to find out whether the documentation or the code is not accurate
and if the former, which particular reincarnation of the code should be
trusted ;-)

> If old code is broken, do not stick to it :-)

I don't as soon as it is indeed broken ;-)

BTW, its the new code for PPC that seems to be broken *for me*. I use EABI
(-meabi -msdata=eabi) heavily, EABI requires r2 to point to '.sdata2' section,
but the raw irq code seems to clobber it :-(

-- 
Sergei.