[4/4] ntdll: Properly set flag which indicates buffer empty state.

[Dmitry Timoshkov]

Dmitry Timoshkov <dmitry at baikal.ru> wrote:

> Wolfgang Walter <wine at stwm.de> wrote:
> 
> > I think that happens:
> > 
> > * application writes data to com port.
> > * all is written, serial buffer is empty
> > * application calls  WaitCommEvent()
> > * wait_on() is called
> > * wait_on() calls get_irq_info()
> > * get_irq_info() sets commio->irq_info->temt = 1
> > * wait_on() calls check_events() and uses sets commio->irq_info for old an new
> > * so old->temt == new->temt and EV_TXEMPTY is not set
> > * if there are no other events (in real world basically EV_RXCHAR):
> > * wait_for_event() is startet with commio->irq_info->temt set to one
> > * wait_for_event() calls get_irq_info() with new_irq_info()
> > * get_irq_info() sets new_irq_info->temt = 1 because the buffer is still empty
> > * wait_for_event() calls check_events() with new_irq_info and commio->irq_info
> > * again check_events() will not set EV_TXEMPTY as both have temt == 1
> > 
> > It seems that we do not recover from that hang.
> > 
> > Please correct me if I misread the code and I'm wrong.
> > 
> > 
> > I think it's better if WaitCommEvent() returns with EV_TXEMPTY even if there 
> > has no new data been sent in between:
> 
> That means that WaitCommEvent(EV_TXEMPTY) without any prior WriteFile
> would always return success and signal the EV_TXEMPTY event. My tests
> show that WaitCommEvent(EV_TXEMPTY) fails with a timeout in this case.
> Since the serial device is very slow it should be unlikely that after
> successful WriteFile() with some data WaitCommEvent() sees an already
> empty transmitter's output queue.

One of the solutions to this would be that WriteFile sets a flag for a
being written handle indicating that there were data pending for this
handle, then WaitCommEvent(EV_TXEMPTY) in the case of empty output queue
would see that flag, clear it and set EV_TXEMPTY flag in the events mask.

That would also explain why SetCommMask() called after WriteFile() makes
WaitCommEvent(EV_TXEMPTY) fail with a timeout in some testbot VMs, in that
case SetCommMask() would also clear the data pending write flag.

-- 
Dmitry.