Pavel Pisa started a new discussion on bsps/arm/tms570/console/tms570-sci.c: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1220#note_149160

 >    bauddiv = ( TMS570_VCLK_HZ + baudrate / 2 ) / baudrate;
 >    ctx->regs->BRS = bauddiv ? bauddiv - 1 : 0;
 >  
 > -  ctx->regs->GCR1 |= TMS570_SCI_GCR1_SWnRST | TMS570_SCI_GCR1_TXENA |
 > -                     TMS570_SCI_GCR1_RXENA;
 > +  // Bring the SCI out of reset and always enable TX
 > +  ctx->regs->GCR1 |= TMS570_SCI_GCR1_SWnRST | TMS570_SCI_GCR1_TXENA;
 > +
 > +  // Check CREAD to see if we should also enable RX
 > +  if ( ( t->c_cflag & CREAD ) != 0 ) {
 > +    ctx->regs->GCR1 |= TMS570_SCI_GCR1_RXENA;
 > +  }

Rx enable only when `CREAD == 1` is correct according to to the authoritative 
spec

https://pubs.opengroup.org/onlinepubs/9799919799/basedefs/termios.h.html

But it should be checked that it is correctly set in BSP and RTEMS termios by 
default and not cleaned by mistake in some of the termios set callers in RTEMS 
source tree. It can be problem for some user if the keeping it one or clear by 
mistake occurs.

So I am for change, but check rest of the RTEMS sources the first.

-- 
View it on GitLab: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1220#note_149160
You're receiving this email because of your account on gitlab.rtems.org.


_______________________________________________
bugs mailing list
[email protected]
http://lists.rtems.org/mailman/listinfo/bugs

Reply via email to