Re: tickle NMI watchdog on serial output.

[Bill Davidsen <davidsen@tmr.com>]

http://newsvote.bbc.co.uk/mpapps/pagetools/print/news.bbc.co.uk/2/hi/health/5232150.stmDave 
Jones wrote:
> Serial is _slow_ sometimes. So slow, that the NMI watchdog kicks in.


> I initially did the patch below a year ago for the Fedora kernel, and have
> been keeping it up to date since.  I recently got the same thing happening
> on a vanilla kernel, so figured it was time to repost this.

Hopefully this will get picked up for mainline. In case of a real hung 
it should still trigger NMI in some reasonable time.
> 
> Signed-off-by: Dave Jones <davej@redhat.com>
> 
> --- linux-2.6/drivers/serial/8250.c~	2005-05-14 02:49:02.000000000 -0400
> +++ linux-2.6/drivers/serial/8250.c	2005-05-14 02:54:30.000000000 -0400
> @@ -2098,9 +2098,11 @@ static inline void wait_for_xmitr(struct
>  	/* Wait up to 1s for flow control if necessary */
>  	if (up->port.flags & UPF_CONS_FLOW) {
>  		tmout = 1000000;
> -		while (--tmout &&
> -		       ((serial_in(up, UART_MSR) & UART_MSR_CTS) == 0))
> +		while (!(serial_in(up, UART_MSR) & UART_MSR_CTS) && --tmout) {
>  			udelay(1);
> +			if ((tmout % 1000) == 0)
> +				touch_nmi_watchdog();
> +		}
>  	}
>  }
>  
> 


-- 
Bill Davidsen <davidsen@tmr.com>
   Obscure bug of 2004: BASH BUFFER OVERFLOW - if bash is being run by a
normal user and is setuid root, with the "vi" line edit mode selected,
and the character set is "big5," an off-by-one errors occurs during
wildcard (glob) expansion.