Re: test6pre1 change for sysrq config-option

[Philipp Rumpf]

On Sat, Aug 05, 2000 at 06:53:41PM -0400, Theodore Y. Ts'o wrote:
>    From: "Juan J. Quintela" <[e-mail]>
>    Date: 03 Aug 2000 18:26:50 +0200
> 
>    SysRq don't work here over serial console (has never worked
>    ok). Phillip Rumpf solved my problems. I thought that you had the
>    patch.  This is the patch actualized to test6-pre1. 

> This patch fixes the problem where the sysrq won't work if the console
> is printing some printk message (since it masks the break key).
> However, if the console is printing some message via printk, the break

"not", I suppose ?

> key should work fine.    So saying that SysRq "doesn't work" is rather
> overstating the case.  That's why I didn't recognize which patch you
> were referring to.

Look harder.  It also moves the "detect a break" logic before

			*status &= info->read_status_mask;

Apparently with the getty quintela uses, BRKINT isn't enabled, so SysRq
"doesn't work" for him.  I'm not sure it's a good idea to depend on
userspace enabling BRKINT - in test6-pre5, sysrq_enabled affects serial
SysRq as well.

	Philipp