Ok, yes, that's the obvious solution! Why didn't I think about that? :-)
That will do the trick, yes. I'll test it soon.

-- Ragge

> On Sun, Jun 03, 2001 at 05:47:19PM +0200, Anders Magnusson wrote:
> 
>  > > Does the VAX's scheme deal with that okay?
>  > >
>  > Not from the slave CPUs (because the chars are sent via IPI).
>  > 
>  > > Also, where do you want
>  > > to put the new #define?  Probably should be in sys/types.h, or something,
>  > > as __NO_KPRINTF_LOCK.
>  > > 
>  > It doesn't matter at all, I just want this to work for some degree of "work".
>  > I am interested in all suggestions as long as this problem is solved.
> 
> Ok, Bill Sommerfeld mentioned something to me which sparked an idea...
> 
> In your machine/lock.h, define a KPRINTF_SLOCK_ACQUIRE like so:
> 
> #define	KPRINTF_SLOCK_ACQUIRE(alp) \
> do { \
> 	struct cpu_info *__ci = curcpu(); \
> 	\
> 	while (__cpu_simple_lock_try(&(alp)->lock_data) == 0) { \
> 		if (__ci->ci_ipis_pending) \
> 			/* go process them */ ; \
> 	} \
> } while (/*CONSTCOND*/0)
> 
> Then in subr_prf.c:
> 
> #ifndef	KPRINTF_SLOCK_ACQUIRE
> #define	KPRINTF_SLOCK_ACQUIRE(alp)	__cpu_simple_lock(&(alp)->lock_data)
> #endif
> 
> #define	KPRINTF_MUTEX_ENTER(s) \
> do { \
> 	(s) = splhigh(); \
> 	KPRINTF_SLOCK_ACQUIRE(&kprintf_slock); \
> } while (/*CONSTCOND*/0)
> 
> -- 
>         -- Jason R. Thorpe <thorpej@zembu.com>
>