Re: [PATCH] Make printk lock when appropriate...

[Keith Owens]

On Sat, 1 Jul 2000 10:17:50 +0200 (CEST), 
Robert de Vries <[e-mail]> wrote:
>On Sat, 1 Jul 2000, Manfred Spraul wrote:
>> Sorry, but the patch is wrong.
>> 
>> "buf" is a static variable, defined in line 28 [printk.c]
>
>However, printk is the only function using it. Why not move it into
>printk? Is 1024 bytes too large for the kernel stack?

You bet your sweet bippy it is too large.  With only ~7K of usable
stack, do you really want to grab 1K?  Especially when you can go
recursive on error and gobble up 2*1K chunks.  What happens when the
stack overflow patch in IKD detects you are running low on stack space
and calls printk to issue a warning?

What about this code instead of spin_lock_irqsave?  Pity we do not have
spin_trylock_irqsave();

local_irq_save(flags);
if (!spin_trylock(&console_lock)) {
	local_irq_restore(flags);
	return(0);
}