Re: IRQ off latency of printk is very high

[Matt Mackall <mpm@selenic.com>]

On Thu, Oct 25, 2007 at 03:52:28PM -0700, Tim Bird wrote:
> Mathieu Desnoyers wrote:
> > It might help to read this thread I posted on LKML in January 2006
> > explaining the problem, which led to some discussion about the issue.
> > 
> > http://lkml.org/lkml/2006/6/3/48
> 
> This is very helpful.  Jon Smirl's answer seems to give the
> rationale for supporting printk output in interrupt context.
> I'm not sure, however, if extending the interrupt off period
> to cover the console output is required.  It didn't until
> Ingo changed it in 2.6.17.

Hmm, I see this at the beginning of the post-BK era (2.6.12-rc2):

	        spin_lock_irqsave(&logbuf_lock, flags);
		...
		spin_unlock(&logbuf_lock);
		call_console_drivers(_con_start, _log_end);
		local_irq_restore(flags);

-- 
Mathematics is the supreme nostalgia of our time.