Re: [patch] Printk kernel version in WARN_ON

[Pavel Machek <pavel@ucw.cz>]

On Sat 2007-11-17 20:42:40, Sam Ravnborg wrote:
> On Sat, Nov 17, 2007 at 11:35:01AM -0800, Arjan van de Ven wrote:
> > On Sat, 17 Nov 2007 10:46:52 -0800
> > Andrew Morton <akpm@linux-foundation.org> wrote:
> > 
> > > > by ... not too much at least, gcc ought to be quite good at merging
> > > > same-strings into one, so it's just one extra pointer argument
> > > > 
> > > 
> > > I think I knew that.  At 1000 callsites.
> > 
> > ok so how about putting the same into dump_stack() instead? (see below)
> > added bonus is that it's now present for all dumps that use
> > dump_stack(), not just WARN_ON()
> > (the format I copied from the exact line used by oopses)
> 
> This solved the "zillion files being rebuild" issue I mentioned.
> So from that angle it is better.
> 
> And I notice you use the namespace aware helpers to access the
> kernelrelease string - I assume this is better than direct use
> of UTS_RELEASE.

I'm not sure... namespace-aware kernel release seems like madness to
me. This stackdump is from 2.6.10, but the other stackdump is from
2.6.20 running at the same machine at the same time?! Why do namespace
helpers touch UTS_RELEASE.

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html