Re: [PATCH 1/2] MCE, AMD: Make MC2 decoding part of amd_decoder_ops as well

[Joe Perches <joe@perches.com>]

On Mon, 2012-12-17 at 14:43 -0600, Jacob Shin wrote:
> On Mon, Dec 17, 2012 at 09:40:11PM +0100, Borislav Petkov wrote:
> > On Mon, Dec 17, 2012 at 12:34:35PM -0800, Joe Perches wrote:
> > > decode_mc2_mce does
> > > 
> > > 	pr_emerg(HW_ERR "MC2 Error: ");
> > > calls
> > > 	fam_ops->mc2_mce()
> > > which may not emit anything
> > > and if it doesn't, does
> > > 	pr_emerg(HW_ERR "Corrupted ...");
> > > on another line leaving a trailing colon without
> > > newline from the pr_emerg above.
> > 
> > Ok, this is in Jacob's patch AFAICT, I'll review those tomorrow but
> > thanks for the heads-up.
> 
> Okay, I'll wait until your feedback before sending V2
> 
> > 
> > > I believe the second pr_emerg should be pr_cont
> > 
> > Actually, it should be
> > 
> > pr_emerg(HW_ERR "\nCorrupted...\n");
> > 
> > to finish the initial pr_emerg.
> 
> Okay, all of the other decode_mc*_mce functions needs to change then.

Hi Jacob.

I don't think that's necessary.

Using any pr_<level>("\n...") is just not good form.
I think the pr_cont would be fine, it's the same
form as all the other decoded continuations.

The new pr_emerg is effectively a termination of the
previous one.  There's no need for 2 consecutive uses
of pr_emerg(HW_ERR ...) here.

cheers, Joe