Re: [RFC PATCH] xfs: consolidate local format inode fork verifiers

[Brian Foster <bfoster@redhat.com>]

On Mon, Jul 24, 2017 at 09:07:52AM -0400, Brian Foster wrote:
> On Mon, Jul 24, 2017 at 03:26:38PM +1000, Dave Chinner wrote:
> > On Sat, Jul 22, 2017 at 08:29:44AM -0400, Brian Foster wrote:
> > > On Sat, Jul 22, 2017 at 09:00:58AM +1000, Dave Chinner wrote:
> > > > On Fri, Jul 21, 2017 at 09:54:05AM -0400, Brian Foster wrote:
> > > > > On Fri, Jul 21, 2017 at 08:49:34AM +1000, Dave Chinner wrote:
...
> 
> [1] A simple test that we can do at the moment is to compare a
> production xfs.ko with one with asserts enabled. My local for-next
> branch builds a module of 40104000 bytes without XFS_WARN and 40540208
> bytes with XFS_WARN enabled. That makes for a difference of ~425k and
> roughly a 1% code size increase (accounting for ~1750 asserts and
> supporting code).
> 

FWIW, the huge size here didn't register to me until Darrick mentioned
it on irc. I'm guessing I have a bunch of debug enabled in my kernel
config (that I don't really want to disable right now). If I strip the
resulting module binaries, I end up with 1076736 bytes vs. 1210624 bytes
(for a 130kb delta and ~12% increase), which I'm guessing is a more
accurate assessment.

Brian

> [2] We may not want to use ASSERT() here, but perhaps define a new
> variant to omit the stack trace and otherwise customize to a verifier
> report.
> 
> > Cheers,
> > 
> > Dave.
> > 
> > -- 
> > Dave Chinner
> > david@fromorbit.com
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html