lkml.org 
[lkml]   [2008]   [Mar]   [4]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
/
DateTue, 4 Mar 2008 08:32:48 +0100
FromIngo Molnar <>
SubjectRe: [patch] x86: phase out forced inlining
* Andrew Morton <akpm@linux-foundation.org> wrote:

> urgh.  This will cause whatever problem 
> 4507a6a59cfc6997e532cd812a8bd244181e6205 fixed five years ago to 
> resurface for incautious gcc-3.x users.

hm, commit 4507a6a59cfc6997e532cd812a8bd244181e6205 does not exist:

  fatal: bad object 4507a6a59cfc6997e532cd812a8bd244181e6205

but i suspect it must be something along the lines of the known problem 
of really old gcc versions creating huge stackframes? Those pristine gcc 
versions were practically unusable for distro kernels anyway (and were 
patched by distros) - but i have no problem with restricting this 
feature to gcc4x. gcc4x creates more compact -Os code too, so it's 
recommended for smaller image sizes.

> I'd suggest that this
> 
> > +#ifndef CONFIG_OPTIMIZE_INLINING
> 
> become something along the lines of
> 
> > +#ifndef CONFIG_OPTIMIZE_INLINING && (__GNUC__ > 3)

good point, fixed that.

> It would be nice to be able to feed the gcc version into the Kconfig 
> logic, really..

yeah, instead of littering our include files with those quirks.

	Ingo


\
 
 \ /
  Last update: 2008-03-04 08:35    [from the cache]
©2003-2008