Re: RFC: remove __read_mostly

[Matt Mackall <mpm@selenic.com>]

On Fri, Dec 14, 2007 at 04:38:04PM +0100, Eric Dumazet wrote:
> Matt Mackall a ?crit :
> >On Thu, Dec 13, 2007 at 11:20:44PM +0100, Adrian Bunk wrote:
> >  
> >>I tried the following patch with a full x86 .config [1]:
> >>
> >>--- a/include/asm-x86/cache.h
> >>+++ b/include/asm-x86/cache.h
> >>-#define __read_mostly __attribute__((__section__(".data.read_mostly")))
> >>+/* #define __read_mostly 
> >>__attribute__((__section__(".data.read_mostly"))) */
> >>
> >>The result [2,3] was:
> >>
> >>-rwxrwxr-x 1 bunk bunk 46607243 2007-12-13 19:50 vmlinux.old
> >>-rwxrwxr-x 1 bunk bunk 46598691 2007-12-13 21:55 vmlinux
> >>
> >>It's not a surprise that the kernel can become bigger when __read_mostly 
> >>gets used, especially in cases where __read_mostly prevents gcc 
> >>optimizations.
> >>
> >>My question is:
> >>Is there anywhere in the kernel a case where __read_mostly brings a 
> >>measurable improvement or can it be removed?
> >>    
> >
> >Yes, but perhaps we can put it under CONFIG_BASE_FULL?
> >
> >  
> Yes, we probably can do something like that (in addition to !CONFIG_SMP)

Excellent point. If either CONFIG_BASE_FULL or CONFIG_SMP are not set,
we should not define __read_mostly.

-- 
Mathematics is the supreme nostalgia of our time.