Re: RFC: remove __read_mostly

[Adrian Bunk <bunk@kernel.org>]

On Thu, Dec 13, 2007 at 11:29:08PM +0100, Andi Kleen wrote:
> Adrian Bunk <bunk@kernel.org> writes:
> >
> > -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
> 
> File sizes are useless -- check size output.

    text    data     bss      dec     hex filename
29268488 3697961 5222400 38188849 246b731 vmlinux.old
29268435 3685565 5228784 38192784 246c690 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.
> 
> What optimizations do you think it prevents?  I don't think it 
> should change the gcc generated code at all; the only difference
> should be to the linker.

What I have seen recently was a static variable marked __read_mostly 
being read-only with some CONFIG_DEBUG_FOO=n.

With __read_mostly gcc feeled obliged to emit a variable.

Without __read_mostly, gcc optimized the variable away.

> -Andi

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed