Re: [PATCH] reduce kallsyms table size on 64-bit arch-s

[Paulo Marques <pmarques@grupopie.com>]

Jan Beulich wrote:
> Some architectures (the patch carries it out for x86-64) can reduce the
> kallsyms pointer table's size by half through the use of relative 32-
> bit pointers. Since the kallsyms tables get created as assembly files
> anyway, the adjustment is strait forward and only requires a little bit
> of abstraction at the consuming side.

I've been wanting to do this for a long time, so I'm really glad someone
stepped up to do it for me :)

I've reviewed the patch and it looks good. I worry that other
architectures have strange uses for addresses that might produce larger
than 32 bit offsets, especially with CONFIG_KALLSYMS_ALL, but since this
is an "opt-in" config option, I guess it's ok.

Acked-by: Paulo Marques <pmarques@grupopie.com>

> Further, independent of architecture, kallsyms_num_syms and
> kallsyms_markers[] can use 32-bit types.
> 
> Signed-off-by: Jan Beulich <jbeulich@novell.com>

-- 
Paulo Marques - www.grupopie.com

"Who is general Failure and why is he reading my disk?"