Re: [PATCH 0/8] x86 boot, pda and gdt cleanups

[Jeremy Fitzhardinge <jeremy@goop.org>]

Rusty Russell wrote:
> Hmm, this invalidated my assumption that write_gdt_entry is always a
> write to this cpu's active gdt.  Better fix is not to call it twice
> anyway...
>   

No, I don't think that's true.  I implemented the write_*_entry
functions with the assumption they could be called either on setup or on
an in-use entry.  I think its good policy to use it all the time anyway,
since the pv_ops backend might want to fiddle with the values on the way
through.

I tried to avoid calling init_gdt twice, but it seemed cleaner to just
let it happen.

> Getting rid of the call in smp_prepare_boot_cpu currently works, but
> it's fragile:  __get_cpu_var(x) && per_cpu(x, smp_processor_id()) will
> differ, and changes made to __get_cpu_var(x) will vanish...
>   

Yes.  I think its definitely a good idea to call init_gdt asap after
doing the percpu setup.

> Fortunately, UP doesn't have to call init_gdt at all, so I think it's
> better to place it in smp_prepare_boot_cpu only and then clean up the UP
> code.  I'll try now...
>   
It doesn't?  The per-cpu gdt is the same as the boot gdt?


    J