Re: [PATCH v2] sh: Implement __get_user_u64() required for 64-bit get_user()

From: Geert Uytterhoeven
Date: Wed Jun 03 2020 - 03:35:28 EST


Hi Adrian,

On Wed, Jun 3, 2020 at 9:20 AM John Paul Adrian Glaubitz
<glaubitz@xxxxxxxxxxxxxxxxxxx> wrote:
> On 5/31/20 12:47 PM, John Paul Adrian Glaubitz wrote:
> > Changes since v1:
> > - Replace single mov instruction for exception handling
> > in case of invalid load
>
> Yutaka Niibe has had a look at my patch and he says, we might have to add
> an entry for the fault handling of the upper word.
>
> Quote:
>
> > (1) I think that there is possibility that the second access to user
> > space fails (while the first access succeeds). IIUC, it's good have
> > an entry in __ex_tables for the second access too, like:
> > ".long 1b+2, 3b\n\t"
> > I don't know if the expression "1b+2" is correct, my intention is
> > detecting the failure in the seccond access.

So just add another numeric label, like is done on m68k.

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds