Re: [PATCH 00/10] mac68k: Miscellaneous fixes, cleanup and modernization

From: Geert Uytterhoeven
Date: Thu Apr 20 2017 - 03:56:38 EST


Hi Finn,

On Mon, Apr 17, 2017 at 3:50 AM, Finn Thain <fthain@xxxxxxxxxxxxxxxxxxx> wrote:
> On Sun, 16 Apr 2017, Geert Uytterhoeven wrote:
>> On Sun, Apr 9, 2017 at 1:51 AM, Finn Thain <fthain@xxxxxxxxxxxxxxxxxxx>
>> wrote:
>> > This series has various patches from several different people. Two
>> > printk modernization patches were originally from Geert Uytterhoeven
>> > and three Nubus patches were originally committed to the Linux/mac68k
>> > CVS by David Huggins-Daines.
>>
>> Thanks, most of them look sane enough to apply and still queue for
>> v4.12.
>>
>> I'm a bit reluctant about the nubus changes (patches 6 and 8), though.
>> Do you think they need more testing?
>>
>> Thanks!
>
> Patch 6 is partly dead code removal. In principle, this patch is a
> reversion to the old code (pre-v2.3.17). The old code was thoroughly
> tested in Debian Sarge. I suppose a reviewer might wonder whether we want
> to keep new code for probing fake slot resources in Apple's on-board ROMs.
> That would be useful if it could eliminate the macintosh_config struct.
> But it can't, and we don't want both mechanisms. Hence the reversion in
> the mac68k CVS.
>
> Patch 8 changes the pointer validation code and although this has been
> tested on the valid path, you are right that it could use some negative
> testing. But that would seem to require cards with flawed ROMs. I don't
> know of any of such cards. So I think that all we can do is more review.
>
> Maybe Michael or Laurent would be willing to review these two patches?

OK.

I've applied and queued all but patches 6 and 8. This required some
small adjustment to e.g. the whitespace patch.

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