Re: [PATCH][0/10] verify_area cleanup

From: Jesper Juhl
Date: Sun Mar 13 2005 - 17:03:28 EST


On Sun, 13 Mar 2005, William Lee Irwin III wrote:

> On Fri, Mar 04, 2005 at 03:47:07AM +0100, Jesper Juhl wrote:
> > Now that 2.6.11 is out the door it's time to try and submit this again.
> > The following patches convert all users of verify_area to access_ok and
> > the final patch then deprecates verify_area acros all archs with the
> > intention of removing it completely later. These patches get rid of 99+%
> > of all users, there's still one or two macros left using it and there are
> > still a few comments left refering to it that could be cleaned up - I'll
> > get to those, but what remains after these patches is extremely little.
> > The reason for doing this is that verify_area is just a wrapper for
> > access_ok anyway, so there's no good reason to keep it around - access_ok
> > also seems more readable anyway with saner return values.
> > Since these patches touch things all over the tree the CC list would be
> > enormous if I CC'd everyone involved on all patches, so I'll just CC this
> > initial mail to a few key people I think are relevant (I hope I got that
> > list right), and the actual patches I'll just send to linux-kernel and
> > Andrew (or directly to people who ask for that).
>
> The execution at first glance appears good.

Thank you - Andrew found a few places where I'd messed up, but corrected
those silently and merged the patches into -mm, so hopefully they can now
get some proper testing.


>Out of curiosity, does this
> serve a larger purpose than eliminating a redundant API?
>
No, it's a redundant API and one that seems harder to get right (judging
from all the comments about fixing up incorrect verify_area use all over
the place) than acces_ok. That's the only reason for getting rid of it.


--
Jesper

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/