Re: [PATCH] utime(s): Honour CAP_FOWNER when times==NULL

[Satyam Sharma]

On Mon, 16 Jul 2007, Linus Torvalds wrote:
> 
> 
> On Tue, 17 Jul 2007, Satyam Sharma wrote:
> >
> > [PATCH] utime(s): Honour CAP_FOWNER when times==NULL
> > 
> > do_utimes() does not honour CAP_FOWNER when times==NULL.
> > Trivial and obvious one-line fix.
> 
> Ahh, ok. Is this old, or was it introduced recently (I'm looking at my 
> recent change to that area, it doesn't seem to introduce this)?

Old, I'd think. Note that this issue was hidden behind the fact that
we would call vfs_permission()->generic_permission() anyway even if
that check failed.

Now generic_permission() returns 0 (success) if capable(CAP_DAC_OVERRIDE)
and -EACCES otherwise (that's for MAY_WRITE tests, which is what this
one is). Now I suspect most (all?) common userspace programs out there
just don't ever have a capable(CAP_DAC_OVERRIDE) && !capable(CAP_FOWNER)
kind of capability mix, so we never saw this before.

[ That does not mean this is not an issue, of course, though. ]

> IOW, how did you even notice this?

Just code inspection. In fact I got interested in following this
codepath precisely after the recent discussion on this list ...

> Also, while your one-liner looks correct, it does make me wonder: the 
> whole
> 
> 	if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
> 
> test is a rather common test, and in fact, arguably, every time you see 
> one part of it, you should probably see the other. Would it make sense to 
> make a helper inline function to do this, and replace all users? Doing a
> 
> 	git grep 'fsuid.*\<i_uid\>'
> 
> seems to show quite a few cases of this pattern..

Yes, I thought of writing a helper function for this myself. The semantics
of CAP_FOWNER sort of justify that, but probably better to get Al's views
on this first.

Satyam
-
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/