On Thu, Oct 04, 2012 at 05:08:48PM +0800, Jeremy Kerr wrote:
> Hi Matt,
>
> >Jeremy, did you want to pick this up as part of your series?
>
> I have this in my series, yes. I'm just working on the authenticated
> delete code, then will send out the next revision.
>
> Speaking of which - Peter: I've realised that doing a GetVariable()
> after the SetVariable is a much cleaner way of checking whether we
> need to drop the inode after an authenticated delete.
>
> Because the inode's size may not be simply updated by the write(),
> we should probably be doing a GetVariable() after an APPEND_WRITE,
> to read the new size. Rather than having a separate (and quite
> complex) code path to check for the delete case, we may as well use
> the same logic to determine if the variable has been deleted.
Yeah, this is fine.
> >Actually, shouldn't the new filesystem be called "efivarfs", or "efifs"
> >to make it more explicit that it is a filesystem?
>
> I'm not too fussed about the name, but +1 for efivarfs.
I like efivarfs for the name, as well.
> >We also need something in Documentation/filesystems/ describing the old
> >EFI variable method, why it's no longer any good, and why the new
> >filesystem is favoured. Does anybody have anything to add to the
> >following?
>
> Looks good to me. We can always elaborate later, if necessary.
Yep.
--
Peter
--
To unsubscribe from this list: send the line "unsubscribe linux-efi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
