[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 02/19] block: Add BDS.backing_overridden
From: |
Kevin Wolf |
Subject: |
Re: [Qemu-devel] [PATCH 02/19] block: Add BDS.backing_overridden |
Date: |
Tue, 3 May 2016 15:34:08 +0200 |
User-agent: |
Mutt/1.5.21 (2010-09-15) |
Am 03.05.2016 um 12:49 hat Max Reitz geschrieben:
> On 02.05.2016 17:35, Kevin Wolf wrote:
> > Am 26.04.2016 um 23:32 hat Max Reitz geschrieben:
> >> If the backing file is overridden, this most probably does change the
> >> guest-visible data of a BDS. Therefore, we will need to consider this in
> >> bdrv_refresh_filename().
> >>
> >> Adding a new field to the BDS is not nice, but it is very simple and
> >> exactly keeps track of whether the backing file has been overridden.
> >>
> >> Signed-off-by: Max Reitz <address@hidden>
> >
> > What about live snapshots or block jobs that manipulate bs->backing?
> > They don't use bdrv_open_backing_file(), but bdrv_append() or
> > bdrv_replace_in_backing_chain() in order to assign a backing file, which
> > may or may not be the same as in the image header.
>
> I'll take a look into each case, thanks.
>
> > Also, do we want to consider a backing file overridden when the user
> > specified it explicitly, but it matches the image header?
>
> For simplicity's sake, I would. I think we've realized before that
> trying to identify whether to file names point to the same image is
> rather difficult.
I was just asking because in that case, we probably must consider things
like the backing file of a live snapshot overlay overridden as well (at
least when using an existing image).
Kevin
pgpByYSJvOaEu.pgp
Description: PGP signature