On Mon, Jan 29, 2018 at 05:01:41PM +0100, Max Reitz wrote:
> On 2017-12-26 03:52, Fam Zheng wrote:
[...]
> So the previous patch makes the use of blank lines consistent and this
> one breaks it again? :-)
>
> > +If specified, @code{qemu-img} will open the image with shared permissions,
> > +which makes it less likely to conflict with a running guest's permissions
> > due
> > +to image locking. For example, this can be used to get the image
> > information
> > +(with 'info' subcommand) when the image is used by a running guest. Note
> > that
> > +this could produce inconsistent results because of concurrent metadata
> > changes,
> > +etc. This option is only allowed when opening images in read-only mode.
>
> I personally don't quite like the "makes it less likely to conflict",
> because that makes it sound like qemu would be stupid and need a nudge
> in the right direction -- when it's actually the user who does something
> a bit risky (and qemu is right in forbidding it by default). But since
> it's only a read-only thing, I won't actually object to it.
>
> (Maybe it should document more exactly what's happening, i.e. that this
> option will allow concurrent writers (as a standard user, I wouldn't
> know what "shared permissions" is supposed to mean).)
Eerie -- Although I reviewed it, I was just mulling over this wording
yesterday (while I was tweaking the wording of '--force-share' on a
slide for a public conference).
Indeed, "concurrent writers" is much clearer.
[...]
--
/kashyap
