Automatic deletion of loopback device upon umount?

[Denys Vlasenko]

On Sunday 15 April 2012 17:18, Florian Fainelli wrote:
> Le 07/04/2012 01:18, Rob Landley a écrit :
> > On 04/06/2012 01:20 AM, ralda at gmx.de wrote:
> >> Hi Florian !
> >>
> >>> 1) loopback mount "foo" to mount /bar
> >>> 2) umount /bar
> >>> 3) append new files and re-generate the "foo" cramfs image
> >>> 4) loopback mount "foo" to mount /bar
> >>> 5) the contents of /bar are the same as in 1) and not 3)
> >>
> >>> Obviously using umount -d in 2) fixes the issue, but I was wondering
> >>> whether it would not be preferable to unconditionnaly delete the
> >>> loopback device upon umount? util-linux does this actually, so other
> >>> users might also be puzzled by such a case.
> >>
> >> I hit that too, some time ago, not cramfs but squashfs and ISO images.
> >> That was the reason I added an "alias umount='umount -d'" to
> >> my /etc/profile and added the "-d" to all umounts in scripts.
> >>
> >> IMHO it would be better to reverse definition of the "-d" option to
> >> umount and do NOT delete the loop device if option gets specified and
> >> drop/delete it in the default case.
> >
> > You mean the way I originally wrote it before this commit broke it?
> >
> > b2e578a1f2c3cf317b391a7d2c059d6a5f5368b8 is the first bad commit
> > commit b2e578a1f2c3cf317b391a7d2c059d6a5f5368b8
> > Author: Denis Vlasenko<vda.linux at googlemail.com>
> > Date:   Thu Feb 14 12:00:21 2008 +0000
> >
> >      umount: instead of non-standard -D, use -d with opposite meaning
> >        (closes bug 1604)
> >
> > I have no idea what bug 1604 was, but leaking loopback devices was
> > wrong.  I had code to automatically clean them up, it ran by default,
> > and now it doesn't.
> 
> Denys, what do you think about this? I agree with Rob here, not having 
> the loopback deleted by default is definitively confusing.

I agree, and I will implement that if util-linux's umount does the same.
Does it?

-- 
vda