Re: The reason to call it 3.0 is the desktop (was Re: [OT] 2.6 not 3.0 - (NUMA)) (fwd)

Re: The reason to call it 3.0 is the desktop (was Re: [OT] 2.6 not 3.0 - (NUMA)) (fwd)

[Alexander Viro]

> Logically, the second /var mount should be "mount --move /initrd/var /var", 
> followed by "umount /initrd" to free up the initrd memory.  Right now it's 
> doing "mount -n --bind /initrd/var /var", because /etc is a symlink into /var 
> (has to remain editable, you see), and this way the information about which 
> partition var actually is can be kept in one place.  (This is an 
> implementation detail: I could have used volume labels instead.)
> 
> The point is, right now I can't free the initial ramdisk because it has an 
> active mount point under it..

umount -l
mount --move

Re: The reason to call it 3.0 is the desktop (was Re: [OT] 2.6 not 3.0 - (NUMA)) (fwd)

[Hans Reiser]

Alexander Viro wrote:

>  
>
>>Logically, the second /var mount should be "mount --move /initrd/var /var", 
>>followed by "umount /initrd" to free up the initrd memory.  Right now it's 
>>doing "mount -n --bind /initrd/var /var", because /etc is a symlink into /var 
>>(has to remain editable, you see), and this way the information about which 
>>partition var actually is can be kept in one place.  (This is an 
>>implementation detail: I could have used volume labels instead.)
>>
>>The point is, right now I can't free the initial ramdisk because it has an 
>>active mount point under it..
>>    
>>
>
>umount -l
>mount --move
>
>
>
>
>  
>
It seems Linux evolves faster than I can track.  These are nice features.:)

Hans

Re: The reason to call it 3.0 is the desktop (was Re: [OT] 2.6 not 3.0 - (NUMA)) (fwd)

[Alexander Viro]

On Mon, 14 Oct 2002, Hans Reiser wrote:

> Alexander Viro wrote:
> 
> >umount -l
> >mount --move
> >
> It seems Linux evolves faster than I can track.  These are nice features.:)

Why, thank you...

They had been there for a year or so, actually - (-l first, then --move).
Both were done as side effects of core cleanups that allowed per-process
namespaces - real work was in massaging infrastructure into the sane
shape, features came pretty much for free...

Re: The reason to call it 3.0 is the desktop (was Re: [OT] 2.6 not 3.0 - (NUMA)) (fwd)

[Rob Landley]

On Sunday 13 October 2002 09:09 pm, Alexander Viro wrote:
> > Logically, the second /var mount should be "mount --move /initrd/var
> > /var", followed by "umount /initrd" to free up the initrd memory.  Right
> > now it's doing "mount -n --bind /initrd/var /var", because /etc is a
> > symlink into /var (has to remain editable, you see), and this way the
> > information about which partition var actually is can be kept in one
> > place.  (This is an implementation detail: I could have used volume
> > labels instead.)
> >
> > The point is, right now I can't free the initial ramdisk because it has
> > an active mount point under it..
>
> umount -l
> mount --move

Cool.  Thanks.

Rob

(Serves me right for still having Red Hat 7.2 on my laptop.  Old man pages.  
Now I've got to find a new project to force myelf to learn VFS internals.  Oh 
well... :)

(Nit-pick:  the man page description of umount -l doesn't look like it'd help 
with the removable media problem, I.E. "umount --gimme_my_cd_back_NOW", but 
the code may disagree, and the discussion's already turned up a 2.4 patch 
from Tirgran via Hugh Dickens, so I'll shut up now. :)