My brain hurts. (Messing with mount.)

[Rob Landley]

On Tuesday 07 March 2006 3:16 am, Bernhard Fischer wrote:
> On Mon, Mar 06, 2006 at 02:55:37PM -0500, Rob Landley wrote:
> >On Monday 06 March 2006 3:31 am, Denis Vlasenko wrote:
> >> > "mount -o remount,rw" does require at least one argument.
> >>
> >> Makes sense too. More to it, passing _two_ arguments would be rather
> >> confusing.
> >
> >But you can.  Probably doesn't do what you expect... :)
>
> Didn't look, just curious if the various -onoac,noacl,xyz_xattr,...
> are supported?

Anything we don't recognize gets passed on to the kernel.

The way the mount syscall works is you feed it your two arguments (source and 
directory), plus a bitfield full of flags (from sys/mounts.h) to tell the VFS 
what to do, and a filesystem options string containing comma-separated 
fields.

To clarify, the bitfield is for the Linux Virtual FileSystem.  It's stuff that 
applies to all filesystems (like "read only" or remount) or else doesn't 
involve any specific filesystem (like bind).  The options string is 
per-filesystem options, the generic parts of the kernel have no idea what it 
means, it just passes that on verbatim to the ext2 driver or whatever.

The way the mount command works blurs the two together.  It recognizes 
specific options from the options string (like "noatime") that are actually 
VFS options rather than filesystem options, splices them out, and sets the 
appropriate flag for the VFS.  Anything that isn't recognized is assumed to 
be a filesystem option.

Rob
-- 
Never bet against the cheap plastic solution.