RE: [RFC PATCH 2/3] add statmnt(2) syscall

[David Laight <David.Laight@ACULAB.COM>]

From: Jeff Layton
> Sent: 18 September 2023 15:30
....
> A bit tangential to this discussion, but one thing we could consider is
> adding something like a mnt_change_cookie field that increments on any
> significant changes on the mount: i.e. remounts with new options,
> changes to parentage or propagation, etc.
> 
> That might make it more palatable to do something with separate syscalls
> for the string-based fields. You could do:
> 
> statmnt(...);
> getmntattr(mnt, "mnt.fstype", ...);
> statmnt(...);
> 
> ...and then if the mnt_change_cookie hasn't changed, you know that the
> string option was stable during that window.

That would also help with the problem of the mount options
being changed while processing a page fault on the user buffer.

Of, with a call to just get the cookie, could find that nothing
has changed so there is no point looking again.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)