[patch] mount fails to call external programs

[Yoshihiro Ota]

On Wed, 31 Jan 2007 23:31:36 -0800
Garrett Cooper <youshi10 at u.washington.edu> wrote:

> Yoshihiro Ota wrote:
> > Hello.
> > 
> > The 'mount' program attempts to maintain external mount programs in order to determine whether
> > it needs to call external programs or handle it internally.
> > 
> > Not all external programs can be listed in mount.c.  Indeed, my mount_md failed after mount
> > trying to handle it itself.
> > 
> > The solution is to maintain interally handlable fs types.
> > 
> > I am not fully sure if I listed all of them correctly; however, it fixes the problem.
> > 
> > Thanks,
> > Hiro
> 
> Unfortunately your patch voids the point of having the filesystems 
> listed there, according to the author of the comment directly above 
> structure you have shown.
> 
> I honestly don't know why anyone would want to statically define that 
> though, because mount_* searching in $PATH should suffice, and the 
> nmount interface could be maintained in each sourcefile. Just my 
> thoughts on the subject though..
> 
> -Garrett

It sounds that you are confused.  You are refering to the comment I copied below, aren't you?

        /* XXX: We need to get away from implementing external mount
         *      programs for every filesystem, and move towards having
         *      each filesystem properly implement the nmount() system call.
         */

This is saying that nmount() system call needs to handle various file system types.  Instead, I am saying that NO ONE can MAINTAIN all external file system names in the list here.

I expect that all of FUSE file systems and mount programs that can be installed from ports will fail because they are not listed in here.

They are on totally diffrent subjects.  My patch does not conflict with what the comment says.

Regards,
Hiro