Re: [PATCH] remove "dev=" mount option special processing for df

[Paul Eggert]

"Paul A. Clarke" <address@hidden> writes:

> I don't think the GPFS use of "dev=" is documented.

That's OK.  Hardly any of this is documented.  What does GPFS use
"dev=" for?  Does the patch I sent work for GPFS?

> why is the "df" command looking for "dev=" in the mount options of
> filesystems which don't even support it?

We don't know what file systems support dev= and what don't.  It's
cheap to look for dev= so we might as well do it.  In the absence of
better info, we assume that people use "dev=" for the same thing
they've been using "dev=" for decades.  This may be wrong, but I
suspect that in practice it's more likely to be right than the other
approach would be.

> Is there a documented standard to which this conforms?

There's no standard that I know of.

> shouldn't the "Irix 6.5" code be conditionally compiled
> only for Irix 6.5?

The feature is likely to leak into other operating systems.  Most
likely, a file system with type "ignore" is one that we'd want to
ignore anyway, even if it was defined independently.  So I'd leave the
code alone for now.

> This won't work if "dev=" is the first option.

It never is, in my experience.  If it happens we could add code to
support it.