[Toybox] [PATCH] df: -a should even show mounts we couldn't stat.

[Rob Landley]

On 2/12/21 7:37 PM, enh via Toybox wrote:
> In xgetmountlist(), keep track of whether we could stat or not, so we
> can distinguish an actual device 0:0 from a failed stat/statvfs.

Is there an actual device 0:0?

> Also switch to FLAG() and capitalize POSIX.

I'm tempted to break that part out and apply it just so I can see what the rest
of the changes are more clearly.

You want to print unstat-able devices with "-" as all of their human_readable()
return values? And in-band signaling is not reliable because you have devices
returning 0 as their major:minor? (This df.c isn't the only thing that's going
to fail on that...)

Rob