Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 8 Aug 1995 02:19:55 -0700 (PDT)
From:      "Rodney W. Grimes" <rgrimes@gndrsh.aac.dev.com>
To:        simon@masi.ibp.fr
Cc:        freebsd-hackers@freebsd.org
Subject:   Re: patch adding the 'noauto' switch to 'mount'
Message-ID:  <199508080919.CAA03196@gndrsh.aac.dev.com>
In-Reply-To: <199508080842.KAA00358@neuromancer.ibp.fr> from "simon@masi.ibp.fr" at Aug 8, 95 10:42:21 am

next in thread | previous in thread | raw e-mail | index | archive | help
> 
> 
> Hi,
> 
> > > +#define 	MNT_NOAUTO	0x00000003	/* don't mount with 'mount -a' */
> > 
> > WRONG VALUE!!!  These are bit fields, not integers.  MNT_NOAUTO is now
> > the same as -o ro,sync :-(.  MNT_NOAUTO does _not_ belong on in this
> > set of bit patterns, the kernel could care less about this!  We are also
> > out of bits for some structures that try to store this value :-(.
> 
> Argh :-O 
> 
> You're right on both points. The kernel doesn't even see that flag, since
> it's processed prior to the actual mounting of the FS, and it would
> be a Bad Thing to waste a precious bit-mask value on it (the first 
> available one is 0x04000000).
> 
> We need to have MNT_NOAUTO defined, though.
> how about :
> 
> #define MNT_NOAUTO      0x00000000

You do not need it defined if it's value is arbitrary, that is just non-sense.
Try again...

> This shouldn't interfere with the existing flags (right ?) and it wouldn't
> really waste a value.

Dangerious thing to define tokens with arbitrary values.  Same day it will
create a bug :-(.

> What do you think ? Any other ideas ?

Not without taking a serious look at the code.



-- 
Rod Grimes                                      rgrimes@gndrsh.aac.dev.com
Accurate Automation Company                 Reliable computers for FreeBSD



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199508080919.CAA03196>