From owner-freebsd-hackers  Tue Mar 11 02:02:49 1997
Return-Path: <owner-hackers>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id CAA10410
          for hackers-outgoing; Tue, 11 Mar 1997 02:02:49 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA10405
          for <freebsd-hackers@FreeBSD.ORG>; Tue, 11 Mar 1997 02:02:44 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id UAA07451; Tue, 11 Mar 1997 20:54:47 +1100
Date: Tue, 11 Mar 1997 20:54:47 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199703110954.UAA07451@godzilla.zeta.org.au>
To: bde@zeta.org.au, terry@lambert.org
Subject: Re: Hard Link Count too small!
Cc: freebsd-hackers@FreeBSD.ORG, jez@netcraft.co.uk
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>>> >They must not be very POSIX dependent, or they would fail from the
>>> >"shall mark for update"/"shall update" discrepancies introduced by
>>> >the async mount.  An async mounted FS is not POSIX compliant.  An
>>> 
>>> Wrong.
>>
>>"SHALL BE UPDATED".
>>
>>Not "SHALL BE WRITTEN TO CACHE AND MAYBE UPDATED SOMETIME".
>
>Right.  Updating consists of converting a mark (which is usually
>implemented as a single bit, e.g., IN_ACCESS in ufs) to a time.  This has
>nothing to do with caching.  POSIX does not specify storage in RAM,

PS: async mounting only has an indirect effect on writing timestamps to
disk anyway.  Timestamps are normally written to the diskwhen the update
daemon syncs everything or as a side effect of writing inodes synchronously
for reasons unrelated to timestamps.

Bruce