From owner-cvs-all  Sat Aug 21  1:20:18 1999
Delivered-To: cvs-all@freebsd.org
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.26.10.9])
	by hub.freebsd.org (Postfix) with ESMTP
	id A4BDC14E20; Sat, 21 Aug 1999 01:19:58 -0700 (PDT)
	(envelope-from bde@godzilla.zeta.org.au)
Received: (from bde@localhost)
	by godzilla.zeta.org.au (8.8.7/8.8.7) id SAA12283;
	Sat, 21 Aug 1999 18:18:44 +1000
Date: Sat, 21 Aug 1999 18:18:44 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199908210818.SAA12283@godzilla.zeta.org.au>
To: green@FreeBSD.org, phk@critter.freebsd.dk
Subject: Re: cvs commit: src/sys/kern kern_physio.c
Cc: cvs-all@FreeBSD.org, cvs-committers@FreeBSD.org
Sender: owner-cvs-all@FreeBSD.ORG
Precedence: bulk

>>>   Revision  Changes    Path
>>>   1.37      +8 -2      src/sys/kern/kern_physio.c
>>
>>-                       bp->b_blkno = btodb(uio->uio_offset);
>>+                       blockno = uio->uio_offset >> DEV_BSHIFT;
>>+                       bp->b_blkno = blockno;
>>+                       if (bp->b_blkno != blockno) {
>>			^- How can this comparison ever fail?
>
>They're different sizes.

They're different types (long and quad_t), but have the same size on some 
machines (e.g., alphas).  The check is written to make it difficult for
compilers to optimise it away on these machines.  It should be written
something like:

	blockno = uio->uio_offset >> DEV_BSHIFT;
	if ((daddr_t)blockno != blockno)
		barf();
	bp->b_blkno = blockno;

or

	blockno = uio->uio_offset >> DEV_BSHIFT;
	if (blockno > DADDR_T_MAX)
		barf();
	bp->b_blkno = blockno;

btodb() probably shouldn't cast to daddr_t.  This mainly prevents its
use in code like the above.  It is not a feature that it may prevent
warnings about truncation when off_t's are assigned to daddr_t's.

Bruce


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message