From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <xfs-bounces@oss.sgi.com>
Received: from relay.sgi.com (relay3.corp.sgi.com [198.149.34.15])
	by oss.sgi.com (Postfix) with ESMTP id D75997F53
	for <xfs@oss.sgi.com>; Fri, 22 Aug 2014 10:01:38 -0500 (CDT)
Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15])
	by relay3.corp.sgi.com (Postfix) with ESMTP id 75CB6AC003
	for <xfs@oss.sgi.com>; Fri, 22 Aug 2014 08:01:35 -0700 (PDT)
Received: from sandeen.net (sandeen.net [63.231.237.45]) by cuda.sgi.com with
	ESMTP id qWL7Cn5esSpvzhhB for <xfs@oss.sgi.com>;
	Fri, 22 Aug 2014 08:01:33 -0700 (PDT)
Message-ID: <53F75B54.8050802@sandeen.net>
Date: Fri, 22 Aug 2014 10:01:40 -0500
From: Eric Sandeen <sandeen@sandeen.net>
MIME-Version: 1.0
Subject: Re: [PATCH 3/4] xfs: combine xfs_rtmodify_summary and
	xfs_rtget_summary
References: <53F6942B.80808@redhat.com> <53F6963D.9090500@sandeen.net>
	<20140822131950.GC3915@laptop.bfoster>
In-Reply-To: <20140822131950.GC3915@laptop.bfoster>
List-Id: XFS Filesystem from SGI <xfs.oss.sgi.com>
List-Unsubscribe: <http://oss.sgi.com/mailman/options/xfs>,
	<mailto:xfs-request@oss.sgi.com?subject=unsubscribe>
List-Archive: <http://oss.sgi.com/pipermail/xfs>
List-Post: <mailto:xfs@oss.sgi.com>
List-Help: <mailto:xfs-request@oss.sgi.com?subject=help>
List-Subscribe: <http://oss.sgi.com/mailman/listinfo/xfs>,
	<mailto:xfs-request@oss.sgi.com?subject=subscribe>
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Errors-To: xfs-bounces@oss.sgi.com
Sender: xfs-bounces@oss.sgi.com
To: Brian Foster <bfoster@redhat.com>
Cc: Eric Sandeen <sandeen@redhat.com>, xfs-oss <xfs@oss.sgi.com>

On 8/22/14, 8:19 AM, Brian Foster wrote:
> On Thu, Aug 21, 2014 at 08:00:45PM -0500, Eric Sandeen wrote:

...

>> @@ -480,15 +484,40 @@ xfs_rtmodify_summary(
>>  		}
>>  	}
>>  	/*
>> -	 * Point to the summary information, modify and log it.
>> +	 * Point to the summary information, modify/log it, and/or copy it out.
>>  	 */
>>  	sp = XFS_SUMPTR(mp, bp, so);
>> -	*sp += delta;
>> -	xfs_trans_log_buf(tp, bp, (uint)((char *)sp - (char *)bp->b_addr),
>> -		(uint)((char *)sp - (char *)bp->b_addr + sizeof(*sp) - 1));
>> +	if (delta) {
>> +		uint first = (uint)((char *)sp - (char *)bp->b_addr);
>> +
>> +		*sp += delta;
>> +		xfs_trans_log_buf(tp, bp, first, first + sizeof(*sp) - 1);
>> +	}
>> +	if (sum) {
>> +		/*
>> +		 * Drop the buffer if we're not asked to remember it.
>> +		 */
>> +		if (!rbpp)
>> +			xfs_trans_brelse(tp, bp);
> 
> This introduces some potentially weird circumstances (e.g., acquire,
> log, release of a buffer), but I think it's resolved by the next patch.
> 
> Reviewed-by: Brian Foster <bfoster@redhat.com>

does it introduce it, or just highlight it?  I thought it was weird too,
but I think it existed before; that's what prompted me to go looking at
callers and drop the rbpp checks, FWIW.

-Eric

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs