[Top] [All Lists]

Re: [XFS updates] XFS public tree - master for latest XFS changes...

To: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Subject: Re: [XFS updates] XFS public tree - master for latest XFS changes...
From: Timothy Shimmin <tes@xxxxxxx>
Date: Tue, 02 Dec 2008 15:01:09 +1100
Cc: Niv Sardi <xaiki@xxxxxxx>, xfs@xxxxxxxxxxx
In-reply-to: <20081201090953.GA31696@xxxxxxxxxxxxx>
References: <200812010015.mB10F1Ab031727@xxxxxxxxxxx> <20081201064949.GL6291@disturbed> <20081201090953.GA31696@xxxxxxxxxxxxx>
User-agent: Thunderbird (Macintosh/20080914)
Christoph Hellwig wrote:
> On Mon, Dec 01, 2008 at 05:49:49PM +1100, Dave Chinner wrote:
>> On Sun, Nov 30, 2008 at 06:15:01PM -0600, Niv Sardi wrote:
>>> - Log -----------------------------------------------------------------
>>> commit 0924b585fc49bf371bc700c23e516a538bf589af
>>> Author: Dave Chinner <david@xxxxxxxxxxxxx>
>>> Date:   Fri Nov 28 14:23:34 2008 +1100
>>>     [XFS] fix uninitialised variable bug in dquot release.
>>>     gcc is warning about an uninitialised variable in xfs_growfs_rt().
>>>     This is a false positive. Fix it by changing the scope of the
>>>     transaction pointer to wholly within the internal loop inside
>>>     the function.
>> The title of that doesn't match the description. I think it
>> was supposed to be:
>> [PATCH 4/5] XFS: fix spurious uninitialised variable warning in xfs_growfs_rt
> Looks like xaiki just sucked in my staging tree where I mistpasted the
> subject line.
>> BTW, can we get a one-line summary of the commits being referenced
>> in the message? The commit hash is less than useful, and having to
>> read through several hundred lines of commit logs to determine
>> what was checked in is not fun.....
> Yeah, the new sort of commit messags aren't too useful.  I would in fact
> prefer to get the old style one mail per commit, maybe even including
> the patch that was commited.  I'm sure this is doable with git as there
> are a lot of projects that do it.
I've changed the script to list the commits with the 1 line description
as Dave suggested (hopefully what he was intending).
However, I thought the msg per push was just fine.
If you and others disagree, then I'll look at a per commit email hook I guess.


<Prev in Thread] Current Thread [Next in Thread>