[Top] [All Lists]

Re: [PATCH 0/10] Perag lookup rework

To: Alex Elder <aelder@xxxxxxx>
Subject: Re: [PATCH 0/10] Perag lookup rework
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Fri, 15 Jan 2010 09:37:25 +1100
Cc: xfs@xxxxxxxxxxx
In-reply-to: <1AB9A794DBDDF54A8A81BE2296F7BDFE012A696E@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
References: <1263210469-22171-1-git-send-email-david@xxxxxxxxxxxxx> <1AB9A794DBDDF54A8A81BE2296F7BDFE012A696E@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
User-agent: Mutt/1.5.18 (2008-05-17)
On Thu, Jan 14, 2010 at 12:42:12PM -0600, Alex Elder wrote:
> Dave Chinner wrote:
> > This is the previous set of patches all put together in the one series
> > for the current xfs-dev tree.
> > 
> > Patch 10 is a new patch - it embeds the busy extent list into the perag
> > structure rather than leaving as a separate allocation.
> All 10 patches look good.  I am adding the following descriptions
> to these patches which lacked them:
> http://patchwork.xfs.org/patch/429/
> [03/10] XFS: Convert filestreams code to use per-ag get/put routines
> Description:
> Use xfs_perag_get() and xfs_perag_put() in the filestreams code.

I'd consier that redundant and therefore unnecssary. I don't put
a descriptio in if the subject line describes the patch
sufficiently. It seems to be a pretty common practise, actually.

> http://patchwork.xfs.org/patch/417/
> [09/10] xfs: handle ENOMEM correctly during initialisation of perag structures
> Description:
> Add proper error handling in case an error occurs while initializing
> new perag structures for a mount point.  The mount structure is
> restored to its previous state by deleting and freeing any perag
> structures added during the call.

That describes what the code does, which I'd also consider
redundant... ;)


Dave Chinner

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