xfs
[Top] [All Lists]

Re: [PATCH] xfs: Improve scalability of busy extent tracking

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH] xfs: Improve scalability of busy extent tracking
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Thu, 22 Apr 2010 13:08:49 -0400
Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>, xfs@xxxxxxxxxxx
In-reply-to: <20100422161626.GE23541@dastard>
References: <1271828835-2094-1-git-send-email-david@xxxxxxxxxxxxx> <20100422110143.GA21867@xxxxxxxxxxxxx> <20100422161626.GE23541@dastard>
User-agent: Mutt/1.5.19 (2009-01-05)
Been looking at this a bit and I have a theory:

 - a tid is not actually unique to a xfs_trans structure, if
   we call xfs_trans_dup a single xlog_ticket, and with that the
   tid is re-used by multiple transaction structure.
 - because of that the major semantic change in the version vs
   the previous one is that we now do not force the synchronous
   transaction for the case where we re-used a block in
   the rolled over transaction.

Still not quite sure about the implications of this.

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