[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 5 May 2011 10:21:15 +0800
From: Shaohua Li <shaohua.li@...el.com>
To: Jeff Garzik <jgarzik@...ox.com>
Cc: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-ide@...r.kernel.org" <linux-ide@...r.kernel.org>,
"jaxboe@...ionio.com" <jaxboe@...ionio.com>,
"htejun@...il.com" <htejun@...il.com>,
"hch@...radead.org" <hch@...radead.org>,
"djwong@...ibm.com" <djwong@...ibm.com>,
"sshtylyov@...sta.com" <sshtylyov@...sta.com>
Subject: Re: [patch v3 1/3] block: add a non-queueable flush flag
On Thu, May 05, 2011 at 10:17:39AM +0800, Jeff Garzik wrote:
> On 05/04/2011 09:59 PM, shaohua.li@...el.com wrote:
> > flush request isn't queueable in some drives. Add a flag to let driver
> > notify block layer about this. We can optimize flush performance with the
> > knowledge.
> >
> > Signed-off-by: Shaohua Li<shaohua.li@...el.com>
> > ---
> > block/blk-settings.c | 6 ++++++
> > include/linux/blkdev.h | 7 +++++++
> > 2 files changed, 13 insertions(+)
>
> hmmm.
>
> This assumes that flush on new hardware, by default, is queueable.
>
> I think the sense should be reversed: don't enable the optimization,
> unless we know the optimization works.
>
> That seems safer than always enabling the optimization, unless we know
> it does not work. That is not a fail-safe mode of operation.
This assumes flush is queueable by default. but I only enable the optimization
for non-queueable flush. So the optimization is off by default, please see
the second patch.
Thanks,
Shaohua
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists