From mboxrd@z Thu Jan 1 00:00:00 1970 From: NeilBrown Subject: Re: [patch 3/3] raid5: relieve lock contention in get_active_stripe() Date: Wed, 11 Sep 2013 11:34:12 +1000 Message-ID: <20130911113412.5ac91a65@notabene.brown> References: <20130905162910.179ea808@notabene.brown> <20130905091822.GA8401@kernel.org> <20130909043318.GA27517@kernel.org> <20130910111318.1d19e8d3@notabene.brown> <20130910023555.GA17907@kernel.org> <20130910140629.702683da@notabene.brown> <20130910042438.GA16797@kernel.org> <20130910152032.48631492@notabene.brown> <20130910065912.GA12038@kernel.org> <20130910172836.43e23cbf@notabene.brown> <20130910073756.GB12038@kernel.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=PGP-SHA1; boundary="Sig_/s/7+HZm3l5mxZXXWA22.oeT"; protocol="application/pgp-signature" Return-path: In-Reply-To: <20130910073756.GB12038@kernel.org> Sender: linux-raid-owner@vger.kernel.org To: Shaohua Li Cc: linux-raid@vger.kernel.org, Dan Williams List-Id: linux-raid.ids --Sig_/s/7+HZm3l5mxZXXWA22.oeT Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Tue, 10 Sep 2013 15:37:56 +0800 Shaohua Li wrote: > Below is my latest patch. >=20 Thanks. It looks good. I have pushed it out to me for-next branch (which a few cosmetic white-space adjustments). I will need to review it again but it is certainly very close to 'right'. One thing I'm a bit concerned about is the md_raid5_congested function. It can return "false", yet a write can still block. That isn't a huge problem, but it could have some negative consequences. Maybe we could have an atomic_t which counts how many hash values as "full" and we report "congested" when any are full. Maybe. Thanks, NeilBrown --Sig_/s/7+HZm3l5mxZXXWA22.oeT Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.19 (GNU/Linux) iQIVAwUBUi/IlTnsnt1WYoG5AQIPBg/9HtcYRjxRVIDwPE5C9nfWdN3M7GZ0RebL TtfVNwGuzEnGWMyqhfvp7+CgNmQN9mOo/8sWT0YhRih48vE2sKcHhIcT0AuBzv7s 6eMOH/3SsHxzDDLOqtEfbfu/JtOjxARf8MORtYQgFpy1xW7qK/oBxivnaqShHSNn FHmc2rRCDK+A12ghBTctuN9vvrGctV+LCe2A011Za4JkW1bpmAPWHKgpJTDROKkj DzEX+tgMMmyen3Sn9NU5nFiK7i5vFNsgh/oSAnmLNS0CZnR5zUxTdqSi64yip7IB MwfPd3gZv6nh8TaHG3YVktsiILf6ipE4XEPJAhmV1NCjbjCCTm47IzOkpT1CQbI2 vWaem3sKRKu1NsEXW1zURKZNmJudx9MSXJIepGitdSyP/e0nLrhGIAH/kBxv0JF0 7M/AwNlFDPdpO8beKfqB0gOVBybVDkAMofvbuYJ5MtfojdKU3zJBxDklti0wR4SD 8USnaYi2QPsRAA0rlM8NCHBExUVx7LQk5KZbK5IYlOFjDJelQ7GwhV+H58JgMOxC JGp5wEBiShGBv4NlSvqIkV0dSWBt/AWSU8S46UqEWhGdLFbn3XTbQSWZsJXOqPQN A3s5Xvl2AijlsncH7OnvzwKnw3wyJquBe84sJ4XlZmMokR3FKH301LT3U0yvAI0k rJc+94HsSCw= =DgKi -----END PGP SIGNATURE----- --Sig_/s/7+HZm3l5mxZXXWA22.oeT--