From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gregory Haskins Subject: Re: [PATCH 3/5] Move irq notifiers lists to its own locking. Date: Mon, 13 Jul 2009 11:32:07 -0400 Message-ID: <4A5B5377.2060607@gmail.com> References: <1247476355-27284-1-git-send-email-gleb@redhat.com> <1247476355-27284-4-git-send-email-gleb@redhat.com> <20090713114550.GA10387@redhat.com> <20090713114844.GF28046@redhat.com> <20090713142320.GJ10402@redhat.com> <20090713143750.GS28046@redhat.com> <20090713144920.GL10402@redhat.com> <20090713152311.GN23086@redhat.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enig27616BF1B52450BBC88D6F42" Cc: "Michael S. Tsirkin" , kvm@vger.kernel.org, avi@redhat.com, mtosatti@redhat.com To: Gleb Natapov Return-path: Received: from wf-out-1314.google.com ([209.85.200.173]:48491 "EHLO wf-out-1314.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755856AbZGMPcQ (ORCPT ); Mon, 13 Jul 2009 11:32:16 -0400 Received: by wf-out-1314.google.com with SMTP id 26so785746wfd.4 for ; Mon, 13 Jul 2009 08:32:15 -0700 (PDT) In-Reply-To: <20090713152311.GN23086@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig27616BF1B52450BBC88D6F42 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Gleb Natapov wrote: > On Mon, Jul 13, 2009 at 05:49:20PM +0300, Michael S. Tsirkin wrote: > =20 >> On Mon, Jul 13, 2009 at 05:37:50PM +0300, Gleb Natapov wrote: >> =20 >>> On Mon, Jul 13, 2009 at 05:23:20PM +0300, Michael S. Tsirkin wrote: >>> =20 >>>> On Mon, Jul 13, 2009 at 02:48:44PM +0300, Gleb Natapov wrote: >>>> =20 >>>>> On Mon, Jul 13, 2009 at 02:45:51PM +0300, Michael S. Tsirkin wrote:= >>>>> =20 >>>>>> On Mon, Jul 13, 2009 at 12:12:33PM +0300, Gleb Natapov wrote: >>>>>> =20 >>>>>>> Signed-off-by: Gleb Natapov >>>>>>> =20 >>>>>> This one is probably better off left as is, >>>>>> =20 >>>>> What do you mean "as is"? >>>>> =20 >>>> This is a slow operation. It seems that we could use irq_lock or sw= itch >>>> to slot lock or kvm lock here. Why do we need another one? >>>> >>>> =20 >>> irq_lock is completely removed. So either we don't remove it and use = it >>> here (and we don't need mutex so we change it to spinlock too), or we= add >>> another lock with the name that actually tell us what its purpose. I = prefer >>> second option. I am not sure you can use kvm lock without deadlock, a= nd >>> slot lock? How this connected to slots management?! >>> >>> And this is not about speed of the operation. It is about making read= er >>> lockless. >>> =20 >> So, to summarize: this patch does not help speed irq injection up, the= >> only reason to change locking here is cosmetical. Is this a fair >> summary? >> >> =20 > The whole series helps to speed irq injection up. This patch is one ste= p > towards the goal. > =20 FWIW: Improving the injection path in the manner Gleb is proposing will pave the way to skip the work-queue deferrment in the irqfd signal path. This is a good thing. Regards, -Greg --------------enig27616BF1B52450BBC88D6F42 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.11 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpbU3cACgkQP5K2CMvXmqFi3QCeMQuIfFXMMbIx5nBgeFr46DDe M98An3KdZWaYqTypUKuj6gcTROqR/sp7 =kUMV -----END PGP SIGNATURE----- --------------enig27616BF1B52450BBC88D6F42--