From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: I/O performance of VirtIO Date: Mon, 26 Oct 2009 09:12:44 +0100 Message-ID: <4AE559FC.4060400@web.de> References: <20091012204901.GA10688@nightfall.luchs.at> <4AD3A38D.3090102@msgid.tls.msk.ru> <4AD41FB8.6000209@web.de> <4AE0885A.1080304@redhat.com> <7A331778-3227-43B1-B2E2-A6BE6F92C834@suse.de> <4AE3E5B0.5080700@redhat.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enigA17E4B536299BB7D42737B23" Cc: Alexander Graf , Michael Tokarev , =?UTF-8?B?UmVuw6kgUGZlaWZmZXI=?= , "kvm@vger.kernel.org" To: Avi Kivity Return-path: Received: from fmmailgate02.web.de ([217.72.192.227]:60318 "EHLO fmmailgate02.web.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755232AbZJZIMo (ORCPT ); Mon, 26 Oct 2009 04:12:44 -0400 In-Reply-To: <4AE3E5B0.5080700@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigA17E4B536299BB7D42737B23 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Avi Kivity wrote: > On 10/23/2009 12:06 AM, Alexander Graf wrote: >> >> Am 22.10.2009 um 18:29 schrieb Avi Kivity : >> >>> On 10/13/2009 08:35 AM, Jan Kiszka wrote: >>>> It can be particularly slow if you use in-kernel irqchips and the >>>> default NIC emulation (up to 10 times slower), some effect I always >>>> wanted to understand on a rainy day. So, when you actually want -net= >>>> user, try -no-kvm-irqchip. >>>> >>> >>> This might be due to a missing SIGIO or SIGALRM; -no-kvm-irqchip >>> generates a lot of extra signals and thus polling opportunities. >> >> Isn't that what dedicated io threads are supposed to solve? >> >=20 > No. Dedicated I/O threads provide parallelism. All latency needs is t= o > have SIGIO sent on all file descriptors (or rather, in qemu-kvm with > irqchip, to have all file descriptors in the poll() call). >=20 > Jan, does slirp add new connections to the select set? >=20 It should do so in slirp_select_fill (it iterates over all TCP&UDP sockets of all instances). I think without doing this, slirp wouldn't receive a single bit at all (no activity without FD_ISSET). Jan --------------enigA17E4B536299BB7D42737B23 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org iEYEARECAAYFAkrlWf8ACgkQitSsb3rl5xRlpQCfSH8dMNx2ldHNFCHqFRTYyon1 C3cAn2zXLtMq0qYZjevqoJZ9/97OnyQW =bSoO -----END PGP SIGNATURE----- --------------enigA17E4B536299BB7D42737B23--