From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: 2.6.27.5 guest boot failure using in-kernel PIT Date: Fri, 21 Nov 2008 08:54:56 +0100 Message-ID: <49266950.40200@web.de> References: <20081120142253.GA30825@blackpad> <20081120220517.GB30825@blackpad> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enigA1E8395BEA2335A1F64C031F" Cc: kvm@vger.kernel.org To: Eduardo Habkost Return-path: Received: from fmmailgate02.web.de ([217.72.192.227]:32864 "EHLO fmmailgate02.web.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752220AbYKUHzw (ORCPT ); Fri, 21 Nov 2008 02:55:52 -0500 In-Reply-To: <20081120220517.GB30825@blackpad> Sender: kvm-owner@vger.kernel.org List-ID: This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigA1E8395BEA2335A1F64C031F Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Eduardo Habkost wrote: > On Thu, Nov 20, 2008 at 12:22:53PM -0200, Eduardo Habkost wrote: >> Hi, >> >> When using a kvm.git kernel as host, I am getting guest boot failures >> when booting Fedora Rawhide kernel (2.6.27.5-117.fc10.x86_64). Guest >> stops booting at: >> >> ENABLING IO-APIC IRQs >> ..TIMER: vector=3D0x30 apic1=3D0 pin1=3D0 apic2=3D-1 pin2=3D-1 >> ..MP-BIOS bug: 8254 timer not connected to IO-APIC >> ...trying to set up timer (IRQ0) through the 8259A ... >> ..... (found apic 0 pin 0) ... >> ....... failed. >> ...trying to set up timer as Virtual Wire IRQ... >> ..... failed. >> ...trying to set up timer as ExtINT IRQ... >=20 > I've just found out this problem happens because the guest has HZ=3D100= 0 > and the host had HZ=3D250 and no CONFIG_HIGH_RES_TIMERS. >=20 > With this setup, the host is not managing to inject enough timer > interrupts during the mdelay() loop on timer_irq_works(). >=20 Interesting, and plausible. My observation so far is a sporadic test failure, often correlating with some raised host OS load. I'm running a high-res kernel, but that cannot prevent that this only 10 ticks long loop of the guest may obtain too few CPU cycles to handle enough of them once in a while (IIRC, it needs 4 out of the 10 ticks to declare the timer routing functional). Maybe Gleb's anti-coalesce patches for the PIC can also deal with your timer resolution conflict. At least worth a try... Jan --------------enigA1E8395BEA2335A1F64C031F 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 iEYEARECAAYFAkkmaVcACgkQniDOoMHTA+kyOgCfZGvfA5TiNypZ+Z+Blgeh8X6S u54An2fJc16BAFUPB/APDODZXnRhXERc =LN+e -----END PGP SIGNATURE----- --------------enigA1E8395BEA2335A1F64C031F--