From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [RFC][PATCH 1/3] Introduce a workqueue to deliver PIT timer interrupts. Date: Wed, 09 Jun 2010 17:16:47 +0300 Message-ID: <4C0FA24F.60204@redhat.com> References: <1276019703-18136-1-git-send-email-clalance@redhat.com> <1276019703-18136-2-git-send-email-clalance@redhat.com> <4C0EFE20.1080508@redhat.com> <20100609132335.GA7396@amt.cnet> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Zachary Amsden , Chris Lalancette , kvm@vger.kernel.org To: Marcelo Tosatti Return-path: Received: from mx1.redhat.com ([209.132.183.28]:55934 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757138Ab0FIOQw (ORCPT ); Wed, 9 Jun 2010 10:16:52 -0400 Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) by mx1.redhat.com (8.13.8/8.13.8) with ESMTP id o59EGpkf023130 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Wed, 9 Jun 2010 10:16:51 -0400 Received: from cleopatra.tlv.redhat.com (cleopatra.tlv.redhat.com [10.35.255.11]) by int-mx01.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with ESMTP id o59EGm7A031871 for ; Wed, 9 Jun 2010 10:16:48 -0400 In-Reply-To: <20100609132335.GA7396@amt.cnet> Sender: kvm-owner@vger.kernel.org List-ID: On 06/09/2010 04:23 PM, Marcelo Tosatti wrote: > >> Is there any chance that using a workqueue might help the problem of >> hrtimers firing too quickly? I wanted to return HR_NORESTART from >> pit_timer_fn always, then restart the hrtimer on delivery, but >> because of unreliable delivery, it wasn't clear how to do that. >> >> Perhaps the workqueue can be used to restart the timer instead, >> avoiding problems of impossibly small timeouts causing hrtimers to >> run amok. >> > It should be rearmed on ACK ideally. How come delivery is unreliable? > > BTW, a massive amount of hrtimers (_single_ non-tickless 32-vcpu guest > with LAPIC) results in: > > hrtimer: interrupt took 122448 ns > > in the host. So don't even need impossibly small timeouts :( > Any idea where that came from? Even looping on vcpus, that took 4 usec per vcpu. -- error compiling committee.c: too many arguments to function