From mboxrd@z Thu Jan 1 00:00:00 1970 From: George Dunlap Subject: Re: [xen-unstable test] 17139: regressions - FAIL Date: Thu, 7 Mar 2013 16:47:54 +0000 Message-ID: <5138C4BA.7020506@eu.citrix.com> References: <51389ABE02000078000C3ECD@nat28.tlf.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <51389ABE02000078000C3ECD@nat28.tlf.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Jan Beulich Cc: xen-devel List-Id: xen-devel@lists.xenproject.org On 07/03/13 12:48, Jan Beulich wrote: >>>> On 07.03.13 at 13:38, xen.org wrote: >> flight 17139 xen-unstable real [real] >> http://www.chiark.greenend.org.uk/~xensrcts/logs/17139/ >> >> Regressions :-( >> >> Tests which did not succeed and are blocking, >> including tests which could not be run: >> test-amd64-i386-xl-credit2 5 xen-boot fail REGR. vs. 16772 > George, > > pretty likely something you want to take a look at: > > Mar 7 02:39:54.552252 (XEN) **************************************** > Mar 7 02:39:54.560215 (XEN) Panic on CPU 0: > Mar 7 02:39:54.560256 (XEN) Assertion 'rt_credit >= 0' failed at sched_credit2.c:1559 > Mar 7 02:39:54.560314 (XEN) **************************************** Very interesting... it seems that under certain circumstances, even if the scheduler sets a timer to pre-empt a VM for 200us, that the VM may run for >11ms. Since the "reset" assumes that your negative credit will never be less than -200us, it merely adds 10ms. So even after the reset, the VM is still in negative credits. Let me see what I want to do about that... -George