From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tianyang Chen Subject: Re: [Bug] sched: credit2: Assertion failed Date: Tue, 2 Feb 2016 13:10:56 -0500 Message-ID: <56B0F130.4050605@seas.upenn.edu> References: <56B04A3D.1040209@seas.upenn.edu> <1454401798.9227.22.camel@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1454401798.9227.22.camel@citrix.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: Dario Faggioli , "xen-devel@lists.xen.org" , George Dunlap , Meng Xu List-Id: xen-devel@lists.xenproject.org On 2/2/2016 3:29 AM, Dario Faggioli wrote: > On Tue, 2016-02-02 at 01:18 -0500, Tianyang Chen wrote: >> The following script caused an unresponsive dom0 and it can not be >> reproduced all the time. The dom0 is using credit2 scheduler. >> >> #!/bin/bash >> xl cpupool-list -c >> xl cpupool-cpu-remove Pool-0 3 >> xl cpupool-cpu-remove Pool-0 2 >> xl cpupool-create name=\"r\" sched=\"credit\" >> xl cpupool-create name=\"t\" sched=\"credit\" >> xl cpupool-cpu-add r 3 >> xl cpupool-cpu-add t 2 >> > Yes, this is known. It is because Credit2 does not support (hard) > affinity. There is a series on the list by someone implementing that, > but it needs refreshing, which I hope to get round to it soon. > >> Another bug can be reproduced(not all the time) if the newly created >> pool is also credit2. There is an assertion failure (see bug.txt), >> which >> is essentially the same thing I got when trying to make the new rtds >> scheduler cpupool safe. >> > This is also known, and it will be fixed when I'll get back and > resubmit this: > > http://lists.xenproject.org/archives/html/xen-devel/2015-09/msg03811.html > > In fact, look at this patch: > http://lists.xenproject.org/archives/html/xen-devel/2015-09/msg03813.html > > And I'll do that right in this days, so I'd say you can ignore this for > RTDS (as it will get fixed by the same series). > > So both are known issues... I'll open bugs in our bugtraker for them to > avoid forgetting or tripping over them again. > Got it, thanks for the heads up. -- Tianyang