From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Tue, 19 Oct 2010 14:52:04 +0100 (BST) From: edward.robbins@domain.hid MIME-Version: 1.0 Content-Type: text/plain;charset=UTF-8 Content-Transfer-Encoding: quoted-printable In-Reply-To: <4CB85C64.9040307@domain.hid> References: <1281709057.650414331@domain.hid> <4C65583D.3030800@domain.hid> <1281960823.030215299@domain.hid> <4C692FC3.5040802@domain.hid> <1283531433.789315367@domain.hid> <1283590285.1709.2153.camel@domain.hid> <1287134519.029617765@domain.hid> <1287143691.028416668@domain.hid> <4CB85C64.9040307@domain.hid> Message-ID: <1287496324.10921156@domain.hid> Subject: Re: [Xenomai-help] Very high latencies under stress testing List-Id: Help regarding installation and common use of Xenomai List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gilles Chanteperdrix Cc: xenomai@xenomai.org Having disabled MTRR (and leaving in Phillipes patch) I am not getting unre= asonable latencies unless I enable graphics acceleration. Can I be sure thi= s has solved the issue?=0A=0AEdward=0A=0AOn Friday, 15 October, 2010 2:51pm= , "Gilles Chanteperdrix" said:=0A=0A> ed= ward.robbins@domain.hid wrote:=0A>> I am also finding that ta= sks are frequently hanging, causing=0A>> ipipetraces with "INFO: task xxx:x= xx blocked for more than 120=0A>> seconds". Ihave just had this with apt-ge= t update, and was previously=0A>> getting itwith grub-install. Could this b= e related?=0A> =0A> Yes. There is obviously an issue in your patched kernel= , we need to find=0A> this first before we draw any conclusion from any oth= er behaviour.=0A> =0A> From the trace you sent, it looks like it happens in= the MTRR code, but=0A> it could also be an SMI, or some other bug. For deb= ugging purposes,=0A> could you try disabling MTRRs in the kernel configurat= ion?=0A> =0A> --=0A> =09=09=09=09=09 Gilles.=0A> =0A