[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1458304756.3600.3.camel@gmail.com>
Date: Fri, 18 Mar 2016 13:39:16 +0100
From: Mike Galbraith <umgwanakikbuti@...il.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Xiong Zhou <jencce.kernel@...il.com>,
Thomas Gleixner <tglx@...utronix.de>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Ingo Molnar <mingo@...nel.org>, Borislav Petkov <bp@...en8.de>,
Andreas Herrmann <aherrmann@...e.com>
Subject: Re: 4.5.0+ panic when setup loop device
On Fri, 2016-03-18 at 11:15 +0100, Peter Zijlstra wrote:
> On Fri, Mar 18, 2016 at 05:11:54AM +0100, Mike Galbraith wrote:
> > On Thu, 2016-03-17 at 10:52 +0100, Peter Zijlstra wrote:
> >
> > > Andreas; Borislav said to Cc you since you wrote all this.
> > > The issue is that Linux assumes:
> > >
> > > > nr_logical_cpus = nr_cores * nr_siblings
> >
> > It also seems to now assume that if SMT is possible, it's enabled.
> >
> > Below is my 8 socket DL980 G7, which has SMT turned off for RT
> > testing,
> > booting NOPREEMPT master tuned for maximum bloat ala distro and
> > getting
> > confused by me telling it (as always) nr_cpus=64. Bad juju ensues.
>
> Ah, did you actually disable HT in the BIOS, or just skip the HT
> enumeration by saying nr_cpus=64 (knowing that all the siblings are
> last)?
It's disabled in BIOS.
> In any case, Thomas has a clue and I'm going to test, but 4 socket
> machine takes forever to boot, so might be a few minutes :/
His one-liner made my DL980 all better.
-Mike
Powered by blists - more mailing lists