[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180817170129.GE24945@kroah.com>
Date: Fri, 17 Aug 2018 19:01:29 +0200
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: Jinpu Wang <jinpu.wang@...fitbricks.com>
Cc: Thomas Gleixner <tglx@...utronix.de>,
stable <stable@...r.kernel.org>, linux-kernel@...r.kernel.org,
Ingo Molnar <mingo@...nel.org>, konrad.wilk@...cle.com
Subject: Re: [BUG]smt sysfs dir missing on 4.4.148 and 4.14.63
On Fri, Aug 17, 2018 at 02:39:00PM +0200, Jinpu Wang wrote:
> On Thu, Aug 16, 2018 at 6:48 PM Jinpu Wang <jinpu.wang@...fitbricks.com> wrote:
> >
> > On Thu, Aug 16, 2018 at 6:39 PM Greg Kroah-Hartman
> > <gregkh@...uxfoundation.org> wrote:
> > >
> > > On Thu, Aug 16, 2018 at 06:28:59PM +0200, Jinpu Wang wrote:
> > > > On Thu, Aug 16, 2018 at 5:51 PM Greg Kroah-Hartman
> > > > <gregkh@...uxfoundation.org> wrote:
> > > > >
> > > > > On Thu, Aug 16, 2018 at 05:43:23PM +0200, Jinpu Wang wrote:
> > > > > > Hi Greg, hi Thomas,
> > > > > >
> > > > > > I noticed /sys/devices/system/cpu/smt dir is missing on 4.4.148 and
> > > > > > 4.14.63, default setting.
> > > > >
> > > > > For 4.4, I would be amazed if anything was there, the backport was done
> > > > > only to handle the local user issue, not the "full" protection. So
> > > > > please try 4.9 or 4.14.
> > > > Thanks for quick reply. Right, just checked. the patch for create the
> > > > sysfs entry is missing.
> > >
> > > What patch is that? What is the git commit id that we missed?
> > >
> > > > > How about 4.17 and 4.18, do those trees work properly for you?
> > > > I guess it's the same.
> > > > >
> > > > > > Tried stable/master branch 31130a16d459 ("Merge tag
> > > > > > 'for-linus-4.19-rc1-tag' of
> > > > > > git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip")
> > > > > > It's the same there.
> > > > >
> > > > > Ah, so we are bug compatible, good :)
> > > > I will try test stable master a bit more to gather the full log.
> > > > >
> > > > > > When boot with 'nosmt' kernel paramter kernel 4.14.63 panic during
> > > > > > boot, 4.4.148 boot fine.
> > > > > > The call trace seem irq related, is it known bug?
> > > > >
> > > > > You are not giving us much to go on here sorry. We need full details,
> > > > > and also please cc: lkml and the other developers who worked on this.
> > > > > THomas is on vacation this week.
> > > > Now, add more developers to cc, also lkml.
> > > >
> > > > I have the config and boot dmesg attached for default setting.
> > >
> > > I see no error in your boot log, what exactly are you worried about?
> > 2 problems here:
> > 1 According to https://www.kernel.org/doc/html/latest/admin-guide/l1tf.html
> > I expect /sys/devices/system/cpu/smt/, but it's not.
> > 2 kernel 4.14.63 panic when I boot with 'nosmt' or 'l1tf=full', I
> > attached the picture here again,
> > I will try to get the full log tomorrow, also test with linus tree.
> > >
> > > confused,
> > >
> > > greg k-h
> >
> > Thanks
> I found the problem, CONFIG_HOTPLUG_SMT is enable due to CONFIG_SMP,
> but I did explicitly enable CONFIG_HOTPLUG_CPU.
>
> That's why the smt dir is missing, and kernel panic when I boot with
> 'nosmt' or 'l1tf=full' probably also other combination.
>
> With CONFIG_HOTPLUG_CPU=y, it works fine so far.
So do we need a new configuration dependancy here to resolve this
problem? Why is this not an issue in Linus's tree?
confused,
greg k-h
Powered by blists - more mailing lists