public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Borislav Petkov <bp@amd64.org>
To: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: "Luck, Tony" <tony.luck@intel.com>, Ingo Molnar <mingo@elte.hu>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"Huang, Ying" <ying.huang@intel.com>, Avi Kivity <avi@redhat.com>
Subject: Re: [PATCH 05/10] MCE: Mask out address mask bits below address granuality
Date: Fri, 10 Jun 2011 11:46:37 +0200	[thread overview]
Message-ID: <20110610094637.GD1621@aftab> (raw)
In-Reply-To: <4DF1D0B1.1020302@jp.fujitsu.com>

On Fri, Jun 10, 2011 at 04:07:13AM -0400, Hidetoshi Seto wrote:
> (2011/06/10 6:33), Luck, Tony wrote:
> > From: Andi Kleen <andi@firstfloor.org>
> > 
> > SER enabled systems report the address granuality for each
> > reported address in a machine check. But the bits below
> > the granuality are undefined. Mask them out before
> > logging the machine check.
> > 
> > Signed-off-by: Andi Kleen <ak@linux.intel.com>
> > Signed-off-by: Tony Luck <tony.luck@intel.com>
> > ---
> >  arch/x86/kernel/cpu/mcheck/mce.c |   12 +++++++++++-
> >  1 files changed, 11 insertions(+), 1 deletions(-)
> > 
> > diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
> > index 0349e87..ffc8d11 100644
> > --- a/arch/x86/kernel/cpu/mcheck/mce.c
> > +++ b/arch/x86/kernel/cpu/mcheck/mce.c
> > @@ -539,8 +539,18 @@ static void mce_read_aux(struct mce *m, int i)
> >  {
> >  	if (m->status & MCI_STATUS_MISCV)
> >  		m->misc = mce_rdmsrl(MSR_IA32_MCx_MISC(i));
> > -	if (m->status & MCI_STATUS_ADDRV)
> > +	if (m->status & MCI_STATUS_ADDRV) {
> >  		m->addr = mce_rdmsrl(MSR_IA32_MCx_ADDR(i));
> > +
> > +		/*
> > +		 * Mask the reported address by the reported granuality.
> > +		 */
> > +		if (mce_ser && (m->status & MCI_STATUS_MISCV)) {
> > +			u8 shift = m->misc & 0x1f;
> > +			m->addr >>= shift;
> > +			m->addr <<= shift;
> > +		}
> > +	}
> >  }
> >  
> >  DEFINE_PER_CPU(unsigned, mce_poll_count);
> 
> Why do you have to mask it out in kernel, why not in user/logger?
> 
> One possible story is:
>   "... the brand-new Xeon XXXX has new MCx_***_VALID bit in ****
>   register, if it is set the lower bits of MCx_ADDR indicates
>   ****, otherwise the bits are undefined ..."
> 
> So I think that kernel should convey the raw value from hardware to
> userland.  Even if it contains some noise on it, user can determine
> whether it is useful or not.  And more, since this is an error record,
> there will be no second chance to retrieve the data afterward.

I think they need the correct address for the poisoning later:

                if (severity == MCE_AO_SEVERITY && mce_usable_address(&m))
                        mce_ring_add(m.addr >> PAGE_SHIFT);


Tony?

-- 
Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Gemeinde Aschheim, Landkreis Muenchen
Registergericht Muenchen, HRB Nr. 43632

  reply	other threads:[~2011-06-10  9:46 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-09 21:25 [RFC] reworked machine check recovery patches Luck, Tony
2011-06-09 21:29 ` [PATCH 01/10] MCE: fixes for mce severity table Luck, Tony
2011-06-09 21:30 ` [PATCH 02/10] MCE: save most severe error information Luck, Tony
2011-06-10  8:06   ` Hidetoshi Seto
2011-06-10 18:08     ` Tony Luck
2011-06-09 21:31 ` [PATCH 03/10] MCE: introduce mce_gather_info() Luck, Tony
2011-06-09 21:32 ` [PATCH 04/10] MCE: Move ADDR/MISC reading code into common function Luck, Tony
2011-06-10  9:33   ` Borislav Petkov
2011-06-10 18:17     ` Tony Luck
2011-06-09 21:33 ` [PATCH 05/10] MCE: Mask out address mask bits below address granuality Luck, Tony
2011-06-10  8:07   ` Hidetoshi Seto
2011-06-10  9:46     ` Borislav Petkov [this message]
2011-06-10 19:06     ` Tony Luck
2011-06-11  0:12       ` Andi Kleen
2011-06-10  9:42   ` Borislav Petkov
2011-06-10 19:09     ` Tony Luck
2011-06-09 21:34 ` [PATCH 06/10] HWPOISON: Handle hwpoison in current process Luck, Tony
2011-06-10  8:07   ` Hidetoshi Seto
2011-06-10 20:36     ` Tony Luck
2011-06-09 21:35 ` [PATCH 07/10] MCE: replace mce.c use of TIF_MCE_NOTIFY with user_return_notifier Luck, Tony
2011-06-10  8:08   ` Hidetoshi Seto
2011-06-10 20:42     ` Tony Luck
2011-06-11 10:24       ` Borislav Petkov
2011-06-12  8:31       ` Avi Kivity
2011-06-12  8:29   ` Avi Kivity
2011-06-12 10:24     ` Borislav Petkov
2011-06-12 10:30       ` Avi Kivity
2011-06-12 13:53         ` Borislav Petkov
2011-06-09 21:36 ` [PATCH 08/10] NOTIFIER: Take over TIF_MCE_NOTIFY and implement task return notifier Luck, Tony
2011-06-12 22:38   ` Borislav Petkov
2011-06-13  5:31     ` Tony Luck
2011-06-13  7:59       ` Avi Kivity
2011-06-13  9:55         ` Borislav Petkov
2011-06-13 11:40           ` Avi Kivity
2011-06-13 12:40             ` Borislav Petkov
2011-06-13 12:47               ` Avi Kivity
2011-06-13 15:12                 ` Borislav Petkov
2011-06-13 16:31                   ` Avi Kivity
2011-06-13 17:13                     ` Tony Luck
2011-06-14  2:50                       ` Hidetoshi Seto
2011-06-14  2:51                         ` [PATCH 1/2] x86, mce: introduce mce_memory_failure_process Hidetoshi Seto
2011-06-14  2:53                         ` [PATCH 2/2] x86, mce: rework use of TIF_MCE_NOTIFY Hidetoshi Seto
2011-06-14 18:02                           ` Tony Luck
2011-06-14 18:28                             ` Tony Luck
2011-06-15  1:29                               ` Hidetoshi Seto
2011-06-15  2:10                                 ` Tony Luck
2011-06-15  3:17                                   ` Hidetoshi Seto
2011-06-14  3:09                         ` [PATCH 08/10] NOTIFIER: Take over TIF_MCE_NOTIFY and implement task return notifier Tony Luck
2011-06-14 11:40                       ` Avi Kivity
2011-06-14 13:33                         ` Borislav Petkov
2011-06-14 13:43                           ` Avi Kivity
2011-06-14 17:13                             ` Luck, Tony
2011-06-15  8:51                               ` Avi Kivity
2011-06-14 16:59                         ` Luck, Tony
2011-06-15  8:52                           ` Avi Kivity
2011-06-13 16:43               ` Tony Luck
2011-06-09 21:37 ` [PATCH 09/10] MCE: run through processors with more severe problems first Luck, Tony
2011-06-10  8:09   ` Hidetoshi Seto
2011-06-10 20:49     ` Tony Luck
2011-06-13 22:03       ` Tony Luck
2011-06-14  1:27         ` Hidetoshi Seto
2011-06-14  3:04           ` Tony Luck
2011-06-09 21:38 ` [PATCH 10/10] MCE: Add Action-Required support Luck, Tony
2011-06-10  8:06   ` Hidetoshi Seto
2011-06-10 21:00     ` Tony Luck

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110610094637.GD1621@aftab \
    --to=bp@amd64.org \
    --cc=avi@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=seto.hidetoshi@jp.fujitsu.com \
    --cc=tony.luck@intel.com \
    --cc=ying.huang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox