[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] Re: [RFC] [PATCH 0/2] Some clean-up to MCA handling



On Tuesday 20 April 2010 10:06:12 Jiang, Yunhong wrote:
> >> It is not "easier" internalhandling. In fact, it makes no difference to
> >> internal handling at all. The reason is: 1) In amd_f10.c, it will only
> >> occupy 4 mc_msr,
> >
> >well, 4 in the generic handler plus 3 MSRs via mcinfo_extended which have
> >been introduced in family10h.
>
> So it means 16 -3 extra mc_msrs be wasted at least, for each
> mcinfo_extended :-)

I consider this as 'reserved space for the future' :-)

>
> >> while in Intel platform, it may occupy 32 mc_msr, that is
> >> sure to cost extra space. The mc_info buffer is quite limited and can't
> >> be expanded in run time, so reduce the size is quite important.
> >> 2) sizeof(void *) is different in 64 hypervisor and 32 bit dom0. I'm not
> >> sure if it is tested in compatibility mode, which might be confused.
> >>
> >> In fact, since we have mc_msrs included in mcinfo_extended already, the
> >> caller can get the size of the buffer quite easy.
> >>
> >> Of course, if you *really* don't care the waste of size in AMD platform,
> >> it's ok for me. After all, in intel platform, either there is no
> >> extended information, or it will occupy all of them, so it really does
> >> not matter to me. But the (void*) issue should be resolved, I suspect.
> >
> >Is it possible to change the internal infrastructure to deal with multiple
> >mc_info's ? The user (Dom0) will keep to see just one because the switch
> >happens underneath.
>
> I think the size limitation happens on two side. Firstly, there is only 10
> mc_info reserved in urgent queue, and 20 in non-urgent queue, and that
> queue size is not adjusted dynamical; Secondly, each mc_info contains 768
> uint64_t.

The API allows the user (Dom0) to fetch a mc_info again and again. So what
didn't fit into the first can be placed into the next.
The reason to have it static is that the MCE implementation in both xen and in 
Dom0 can be lockless (xmalloc/xfree use locks).

>
> >What does not fit into one mc_info will be put into the next.
> >
> >In xen you will need some operations:
> >lowlevel: allocate, free, switch, read, write
> >highlevel: get and put
> >
> >The mce code itself should just use the highlevel operations and also just
> >see one mc_info. The highlevel operations see as many mc_info as needed
> > and use the lowlevel operations which work on mc_info directly.
> >
> >Does that make sense to you ?
>
> I suspect this will cause much more complex. Also will this cause trouble
> to ABI also, since the mc_info is defined in public?

Yes, that is to make mc_info look dynamic from the internal side.

> As I have no data for MCE/CMCI trigger model in run time, maybe we can
> postpone this changes, unless some one raise this issue?

You can ask back within your company to get a known broken but reasonable 
usable hardware (i.e. CPU L2 cache produces certain errors or DIMM plugged
in slot 3 produces certain errors).
Then you know what error types you can expect on such a machine and how
to deal with.

> Attached is the new patch, which does not change the interface anymore.

This patch is good.

>
> --jyh
>
> >> How about your option to the other patch?
> >
> >Still need to have a look at it.
> >
> >> Thanks
> >> --jyh
> >>
> >> >Christoph
> >> >
> >> >
> >> >--
> >> >---to satisfy European Law for business letters:
> >> >Advanced Micro Devices GmbH
> >> >Karl-Hammerschmidt-Str. 34, 85609 Dornach b. Muenchen
> >> >Geschaeftsfuehrer: Andrew Bowd, Thomas M. McCoy, Giuliano Meroni
> >> >Sitz: Dornach, Gemeinde Aschheim, Landkreis Muenchen
> >> >Registergericht Muenchen, HRB Nr. 43632
> >
> >--
> >---to satisfy European Law for business letters:
> >Advanced Micro Devices GmbH
> >Karl-Hammerschmidt-Str. 34, 85609 Dornach b. Muenchen
> >Geschaeftsfuehrer: Andrew Bowd, Thomas M. McCoy, Giuliano Meroni
> >Sitz: Dornach, Gemeinde Aschheim, Landkreis Muenchen
> >Registergericht Muenchen, HRB Nr. 43632



-- 
---to satisfy European Law for business letters:
Advanced Micro Devices GmbH
Karl-Hammerschmidt-Str. 34, 85609 Dornach b. Muenchen
Geschaeftsfuehrer: Andrew Bowd, Thomas M. McCoy, Giuliano Meroni
Sitz: Dornach, Gemeinde Aschheim, Landkreis Muenchen
Registergericht Muenchen, HRB Nr. 43632


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.