All of lore.kernel.org
 help / color / mirror / Atom feed
From: Carsten Schiers <carsten@schiers.de>
To: xen-devel <xen-devel@lists.xensource.com>
Subject: AW: AW: Re: No C-States any longer...
Date: Wed, 8 Jun 2011 15:51:39 +0200	[thread overview]
Message-ID: <27587646.81307541099106.JavaMail.root@uhura> (raw)
In-Reply-To: <15131694.51307540285845.JavaMail.root@uhura>

Isn't it rather in xen/arch/x86/acpi/cpu_idle.c? I ask because a xm debug-key "c" 
doesn't print anything, but C-States should be recognized and managed by xen and 
not by the Dom0 kernel, shouldn't it?

I found a line like this in function set_cx_pminfo by the way:

/* FIXME: C-state dependency is nut supported by far */

But maybe it's the same Intel crew that has done it...

BR,
Carsten.

----- Originalnachricht -----
Von: Ian Campbell <Ian.Campbell@eu.citrix.com>
Gesendet: Mit, 8.6.2011 11:20
An: Carsten Schiers <carsten@schiers.de>
Cc: xen-devel <xen-devel@lists.xensource.com> ; yunhong.jiang@intel.com ; kevin.tian@intel.com ; ke.yu@intel.com
Betreff: Re: AW: Re: [Xen-devel] No C-States any longer...

I'm really not sure how this stuff is supposed to work.

I've CC'd some Intel folks who authored/touched
driver/acpi/processor_xen.c and drivers/xen/acpi_processor.c which I
think is the glue which is supposed to make this stuff work.

Ian.

On Tue, 2011-06-07 at 21:16 +0100, Carsten Schiers wrote:
> I tested with Xen 4.1.0 and latest xen.git#xen/stable-2.6.32: still not 
> working.
> Find attached some files. I currently have no working system to produce 
> equivalents.
> 
> BR,
> Carsten.
> 
> -----Ursprüngliche Nachricht-----
> Von: Ian Campbell [mailto:Ian.Campbell@citrix.com] 
> Gesendet: Montag, 6. Juni 2011 10:31
> An: Carsten Schiers
> Cc: xen-devel
> Betreff: Re: [Xen-devel] No C-States any longer...
> 
> On Sat, 2011-06-04 at 17:18 +0100, Carsten Schiers wrote:
> > after my move to Xen 4.1.0 and Debian 2.6.32-5-xen-amd64 pvops Dom0. 
> > Seems to be a problem with ACPI.
> > Can it be a BIOS problem (although it worked up to now) or is this not 
> 
> > implemented? 
> 
> It may not have been implemented when Debian took their snapshot of
> pvops for the Squeeze release or there may have been missing bugfixes.
> Can you try the latest xen.git#xen/stable-2.6.32?
> 
> > Some infos:
> 
> It might be interesting to see the equivalent on a working system. Also
> a complete dmesg from each case would be useful.
> 
> Ian.
> 
> > 
> > xm dmesg after xm debug-key c:
> > --------------------------------
> > (XEN) 'c' pressed -> printing ACPI Cx structures
> > 
> > xenpm get-cpuidle-states 0
> > --------------------------
> > Max C-state: C7
> > 
> > cpu id               : 0
> > total C-states       : 0
> > idle time(ms)        : 0
> > pc3                  : [00000000000000000018 ms]
> > pc6                  : [00000000000000004294 ms]
> > pc7                  : [00000000000140733193 ms]
> > cc3                  : [00000000000000000000 ms]
> > cc6                  : [00000000000000000006 ms]
> > 
> > xm info
> > -------
> > host                   : data
> > release                : 2.6.32-5-xen-amd64
> > version                : #1 SMP Thu May 19 01:16:47 UTC 2011
> > machine                : x86_64
> > nr_cpus                : 3
> > nr_nodes               : 1
> > cores_per_socket       : 3
> > threads_per_core       : 1
> > cpu_mhz                : 2210
> > hw_caps                : 
> > 
> 178bf3ff:efd3fbff:00000000:00001310:00802001:00000000:000037ff:00000000
> > virt_caps              : hvm
> > total_memory           : 4094
> > free_memory            : 373
> > free_cpus              : 0
> > xen_major              : 4
> > xen_minor              : 1
> > xen_extra              : .0
> > xen_caps               : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32 
> 
> > hvm-3.0-x86_32p hvm-3.0-x86_64
> > xen_scheduler          : credit
> > xen_pagesize           : 4096
> > platform_params        : virt_start=0xffff800000000000
> > xen_changeset          : unavailable
> > xen_commandline        : dom0_mem=256M dom0_vcpus_pin cpuidle 
> > cpufreq=xen
> > cc_compiler            : gcc version 4.4.5 (Debian 4.4.5-8)
> > cc_compile_by          : root
> > cc_compile_domain      :
> > cc_compile_date        : Fri Jun  3 17:03:43 CEST 2011
> > xend_config_format     : 4
> > 
> > 
> > 
> > 
> > _______________________________________________
> > Xen-devel mailing list
> > Xen-devel@lists.xensource.com
> > http://lists.xensource.com/xen-devel
> 
> 
> 
> 

       reply	other threads:[~2011-06-08 13:51 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <15131694.51307540285845.JavaMail.root@uhura>
2011-06-08 13:51 ` Carsten Schiers [this message]
2011-06-08 13:56   ` AW: AW: Re: No C-States any longer Ian Campbell
2011-06-09 11:41 AW: AW: " Carsten Schiers
2011-06-09 11:45 ` Ian Campbell

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=27587646.81307541099106.JavaMail.root@uhura \
    --to=carsten@schiers.de \
    --cc=xen-devel@lists.xensource.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.