Strange things on 2.6.19/20 for a dual-core CPU

Strange things on 2.6.19/20 for a dual-core CPU

[Pavel Troller]

Hi!
  I posted the following question, when 2.6.19 was freshly out. However, nobody
has answered. OK, I told myself, let's get things to stabilize, and I waited
patiently for 2.6.20. Now, the things are absolutely the same, and IMHO wrong.
Could anybody look at this and decide, whether it is a real bug, which has to
be fixed, or not ?
      With regards, Pavel Troller
      
----- Forwarded message from Pavel Troller <patrol@sinus.cz> -----

From: Pavel Troller <patrol@sinus.cz>
To: linux-acpi@vger.kernel.org
Subject: Strange things on 2.6.19 for a dual-core CPU
Mail-Followup-To: Pavel Troller <patrol@sinus.cz>,
	linux-acpi@vger.kernel.org
Content-Disposition: inline
User-Agent: Mutt/1.5.9i

Hi!
  I've updated to vanilla 2.6.19 on my Pentium-D (dual-core x86_64) box.
Now I can't see even C1 in the /proc/acpi/processor/*/power output:
patrol@arcus:~$ cat /proc/acpi/processor/CPU1/power
active state:            C0
max_cstate:              C8
bus master activity:     00000000
maximum allowed latency: 2000 usec
states:
patrol@arcus:~$ cat /proc/acpi/processor/CPU2/power
active state:            C0
max_cstate:              C8
bus master activity:     00000000
maximum allowed latency: 2000 usec
states:

  Another interesting thing is shown here:
patrol@arcus:~$ cat /proc/acpi/processor/CPU1/info
processor id:            0
acpi id:                 1
bus mastering control:   no
power management:        no
throttling control:      yes
limit interface:         yes
patrol@arcus:~$ cat /proc/acpi/processor/CPU2/info
processor id:            1
acpi id:                 2
bus mastering control:   no
power management:        no
throttling control:      no
limit interface:         no

  As I remember, both cores were showing the same things formerly.
  The only line referring to CPUs during boot is
ACPI: Processor [CPU1] (supports 8 throttling states)
  and CPU2 is not mentioned at all.

  The last (but maybe not acpi-related) strange thing is that in /proc/cpuinfo,
CPU1 reports 6403.56 bogomips (as always, approximately twice the clock) and CPU2
8314.32 ones (too much). It's also very suspicious. Formerly the difference was
very small.

  Should I provide more info to debug these things, or is it OK ?
         With regards, Pavel Troller

----- End forwarded message -----

Re: Strange things on 2.6.19/20 for a dual-core CPU

[Luming Yu]

Well, enter a acpi bug on bugzilla with acpidump output and dmesg attached.

>   Another interesting thing is shown here:
> patrol@arcus:~$ cat /proc/acpi/processor/CPU1/info
> processor id:            0
> acpi id:                 1
> bus mastering control:   no
> power management:        no
> throttling control:      yes
> limit interface:         yes
> patrol@arcus:~$ cat /proc/acpi/processor/CPU2/info
> processor id:            1
> acpi id:                 2
> bus mastering control:   no
> power management:        no
> throttling control:      no
> limit interface:         no
>
>   As I remember, both cores were showing the same things formerly.
>   The only line referring to CPUs during boot is
> ACPI: Processor [CPU1] (supports 8 throttling states)
>   and CPU2 is not mentioned at all.
>
>   The last (but maybe not acpi-related) strange thing is that in /proc/cpuinfo,
> CPU1 reports 6403.56 bogomips (as always, approximately twice the clock) and CPU2
> 8314.32 ones (too much). It's also very suspicious. Formerly the difference was
> very small.
>
>   Should I provide more info to debug these things, or is it OK ?
>          With regards, Pavel Troller
>
> ----- End forwarded message -----
> -
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

Re: Strange things on 2.6.19/20 for a dual-core CPU

[Sergio Monteiro Basto]

[-- Attachment #1: Type: text/plain, Size: 414 bytes --]

On Mon, 2007-02-05 at 14:34 +0800, Luming Yu wrote:
> Well, enter a acpi bug on bugzilla with acpidump output and dmesg
> attached.

I put my acpidump here 
http://bugzilla.kernel.org/show_bug.cgi?id=6419#c65
this is the correct way to post acpidump , or do you prefer binary ?

acpidump send to stderr this message: "Wrong checksum for generic
table!"
is this a problem ? 

Thanks,
-- 
Sérgio M.B.

[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 2192 bytes --]

Re: Strange things on 2.6.19/20 for a dual-core CPU

[Sergio Monteiro Basto]

[-- Attachment #1: Type: text/plain, Size: 2924 bytes --]

On Mon, 2007-02-05 at 07:04 +0100, Pavel Troller wrote:
> Hi!
>   I posted the following question, when 2.6.19 was freshly out. However, nobody
> has answered. OK, I told myself, let's get things to stabilize, and I waited
> patiently for 2.6.20. Now, the things are absolutely the same, and IMHO wrong.
> Could anybody look at this and decide, whether it is a real bug, which has to
> be fixed, or not ?

Hi , I also have a Pentium D with a very strange things 
http://bugzilla.kernel.org/show_bug.cgi?id=6419

have you test yours in older kernels? 


>       With regards, Pavel Troller
>       
> ----- Forwarded message from Pavel Troller <patrol@sinus.cz> -----
> 
> From: Pavel Troller <patrol@sinus.cz>
> To: linux-acpi@vger.kernel.org
> Subject: Strange things on 2.6.19 for a dual-core CPU
> Mail-Followup-To: Pavel Troller <patrol@sinus.cz>,
> 	linux-acpi@vger.kernel.org
> Content-Disposition: inline
> User-Agent: Mutt/1.5.9i
> 
> Hi!
>   I've updated to vanilla 2.6.19 on my Pentium-D (dual-core x86_64) box.
> Now I can't see even C1 in the /proc/acpi/processor/*/power output:
> patrol@arcus:~$ cat /proc/acpi/processor/CPU1/power
> active state:            C0
> max_cstate:              C8
> bus master activity:     00000000
> maximum allowed latency: 2000 usec
> states:
> patrol@arcus:~$ cat /proc/acpi/processor/CPU2/power
> active state:            C0
> max_cstate:              C8
> bus master activity:     00000000
> maximum allowed latency: 2000 usec
> states:
> 
>   Another interesting thing is shown here:
> patrol@arcus:~$ cat /proc/acpi/processor/CPU1/info
> processor id:            0
> acpi id:                 1
> bus mastering control:   no
> power management:        no
> throttling control:      yes
> limit interface:         yes
> patrol@arcus:~$ cat /proc/acpi/processor/CPU2/info
> processor id:            1
> acpi id:                 2
> bus mastering control:   no
> power management:        no
> throttling control:      no
> limit interface:         no
> 
>   As I remember, both cores were showing the same things formerly.
>   The only line referring to CPUs during boot is
> ACPI: Processor [CPU1] (supports 8 throttling states)
>   and CPU2 is not mentioned at all.
> 
>   The last (but maybe not acpi-related) strange thing is that in /proc/cpuinfo,
> CPU1 reports 6403.56 bogomips (as always, approximately twice the clock) and CPU2
> 8314.32 ones (too much). It's also very suspicious. Formerly the difference was
> very small.
> 
>   Should I provide more info to debug these things, or is it OK ?
>          With regards, Pavel Troller
> 
> ----- End forwarded message -----
> -
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
-- 
Sérgio M.B.

[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 2192 bytes --]