ERR in /proc/interrupts

ERR in /proc/interrupts

[Andreas Schwarz]

I get a very high ERR count in /proc/interrupts. If I move my USB mouse
the number increases.

What does ERR mean? Nothing good, I suppose?

Kernel: 2.6.0-test6-mm1

/proc/interrupts:
====================================================

           CPU0       
  0:    1924793          XT-PIC  timer
  1:       9571          XT-PIC  i8042
  2:          0          XT-PIC  cascade
  5:       1561          XT-PIC  ohci-hcd, eth1
  8:          2          XT-PIC  rtc
  9:          0          XT-PIC  acpi
 10:          0          XT-PIC  ehci_hcd, ohci-hcd, eth0
 11:      31910          XT-PIC  EMU10K1, uhci-hcd, uhci-hcd
 14:      12483          XT-PIC  ide0
 15:         41          XT-PIC  ide1
NMI:          0 
LOC:    1924399 
ERR:      15255
MIS:          0

-- 
AVR-Tutorial, über 350 Links
Forum für AVRGCC und MSPGCC
-> http://www.mikrocontroller.net

Re: ERR in /proc/interrupts

[Christian]

Andreas Schwarz wrote:
> I get a very high ERR count in /proc/interrupts. If I move my USB mouse
> the number increases.
> 
> What does ERR mean? Nothing good, I suppose?
> 

../Documentation/filesystems/proc.txt says:

ERR is incremented in the case of errors in the IO-APIC bus (the bus 
that connects the CPUs in a SMP system. This means that an error has 
been detected, the IO-APIC automatically retry the transmission, so it 
should not be a big problem, but you should read the SMP-FAQ.

Christian, knowing nothing more than this too :-)

Re: ERR in /proc/interrupts

[Andreas Schwarz]

Christian wrote:
> Andreas Schwarz wrote:
>> I get a very high ERR count in /proc/interrupts. If I move my USB mouse
>> the number increases.
>> 
>> What does ERR mean? Nothing good, I suppose?
>> 
>
> ../Documentation/filesystems/proc.txt says:
>
> ERR is incremented in the case of errors in the IO-APIC bus (the bus 
> that connects the CPUs in a SMP system. This means that an error has 
> been detected, the IO-APIC automatically retry the transmission, so it 
> should not be a big problem, but you should read the SMP-FAQ.

I don't have a SMP kernel. I compiled the kernel with IO APIC support,
but disabled it later with the noapic kernel boot option.

-- 
AVR-Tutorial, über 350 Links
Forum für AVRGCC und MSPGCC
-> http://www.mikrocontroller.net

Re: ERR in /proc/interrupts

[Andreas Schwarz]

Andreas Schwarz wrote:
> Christian wrote:
>> Andreas Schwarz wrote:
>>> I get a very high ERR count in /proc/interrupts. If I move my USB mouse
>>> the number increases.
>>> 
>>> What does ERR mean? Nothing good, I suppose?
>>> 
>>
>> ../Documentation/filesystems/proc.txt says:
>>
>> ERR is incremented in the case of errors in the IO-APIC bus (the bus 
>> that connects the CPUs in a SMP system. This means that an error has 
>> been detected, the IO-APIC automatically retry the transmission, so it 
>> should not be a big problem, but you should read the SMP-FAQ.
>
> I don't have a SMP kernel. I compiled the kernel with IO APIC support,
> but disabled it later with the noapic kernel boot option.

I recompiled the kernel with both APIC options disabled. Result: ERR is
0, and my USB mouse is no longer choppy.

APIC = evil (at least on my K7S5A Pro)!

-- 
AVR-Tutorial, über 350 Links
Forum für AVRGCC und MSPGCC
-> http://www.mikrocontroller.net

Re: ERR in /proc/interrupts

[Raj]

> ERR is incremented in the case of errors in the IO-APIC bus (the bus

I have observed that ERR is incremented in case of spurious interrupts
too.

/Raj

ERR in /proc/interrupts

[Jorge Boncompte (DTI2)]

    What does ERR mean in /proc/interrupts? I have a computer running
2.4.0test12 that has a lot of this ERR's?

           CPU0
  0:  116445752          XT-PIC  timer
  1:     389614          XT-PIC  keyboard
  2:          0          XT-PIC  cascade
  5:   34298837          XT-PIC  eth1
  8:          1          XT-PIC  rtc
 10:  400182075          XT-PIC  eth0
 11:   23181909          XT-PIC  ide0, ide1
 14:          4          XT-PIC  ide2
 15:     692215          XT-PIC  ide3
 64:          0            none  acpi
NMI:          0
ERR:    1586756

This is an AMD 800 + Tyan K7 mobo.

    -Jorge

==============================================================
Jorge Boncompte - Técnico de sistemas
DTI2 - Desarrollo de la Tecnología de las Comunicaciones
--------------------------------------------------------------
C/ Abogado Enriquez Barrios, 5   14004 CORDOBA (SPAIN)
Tlf: +34 957 761395 / FAX: +34 957 450380
--------------------------------------------------------------
jorge@dti2.net _-_-_-_-_-_-_-_-_-_-_-_-_-_ http://www.dti2.net
==============================================================
Without wicker a basket cannot be done.
==============================================================


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: ERR in /proc/interrupts

[Tim Walberg]

[-- Attachment #1: Type: text/plain, Size: 2232 bytes --]

A quick perusal of the 2.2.18 source code (I don't have
a copy of 2.4.x handy) shows that it's directly
related to the number of IPIs (inter-processor
interrupts) the system has taken. I'm not real
sure under what conditions this occurs, but it's
someplace to start...


			tw


On 01/18/2001 22:54 +0100, Jorge Boncompte (DTI2) wrote:
>>	    What does ERR mean in /proc/interrupts? I have a computer running
>>	2.4.0test12 that has a lot of this ERR's?
>>	
>>	           CPU0
>>	  0:  116445752          XT-PIC  timer
>>	  1:     389614          XT-PIC  keyboard
>>	  2:          0          XT-PIC  cascade
>>	  5:   34298837          XT-PIC  eth1
>>	  8:          1          XT-PIC  rtc
>>	 10:  400182075          XT-PIC  eth0
>>	 11:   23181909          XT-PIC  ide0, ide1
>>	 14:          4          XT-PIC  ide2
>>	 15:     692215          XT-PIC  ide3
>>	 64:          0            none  acpi
>>	NMI:          0
>>	ERR:    1586756
>>	
>>	This is an AMD 800 + Tyan K7 mobo.
>>	
>>	    -Jorge
>>	
>>	==============================================================
>>	Jorge Boncompte - Técnico de sistemas
>>	DTI2 - Desarrollo de la Tecnología de las Comunicaciones
>>	--------------------------------------------------------------
>>	C/ Abogado Enriquez Barrios, 5   14004 CORDOBA (SPAIN)
>>	Tlf: +34 957 761395 / FAX: +34 957 450380
>>	--------------------------------------------------------------
>>	jorge@dti2.net _-_-_-_-_-_-_-_-_-_-_-_-_-_ http://www.dti2.net
>>	==============================================================
>>	Without wicker a basket cannot be done.
>>	==============================================================
>>	
>>	
>>	-
>>	To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>>	the body of a message to majordomo@vger.kernel.org
>>	Please read the FAQ at http://www.tux.org/lkml/
End of included message



-- 
+--------------------------+------------------------------+
| Tim Walberg              | tewalberg@mediaone.net       |
| 828 Marshall Ct.         | www.concentric.net/~twalberg |
| Palatine, IL 60074       |                              |
+--------------------------+------------------------------+

[-- Attachment #2: Type: application/pgp-signature, Size: 175 bytes --]

RE: ERR in /proc/interrupts

[Jorge Boncompte (DTI2)]

    Are IPI related to SMP machines? This is not an SMP machine nor kernel.

    Regards.

    -Jorge

==============================================================
Jorge Boncompte - Técnico de sistemas
DTI2 - Desarrollo de la Tecnología de las Comunicaciones
--------------------------------------------------------------
C/ Abogado Enriquez Barrios, 5   14004 CORDOBA (SPAIN)
Tlf: +34 957 761395 / FAX: +34 957 450380
--------------------------------------------------------------
jorge@dti2.net _-_-_-_-_-_-_-_-_-_-_-_-_-_ http://www.dti2.net
==============================================================
Without wicker a basket cannot be done.
==============================================================

----- Mensaje original -----
De: "Tim Walberg" <tewalberg@mediaone.net>
Para: "Jorge Boncompte (DTI2)" <jorge@dti2.net>
CC: <linux-kernel@vger.kernel.org>
Enviado: jueves, 18 de enero de 2001 23:10
Asunto: Re: ERR in /proc/interrupts

A quick perusal of the 2.2.18 source code (I don't have
a copy of 2.4.x handy) shows that it's directly
related to the number of IPIs (inter-processor
interrupts) the system has taken. I'm not real
sure under what conditions this occurs, but it's
someplace to start...




-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: ERR in /proc/interrupts

[Timur Tabi]

** Reply to message from "Jorge Boncompte \(DTI2\)" <jorge@dti2.net> on Thu, 18
Jan 2001 23:31:38 +0100


> Are IPI related to SMP machines? This is not an SMP machine nor kernel.

Yes, Inter-Process Interrupts are an SMP thing.  I know you need to have an SMP
kernel for IPI's to be issued, but I don't know if you actually need to have
more than one processor.

Obviously, someone here is confused.  I just hope it's not me!


-- 
Timur Tabi - ttabi@interactivesi.com
Interactive Silicon - http://www.interactivesi.com

When replying to a mailing-list message, please direct the reply to the mailing list only.  Don't send another copy to me.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

RE: ERR in /proc/interrupts

[Jorge Boncompte (DTI2)]

    Timur,

     This isn't a SMP system. I have no plans to upgrade this box to SMP
unless AMD has a processor ready for it ;-)

    Kernel 2.4.0test12 + reiserfs 3.6.23
    Tyan K7 mobo
    AMD Athlon 800 (kernel built for AMD Athlon no SMP)
    d-link 530tx (via-rhine module)
    d-link 550tx (sundance module)
    cmd-649 ide controller (cmd driver)

     -Jorge

==============================================================
Jorge Boncompte - Técnico de sistemas
DTI2 - Desarrollo de la Tecnología de las Comunicaciones
--------------------------------------------------------------
C/ Abogado Enriquez Barrios, 5   14004 CORDOBA (SPAIN)
Tlf: +34 957 761395 / FAX: +34 957 450380
--------------------------------------------------------------
jorge@dti2.net _-_-_-_-_-_-_-_-_-_-_-_-_-_ http://www.dti2.net
==============================================================
Without wicker a basket cannot be done.
==============================================================

----- Mensaje original -----
De: "Timur Tabi" <ttabi@interactivesi.com>
Para: <linux-kernel@vger.kernel.org>
Enviado: jueves, 18 de enero de 2001 23:41
Asunto: Re: ERR in /proc/interrupts


> ** Reply to message from "Jorge Boncompte \(DTI2\)" <jorge@dti2.net> on
Thu, 18
> Jan 2001 23:31:38 +0100
>
>
> > Are IPI related to SMP machines? This is not an SMP machine nor kernel.
>
> Yes, Inter-Process Interrupts are an SMP thing.  I know you need to have
an SMP
> kernel for IPI's to be issued, but I don't know if you actually need to
have
> more than one processor.
>
> Obviously, someone here is confused.  I just hope it's not me!
>
>
> --
> Timur Tabi - ttabi@interactivesi.com
> Interactive Silicon - http://www.interactivesi.com
>
> When replying to a mailing-list message, please direct the reply to the
mailing list only.  Don't send another copy to me.
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> Please read the FAQ at http://www.tux.org/lkml/
>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: ERR in /proc/interrupts

[Timur Tabi]

** Reply to message from Tim Walberg <tewalberg@mediaone.net> on Thu, 18 Jan
2001 16:10:00 -0600


> A quick perusal of the 2.2.18 source code (I don't have
> a copy of 2.4.x handy) shows that it's directly
> related to the number of IPIs (inter-processor
> interrupts) the system has taken. I'm not real
> sure under what conditions this occurs, but it's
> someplace to start...

I thought one reason why IPI's were used was to synchronize the CPUs for a
cache flush.  That hardly sounds like an error condition to me.


-- 
Timur Tabi - ttabi@interactivesi.com
Interactive Silicon - http://www.interactivesi.com

When replying to a mailing-list message, please direct the reply to the mailing list only.  Don't send another copy to me.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/