2.4.1-pre10 -> 2.4.1 klogd at 100% CPU ; 2.4.0 OK

2.4.1-pre10 -> 2.4.1 klogd at 100% CPU ; 2.4.0 OK

[David Rees]

I've been trying different 2.4.1-pre kernels trying to find one that 
doesn't end up with klogd pegging the CPU.  2.4.0 is OK, but 
2.4.1-pre10 to 2.4.1 all leave klogd sitting at 100% CPU.

The machine in question is a Gateway E-3200, a basic PIII-500 running RH 
7.0 with all the latest updates as well as the recommended updates 
documented in the Changes file.  The kernel is compiled with kgcc 
(egcs-1.1.2).

Below is the output from various files to hopefully give some hits as to 
what may be the problem:

> sh scripts/ver_linux
-- Versions installed: (if some fields are empty or look
-- unusual then possibly you have very old versions)
Linux shooter.ebetinc.com 2.4.0-c1 #1 Tue Jan 30 02:08:19 PST 2001 i686 
unknown
Kernel modules         2.4.1
Gnu C                  2.96
Gnu Make               3.79.1
Binutils               2.10.0.33
Linux C Library        > libc.2.2
Dynamic linker         ldd (GNU libc) 2.2
Procps                 2.0.7
Mount                  2.10m
Net-tools              1.56
Console-tools          0.3.3
Sh-utils               2.0
Modules Loaded         


.config is at http://spoke.nols.com/~drees/config.txt
cpuinfo is at http://spoke.nols.com/~drees/cpuinfo.txt
dmesg is at   http://spoke.nols.com/~drees/dmesg.txt

Thanks,
Dave
-
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/

2.4.1-pre10 -> 2.4.1 klogd at 100% CPU ; 2.4.0 OK

[Chris Hanson]

   I've been trying different 2.4.1-pre kernels trying to find one
   that doesn't end up with klogd pegging the CPU. 2.4.0 is OK, but
   2.4.1-pre10 to 2.4.1 all leave klogd sitting at 100% CPU.

   The machine in question is a Gateway E-3200, a basic PIII-500
   running RH 7.0 with all the latest updates as well as the
   recommended updates documented in the Changes file. The kernel is
   compiled with kgcc (egcs-1.1.2).

I'm seeing exactly the same problem with 2.4.1 on a laptop (HP
OmniBook 6000, 700 MHz PIII) running Debian 2.2, patched to have
modutils 2.4.1 and linux-utils 2.10q.  The kernel was compiled with
gcc 2.95.2.  Again, 2.4.0 worked fine.

BTW, in my case, stopping and restarting klogd makes the problem go
away.  But every time I suspend/resume the machine, it comes back.

On the other hand, a desktop machine (ASUS P2B-LS, 600 MHz PIII)
running Debian woody and Linux 2.4.1 is _not_ having this problem.
-
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: 2.4.1-pre10 -> 2.4.1 klogd at 100% CPU ; 2.4.0 OK

[Padraig Brady]

Chris Hanson wrote:

>    Date: Wed, 31 Jan 2001 17:48:50 +0000
>    From: Padraig Brady <Padraig@AnteFacto.com>
> 
>    Are you using the 3c59x driver?
> 
> Yes.

Can we sort this out once and for all? There are a few emails
everyday relating to this bug.

The following patch posted by "Troels Walsted Hansen" <troels@thule.no>
on Jan 11th fixes this. The problem is that when 2 consequtive
NULLs are sent to klogd it goes into a busy loop. Andrew Mortons
3c59x driver does this, but also on Jan 11th he replied that he had
fixed it. I'm using 2.4ac4 with no problems, so I presume some
of these patches have been lost along the way?

--- sysklogd-1.4.orig/klogd.c    Mon Sep 18 09:34:11 2000
+++ sysklogd-1.4/klogd.c    Thu Jan 11 09:26:10 2001
@@ -739,6 +758,13 @@
            break;  /* full line_buff or end of input buffer */
                 }

+               if( *ptr == '\0' ) /* zero byte */
+               {
+                  ptr++;    /* skip zero byte */
+                  space -= 1;
+                  len   -= 1;
+                  break;
+               }
                 if( *ptr == '\n' )  /* newline */
                 {
                    ptr++;    /* skip newline */

-
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/

2.4.1-pre10 -> 2.4.1 klogd at 100% CPU ; 2.4.0 OK

[Chris Hanson]

   Date: Wed, 31 Jan 2001 18:00:09 +0000
   From: Padraig Brady <Padraig@AnteFacto.com>

   Can we sort this out once and for all? There are a few emails
   everyday relating to this bug.

   The following patch posted by "Troels Walsted Hansen" <troels@thule.no>
   on Jan 11th fixes this. The problem is that when 2 consequtive
   NULLs are sent to klogd it goes into a busy loop. Andrew Mortons
   3c59x driver does this, but also on Jan 11th he replied that he had
   fixed it. I'm using 2.4ac4 with no problems, so I presume some
   of these patches have been lost along the way?

Yes, that did it.  It also explains why my desktop works fine, since
it uses an eepro100 network card.
-
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: 2.4.1-pre10 -> 2.4.1 klogd at 100% CPU ; 2.4.0 OK

[David Rees]

On Wed, Jan 31, 2001 at 06:00:09PM +0000, Padraig Brady wrote:
> Chris Hanson wrote:
> 
> >    Date: Wed, 31 Jan 2001 17:48:50 +0000
> >    From: Padraig Brady <Padraig@AnteFacto.com>
> > 
> >    Are you using the 3c59x driver?
> > 
> > Yes.
> 
> Can we sort this out once and for all? There are a few emails
> everyday relating to this bug.
> 
> The following patch posted by "Troels Walsted Hansen" <troels@thule.no>
> on Jan 11th fixes this. The problem is that when 2 consequtive
> NULLs are sent to klogd it goes into a busy loop. Andrew Mortons
> 3c59x driver does this, but also on Jan 11th he replied that he had
> fixed it. I'm using 2.4ac4 with no problems, so I presume some
> of these patches have been lost along the way?
<snip of patch>

And if you are using RedHat, you can also grab the latest sysklogd
from rawhide which includes the patch:
ftp://ftp.redhat.com/rawhide/i386/RedHat/RPMS/sysklogd-1.4-5.i386.rpm

-Dave
-
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/