All of lore.kernel.org
 help / color / mirror / Atom feed
* 3c59x very slow with 2.6.X
@ 2004-08-04 20:03 Bernd Schubert
  2004-08-05  6:00 ` Frank Steiner
  0 siblings, 1 reply; 4+ messages in thread
From: Bernd Schubert @ 2004-08-04 20:03 UTC (permalink / raw)
  To: Kernel Mailing List

Hello,

somehow the network speed is reduced to 14MBit with 2.6.X on my system, with 
2.4.X it has full 100MBit.
As all our systems boot diskless this is really annoying and I will reboot 
2.4.27 soon.

euklid:~# nttcp -T hamilton2
     Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
l  8388608    4.73    4.60     14.2004     14.5911    2048    433.36     445.3
1  8388608    4.73    0.03     14.1984   2236.9621    6145   1300.12  204833.3


from dmesg:

3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
0000:00:0a.0: 3Com PCI 3c905B Cyclone 100baseTx at 0xb800. Vers LK1.1.19


euklid:~# mii-tool 
eth0: 100 Mbit, full duplex, link ok


Any ideas whats going on?



Please tell me if you need further information.

Cheers,
	Bernd


PS: Happens with 2.6.7 and 2.6.8-rc2-mm2

-- 
Bernd Schubert
Physikalisch Chemisches Institut / Theoretische Chemie
Universität Heidelberg
INF 229
69120 Heidelberg
e-mail: bernd.schubert@pci.uni-heidelberg.de

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: 3c59x very slow with 2.6.X
  2004-08-04 20:03 3c59x very slow with 2.6.X Bernd Schubert
@ 2004-08-05  6:00 ` Frank Steiner
  2004-08-05  7:41   ` Måns Rullgård
  2004-08-05  9:00   ` Bernd Schubert
  0 siblings, 2 replies; 4+ messages in thread
From: Frank Steiner @ 2004-08-05  6:00 UTC (permalink / raw)
  To: Bernd Schubert; +Cc: Kernel Mailing List

Bernd Schubert wrote:
> Hello,
> 
> somehow the network speed is reduced to 14MBit with 2.6.X on my system, with 
> 2.4.X it has full 100MBit.
> As all our systems boot diskless this is really annoying and I will reboot 
> 2.4.27 soon.
> 
> euklid:~# nttcp -T hamilton2
>      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
> l  8388608    4.73    4.60     14.2004     14.5911    2048    433.36     445.3
> 1  8388608    4.73    0.03     14.1984   2236.9621    6145   1300.12  204833.3


No problem here: both machines have a 3com 3c905CX-TX and run kernel 2.6.7.

zassenhaus /root# nttcp -T riemann
      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
l  8388608    0.71    0.01     94.9903   7457.3690    2048   2898.87  227580.8
1  8388608    0.71    0.04     93.9984   1598.0964    5794   8115.57  137975.4

The driver is compiled into the kernel, not as module. Could that make a
difference for you?

cu,
Frank

-- 
Dipl.-Inform. Frank Steiner   Web:  http://www.bio.ifi.lmu.de/~steiner/
Lehrstuhl f. Bioinformatik    Mail: http://www.bio.ifi.lmu.de/~steiner/m/
LMU, Amalienstr. 17           Phone: +49 89 2180-4049
80333 Muenchen, Germany       Fax:   +49 89 2180-99-4049
* Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: 3c59x very slow with 2.6.X
  2004-08-05  6:00 ` Frank Steiner
@ 2004-08-05  7:41   ` Måns Rullgård
  2004-08-05  9:00   ` Bernd Schubert
  1 sibling, 0 replies; 4+ messages in thread
From: Måns Rullgård @ 2004-08-05  7:41 UTC (permalink / raw)
  To: linux-kernel

Frank Steiner <fsteiner-mail@bio.ifi.lmu.de> writes:

> Bernd Schubert wrote:
>> Hello,
>> somehow the network speed is reduced to 14MBit with 2.6.X on my
>> system, with 2.4.X it has full 100MBit.
>> As all our systems boot diskless this is really annoying and I will
>> reboot 2.4.27 soon.
>> euklid:~# nttcp -T hamilton2
>>      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
>> l  8388608    4.73    4.60     14.2004     14.5911    2048    433.36     445.3
>> 1  8388608    4.73    0.03     14.1984   2236.9621    6145   1300.12  204833.3
>
> No problem here: both machines have a 3com 3c905CX-TX and run kernel 2.6.7.
>
> zassenhaus /root# nttcp -T riemann
>       Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
> l  8388608    0.71    0.01     94.9903   7457.3690    2048   2898.87  227580.8
> 1  8388608    0.71    0.04     93.9984   1598.0964    5794   8115.57  137975.4
>
> The driver is compiled into the kernel, not as module. Could that make a
> difference for you?

FWIW, I'm using a 3c905C-TX with kernel 2.6.6 and 3c59x as a module,
without any problems at all.

-- 
Måns Rullgård
mru@kth.se


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: 3c59x very slow with 2.6.X
  2004-08-05  6:00 ` Frank Steiner
  2004-08-05  7:41   ` Måns Rullgård
@ 2004-08-05  9:00   ` Bernd Schubert
  1 sibling, 0 replies; 4+ messages in thread
From: Bernd Schubert @ 2004-08-05  9:00 UTC (permalink / raw)
  To: Frank Steiner; +Cc: Kernel Mailing List

On Thursday 05 August 2004 08:00, you wrote:
> Bernd Schubert wrote:
> > Hello,
> >
> > somehow the network speed is reduced to 14MBit with 2.6.X on my system,
> > with 2.4.X it has full 100MBit.
> > As all our systems boot diskless this is really annoying and I will
> > reboot 2.4.27 soon.
> >
> > euklid:~# nttcp -T hamilton2
> >      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s  
> > CPU-C/s l  8388608    4.73    4.60     14.2004     14.5911    2048   
> > 433.36     445.3 1  8388608    4.73    0.03     14.1984   2236.9621   
> > 6145   1300.12  204833.3
>
> No problem here: both machines have a 3com 3c905CX-TX and run kernel 2.6.7.
>
> zassenhaus /root# nttcp -T riemann
>       Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s  
> CPU-C/s l  8388608    0.71    0.01     94.9903   7457.3690    2048  
> 2898.87  227580.8 1  8388608    0.71    0.04     93.9984   1598.0964   
> 5794   8115.57  137975.4
>
> The driver is compiled into the kernel, not as module. Could that make a
> difference for you?

Its also compiled into the kernel on our systems. Here are the numbers with 
2.4.27-rc5:

euklid:~# nttcp -T hamilton2

     Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
l  8388608    0.71    0.04     94.9401   1677.7216    2048   2897.34   51200.0
1  8388608    0.71    0.03     94.1146   2236.9621    5797   8129.81  193233.3

Cheers,
	Bernd

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2004-08-05  9:00 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-08-04 20:03 3c59x very slow with 2.6.X Bernd Schubert
2004-08-05  6:00 ` Frank Steiner
2004-08-05  7:41   ` Måns Rullgård
2004-08-05  9:00   ` Bernd Schubert

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.