network speed extremly slowed down

network speed extremly slowed down

[Bernd Schubert]

Hi,

until one hour ago I thought this bug affects only 3com 3c509B cards. Now we 
go on switching more and more systems from 2.4.X to 2.6.X and it seems that 
many cards work much slower with 2.6.X

with 2.4.27:
=========
bernd@pc-brenn bernd>nttcp -T hamilton2
     Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
l  8388608    0.72    0.02     92.7710   3355.4432    2048   2831.15  102400.0
1  8388608    0.73    0.04     92.0581   1864.3941    5783   7932.96  160661.2

with 2.6.11.3:
==========

bernd@pc-brenn bernd>nttcp -T hamilton2
     Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
l  8388608    3.29    3.28     20.3957     20.4569    2048    622.43     624.3
1  8388608    3.29    0.04     20.3923   1637.0411    6143   1866.67  149851.2


Cards that show this behaviour:
=======================
Yukon Gigabit Ethernet 10/100/1000Base-T Adapter (sk98lin)
3Com PCI 3c905B Cyclone 100baseTx a
eth0: VIA VT6102 Rhine-II 

Cards that work fine with 2.4.X and 2.6.X:
==============================
3Com PCI 3c905C Tornado


The drivers always claim they are using full duplex, here e.g.:

eth0: network connection up using port A
    speed:           100
    autonegotiation: yes
    duplex mode:     full
    flowctrl:        none
    irq moderation:  disabled
    scatter-gather:  enabled


I already reported this issue about the 3C509B on LKML several month ago, but 
nobody could help. Ever since that mail I got several reports from other 
people having the same trouble with their cards.

Initially I wanted to write a bug report following the instructions 
in /usr/src/linux/Documentation/networking/vortex.txt, but since this seems 
to be more or less a general problem I don't know how to proceed now.


Thanks in advance,
 Bernd


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

Re: network speed extremly slowed down

[Stephen Hemminger]

On Mon, 14 Mar 2005 17:45:06 +0100
Bernd Schubert <bernd.schubert@pci.uni-heidelberg.de> wrote:

> Hi,
> 
> until one hour ago I thought this bug affects only 3com 3c509B cards. Now we 
> go on switching more and more systems from 2.4.X to 2.6.X and it seems that 
> many cards work much slower with 2.6.X
> 
> with 2.4.27:
> =========
> bernd@pc-brenn bernd>nttcp -T hamilton2
>      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
> l  8388608    0.72    0.02     92.7710   3355.4432    2048   2831.15  102400.0
> 1  8388608    0.73    0.04     92.0581   1864.3941    5783   7932.96  160661.2
> 
> with 2.6.11.3:
> ==========
> 
> bernd@pc-brenn bernd>nttcp -T hamilton2
>      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
> l  8388608    3.29    3.28     20.3957     20.4569    2048    622.43     624.3
> 1  8388608    3.29    0.04     20.3923   1637.0411    6143   1866.67  149851.2
> 

What type is the other host (hamilton2) and what is the round trip delay?
Is it on the same local network?

Re: network speed extremly slowed down

[Bernd Schubert]

On Monday 14 March 2005 18:21, Stephen Hemminger wrote:
> On Mon, 14 Mar 2005 17:45:06 +0100
>
> Bernd Schubert <bernd.schubert@pci.uni-heidelberg.de> wrote:
> > Hi,
> >
> > until one hour ago I thought this bug affects only 3com 3c509B cards. Now
> > we go on switching more and more systems from 2.4.X to 2.6.X and it seems
> > that many cards work much slower with 2.6.X
> >
> > with 2.4.27:
> > =========
> > bernd@pc-brenn bernd>nttcp -T hamilton2
> >      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s  
> > CPU-C/s l  8388608    0.72    0.02     92.7710   3355.4432    2048  
> > 2831.15  102400.0 1  8388608    0.73    0.04     92.0581   1864.3941   
> > 5783   7932.96  160661.2
> >
> > with 2.6.11.3:
> > ==========
> >
> > bernd@pc-brenn bernd>nttcp -T hamilton2
> >      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s  
> > CPU-C/s l  8388608    3.29    3.28     20.3957     20.4569    2048   
> > 622.43     624.3 1  8388608    3.29    0.04     20.3923   1637.0411   
> > 6143   1866.67  149851.2
>
> What type is the other host (hamilton2) and what is the round trip delay?
> Is it on the same local network?

Yes of course, hamilton2 is on our local network. And of course, this is 
completely independet of hamilton2, it can be reproduced with any local 
system (as long as this system has full network speed, of course ;) ).


Round trip delay, e.g. the ping time?

Hmm, interesting!


2.4.27:
=====
bernd@wigner bernd>ping hamilton2
PING hamilton2.pci.uni-heidelberg.de (129.206.21.202): 56 data bytes
64 bytes from 129.206.21.202: icmp_seq=0 ttl=64 time=0.1 ms
64 bytes from 129.206.21.202: icmp_seq=1 ttl=64 time=0.1 ms
64 bytes from 129.206.21.202: icmp_seq=2 ttl=64 time=0.1 ms
64 bytes from 129.206.21.202: icmp_seq=3 ttl=64 time=0.1 ms
64 bytes from 129.206.21.202: icmp_seq=4 ttl=64 time=0.1 ms
64 bytes from 129.206.21.202: icmp_seq=5 ttl=64 time=0.1 ms


2.6.11.3
======
bernd@wigner bernd>ping hamilton2
PING hamilton2.pci.uni-heidelberg.de (129.206.21.202): 56 data bytes
64 bytes from 129.206.21.202: icmp_seq=0 ttl=64 time=0.9 ms
64 bytes from 129.206.21.202: icmp_seq=1 ttl=64 time=0.4 ms
64 bytes from 129.206.21.202: icmp_seq=2 ttl=64 time=0.5 ms
64 bytes from 129.206.21.202: icmp_seq=3 ttl=64 time=0.4 ms
64 bytes from 129.206.21.202: icmp_seq=4 ttl=64 time=0.4 ms
64 bytes from 129.206.21.202: icmp_seq=5 ttl=64 time=0.4 ms



Thanks,
 Bernd

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

Re: network speed extremly slowed down

[Stephen Hemminger]

On Mon, 14 Mar 2005 17:45:06 +0100
Bernd Schubert <bernd.schubert@pci.uni-heidelberg.de> wrote:

> Hi,
> 
> until one hour ago I thought this bug affects only 3com 3c509B cards. Now we 
> go on switching more and more systems from 2.4.X to 2.6.X and it seems that 
> many cards work much slower with 2.6.X
> 
> with 2.4.27:
> =========
> bernd@pc-brenn bernd>nttcp -T hamilton2
>      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
> l  8388608    0.72    0.02     92.7710   3355.4432    2048   2831.15  102400.0
> 1  8388608    0.73    0.04     92.0581   1864.3941    5783   7932.96  160661.2
> 
> with 2.6.11.3:
> ==========
> 
> bernd@pc-brenn bernd>nttcp -T hamilton2
>      Bytes  Real s   CPU s Real-MBit/s  CPU-MBit/s   Calls  Real-C/s   CPU-C/s
> l  8388608    3.29    3.28     20.3957     20.4569    2048    622.43     624.3
> 1  8388608    3.29    0.04     20.3923   1637.0411    6143   1866.67  149851.2
> 
> 
> Cards that show this behaviour:
> =======================
> Yukon Gigabit Ethernet 10/100/1000Base-T Adapter (sk98lin)
> 3Com PCI 3c905B Cyclone 100baseTx a
> eth0: VIA VT6102 Rhine-II 
> 
> Cards that work fine with 2.4.X and 2.6.X:
> ==============================
> 3Com PCI 3c905C Tornado
> 
> 
> The drivers always claim they are using full duplex, here e.g.:
> 
> eth0: network connection up using port A
>     speed:           100
>     autonegotiation: yes
>     duplex mode:     full
>     flowctrl:        none
>     irq moderation:  disabled
>     scatter-gather:  enabled
>
Are there any errors reported? Cpu time sees high perhaps the console log
is filling with messages or something.

Re: network speed extremly slowed down

[David S. Miller]

On Mon, 14 Mar 2005 11:57:57 -0800
Stephen Hemminger <shemminger@osdl.org> wrote:

> Are there any errors reported? Cpu time sees high perhaps the console log
> is filling with messages or something.

Or he has CONFIG_DEBUG_SLAB enabled in his kernel config.
So many people do this unknowingly.

Re: network speed extremly slowed down

[Bernd Schubert]

On Mon, Mar 14, 2005 at 12:15:09PM -0800, David S. Miller wrote:
> On Mon, 14 Mar 2005 11:57:57 -0800
> Stephen Hemminger <shemminger@osdl.org> wrote:
> 
> > Are there any errors reported? Cpu time sees high perhaps the console log
> > is filling with messages or something.
> 
> Or he has CONFIG_DEBUG_SLAB enabled in his kernel config.
> So many people do this unknowingly.
> 

No, no unusual error reports, but full debugging is enabled, so also
CONFIG_DEBUG_SLAB.
Recompiling now with only CONFIG_MAGIC_SYSRQ support.

Thanks,
	Bernd 

Re: network speed extremly slowed down

[Bernd Schubert]

On Monday 14 March 2005 21:15, David S. Miller wrote:
> On Mon, 14 Mar 2005 11:57:57 -0800
>
> Stephen Hemminger <shemminger@osdl.org> wrote:
> > Are there any errors reported? Cpu time sees high perhaps the console log
> > is filling with messages or something.
>
> Or he has CONFIG_DEBUG_SLAB enabled in his kernel config.
> So many people do this unknowingly.

Thanks a lot, disabling debugging support solved our network speed problems.

Thanks again,
 Bernd