|
Peter Mueller wrote:
>
> what kind of load are we talking about? My valinux boxes utilize
> eepro100's, are the valinux-version (standard kernel?) drivers effected?
> can you point me to some documentation regarding this subject? thanks!
>
Hi Peter,
I've seen it pop up intermittently whenever the card is handling a
mostly-saturated 100Mbps line (i.e. during stress testing, extremely
heavy load) but only logs an occasional message without apparently
affecting anything. I've seen it in all recent "stock" versions of the
eepro100 driver (Since about kernel 2.2.14 or so) although absolute most
recent versions seem to experience it much less.
You can be virtually guaranteed of experiencing an aspect of the bug
that will lock up your machine by running this shell script; on [2.2.14
< kernel version < 2.2.19] it only takes about twenty iterations for me
to lock up a box:
#!/bin/sh
while true
do
ifdown eth0
rmmod eepro100
ifup eth0
sleep 1
done
It has something to do with the initialization of the card itself, and
the setup of the DMA ring buffers, so bringing interfaces up and down
really stresses that bit of code. (If I knew more about network driver
hacking, I could tell you more, or try to fix it myself, alas I don't.)
This is also why generally speaking, increasing the 'txfifo' and
'rxfifo' - which increase the size of the ring buffers - seems to
alleviate the problem in most cases.
For further information, just do either a google search or search the
eepro100 mailing lists for "card reports no resources". You'll find
tons of hits.
Apologies for the somewhat off-topic-ness of this thread. (But not
entirely, since our 2.2 kernel based LVS boxes experience this error a
lot if I don't replace the stock kernel driver with Donald Becker's
driver, which has been an absolute rock under heavy LVS use.)
> >
> > Kim Le wrote:
> > >
> > > Hi All,
> > >
> > > I am getting error message "Eth0: no resource" when trying
> > to do stress test
> > > on linux 2.4.3. Could anyone give me a hint on what is the
> > reason. Thanks
> > > in advance for any help.
> >
> > Let me guess - you are using Intel EEPro100 cards right? (The full
> > error message would be "card reports no resources")
> >
> > This is a fairly well-known bug in the eepro100 driver that refuses to
> > be fixed. (I have a whole rant about that, let's just say that the
--
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
#!/usr/bin/perl -w
$_='while(read+STDIN,$_,2048){$a=29;$b=73;$c=142;$t=255;@t=map
{$_%16or$t^=$c^=($m=(11,10,116,100,11,122,20,100)[$_/16%8])&110;
$t^=(72,@z=(64,72,$a^=12*($_%16-2?0:$m&17)),$b^=$_%64?12:0,@z)
[$_%8]}(16..271);if((@a=unx"C*",$_)[20]&48){$h=5;$_=unxb24,join
"",@b=map{xB8,unxb8,chr($_^$a[--$h+84])}@ARGV;s/...$/1$&/;$d=
unxV,xb25,$_;$e=256|(ord$b[4])<<9|ord$b[3];$d=$d>>8^($f=$t&($d
>>12^$d>>4^$d^$d/8))<<17,$e=$e>>8^($t&($g=($q=$e>>14&7^$e)^$q*
8^$q<<6))<<9,$_=$t[$_]^(($h>>=8)+=$f+(~$g&$t))for@a[128..$#a]}
print+x"C*",@a}';s/x/pack+/g;eval
usage: qrpff 153 2 8 105 225 < /mnt/dvd/VOB_FILENAME \
| extract_mpeg2 | mpeg2dec -
http://www.eff.org/ http://www.opendvd.org/
http://www.cs.cmu.edu/~dst/DeCSS/Gallery/
|
