Hello,

I've never seen such a strange behavior.
I suppose you've ensured that it's not a cable problem, haven't you?
If you really want to investigate what happen you may turn on full debugging
in the driver and add show_state() call with fine grained timestamps
for each interrupt to control the progress of the packet sending in the
driver.

Best regards
                                        Andrey V.
                                        Savochkin

On Wed, Apr 05, 2000 at 01:42:25PM -0700, Cacophonix wrote:
> Hello,
> I have the following weirdosity with some of my eepro100 cards, on 
> 2.3.99-preX. Specifically, the problem only seems to occur on the
> cards with addresses in the range 00:A0:C9:xx:xx:xx, and not the
> cards with mac addresses in the range 00:90:27:xx:xx:xx.
> 
> The pings are from 15.5.2.43 to 15.5.2.42 (the problem is on the
> latter). The tcpdump was captured on another system on the same
> ethernet.
[snip]
> [root@pc43 /root]# ping 15.5.2.42
> PING 15.5.2.42 (15.5.2.42): 56 data bytes
> 64 bytes from 15.5.2.42: icmp_seq=10 ttl=255 time=28.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=11 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=28 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=29 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=30 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=31 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=32 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=40 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=57 ttl=255 time=991.6 ms
> 64 bytes from 15.5.2.42: icmp_seq=58 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=59 ttl=255 time=991.6 ms
> 64 bytes from 15.5.2.42: icmp_seq=60 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=61 ttl=255 time=991.6 ms
> 64 bytes from 15.5.2.42: icmp_seq=62 ttl=255 time=0.3 ms
> 64 bytes from 15.5.2.42: icmp_seq=64 ttl=255 time=991.6 ms
> 64 bytes from 15.5.2.42: icmp_seq=64 ttl=255 time=3000.2 ms (DUP!)
> 64 bytes from 15.5.2.42: icmp_seq=64 ttl=255 time=4991.7 ms (DUP!)
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]

Reply via email to