* Problem with 3c905B nic
@ 2002-03-06 12:41 Denis Vlasenko
2002-03-06 8:47 ` Zwane Mwaikambo
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: Denis Vlasenko @ 2002-03-06 12:41 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-kernel
Hi,
I have a NFS client which was connected to the server directly
by crossover cable. 100mbit Enternet was working as expected
(~10mbytes/sec peak). Recently I had to move to different
location and now I'm connected to the same server through
stack of four HP ProCurve 4000M switches.
Now I'm getting ~2mbytes/sec peak.
My NIC is 3c905B, seems to working fine under Win NT in 100mbit
full-duplex mode.
Since I boot from network I have NIC drivers compiled in,
tried to instruct 3c59x.c to be more verbose with
ether=0,0,0x8200,eth0 with no success... why?
I put debug printk in the source, it does not print:
if (dev->mem_start) {
/*
* The 'options' param is passed in as the third arg to the
* LILO 'ether=' argument for non-modular use
*/
option = dev->mem_start;
===> printk(KERN_DEBUG "VDA: ether=xx,xx,0x%08x,xxx\n",dev->mem_start);
}
Ok, I have recompiled drivers/net/3c59x.c with vortex_debug=4
set manually and now I see I'm having problems.
Do someone know what's up here?
/var/log/syslog.7.debug (abridged):
===================================
Mar 6 10:10:14 manta kernel: Kernel command line: root=/dev/nfs nfsroot=172.16.42.75:/.rootfs/.std,hard,intr ip=:172.16.42.75:::manta:eth0:dhcp devfs=mount ether=0,0,0x8200,eth0
Mar 6 10:10:14 manta kernel: Initializing CPU#0
Mar 6 10:10:14 manta kernel: Memory: 158480k/163840k available (1403k kernel code, 4972k reserved, 403k data, 152k init, 0k highmem)
Mar 6 10:10:14 manta kernel: PCI: Found IRQ 10 for device 00:06.0
Mar 6 10:10:14 manta kernel: 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
Mar 6 10:10:14 manta kernel: See Documentation/networking/vortex.txt
Mar 6 10:10:14 manta kernel: 00:06.0: 3Com PCI 3c905B Cyclone 100baseTx at 0xfc00. Vers LK1.1.16
Mar 6 10:10:14 manta kernel: 00:01:02:1a:ff:5f, IRQ 10
Mar 6 10:10:14 manta kernel: product code 434d rev 00.9 date 03-31-00
Mar 6 10:10:14 manta kernel: Internal config register is 1800000, transceivers 0xa.
Mar 6 10:10:14 manta kernel: 8K byte-wide RAM 5:3 Rx:Tx split, autoselect/Autonegotiate interface.
Mar 6 10:10:14 manta kernel: MII transceiver found at address 24, status 786d.
Mar 6 10:10:14 manta kernel: Enabling bus-master transmits and whole-frame receives.
Mar 6 10:10:14 manta kernel: 00:06.0: scatter/gather enabled. h/w checksums enabled
Mar 6 10:10:14 manta kernel: NET4: Linux TCP/IP 1.0 for NET4.0
Mar 6 10:10:14 manta kernel: IP Protocols: ICMP, UDP, TCP, IGMP
Mar 6 10:10:14 manta kernel: IP: routing cache hash table of 1024 buckets, 8Kbytes
Mar 6 10:10:14 manta kernel: TCP: Hash tables configured (established 16384 bind 16384)
Mar 6 10:10:14 manta kernel: eth0: Filling in the Rx ring.
Mar 6 10:10:14 manta kernel: eth0: using NWAY device table, not 8
Mar 6 10:10:14 manta kernel: eth0: Initial media type Autonegotiate.
Mar 6 10:10:14 manta kernel: eth0: MII #24 status 786d, link partner capability 45e1, info1 0010, setting full-duplex.
Mar 6 10:10:14 manta kernel: eth0: vortex_up() InternalConfig 01800000.
Mar 6 10:10:14 manta kernel: eth0: vortex_up() irq 10 media status 8880.
Mar 6 10:10:14 manta kernel: Sending DHCP requests .<7>eth0: Media selection timer tick happened, Autonegotiate.
Mar 6 10:10:14 manta kernel: dev->watchdog_timeo=500
Mar 6 10:10:14 manta kernel: eth0: MII transceiver has status 786d.
Mar 6 10:10:14 manta kernel: eth0: Media selection timer finished, Autonegotiate.
Mar 6 10:10:14 manta kernel: ., OK
Mar 6 10:10:14 manta kernel: IP-Config: Got DHCP answer from 255.255.255.255, my address is 172.16.42.211
Mar 6 10:10:14 manta kernel: IP-Config: Complete:
Mar 6 10:10:14 manta kernel: device=eth0, addr=172.16.42.211, mask=255.255.255.0, gw=172.16.42.98,
Mar 6 10:10:15 manta kernel: host=manta, domain=, nis-domain=(none),
Mar 6 10:10:15 manta kernel: bootserver=255.255.255.255, rootserver=172.16.42.75, rootpath=
Mar 6 10:10:15 manta kernel: NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
Mar 6 10:10:15 manta kernel: Looking up port of RPC 100003/2 on 172.16.42.75
Mar 6 10:10:15 manta kernel: Looking up port of RPC 100005/1 on 172.16.42.75
Mar 6 10:10:15 manta kernel: VFS: Mounted root (nfs filesystem).
Mar 6 10:10:15 manta kernel: Mounted devfs on /dev
Mar 6 10:10:15 manta kernel: Freeing unused kernel memory: 152k freed
Mar 6 10:10:15 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:15 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:15 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:15 manta last message repeated 3 times
Mar 6 10:10:15 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:15 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:18 manta last message repeated 3 times
Mar 6 10:10:18 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:19 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:20 manta last message repeated 2 times
Mar 6 10:10:21 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:21 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:23 manta last message repeated 3 times
Mar 6 10:10:24 manta dhcpcd[1576]: broadcasting DHCP_DISCOVER
Mar 6 10:10:24 manta dhcpcd[1576]: broadcastAddr option is missing in DHCP server response. Assuming 172.16.42.255
Mar 6 10:10:24 manta dhcpcd[1576]: broadcasting second DHCP_DISCOVER
Mar 6 10:10:24 manta dhcpcd[1576]: DHCP_OFFER received from (172.16.42.102)
Mar 6 10:10:24 manta dhcpcd[1576]: broadcasting DHCP_REQUEST for 172.16.42.211
Mar 6 10:10:24 manta dhcpcd[1576]: DHCP_ACK received from (172.16.42.102)
Mar 6 10:10:24 manta dhcpcd[1576]: dhcpConfig: ioctl SIOCADDRT: File exists
Mar 6 10:10:25 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:26 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:27 manta last message repeated 2 times
Mar 6 10:10:29 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:31 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:32 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:33 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:33 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:45 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:47 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:50 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:50 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:52 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:53 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:55 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:10:57 manta kernel: eth0: vortex_error(), status=0xe081
Mar 6 10:10:59 manta kernel: eth0: vortex_error(), status=0xe481
Mar 6 10:11:04 manta kernel: eth0: Media selection timer tick happened, Autonegotiate.
Mar 6 10:11:04 manta kernel: dev->watchdog_timeo=500
Mar 6 10:11:04 manta kernel: eth0: MII transceiver has status 786d.
Mar 6 10:11:04 manta kernel: eth0: Media selection timer finished, Autonegotiate.
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: Problem with 3c905B nic
2002-03-06 12:41 Problem with 3c905B nic Denis Vlasenko
@ 2002-03-06 8:47 ` Zwane Mwaikambo
2002-03-06 9:01 ` Same problem " Dimtiriu Vlad
2002-03-06 9:29 ` Problem " Andrew Morton
2 siblings, 0 replies; 4+ messages in thread
From: Zwane Mwaikambo @ 2002-03-06 8:47 UTC (permalink / raw)
To: Denis Vlasenko; +Cc: Andrew Morton, linux-kernel
On Wed, 6 Mar 2002, Denis Vlasenko wrote:
> Hi,
>
> I have a NFS client which was connected to the server directly
> by crossover cable. 100mbit Enternet was working as expected
> (~10mbytes/sec peak). Recently I had to move to different
> location and now I'm connected to the same server through
> stack of four HP ProCurve 4000M switches.
> Now I'm getting ~2mbytes/sec peak.
Can you verify your negotiated link status?
> Since I boot from network I have NIC drivers compiled in,
> tried to instruct 3c59x.c to be more verbose with
> ether=0,0,0x8200,eth0 with no success... why?
Out of interest, where does 0x8200 come from?
> I put debug printk in the source, it does not print:
> if (dev->mem_start) {
> /*
> * The 'options' param is passed in as the third arg to the
> * LILO 'ether=' argument for non-modular use
> */
> option = dev->mem_start;
> ===> printk(KERN_DEBUG "VDA: ether=xx,xx,0x%08x,xxx\n",dev->mem_start);
> }
>
> Ok, I have recompiled drivers/net/3c59x.c with vortex_debug=4
> set manually and now I see I'm having problems.
>
> Do someone know what's up here?
Could you elaborate more on which particular problem you're having. This
looks like a mixed bag.
Regards,
Zwane
^ permalink raw reply [flat|nested] 4+ messages in thread* Same problem with 3c905B nic
2002-03-06 12:41 Problem with 3c905B nic Denis Vlasenko
2002-03-06 8:47 ` Zwane Mwaikambo
@ 2002-03-06 9:01 ` Dimtiriu Vlad
2002-03-06 9:29 ` Problem " Andrew Morton
2 siblings, 0 replies; 4+ messages in thread
From: Dimtiriu Vlad @ 2002-03-06 9:01 UTC (permalink / raw)
To: vda; +Cc: Andrew Morton, linux-kernel
The same problem here:
* 3Com Corporation 3cSOHO100-TX Hurricane
* /etc/modules.conf
options 3c59x debug=3 rx_copybreak=300
Mar 6 11:02:54 lambda kernel: dev->watchdog_timeo=500
Mar 6 11:02:54 lambda kernel: eth1: MII transceiver has status 786d.
Mar 6 11:02:54 lambda kernel: eth1: Media selection timer finished,
Autonegotiate.
Mar 6 11:02:54 lambda kernel: eth2: Media selection timer
tick happened, Autonegotiate.
Mar 6 11:02:54 lambda kernel: dev->watchdog_timeo=500
Mar 6 11:02:54 lambda kernel: eth2: MII transceiver has status 786d.
Mar 6 11:02:54 lambda kernel: eth2: Media selection timer finished,
Autonegotiate.
Mar 6 11:03:10 lambda kernel: eth2: vortex_error(), status=0xe481
Those twho NICs are connected into a CenterCom
3016SL 10MB/s HUB.
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: Problem with 3c905B nic
2002-03-06 12:41 Problem with 3c905B nic Denis Vlasenko
2002-03-06 8:47 ` Zwane Mwaikambo
2002-03-06 9:01 ` Same problem " Dimtiriu Vlad
@ 2002-03-06 9:29 ` Andrew Morton
2 siblings, 0 replies; 4+ messages in thread
From: Andrew Morton @ 2002-03-06 9:29 UTC (permalink / raw)
To: vda; +Cc: linux-kernel
Denis Vlasenko wrote:
>
> Mar 6 10:10:31 manta kernel: eth0: vortex_error(), status=0xe481
That's actually not an error (sorry). It just means that the NIC's
statistics accumulators are getting full and it's time for software to
unload them.
Probably a duplex problem. Check your statistics counters in /proc/net/dev
for errors and take a look at the mii-diag and vortex-diag output. There
are pointers to these in Documentation/networking/vortex.txt.
-
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2002-03-06 9:31 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-03-06 12:41 Problem with 3c905B nic Denis Vlasenko
2002-03-06 8:47 ` Zwane Mwaikambo
2002-03-06 9:01 ` Same problem " Dimtiriu Vlad
2002-03-06 9:29 ` Problem " Andrew Morton
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox