[PATCH 2.4] Tulip 21041 port breakage
From: Geert Uytterhoeven
Date: Thu Feb 26 2004 - 11:21:46 EST
Hi Jeff,
I still (2.4.23, but the Tulip driver wasn't changed in the mean time) need the
following patch in my tree:
--- linux-2.4.26-pre1/drivers/net/tulip/tulip_core.c.orig 2004-02-26 17:06:50.000000000 +0100
+++ linux-2.4.26-pre1/drivers/net/tulip/tulip_core.c 2004-02-26 17:09:54.000000000 +0100
@@ -579,7 +579,7 @@
dev->if_port = 2 - dev->if_port;
} else
dev->if_port = 0;
- else
+ else if (dev->if_port != 0 || (csr12 & 0x0004) != 0)
dev->if_port = 1;
tulip_select_media(dev, 0);
}
Without this patch, on heavy network activity the driver switches from 10-baseT
to 10-base2, while there's still a 10-baseT link beat, and the network is lost,
requiring manual intervention (ifdown/ifup).
With the patch, the driver automatically recovers after a few minutes, as
evidenced from this log excerpt:
| Feb 3 17:48:50 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:48:50 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:48:50 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:48:58 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:48:58 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:48:58 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:49:18 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:49:18 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:49:18 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:49:26 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:49:26 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:49:26 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:49:34 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:49:34 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:49:34 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:49:42 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:49:42 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:49:42 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:49:50 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:49:50 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:49:50 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:49:58 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:49:58 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:49:58 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:50:06 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:50:06 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:50:06 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:50:14 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:50:14 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:50:14 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:50:22 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:50:22 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:50:22 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:50:30 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:50:30 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:50:30 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:50:42 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:50:42 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:50:42 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:50:50 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:50:50 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:50:50 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:50:58 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:50:58 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:50:58 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:51:06 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:51:06 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:51:06 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:51:14 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:51:14 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:51:14 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:51:22 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:51:22 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:51:22 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:51:30 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:51:30 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:51:30 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:51:38 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:51:38 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:51:38 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:51:46 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:51:46 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:51:46 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:52:10 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:52:10 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:52:10 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:52:26 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:52:26 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:52:26 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:52:46 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:52:46 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:52:46 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:52:54 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:52:54 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:52:54 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:53:10 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:53:10 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:53:10 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:53:18 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:53:18 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:53:18 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:55:38 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:55:38 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:55:38 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:55:54 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:55:54 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:55:54 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:56:02 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:56:02 callisto kernel: eth0: 21041 transmit timed out, status fc660000, CSR12 000051c8, CSR13 ffffef01, CSR14 ffffffff, resetting...
| Feb 3 17:56:02 callisto kernel: tulip: old driver would switch to 10base2, but we don't
| Feb 3 17:56:02 callisto kernel: eth0: No 21041 10baseT link beat, Media switched to 10base2.
| Feb 3 17:56:10 callisto kernel: NETDEV WATCHDOG: eth0: transmit timed out
| Feb 3 17:56:10 callisto kernel: eth0: 21041 transmit timed out, status fc260010, CSR12 000050c8, CSR13 ffffef09, CSR14 fffff7fd, resetting...
| Feb 3 17:56:12 callisto kernel: eth0: Out-of-sync dirty pointer, 76318 vs. 76335.
Machine is CHRP LongTrail with a PPC604e and a 21041, connected to a 10 Mbps
hub:
00:04.0 Ethernet controller: Digital Equipment Corporation DECchip 21041 [Tulip Pass 3] (rev 21)
Subsystem: D-Link System Inc DE-530+
Flags: bus master, medium devsel, latency 0, IRQ 29
I/O ports at 1080 [size=128]
Memory at c1080000 (32-bit, non-prefetchable) [size=128]
Expansion ROM at c11c0000 [disabled] [size=256K]
Kernel is 2.4.23, with the following patch applied to the tulip driver:
--- linux-ppc-2.4.23/drivers/net/tulip/tulip_core.c.orig 2003-11-28 21:04:35.000000000 +0100
+++ linux-ppc-2.4.23/drivers/net/tulip/tulip_core.c 2003-11-30 11:37:45.000000000 +0100
@@ -580,7 +580,15 @@
} else
dev->if_port = 0;
else
+ {
+printk("tulip: old driver would switch to 10base2, ");
+ if (dev->if_port != 0 || (csr12 & 0x0004) != 0) {
+printk("and we do\n");
dev->if_port = 1;
+ } else {
+printk("but we don't\n");
+ }
+ }
tulip_select_media(dev, 0);
}
} else if (tp->chip_id == DC21140 || tp->chip_id == DC21142
BTW, recently I noticed a different problem:
(tux is my Vaio laptop with i8255x Ethernet, callisto is the PPC box with a
21041):
| tux$ ping callisto
| PING callisto.of.borg (192.168.0.1): 56 data bytes
| 64 bytes from 192.168.0.1: icmp_seq=5 ttl=64 time=0.1 ms
| 64 bytes from 192.168.0.1: icmp_seq=0 ttl=64 time=4998.4 ms
| 64 bytes from 192.168.0.1: icmp_seq=1 ttl=64 time=4002.3 ms
| 64 bytes from 192.168.0.1: icmp_seq=2 ttl=64 time=3002.4 ms
| 64 bytes from 192.168.0.1: icmp_seq=3 ttl=64 time=2002.5 ms
| 64 bytes from 192.168.0.1: icmp_seq=4 ttl=64 time=1002.6 ms
| 64 bytes from 192.168.0.1: icmp_seq=31 ttl=64 time=0.1 ms
| 64 bytes from 192.168.0.1: icmp_seq=6 ttl=64 time=25032.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=7 ttl=64 time=24032.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=8 ttl=64 time=23032.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=9 ttl=64 time=22032.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=10 ttl=64 time=21032.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=11 ttl=64 time=20022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=12 ttl=64 time=19022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=13 ttl=64 time=18022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=14 ttl=64 time=17022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=15 ttl=64 time=16022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=16 ttl=64 time=15022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=17 ttl=64 time=14022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=18 ttl=64 time=13022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=19 ttl=64 time=12022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=20 ttl=64 time=11022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=21 ttl=64 time=10022.0 ms
| 64 bytes from 192.168.0.1: icmp_seq=22 ttl=64 time=9021.9 ms
| 64 bytes from 192.168.0.1: icmp_seq=23 ttl=64 time=8021.9 ms
| 64 bytes from 192.168.0.1: icmp_seq=24 ttl=64 time=7021.9 ms
| 64 bytes from 192.168.0.1: icmp_seq=25 ttl=64 time=6021.9 ms
| 64 bytes from 192.168.0.1: icmp_seq=26 ttl=64 time=5021.9 ms
| 64 bytes from 192.168.0.1: icmp_seq=27 ttl=64 time=4021.9 ms
| 64 bytes from 192.168.0.1: icmp_seq=28 ttl=64 time=3021.9 ms
| 64 bytes from 192.168.0.1: icmp_seq=29 ttl=64 time=2021.8 ms
| 64 bytes from 192.168.0.1: icmp_seq=30 ttl=64 time=1021.9 ms
| 64 bytes from 192.168.0.1: icmp_seq=55 ttl=64 time=0.1 ms
| 64 bytes from 192.168.0.1: icmp_seq=32 ttl=64 time=23016.2 ms
| 64 bytes from 192.168.0.1: icmp_seq=33 ttl=64 time=22016.2 ms
| 64 bytes from 192.168.0.1: icmp_seq=34 ttl=64 time=21016.2 ms
| 64 bytes from 192.168.0.1: icmp_seq=35 ttl=64 time=20016.2 ms
| 64 bytes from 192.168.0.1: icmp_seq=36 ttl=64 time=19016.3 ms
| 64 bytes from 192.168.0.1: icmp_seq=37 ttl=64 time=18016.2 ms
| 64 bytes from 192.168.0.1: icmp_seq=38 ttl=64 time=17016.2 ms
| 64 bytes from 192.168.0.1: icmp_seq=39 ttl=64 time=16016.3 ms
| ...
|
| --- callisto.of.borg ping statistics ---
| 452 packets transmitted, 420 packets received, 7% packet loss
| round-trip min/avg/max = 0.1/10917.2/25032.0 ms
However, according to dmesg the case to be adressed by the patch above was not
triggered, so this is a different problem.
ifdown/ifup fixed the problem.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/