ACPI problems with 2.6.6
Hi,
I've got some serious strangeness being caused by ACPI on 2.6.6.
2.6.3 works perfectly on the same machine. I don't use any acpi option on the kernel command line so it's just using the default options.
The first problem is that a strange device appears as eth0:
eth0 Link encap:UNSPEC HWaddr
00-90-F5-00-00-22-91-25-00-00-00-00-00-00-00-00
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
It seems to be some incarnation of the real ethernet adaptor, but even when I manually configure it it is unable to actually talk to the network. I can configure the real ethernet adaptor (8139too) as eth1 with no problems.
Also, for some reason the loopback device is not added to the routing table which in turn causes a stack more problems.
Lots of other stuff also doesn't work including X.
But if I boot 2.6.3 or set acpi=off for 2.6.6 everything works perfectly. The same problems exist in 2.6.5 but I haven't tried 2.6.4.
I've attached my dmesg and .config - let me know what else is needed.
Thanks,
David.
Signature · 2 lines
David Johnson http://www.david-web.co.uk/
Quoted 25 lines · David Johnson
Hi,
I've got some serious strangeness being caused by ACPI on 2.6.6.
2.6.3 works perfectly on the same machine. I don't use any acpi option on the kernel command line so it's just using the default options.
The first problem is that a strange device appears as eth0:
eth0 Link encap:UNSPEC HWaddr
00-90-F5-00-00-22-91-25-00-00-00-00-00-00-00-00
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)It seems to be some incarnation of the real ethernet adaptor, but even when I manually configure it it is unable to actually talk to the network. I can configure the real ethernet adaptor (8139too) as eth1 with no problems.
Also, for some reason the loopback device is not added to the routing table which in turn causes a stack more problems.
Lots of other stuff also doesn't work including X.
Quoted 7 lines
But if I boot 2.6.3 or set acpi=off for 2.6.6 everything works perfectly. The same problems exist in 2.6.5 but I haven't tried 2.6.4.
I've attached my dmesg and .config - let me know what else is needed.
Thanks,
David.
This is not a ACPI problem. Here's the culprit:
CONFIG_IEEE1394_ETH1394=m
The IEEE1394 ethernet module is loaded and is now eth0. Tell tail sings are the long MAC address and the Link encap. which is set to unspecified.
Jurgen
Quoted 30 lines · Jurgen Kramer
Quoted 21 lines · David Johnson
Hi,
I've got some serious strangeness being caused by ACPI on 2.6.6.
2.6.3 works perfectly on the same machine. I don't use any acpi option on the kernel command line so it's just using the default options.
The first problem is that a strange device appears as eth0:
eth0 Link encap:UNSPEC HWaddr 00-90-F5-00-00-22-91-25-00-00-00-00-00-00-00-00
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)It seems to be some incarnation of the real ethernet adaptor, but even when I manually configure it it is unable to actually talk to the network. I can configure the real ethernet adaptor (8139too) as eth1 with no problems.
Also, for some reason the loopback device is not added to the routing table which in turn causes a stack more problems. Lots of other stuff also doesn't work including X.
Quoted 6 lines
But if I boot 2.6.3 or set acpi=off for 2.6.6 everything works perfectly. The same problems exist in 2.6.5 but I haven't tried 2.6.4.
I've attached my dmesg and .config - let me know what else is needed.
Thanks,
David.
Quoted 7 lines
This is not a ACPI problem. Here's the culprit:
CONFIG_IEEE1394_ETH1394=mThe IEEE1394 ethernet module is loaded and is now eth0. Tell tail sings are the long MAC address and the Link encap. which is set to unspecified.
Could you clarify that a bit? Why is this correct behaviour on 2.6.6 and not 2.6.3? Actually, if there's no adaptor I wouldn't think the module would be loaded, and if there is I would hope it would be configured usefully.
I do understand why loading would cause problems, my question is why loading is correct on this system, and why ACPI is doing so now and not in 2.6.3.
Signature · 3 lines
-bill davidsen ([email protected]) "The secret to procrastination is to put things off until the last possible moment - but no longer" -me