Summary: | Shutting down a ipw3945 interface causes kernel panic in baselayout 1.12.4 | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Antek Grzymała (antoszka) <antoni> |
Component: | [OLD] Core system | Assignee: | Christian Heim (RETIRED) <phreak> |
Status: | RESOLVED FIXED | ||
Severity: | critical | CC: | beatryder, pacho, uberlord, zaai |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | |||
Bug Blocks: | 144431 |
Description
Antek Grzymała (antoszka)
2006-08-14 07:16:04 UTC
The second screenshot link is for my compy :D I have found that shutting down services in this order works: net.eth1 ipw3945d the rest of the system. Add this to /etc/conf.d/net.eth1 RC_NEED="ipw3945d" That seemed to do it. I'll take care of that when I find some time to fix all pending ipw3945 bugs. > Add this to /etc/conf.d/net.eth1 > RC_NEED="ipw3945d" This causes errors (about missing ipw3945 service) on my setup. And doesn't even start the card properly. How is it supposed to work? @Ryan Neufeld: > That seemed to do it. Strange. What versions of related packages are you using? No errors upon starting your system up? I am using the service listed here: http://forums.gentoo.org/viewtopic-p-3351901.html#3351901 Well, that's not a perfectly kosher solution I think. Will stick with baselayout-1.12.1 until this gets resolved properly. Hey, can I be of any help in solving this bug? I have the hardware so I can test possible solutions and report. This is pretty crucial to me and I'd be willing not to wait indefinitely for a solution. (In reply to comment #8) > Hey, can I be of any help in solving this bug? I have the hardware so I can > test possible solutions and report. This is pretty crucial to me and I'd be > willing not to wait indefinitely for a solution. The solution offered is the best I have at the moment. AFAIK the ebuild doesn't provide an init script or anything like that atm and I don't have the hardware to test. This should be fixed now with 1.7.22-r2 in portage. Feel free to reopen if the issue isn't fixed for you. Problem is not solved. The new init script improperly starts the daemon. Osiris ryan # modprobe -r ipw3945 Osiris ryan # modprobe ipw3945 Osiris ryan # /etc/init.d/ipw3945d restart * Service ipw3945d starting ipw3945d - regulatory daemon Copyright (C) 2005-2006 Intel Corporation. All rights reserved. version: 1.7.22 2006-09-06 10:44:44: ERROR: Could not find Intel PRO/Wireless 3945ABG Network Connection [ !! ] * ERROR: ipw3945d failed to start Osiris ryan # ipw3945d ipw3945d - regulatory daemon Copyright (C) 2005-2006 Intel Corporation. All rights reserved. version: 1.7.22 Intel PRO/Wireless 3945ABG Network Connection found at: /sys/bus/pci/drivers/ipw3945/0000:0c:00.0 Daemon launched as pid 21026. Exiting. No its not fixed. I'm using baselayout 1.12.6 and the kernel panic is still there. Might be a slightly different use case since I'm using wpa_supplication. I've added the line RC_NEED="ipw3945d" to /etc/conf.d/net.eth1 (is this correct?), but it has no effect. I notice that eth1 is stopped before ipw3945. During shutdown: --- * Stopping eth1 * Bringing down eth1 * Stopping dhclient on eth1 ... * Shutting down eth1 ... * Stopping wpa_cli on eth1 * Stopping wpa_supplicant on eth1 * Stopping ipw3945d ... * Stopping Hardware abstraction Layer daemon ... # BUG: unable to handler kernel paging request at virtual address 00010000 blah blah <0>Kernel panic - ... |