Re: [gentoo-user] new udev (?) loading ipw3945 without starting ipw3945d
On Fri, 1 Dec 2006, jak gentoo wrote: if you use the testing versoin of ipw3945d it comes with an rc startscript, maybe that helps. I 'm using ipw3945d testing an I didn't had problems after updating udev to 103 That solved it for me. That's probably a better design overall, anyway. Although it would be neat if the initscripts were clever enough to have net.* check whether the device it's starting needs anything, rather than simply making sure you do the necessary stuff first. -Daniel *This .sig left intentionally blank* -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] new udev (?) loading ipw3945 without starting ipw3945d
On Mon, 4 Dec 2006 15:41:27 -0500 (EST), Daniel Barkalow wrote: That solved it for me. That's probably a better design overall, anyway. Although it would be neat if the initscripts were clever enough to have net.* check whether the device it's starting needs anything, rather than simply making sure you do the necessary stuff first. You could add something to preup() in /etc/conf.d/net to run the program before the interface comes up. -- Neil Bothwick There are 10 types of people in this world, those who understand binary notation and those who don't. signature.asc Description: PGP signature
Re: [gentoo-user] new udev (?) loading ipw3945 without starting ipw3945d
On 11/30/06, Daniel Barkalow [EMAIL PROTECTED] wrote: On Wed, 29 Nov 2006, Richard Fish wrote: On 11/29/06, Daniel Barkalow [EMAIL PROTECTED] wrote: I switched to udev-103 recently, and now when I boot I find that ipw3945d is not getting started, which causes my wireless card to not appear at all. rmmod ipw3945; modprobe ipw3945 once the system has started works. Any advice? my /etc/modules.d/ipw3945 file contains the following: ipw3945d, I assume? install ipw3945 /sbin/modprobe --ignore-install ipw3945 ; sleep 0.2; /sbin/ipw3945d --quiet remove ipw3945 /sbin/ipw3945d --kill ; sleep 0.2; /sbin/modprobe -r --ignore-remove ipw3945 alias pci:v8086d4222sv*sd*bc*sc*i* off alias pci:v8086d4227sv*sd*bc*sc*i* off The alias lines are necessary to prevent udev from coldplugging the driver which otherwise would occur at a very early point in the boot sequence...in fact before /var is mounted on my system. Without /var mounted and read-write, ipw3945d cannot start. I then /sbin/modprobe ipw3945 in /etc/conf.d/local.start to load the module near the end of the boot sequence. Perhaps you need to do something similar? That's probably the same thing I need (except / rw should be sufficient for me, so I think I can use /etc/modules.autoload). I wonder if udev can be configured not to load the module. Probably the right thing is really to have ipw3945d run as regular service. It'd be nice if ipw3945 produced class net node for the device when the daemon isn't running, and just required the daemon to actually turn it on. -Daniel *This .sig left intentionally blank* -- gentoo-user@gentoo.org mailing list if you use the testing versoin of ipw3945d it comes with an rc startscript, maybe that helps. I 'm using ipw3945d testing an I didn't had problems after updating udev to 103
[gentoo-user] new udev (?) loading ipw3945 without starting ipw3945d
I switched to udev-103 recently, and now when I boot I find that ipw3945d is not getting started, which causes my wireless card to not appear at all. rmmod ipw3945; modprobe ipw3945 once the system has started works. Any advice? -Daniel *This .sig left intentionally blank* -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] new udev (?) loading ipw3945 without starting ipw3945d
On 11/29/06, Daniel Barkalow [EMAIL PROTECTED] wrote: I switched to udev-103 recently, and now when I boot I find that ipw3945d is not getting started, which causes my wireless card to not appear at all. rmmod ipw3945; modprobe ipw3945 once the system has started works. Any advice? my /etc/modules.d/ipw3945 file contains the following: install ipw3945 /sbin/modprobe --ignore-install ipw3945 ; sleep 0.2; /sbin/ipw3945d --quiet remove ipw3945 /sbin/ipw3945d --kill ; sleep 0.2; /sbin/modprobe -r --ignore-remove ipw3945 alias pci:v8086d4222sv*sd*bc*sc*i* off alias pci:v8086d4227sv*sd*bc*sc*i* off The alias lines are necessary to prevent udev from coldplugging the driver which otherwise would occur at a very early point in the boot sequence...in fact before /var is mounted on my system. Without /var mounted and read-write, ipw3945d cannot start. I then /sbin/modprobe ipw3945 in /etc/conf.d/local.start to load the module near the end of the boot sequence. Perhaps you need to do something similar? -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] new udev (?) loading ipw3945 without starting ipw3945d
On Wed, 29 Nov 2006, Richard Fish wrote: On 11/29/06, Daniel Barkalow [EMAIL PROTECTED] wrote: I switched to udev-103 recently, and now when I boot I find that ipw3945d is not getting started, which causes my wireless card to not appear at all. rmmod ipw3945; modprobe ipw3945 once the system has started works. Any advice? my /etc/modules.d/ipw3945 file contains the following: ipw3945d, I assume? install ipw3945 /sbin/modprobe --ignore-install ipw3945 ; sleep 0.2; /sbin/ipw3945d --quiet remove ipw3945 /sbin/ipw3945d --kill ; sleep 0.2; /sbin/modprobe -r --ignore-remove ipw3945 alias pci:v8086d4222sv*sd*bc*sc*i* off alias pci:v8086d4227sv*sd*bc*sc*i* off The alias lines are necessary to prevent udev from coldplugging the driver which otherwise would occur at a very early point in the boot sequence...in fact before /var is mounted on my system. Without /var mounted and read-write, ipw3945d cannot start. I then /sbin/modprobe ipw3945 in /etc/conf.d/local.start to load the module near the end of the boot sequence. Perhaps you need to do something similar? That's probably the same thing I need (except / rw should be sufficient for me, so I think I can use /etc/modules.autoload). I wonder if udev can be configured not to load the module. Probably the right thing is really to have ipw3945d run as regular service. It'd be nice if ipw3945 produced class net node for the device when the daemon isn't running, and just required the daemon to actually turn it on. -Daniel *This .sig left intentionally blank* -- gentoo-user@gentoo.org mailing list