Re: 2.6.14 USB vs. sleep issues
On Tue, 2005-11-15 at 08:05 +1100, Benjamin Herrenschmidt wrote:
> On Mon, 2005-11-14 at 13:07 -0600, drag sidious wrote:
>
> > By rmmod'ing the ohci_hcd modules the script it disables the mouse and
> > seems to avoid the kernel-panic-on-resume issues with it. Of course if
> > you have other USB devices you'll probably have to modify the scripts
> > for that. With the mouse the usbhid module is loaded automaticly after
> > the ohci_hcd is loaded after resume. For X it works fine because, but
> > other people may have to do ctrl-alt-F1 to switch out of X and then
> > ctrl-alt-whatever to get back in.
> >
> > Also should be handy for other things that may get goofy after a
> > suspend-resume cycle.
> >
> > That way I got a temporary work around without having to recompile the
> > kernel for now.
>
> It's generally not a good idea to rmmod the USB controllers though, that
> prevent them from suspending which isn't a good idea
>
> Ben.
>
>
Ok. I was afraid that it would be a problem. I knew it was a hack from
the beginning.
I found a nice howto on building packages from deb-src files and being
able to patch them.
http://www.togaware.com/linux/survivor/Building_Packages0.shtml
So I followed that and was able to successfully apply your patch by
going (after using wajig to download sources, and using wget to fetch
your patch):
cd linux-2.6-2.6.14
patch -p1 < ../fix-ohci-sleep.diff
I am using that and I disabled my script and it works pretty well. No
crashes or anything like that anymore with just my USB mouse (haven't
tried anything else). Very good job.
Reply to: