| 1 |
On Sun, Nov 18, 2012 at 07:42:11PM -0800, Diego Elio Pettenò wrote: |
| 2 |
> On 18/11/2012 19:38, Joshua Kinard wrote: |
| 3 |
> > Correct me if wrong, but didn't the issue start with udev wanting to put the |
| 4 |
> > PCI ID database/file into /usr/share from /etc? Then kmod was changed to |
| 5 |
> > link against libs in /usr/lib, and then udev made dependent on kmod? I |
| 6 |
> > think that led to a scenario where openrc starts udev up before localmount |
| 7 |
> > has run, and then things fall apart. |
| 8 |
> |
| 9 |
> I honestly can't remember if pci.ids was ever in /etc — I always knew it |
| 10 |
> in /usr/share/misc... |
| 11 |
|
| 12 |
Yes, it was always in /usr/somewhere. |
| 13 |
|
| 14 |
And the pci.ids file came from the pciutils package, not udev. |
| 15 |
|
| 16 |
But note, we are moving that file out of pciutils (and the usb.ids file |
| 17 |
out of usbutils) and they will eventually be generated from the udev |
| 18 |
package itself, as it holds the master hardware database. But that's a |
| 19 |
totally different topic than the one at hand, and is still being worked |
| 20 |
on by the developers of the different upstream packages. |
| 21 |
|
| 22 |
thanks, |
| 23 |
|
| 24 |
greg k-h |
Archives