| 1 |
On Sonntag, 24. April 2011, Michał Górny wrote: |
| 2 |
> On Sun, 24 Apr 2011 21:43:16 +0200 |
| 3 |
> |
| 4 |
> Matthias Schwarzott <zzam@g.o> wrote: |
| 5 |
> > Sounds like we should fix udev ebuild and some ebuilds installing |
| 6 |
> > udev rules to not use /$(get_libdir)/udev, but plain /lib/udev. |
| 7 |
> > |
| 8 |
> > I used that in believe that /lib is identical or links |
| 9 |
> > to /$(get_libdir) and multilib-strict requires it, but it seems to be |
| 10 |
> > intelligent enough to only deny 64-bit libs to go to /lib. |
| 11 |
> > |
| 12 |
> > So proper udev should use /lib/udev, correct? |
| 13 |
> |
| 14 |
> Do you really think it'd be fine for some systems to possibly |
| 15 |
> have /lib64 and /lib with random different contents? |
| 16 |
|
| 17 |
Regarding /lib64/udev vs. /lib/udev: I think it is fine for some time. |
| 18 |
Having some rules only in /lib64/udev when udevd looks info /lib/udev will |
| 19 |
make only these things break that depend on the extra rules. |
| 20 |
|
| 21 |
The main question is: How many systems are affected by this /lib64 is not the |
| 22 |
same as /lib ? |
| 23 |
|
| 24 |
Matthias |
Archives