| 1 |
On 11/08/13 08:36, Walter Dnes wrote: |
| 2 |
> On Sat, Aug 10, 2013 at 09:57:52AM +0300, Samuli Suominen wrote |
| 3 |
> |
| 4 |
>> I expect it to happen around every new udev release that causes slight |
| 5 |
>> incompability; the default of the virtual/udev, sys-fs/udev, doesn't |
| 6 |
>> have to wait for the alternative providers. |
| 7 |
> |
| 8 |
> The elegant solution is outlined in my post... |
| 9 |
> http://www.gossamer-threads.com/lists/gentoo/user/275977#275977 |
| 10 |
> |
| 11 |
> I.e. *UNTIL SUCH TIME AS EUDEV HITS STABLE* (on whatever arch you're |
| 12 |
> using), add the entry |
| 13 |
> |
| 14 |
> <sys-fs/eudev-9999 ~amd64 |
| 15 |
> |
| 16 |
> to package.keywords (replace amd64 with your arch if necessary). |
| 17 |
> Basically, if you keyword a specific version, and the ebuild gets |
| 18 |
> removed by "emerge --sync", there are no eudev ebuilds to satisfy |
| 19 |
> virtual/udev. So portage falls back to udev. My solution isn't |
| 20 |
> hard-coded to any one version, and is immune to to version bumps and |
| 21 |
> removals. |
| 22 |
> |
| 23 |
|
| 24 |
bad idea to unmask "the new multilib eudev" on stable, regarding |
| 25 |
blockers it has like |
| 26 |
"!<=app-emulation/emul-linux-x86-baselibs-20130224-r7" on amd64 multilib |
| 27 |
when ABI_X86="32" is enabled |
| 28 |
|
| 29 |
as in, unresolvable dependencies |
Archives