| 1 |
Apparently, though unproven, at 16:11 on Friday 22 October 2010, Allan |
| 2 |
Gottlieb did opine thusly: |
| 3 |
|
| 4 |
[snip] |
| 5 |
|
| 6 |
> > package.mask is the right place, but you should add the specific version. |
| 7 |
> > Then the system will only upgrade when a newer (hopefully fixed) version |
| 8 |
> > arrives. |
| 9 |
> > |
| 10 |
> > =media-libs/mesa-7.8.2 |
| 11 |
> |
| 12 |
> I tried this yesterday with great success. As mentioned in b.g.o. 7.8.2 |
| 13 |
> causes slowdowns for many people (including me). |
| 14 |
> This system is ~amd64 |
| 15 |
> |
| 16 |
> But this morning after an eix-sync, my normal update failed |
| 17 |
|
| 18 |
[snip] |
| 19 |
|
| 20 |
> !!! All ebuilds that could satisfy ">=media-libs/mesa-7.8_rc[nptl=]" have |
| 21 |
> been masked. !!! One of the following masked packages is required to |
| 22 |
> complete your request: - media-libs/mesa-7.8.2 (masked by: package.mask) |
| 23 |
> /etc/portage/package.mask: |
| 24 |
> # This version of mesa 7.8.2 is rummored to cause slowdown |
| 25 |
> # The previous version 7.7.1 is rummored to be much better |
| 26 |
> # Masking only 7.8.2 so that future (fixed??) versions can be installed |
| 27 |
> |
| 28 |
> (dependency required by "x11-base/xorg-server-1.9.0.902" [ebuild]) |
| 29 |
> For more information, see the MASKED PACKAGES section in the emerge |
| 30 |
> man page or refer to the Gentoo Handbook. |
| 31 |
> |
| 32 |
> Would you like to merge these packages? [Yes/No] no |
| 33 |
> |
| 34 |
> Quitting. |
| 35 |
> |
| 36 |
> ajglap gottlieb # |
| 37 |
> |
| 38 |
> Perhaps I should be downgrading xorg-server as well. |
| 39 |
|
| 40 |
Masking mesa-7.8.2 means (per the ebuilds) you will have to drop back to |
| 41 |
xorg-server-1.7.7-r1. Both are latest stable versions. Or, you could always |
| 42 |
install the xorg overlay and build the latest git copy of mesa... |
| 43 |
|
| 44 |
> |
| 45 |
> Since I would rather have a slow X than an angry portage, |
| 46 |
> I removed the package mask and expected all to be well, but was |
| 47 |
> surprised by the following. In particular at the end it says there is |
| 48 |
> one block but I don't see any. |
| 49 |
|
| 50 |
[snip] |
| 51 |
|
| 52 |
Here you go: |
| 53 |
|
| 54 |
|
| 55 |
> [blocks b ] <sys-apps/shadow-4.1.4.2-r6 |
| 56 |
> ("<sys-apps/shadow-4.1.4.2-r6" is blocking sys-auth/consolekit-0.4.2-r3) |
| 57 |
> [ebuild U ] sys-apps/shadow-4.1.4.2-r6 [4.1.4.2-r5] USE="cracklib |
| 58 |
> nls pam -audit (-selinux) -skey" 1,749 kB |
| 59 |
|
| 60 |
Don't worry about it. I'm not sure if portage-2.1.9.20 will deal with this |
| 61 |
automagically (I *think* it does these days and 2.2 definitely does) but if |
| 62 |
not just |
| 63 |
|
| 64 |
emerge -C shadow ; emerge -1 shadow |
| 65 |
|
| 66 |
then emerge -avuND world. |
| 67 |
|
| 68 |
No good technical reason for doing shadow first apart from getting it over and |
| 69 |
done with while you watch and confirm it works fine. Then do world and wander |
| 70 |
over to the kettle letting portage go on with doing it's thing unattended |
| 71 |
|
| 72 |
|
| 73 |
-- |
| 74 |
alan dot mckinnon at gmail dot com |
Archives