1 |
Joel Wiramu Pauling posted <1120725375.18318.2.camel@localhost>, excerpted |
2 |
below, on Thu, 07 Jul 2005 20:36:15 +1200: |
3 |
|
4 |
> I did an emerge deep world last night, and noticed it pick up new kde |
5 |
> ebuilds. This morning KDE and almost any KDE app (with the exception of |
6 |
> LICQ's kde-gui plugin... which doesn't really use the framework very |
7 |
> much) either hangs, or hardlocks the system. |
8 |
> |
9 |
> Has anyone else doing a deep world upgrade with kde suffered this |
10 |
> problem? |
11 |
> |
12 |
> I am attempting to rebuild qt and kdelibs in the hope that it solves it |
13 |
> but this has really put a dampner on any development work I was hoping |
14 |
> to do before next week. |
15 |
> |
16 |
> BTW I am running ~amd64 |
17 |
|
18 |
No issues of that nature here, and the only DE I run is KDE, so if it hit, |
19 |
I'd DEFINITELY notice it and be scrambling to figure out the issue and get |
20 |
things fixed. |
21 |
|
22 |
I just resynced to be sure, altho I'd synced about 14 hours ago and |
23 |
updated everything. The only things a -puD lists as not updated in the |
24 |
~14 hours is dev-libs/libxslt-1.1.14 (1.1.13-r1 currently merged), and |
25 |
sys-fs/udev-061 (060 currently merged). |
26 |
|
27 |
>From previous updates, I still show x11-base/xorg-x11-6.8.99.14 as |
28 |
unmerged (.13 currently merged). I've unmasked these snapshots locally, |
29 |
but .14 scrambled the second monitor in xinerama mode so I reverted to |
30 |
.13, and haven't reset masking to mask .14 only, yet. |
31 |
|
32 |
One POSSIBLE culprit COULD be the qt update from a few days ago. The |
33 |
changelog for it said the changes were only for the database folks, and I |
34 |
run KDE off it not any databases, so that shouldn't affect me and I just |
35 |
masked the new version (x11-libs/qt-3.3.4-r6, I've still -r5 merged). |
36 |
I've found that after merging a new Qt, KDE still works, but with a couple |
37 |
quirks (mainly, menus are the wrong color, but it's possible/likely there |
38 |
are other quirks I haven't noticed), until I remerge kdelibs as well. |
39 |
Since both Qt and kdelibs are rather long merges, I didn't feel like |
40 |
remerging both of them for a Qt version bump for database stuff that |
41 |
shouldn't affect me anyway, so I simply masked the bumped Qt. It's |
42 |
possible but not likely that this particular Qt causes other issues as |
43 |
well, and that's what you are observing. |
44 |
|
45 |
Do note, however, that in addition to running ~amd64, I'm running |
46 |
testing-masked versions of gcc (now 4.0.1_pre20050702) and binutils |
47 |
(2.16.91.0.1) to support gcc-4.x. I run the 4.0 gcc-config profile as my |
48 |
default, and have been compiling anything that will with it for some time, |
49 |
tho I still have gcc-3.4.x merged (in its own parallel slot, naturally) |
50 |
and gcc-config to it for stuff (such as xorg) that won't yet compile with |
51 |
gcc-4.x. Thus, much of my system, including Qt and all of KDE-3.4.1, is |
52 |
now gcc-4 compiled. If you are running a straight ~amd64 system, you will |
53 |
have everything gcc-3.x compiled, which /might/ make a difference, tho in |
54 |
theory my gcc-4.x compiled system should be less stable. |
55 |
|
56 |
Also, you mention that you run ~amd64, but don't mention how often you |
57 |
update. It's quite possible if you update only about once a week or so, |
58 |
compared to my usual 1-3 day updates, that my updates were processed in a |
59 |
different order than yours, and some dependency (qt and kdelibs as you |
60 |
mentioned, but also arts) needs recompiled in your case. Have you tried a |
61 |
revdep-rebuild lately? That /might/ catch it. |
62 |
|
63 |
-- |
64 |
Duncan - List replies preferred. No HTML msgs. |
65 |
"Every nonfree program has a lord, a master -- |
66 |
and if you use the program, he is your master." Richard Stallman in |
67 |
http://www.linuxdevcenter.com/pub/a/linux/2004/12/22/rms_interview.html |
68 |
|
69 |
|
70 |
-- |
71 |
gentoo-amd64@g.o mailing list |