1 |
On 02/11/2013 14:04, hasufell wrote: |
2 |
> Another round of questioning the users here. |
3 |
> |
4 |
> more specifically: |
5 |
> * how often do you experience useless rebuilds? |
6 |
|
7 |
Let's see, that depends. |
8 |
|
9 |
It mostly mostly on what is happening with poppler and icu today. |
10 |
|
11 |
Other than those, I don't recall any long-term problems relating to |
12 |
useless rebuilds; by and large they seem to be valid |
13 |
|
14 |
> * do you really have a problem with running |
15 |
> revdep-rebuild/haskell-updater/perl-cleaner etc after every emerge? |
16 |
|
17 |
No, no problem whatsoever. emerge @preserved -rebuild is my preferred |
18 |
method, I find it vastly superior to sub-slot operators which |
19 |
|
20 |
a) involve weird magic in the ebuild and |
21 |
b) I still don't understand despite reading all the docs I can find on |
22 |
the matter. |
23 |
|
24 |
The problem seems to be that preserved-rebuild and revdep-rebuild detect |
25 |
actual breakage and fix what is really wrong right now. |
26 |
|
27 |
Subslots seem to try and avoid breakage and depend heavily on amount of |
28 |
clue from the dev (a highly variable quantity) |
29 |
|
30 |
> * do you think it's worth the effort to add more stuff to the PM, so |
31 |
> that you don't have to run revdep-rebuild that often? |
32 |
|
33 |
No, subslots should be obsoleted, I consider them a failed experiment in |
34 |
unnecessary complexity. emerge world is already long-running, if I had a |
35 |
problem with a few more steps after it completes I would write a wrapper |
36 |
script |
37 |
|
38 |
> * do you trust the other methods like subslots or preserved-rebuild to |
39 |
> work reliably? (as in: do you still use revdep-rebuild?) |
40 |
|
41 |
yes, revdep-rebuild is my plan C. Occasionally it finds something, |
42 |
usually itcompletes in about 40 secodns and is clean. I consider those |
43 |
40 seconds well spent, a final confirmation step |
44 |
|
45 |
> |
46 |
> If you want my opinion on subslots: |
47 |
> # grep EMERGE_DEFAULT_OPTS /etc/portage/make.conf |
48 |
> EMERGE_DEFAULT_OPTS="--ignore-built-slot-operator-deps=y" |
49 |
> |
50 |
|
51 |
-- |
52 |
Alan McKinnon |
53 |
Systems Engineer^W Technician |
54 |
Infrastructure Services |
55 |
Internet Solutions |
56 |
|
57 |
+27 11 575 7585 |
58 |
|
59 |
|
60 |
-- |
61 |
Alan McKinnon |
62 |
alan.mckinnon@×××××.com |