1 |
On Wed, Dec 21, 2016 at 10:06:00PM +0200, Alan McKinnon wrote: |
2 |
> On 21/12/2016 21:51, Rich Freeman wrote: |
3 |
> > On Wed, Dec 21, 2016 at 2:46 PM, Rich Freeman <rich0@g.o> wrote: |
4 |
> >> On Wed, Dec 21, 2016 at 2:20 PM, <Meino.Cramer@×××.de> wrote: |
5 |
> >> |
6 |
> >>> The following USE changes are necessary to proceed: |
7 |
> >>> (see "package.use" in the portage(5) man page for more details) |
8 |
> >>> # required by kde-plasma/kwin-5.8.3::gentoo |
9 |
> >>> # required by kde-plasma/plasma-workspace-5.8.3-r4::gentoo |
10 |
> >>> # required by net-p2p/ktorrent-5.0.1::gentoo[shutdown] |
11 |
> >>> # required by @selected |
12 |
> >>> # required by @world (argument) |
13 |
> >>>> =media-libs/mesa-12.0.1 wayland |
14 |
> >> |
15 |
> >> |
16 |
> >> I suggest ignoring this for the moment and see if the info above |
17 |
> >> resolves your systemd issues. I'm not sure why kwin has the |
18 |
> >> dependency that it does, but it looks to me like it is set up as a |
19 |
> >> hard dependency that you can't avoid without modifying the ebuild. |
20 |
> >> I'll see if I can figure out more. The changes above should at least |
21 |
> >> get rid of whatever is pulling in systemd. |
22 |
> >> |
23 |
> >> Installing wayland shouldn't actually hurt anything. I noticed that I |
24 |
> >> have it installed likely for the same reason, and it isn't like it |
25 |
> >> will start running on its own. But, I'm not sure yet whether you can |
26 |
> >> avoid it. |
27 |
> >> |
28 |
> > |
29 |
> > Well, I should have just waited to reply, but here is the issue: |
30 |
> > https://mail.kde.org/pipermail/release-team/2015-July/008725.html |
31 |
> > |
32 |
> > kwin does in fact have a non-conditional dependency on wayland, so you |
33 |
> > need to install it. It won't do anything if you don't run it, but it |
34 |
> > is not possible to build kwin without wayland support. Judging by the |
35 |
> > claim in the email that it used to take 100 conditionals in the source |
36 |
> > to make it optional, I doubt anybody in Gentoo will be patching this |
37 |
> > anytime soon. I guess you could always fork it if you wanted to. |
38 |
> > |
39 |
> > So, sorry, not what you wanted to hear, and not really what I care to |
40 |
> > hear either since I don't use wayland, but at least it doesn't need to |
41 |
> > be running in this case. I wouldn't be surprised if that changes in |
42 |
> > the future, but everybody knows that xorg is on borrowed time right |
43 |
> > now. |
44 |
> > |
45 |
> > Well, if nothing else at least this splits the thread so that you can |
46 |
> > reply to the systemd and the wayland issues separately... |
47 |
> > |
48 |
> |
49 |
> |
50 |
> Doesn't it strike you as curious that the 4 extra wayland packages |
51 |
> consume 8.5M installed (sans size of sources in distfiles) and for 18 |
52 |
> months no-one has raised nary a whimper about it, whereas recall the |
53 |
> giant whinge-fest a while back about a few 10s of harmless unit files |
54 |
> (text), each less than one fs block? |
55 |
|
56 |
How does a file take up less than a single FS block? An inode has to be |
57 |
allocated _somewhere_, does it not? |
58 |
|
59 |
As for the KDE <-> wayland thing, it's possible KDE users don't care |
60 |
about something like that. In fact I would argue that the average person |
61 |
who wants a desktop environment cares little about the dependency tree |
62 |
for said environment, because they care more about their DE than what it |
63 |
takes to run it. It's also an exercise in insanity, given the size of |
64 |
DEs. I applaud the teams working on packaging them; it's a huge effort. |
65 |
|
66 |
Exceptions are obviously GNOME + systemd, which caused a large upset. |
67 |
If/when KDE starts requiring wayland, I expect a similar, though |
68 |
somewhat smaller outcry. It'll come down to having a quality Wayland |
69 |
setup that's as smooth to get up and running as Xorg. |
70 |
|
71 |
One of the issues with wayland adoption is 1) the sheer amount of |
72 |
software that'd need to be ported (there's apparently an X-compatible |
73 |
mode in Wayland or something, but I've heard nothing regarding its |
74 |
quality or its interoperability), 2) its structure is not exactly clear, |
75 |
3) afaik there is no known compositor that supports all the features |
76 |
you'd want to see in a modern display manager, and 4) writing tools |
77 |
for the display server basically depends on a competent and extensible |
78 |
compositor since everything goes through it first. |
79 |
|
80 |
What I see more likely in the future for Wayland are different levels of |
81 |
"rich" experience, all using a common Wayland protocol to be |
82 |
mix-and-match. So someone can have a barebones system with Fluxbox (like |
83 |
me), but another Wayland implementation can have all the bells and |
84 |
whistles the DE crowd is used to. Who knows? |
85 |
|
86 |
> |
87 |
> For the record, openrc user here on Gentoo; systemd on Ubuntu at work |
88 |
> (no feasible choice with Ubuntu) |
89 |
> |
90 |
> -- |
91 |
> Alan McKinnon |
92 |
> alan.mckinnon@×××××.com |
93 |
> |
94 |
> |