Gentoo Archives: gentoo-dev

From: "Michał Górny" <mgorny@g.o>
To: William Hubbs <williamh@g.o>
Cc: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] new profile layout with flavors and mix-ins
Date: Wed, 02 Jul 2014 17:54:55
Message-Id: 20140702195437.09c8efdb@pomiot.lan
In Reply to: [gentoo-dev] new profile layout with flavors and mix-ins by William Hubbs
1 Dnia 2014-07-02, o godz. 10:44:16
2 William Hubbs <williamh@g.o> napisał(a):
3
4 > All,
5 >
6 > I'm moving to a new thread since the discussion has moved away from just
7 > a sub profile for no-multilib.
8 >
9 > On Wed, Jul 02, 2014 at 09:30:50AM -0400, Rich Freeman wrote:
10 > > On Wed, Jun 25, 2014 at 4:01 PM, Andreas K. Huettel
11 > > <dilfridge@g.o> wrote:
12 > > > Am Mittwoch 25 Juni 2014, 15:11:40 schrieb Rich Freeman:
13 > > >> On Wed, Jun 25, 2014 at 2:44 PM, Michał Górny <mgorny@g.o> wrote:
14 > > >> > Long story short, doing anything to Gentoo profiles is utter pain
15 > > >> > and comes with random breakage guarantee. Therefore, I'm asking -- nuke
16 > > >> > those damn profiles, and start over! The current situation is
17 > > >> > completely unmaintainable.
18 > > >>
19 > > >> ++
20 > > >>
21 > > >> But, would it make sense to just go the Funtoo route with "mix-ins."
22 > > > ++
23 > > >
24 > > > this is what we've been just discussing on the irc channel
25 > >
26 > > So, not wanting this to die on the vine.
27 > >
28 > > If we did the mix-in approach, would we just follow the example of Funtoo?
29 > >
30 > > They use an arch profile, a stability profile (~arch vs arch), a
31 > > "flavor" profile (core, minimal, desktop), and then users can layer as
32 > > much other stuff on top of that as they want (gnome, kde, multimedia,
33 > > etc).
34 >
35 > I think this could work for us as well, or something similar anyway.
36 >
37 > For those who are curious, I am including the link to the flavors and
38 > mix-ins descriptions from the funtoo site. [1]
39
40 It's not that easy. As you can see on that site, they're supporting
41 much less variants than Gentoo does. In particular, they don't seem to
42 support non-GNU/Linux at all. No Prefix, no Hardened, no FreeBSD.
43
44 I was thinking about modularization a bit and the main issue is
45 handling intersecting profiles. As you can see in Funtoo, it already
46 starts with the 'build' flavor -- they're pretty much applying a cheap
47 hack (ACCEPT_KEYWORDS="${ARCH}") but such hacks can't cover all our
48 needs.
49
50 A simple example is CHOST. The value of CHOST depends on the arch,
51 often ABI, kernel, libc. Of course, we could hack this around by
52 creating some intermediate variables and merging them afterwards.
53 But not everything can be hacked around like this.
54
55 I don't feel like we ought to vote on something like this without
56 understanding most of the current profiles. And I'm afraid there are
57 only few people who have any idea about the current profile
58 structure...
59
60 > > Do we want to do things the same way?
61 > >
62 > > Some things to think about include multilib (just another arch?),
63 > > systemd, and usr-merge. I'm not saying that we need to implement any
64 > > of that stuff completely - but when planning the profile layout we
65 > > should at least consider whether it will handle things like this in
66 > > the future. Should some types of profiles be only additive? Etc...
67 >
68 > I see systemd and multilib as mix-ins, like the ones you mentioned
69 > above.
70
71 Multilib won't work as Funtoo-style mix-in for the simple reason it
72 relies on heavily on the architecture in use.
73
74 --
75 Best regards,
76 Michał Górny

Attachments

File name MIME type
signature.asc application/pgp-signature

Replies