Gentoo Archives: gentoo-mips

From: Markos Chandras <hwoarang@g.o>
To: gentoo-mips@l.g.o
Subject: Re: [gentoo-mips] On MIPS using the same CHOST for all multilib ABIs
Date: Fri, 17 Jan 2014 18:21:15
Message-Id: 52D9746E.4020200@gentoo.org
In Reply to: Re: [gentoo-mips] On MIPS using the same CHOST for all multilib ABIs by "Michał Górny"
1 On 01/17/2014 04:47 AM, Michał Górny wrote:
2 > Dnia 2014-01-16, o godz. 17:29:43 "Anthony G. Basile"
3 > <blueness@g.o> napisał(a):
4 >
5 >> On 01/16/2014 04:24 PM, Michał Górny wrote:
6 >>> Because AC_PATH_TOOL uses CHOST and some random Gentoo
7 >>> invention.
8 >>
9 >> I got that AC_PATH_TOOL and AC_CHECK_TOOL prefix whatever utility
10 >> they search for with the canonicalized chost (usually from
11 >> config.guess), but I still don't see why we need this to avoid
12 >> hackery? Can you give me a practial example because right now I
13 >> just don't see a serious problem.
14 >
15 > libgpg-error installs ${CHOST}-gpg-error-config.
16 >
17 > Now libgcrypt (and possibly other tools) are using AC_PATH_TOOL to
18 > find it. If we have proper CHOSTs, they find the right
19 > gpg-error-config and we don't have to put any more effort into
20 > that. Then libgcrypt installs ${CHOST}-libgcrypt-config.
21 >
22 > Now other tools are using AC_PATH_TOOL to find proper
23 > libgcrypt-config. If we have proper CHOSTs, it just works and we
24 > don't have to put any more effort into that.
25 >
26 > Same goes for LLVM & Mesa.
27 >
28 > If we play by the rules nicely, all pieces fit together nicely and
29 > we don't have to worry. If we don't, we ask the developers to spit
30 > Gentoo- specific hackery all over the place.
31 >
32 You need to consider that besides changing CHOST to new stages (which
33 is a lengthy and tiring process), you somehow need to migrate existing
34 users to the new CHOST (no?) otherwise the multilib eclass (or any
35 other eclass/package) that depends on CHOST will be broken as soon as
36 they update their tree and try to install package updates.
37 This is definitely not a pleasant user experience.
38
39 Apologies for the stupid question (still trying to understand how our
40 multilib works) but if our CHOST are being a problem, how come our
41 multilib profiles work fine?
42
43 --
44 Regards,
45 Markos Chandras

Replies