Openembedded Devel Discussions
 help / color / mirror / Atom feed
From: Roman I Khimov <roman@khimov.ru>
To: openembedded-devel@lists.openembedded.org
Subject: Re: [PATCH 3/3] uClibc: redo configuration
Date: Tue, 6 Apr 2010 00:22:51 +0400	[thread overview]
Message-ID: <201004060022.54403.roman@khimov.ru> (raw)
In-Reply-To: <1268415992-19603-4-git-send-email-rep.dot.nop@gmail.com>

[-- Attachment #1: Type: Text/Plain, Size: 2940 bytes --]

В сообщении от Пятница 12 марта 2010 20:46:32 автор Bernhard Reutner-Fischer 
написал:
> +# Map distro and machine features to config settings
> +def features_to_uclibc_settings(d):
> +	cnf, rem = ([], [])
> +	features = bb.data.getVar('DISTRO_FEATURES', d).split()
> +	uclibc_cfg('ipv4', features, 'UCLIBC_HAS_IPV4', cnf, rem)
> +	uclibc_cfg('ipv6', features, 'UCLIBC_HAS_IPV6', cnf, rem)
> +	uclibc_cfg('nls',  features, 'UCLIBC_HAS_LOCALE', cnf, rem)
> +	uclibc_cfg('bx',   features, 'USE_BX', cnf, rem)
> +	return "\n".join(cnf), "\n".join(rem)

Last time I've tried to do something similar in recipes it was considered as 
BAD (USE-flag like) thing. Although I'm fine with it.

But in general, I don't see how it makes life any easier. It adds logic to 
determine some of options, but then there are still lots of other ones. And 
they would still come through machine and distro config which would still be 
carried around and stored for each version. If this dynamic things somehow 
eliminated any need in those, than it would be excellent, but from what I see 
it's not the case and probably there is no way to sanely implement it.

What could make sense is to improve those static configs we have. From what I 
understand, most of machine config parts are not changing from version to 
version and even if they do, it should mostly be adding things, which should 
be harmless for old versions. So we can try to move this configs from 
uclibc-${PV}/ to files/ or uclibc/, so that they would be reused in all 
versions.

Distro configs do change between versions, but than there are not that much 
distros that care about uclibc and each of those could store their version of 
config if there is any need for that.

>  b/recipes/uclibc/uclibc-initial_git.bb index 00330e2..3a33bd8 100644
> --- a/recipes/uclibc/uclibc-initial_git.bb
> +++ b/recipes/uclibc/uclibc-initial_git.bb
> @@ -7,19 +7,16 @@ PACKAGES = ""
> 
>  do_install() {
>  	# Install initial headers into the cross dir
> -	make V=1 CC="${CC}" PREFIX=${D} DEVEL_PREFIX=${prefix}/ RUNTIME_PREFIX=/
>  \ +	make PREFIX=${D} DEVEL_PREFIX=${prefix}/ RUNTIME_PREFIX=/ \
>  		install_headers
>  	#ln -sf include ${CROSS_DIR}/${TARGET_SYS}/sys-include
> 
> -	# This conflicts with the c++ version of this header
> -	rm -f ${D}${includedir}/bits/atomicity.h
>  	install -d ${D}${libdir}/
.....
>  do_compile () {
> -	make V=1 CC="${CC}" PREFIX=${D} DEVEL_PREFIX=${prefix}/ RUNTIME_PREFIX=/
>  \ +	make PREFIX=${D} DEVEL_PREFIX=${prefix}/ RUNTIME_PREFIX=/ \


I guess these are things that just not needed with newer uClibc versions? If 
so, this would justify moving old uclibc.inc to uclibc-old.inc and adopting 
newer simpler uclibc.inc (with recipes clean up) IMO.

-- 
 http://roman.khimov.ru
mailto: roman@khimov.ru
gpg --keyserver hkp://subkeys.pgp.net --recv-keys 0xE5E055C3

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

  reply	other threads:[~2010-04-05 20:29 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-02-21 15:31 [PATCH] uClibc: rename main include Bernhard Reutner-Fischer
2010-02-22  5:54 ` Khem Raj
2010-03-12 17:46   ` [RFC, PATCH 0/3] uClibc recipe touchup Bernhard Reutner-Fischer
2010-03-12 17:46   ` [PATCH 1/3] uClibc: rename include file of old releases Bernhard Reutner-Fischer
2010-03-12 17:46   ` [PATCH 2/3] sane-toolchain: add PREFERRED_UCLIBC_VERSION Bernhard Reutner-Fischer
2010-04-05 19:49     ` Roman I Khimov
2010-04-08  9:30       ` Stefan Schmidt
2010-06-09 17:11         ` Bernhard Reutner-Fischer
2010-03-12 17:46   ` [PATCH 3/3] uClibc: redo configuration Bernhard Reutner-Fischer
2010-04-05 20:22     ` Roman I Khimov [this message]
2010-06-09 17:10   ` [PATCH 0/3] uClibc configury touchup; RFC WRT feature/dep picking heuristics Bernhard Reutner-Fischer
2010-06-09 21:26     ` RFC " Bernhard Reutner-Fischer
2010-06-09 17:10   ` [PATCH 1/3] uClibc: redo configuration Bernhard Reutner-Fischer
2010-06-09 17:10   ` [PATCH 2/3] uclibc_git: keep PV at "git" Bernhard Reutner-Fischer
2010-06-09 17:10   ` [PATCH 3/3] busybox: picking IPv6 per default is not up to the package Bernhard Reutner-Fischer
2010-06-09 18:44     ` Phil Blundell
2010-06-09 18:52       ` Bernhard Reutner-Fischer
2010-06-09 19:18         ` Phil Blundell
2010-06-09 19:32           ` Bernhard Reutner-Fischer
2010-06-09 20:22             ` Phil Blundell
2010-06-10 19:46               ` [PATCH 1/2] uclibc: handle DISTRO_FEATURE="largefile" Bernhard Reutner-Fischer
2010-06-10 19:46               ` [PATCH 2/2] busybox: configure according to {MACHINE, DISTRO}_FEATURES Bernhard Reutner-Fischer
2010-06-10 19:55                 ` Chris Larson
2010-06-10 20:22                   ` Phil Blundell
2010-06-10 20:27                     ` Chris Larson
2010-06-10 20:50                       ` Bernhard Reutner-Fischer
2010-06-10 21:06                         ` Phil Blundell
2010-06-10 21:20                           ` Bernhard Reutner-Fischer
2010-06-11 12:59                             ` Phil Blundell
2010-06-22 20:39                               ` Bernhard Reutner-Fischer
2010-06-28 18:23                                 ` ping**2 [was: Re: [PATCH 2/2] busybox: configure according to {MACHINE, DISTRO}_FEATURES] Bernhard Reutner-Fischer
2010-06-10 22:56                       ` [PATCH 2/2] busybox: configure according to {MACHINE, DISTRO}_FEATURES Khem Raj
2010-06-11  7:16                         ` Bernhard Reutner-Fischer
2010-06-10 20:44                     ` Bernhard Reutner-Fischer
2010-06-10 21:09                       ` Phil Blundell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=201004060022.54403.roman@khimov.ru \
    --to=roman@khimov.ru \
    --cc=openembedded-devel@lists.openembedded.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox