From: Joe MacDonald <Joe_MacDonald@mentor.com>
To: "Bian, Naimeng" <biannm@cn.fujitsu.com>
Cc: "openembedded-devel@lists.openembedded.org"
<openembedded-devel@lists.openembedded.org>
Subject: Re: [PATCH 6/6] [meta-oe] Port corosync from openembedded and upgrade to 1.4.7
Date: Wed, 26 Nov 2014 09:08:21 -0500 [thread overview]
Message-ID: <20141126140819.GA3886@mentor.com> (raw)
In-Reply-To: <30DC4EDFD798C94083F3E083754532CC73DDDAC2@G08CNEXMBPEKD02.g08.fujitsu.local>
[-- Attachment #1: Type: text/plain, Size: 11063 bytes --]
[RE: [oe] [PATCH 6/6] [meta-oe] Port corosync from openembedded and upgrade to 1.4.7] On 14.11.26 (Wed 03:36) Bian, Naimeng wrote:
> > -----Original Message-----
> > From: openembedded-devel-bounces@lists.openembedded.org
> > [mailto:openembedded-devel-bounces@lists.openembedded.org] On Behalf Of
> > Alexandru Vaduva
> > Sent: Tuesday, November 25, 2014 6:19 PM
> > To: openembedded-devel@lists.openembedded.org; Joe MacDonald
> > Subject: Re: [oe] [PATCH 6/6] [meta-oe] Port corosync from openembedded and
> > upgrade to 1.4.7
> >
> > It is my opinion that meta-cgl is the place for your patch being the fact that
> > pacemaker and the rest of the HA cluster resources are available here.
>
> Thanks.
> I confirmed that corosync-1.2.8 and pacemaker are available inside meta-cgl.
>
> > Also I believe this patch should be available inside the open embedded
> > repository.
> >
>
> Actually, I do not know what kinds of recipes meta-openembedded likes to merge.
> In another hand, what kinds of recipes are no good.
> (Of course the unmaintained one such as printproto is no good.)
>
> If there are some rules, would you tell me what's that.
> Maybe it's helpful to a newcomer such as me. ^_^
I can't speak for everyone but in all honesty, where I do networking and
CGL stuff I see a lot of things that go 'unmaintained' for years and are
still perfectly functional and widely used. Time since last commit
isn't an immediate disqualifier in my view. That said, as I think was
mentioned with some of your other submissions, if you have a recipe for
a package that hasn't been maintained in several years, you should
expect to be asked "but why?" There may be a very good answer to that
that convinces a layer maintainer to integrate your recipe. You just
need to be prepared for the question.
For what it's worth, that can also apply to brand new packages. :-)
> I got some points from the following document, but it's not detailed.
> http://www.openembedded.org/wiki/OpenEmbedded-Core#meta-openembedded
>
> ----------------------------------------------------------------------------------
> meta-openembedded
>
> For items shared amongst multiple layers that do not fit into OE-Core or any other
> existing layer, there is the meta-oe layer. This exists in a repository, also called
> meta-openembedded, which contains a number of other more focused layers (meta-efl,
> meta-gnome, etc.). This layer will need to be managed carefully over time to avoid
> it turning into just a newer version of OE-classic.
> ----------------------------------------------------------------------------------
This last is the crux of the matter, I think. I personally want to
avoid collecting a bunch of recipes that aren't regularly used and fall
out of date (either with the needs of the community or with the
upstream).
> Maybe we should update this document, add some suggestions like following.
> ----------------------------------------------------------------------------------
> Attention
> Please avoid send a patch such as following.
> 1. have not been unmaintained by upstream long time
> 2. ... ...
> ----------------------------------------------------------------------------------
>
> > If it would be ok with you I would suggest to have two patches: one sent for
> > the OpenEmbedded and one for the meta-cgl.
> > If the porting for meta-cgl would not be possible please let me know for assuring
> > that the patch is ported inside meta-cgl.
>
> I confirmed that corosync-1.2.8 is available inside meta-cgl now.
> I will send the patch for corosync-1.4.7 to meta-cgl later.
That is probably best. I have no objection to it being in
meta-networking, but I'm not sure it is broadly applicable and if Alex
prefers to keep a corosync recipe in meta-cgl, I'm all for that.
> > Another extra reason for which I would prefer this patch to be also available
> > inside meta-cgl would be the extra testing that will be done for it and the
> > fact that corosync messaging layer will be used with pacemaker.
> >
>
> Looks good.
> And the same question. ^_^
> Would you tell me what kinds of recipes meta-cgl likes to merge.
Alex gave you a pointer, I would also say if you're just asking in
general, you can have a look at what software packages are used by CGL
registered distributions here:
http://www.linuxfoundation.org/collaborate/workgroups/cgl/registered-distributions
and the specs, linked from the same page, if you really want to dig into
it. The stated goal of meta-cgl is to provide a CGL layer for Yocto, so
I'd hazard a guess that anything that gets it closer to that goal in a
sensible fashion is welcome.
-J.
>
> Regards
> Bian
>
> >
> > Thank you,
> > Alex Vaduva
> >
> > -----Original Message-----
> > From: openembedded-devel-bounces@lists.openembedded.org
> > [mailto:openembedded-devel-bounces@lists.openembedded.org] On Behalf Of Bian,
> > Naimeng
> > Sent: Tuesday, November 25, 2014 2:29 AM
> > To: Joe MacDonald
> > Cc: openembedded-devel@lists.openembedded.org
> > Subject: Re: [oe] [PATCH 6/6] [meta-oe] Port corosync from openembedded and
> > upgrade to 1.4.7
> >
> > Hi Joe
> >
> > Thanks for your advices.
> > I will look at meta-cgl.
> >
> > Regards
> > Bian
> >
> > > -----Original Message-----
> > > From: Joe MacDonald [mailto:Joe_MacDonald@mentor.com]
> > > Sent: Tuesday, November 25, 2014 3:40 AM
> > > Cc: openembedded-devel@lists.openembedded.org
> > > Subject: Re: [oe] [PATCH 6/6] [meta-oe] Port corosync from
> > > openembedded and upgrade to 1.4.7
> > >
> > > Hi Bian,
> > >
> > > [[oe] [PATCH 6/6] [meta-oe] Port corosync from openembedded and
> > > upgrade to 1.4.7] On 14.11.24 (Mon 14:45) Bian Naimeng wrote:
> > >
> > > > Signed-off-by: Bian Naimeng <biannm@cn.fujitsu.com>
> > > > ---
> > > > meta-oe/recipes-support/corosync/corosync_1.4.7.bb | 31
> > > ++++++++++++++++++++++
> > > > .../corosync/files/config-site.corosync | 4 +++
> > > > meta-oe/recipes-support/corosync/files/groff.patch | 12 +++++++++
> > > > 3 files changed, 47 insertions(+)
> > > > create mode 100644
> > > > meta-oe/recipes-support/corosync/corosync_1.4.7.bb
> > > > create mode 100644
> > > > meta-oe/recipes-support/corosync/files/config-site.corosync
> > > > create mode 100644
> > > > meta-oe/recipes-support/corosync/files/groff.patch
> > >
> > > This probably belongs in meta-networking (unless there's strong
> > > opinion it belongs somewhere else). Did you have a look at what has
> > > been done in meta-cgl (git://git.enea.com/linux/meta-cgl) with the
> > > recipe to see if there was anything useful there? I know meta-cgl
> > > isn't very well publicized yet, but it is in the layer index. If you
> > > haven't looked at what's there yet, it may be worth having a look to
> > > see if any of it can save you some time.
> > >
> > > For what it's worth, there does seem to be rather a lot more in
> > > meta-cgl and I think at least some of it still applies (eg. the volatiles
> > file).
> > >
> > > > diff --git a/meta-oe/recipes-support/corosync/corosync_1.4.7.bb
> > > > b/meta-oe/recipes-support/corosync/corosync_1.4.7.bb
> > > > new file mode 100644
> > > > index 0000000..a3b8864
> > > > --- /dev/null
> > > > +++ b/meta-oe/recipes-support/corosync/corosync_1.4.7.bb
> > > > @@ -0,0 +1,31 @@
> > > > +SUMMARY = "The Corosync Cluster Engine and Application Programming
> > > Interfaces"
> > > > +DESCRIPTION = "This package contains the Corosync Cluster Engine
> > > > +Executive, several default ¥ APIs and libraries, default
> > > > +configuration
> > > files, and an init script."
> > > > +HOMEPAGE = "http://corosync.github.io/corosync/"
> > > > +
> > > > +inherit autotools pkgconfig
> > > > +
> > > > +SRC_URI =
> > > "http://build.clusterlabs.org/corosync/releases/${BPN}-${PV}.tar.gz ¥
> > > > + file://config-site.corosync ¥
> > > > + file://groff.patch ¥
> > > > + "
> > >
> > > Please remove the tab before space above.
> > >
> > > > +
> > > > +SRC_URI[md5sum] = "da9b2cf0b55f08ac4cf7cd82ac2f977a"
> > > > +SRC_URI[sha256sum] =
> > > "c1d005b2093e9a725abd1dfb2be936114b561c0b3145aca11f58c1a733a05af8"
> > > > +
> > > > +LICENSE = "BSD"
> > >
> > > I think this is actually a BSD-3-Clause license, isn't it?
> > >
> > > > +LIC_FILES_CHKSUM =
> > > "file://LICENSE;md5=25656171d1e4054c636a9893067f8c30"
> > > > +
> > > > +CONFIG_SITE = "${WORKDIR}/config-site.corosync"
> > > > +
> > > > +DEPENDS = "groff-native nss"
> > > > +
> > > > +EXTRA_OECONF += " --enable-nss "
> > > > +CFLAGS_append += " -fPIC "
> > > > +
> > > > +do_compile() {
> > > > + DESCPATH="${STAGING_DATADIR_NATIVE}/groff/`groff -v | awk
> > > '{if(NR==1)print $4}'`/font"
> > > > + oe_runmake DESCPATH=${DESCPATH} }
> > > > +
> > > > +FILES_${PN}-dbg += "${libexecdir}/lcrso/.debug"
> > > > diff --git
> > > > a/meta-oe/recipes-support/corosync/files/config-site.corosync
> > > > b/meta-oe/recipes-support/corosync/files/config-site.corosync
> > > > new file mode 100644
> > > > index 0000000..3472dec
> > > > --- /dev/null
> > > > +++ b/meta-oe/recipes-support/corosync/files/config-site.corosync
> > > > @@ -0,0 +1,4 @@
> > > > +ac_cv_func_fnmatch_works=yes
> > > > +ac_cv_func_malloc_0_nonnull=yes
> > > > +ac_cv_func_realloc_0_nonnull=yes
> > > > +
> > > > diff --git a/meta-oe/recipes-support/corosync/files/groff.patch
> > > > b/meta-oe/recipes-support/corosync/files/groff.patch
> > >
> > > Can you add a header to this patch please?
> > >
> > > Thanks.
> > >
> > > -J.
> > >
> > > > new file mode 100644
> > > > index 0000000..da1378a
> > > > --- /dev/null
> > > > +++ b/meta-oe/recipes-support/corosync/files/groff.patch
> > > > @@ -0,0 +1,12 @@
> > > > +diff -Nurp corosync-1.3.1.org/man/Makefile.am
> > > corosync-1.3.1/man/Makefile.am
> > > > +--- corosync-1.3.1.org/man/Makefile.am 2011-04-25
> > > 10:37:50.000000000 +0800
> > > > ++++ corosync-1.3.1/man/Makefile.am 2014-11-24 10:33:36.583426707
> > > +0800
> > > > +@@ -126,7 +126,7 @@ HTML_DOCS = $(dist_man_MANS:%=%.html)
> > > > +INDEX_HTML = index.html
> > > > +
> > > > + %.html: %
> > > > +- $(GROFF) -mandoc -Thtml $^ > $@
> > > > ++ $(GROFF) -F $(DESCPATH) -mandoc -Thtml $^ > $@
> > > > +
> > > > + install-data-local:
> > > > + $(INSTALL) -d $(DESTDIR)/${docdir}/html
> > > > --
> > > > 1.9.1
> > > >
> > > --
> > > -Joe MacDonald.
> > > :wq
> > --
> > _______________________________________________
> > Openembedded-devel mailing list
> > Openembedded-devel@lists.openembedded.org
> > http://lists.openembedded.org/mailman/listinfo/openembedded-devel
> > --
> > _______________________________________________
> > Openembedded-devel mailing list
> > Openembedded-devel@lists.openembedded.org
> > http://lists.openembedded.org/mailman/listinfo/openembedded-devel
--
-Joe MacDonald.
:wq
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 501 bytes --]
next prev parent reply other threads:[~2014-11-26 14:08 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-24 6:45 [PATCH 0/6] Port some recipes from openembedded and upgrade Bian Naimeng
2014-11-24 6:45 ` [PATCH 1/6] [meta-oe] evieext: Port evieext from openembedded Bian Naimeng
2014-11-24 10:32 ` Burton, Ross
2014-11-24 6:45 ` [PATCH 2/6] [meta-oe] fontcacheproto: Port fontcacheproto " Bian Naimeng
2014-11-24 10:39 ` Burton, Ross
2014-11-24 6:45 ` [PATCH 3/6] [meta-oe] libxfontcache: Port libxfontcache " Bian Naimeng
2014-11-24 6:45 ` [PATCH 4/6] [meta-oe] printproto: Port printproto " Bian Naimeng
2014-11-24 10:30 ` Burton, Ross
2014-11-25 1:14 ` Bian, Naimeng
2014-11-25 9:13 ` Burton, Ross
2014-11-25 9:31 ` Bian, Naimeng
2014-11-24 6:45 ` [PATCH 5/6] [meta-oe] libxp: Port libxp " Bian Naimeng
2014-11-24 6:45 ` [PATCH 6/6] [meta-oe] Port corosync from openembedded and upgrade to 1.4.7 Bian Naimeng
2014-11-24 19:39 ` Joe MacDonald
2014-11-25 0:29 ` Bian, Naimeng
2014-11-25 10:18 ` Alexandru Vaduva
2014-11-26 3:36 ` Bian, Naimeng
2014-11-26 9:17 ` Alexandru Vaduva
2014-11-26 14:08 ` Joe MacDonald [this message]
2014-11-26 14:23 ` Alexandru Vaduva
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=20141126140819.GA3886@mentor.com \
--to=joe_macdonald@mentor.com \
--cc=biannm@cn.fujitsu.com \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.