From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail.windriver.com ([147.11.1.11]) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1RuKny-0003Yd-Dd for openembedded-core@lists.openembedded.org; Mon, 06 Feb 2012 10:24:10 +0100 Received: from ALA-HCA.corp.ad.wrs.com (ala-hca [147.11.189.40]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id q169G7t1029567 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for ; Mon, 6 Feb 2012 01:16:08 -0800 (PST) Received: from [128.224.162.223] (128.224.162.223) by ALA-HCA.corp.ad.wrs.com (147.11.189.50) with Microsoft SMTP Server (TLS) id 14.1.255.0; Mon, 6 Feb 2012 01:16:07 -0800 Message-ID: <4F2F9A58.3030300@windriver.com> Date: Mon, 6 Feb 2012 17:16:08 +0800 From: Xiaofeng Yan User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.23) Gecko/20110921 Thunderbird/3.1.15 MIME-Version: 1.0 To: References: <20120206070513.GC3872@jama.jama.net> <4F2F8344.5000509@windriver.com> In-Reply-To: <4F2F8344.5000509@windriver.com> X-Originating-IP: [128.224.162.223] Subject: Re: [PATCH 0/2] glib-2.0: Don't check function "qsort_r" in configure.ac X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list Reply-To: Patches and discussions about the oe-core layer List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 06 Feb 2012 09:24:10 -0000 Content-Type: multipart/alternative; boundary="------------000105040209030609000502" --------------000105040209030609000502 Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: quoted-printable X-MIME-Autoconverted: from 8bit to quoted-printable by mail.windriver.com id q169G7t1029567 On 2012=E5=B9=B402=E6=9C=8806=E6=97=A5 15:37, Xiaofeng Yan wrote: > On 2012=E5=B9=B402=E6=9C=8806=E6=97=A5 15:05, Martin Jansa wrote: >> On Mon, Feb 06, 2012 at 02:41:09PM +0800, Xiaofeng Yan wrote: >>> From: Xiaofeng Yan >>> >>> The function "g_qsort_with_data" has two kinds of realization. One=20 >>> calls qsort_r >>> from libglib, the other realize itself. the realization method is=20 >>> controlled by macro "HAVE_QSORT_R". >>> qsort_r from libglib cause sort error. Add no-check-qsort_r.patch=20 >>> for fixing this problem. >>> This patch is to fix bug 1959. After this patch is merged into=20 >>> OE-core I will open a new bug to >>> track the reason why qsort_r report normally. For finishing lsb test=20 >>> in YOCTO 1.2 M3 this patch >>> is applied temporarily. >>> >>> Pull URL: git://git.pokylinux.org/poky-contrib.git >>> Branch: xiaofeng/1959 >>> Browse:=20 >>> http://git.pokylinux.org/cgit.cgi/poky-contrib/log/?h=3Dxiaofeng/1959 >>> >>> Thanks, >>> Xiaofeng Yan >>> --- >> no-check-qsort_r.patch is the same as remove.test.for.qsort_r.patch fr= om >> http://git.openembedded.org/openembedded-core/commit/?id=3D74f43e3b269= c01dba7cd5215a8e825229bd5ecff=20 >> >> >> which was later removed from glib-2.0-native in >> http://git.openembedded.org/openembedded-core/commit/?id=3D37579d7d74d= 127c90c1e078d05c5bf4ba0b3f755=20 >> >> >> and later removed completely in >> http://git.openembedded.org/openembedded-core/commit/?id=3D812342e44e9= e361a0fa93cfcfe65cf4dd6f2fb57=20 >> >> >> Are you sure this patch is right fix? And no need to split adding patc= h >> itself and adding it to SRC_URI to 2 patches. >> >> Cheers, >> > Hi Martin Jansa, > I reviewed the above link, no-check-qsort_r.patch is the same as=20 > remove.test.for.qsort_r.patch really. Please check attachment for the=20 > test program from LSB 4.1. I think we need this patch because it is an=20 > error really. LSB report the next information: > > After sorting array, the 0-th element of array should be 1, but is 2 > After sorting array, the 1-th element of array should be 2, but is 1 > After sorting array, the 2-th element of array should be 3, but is 5 > After sorting array, the 4-th element of array should be 5, but is 3 > > > I can reuse this patch which was removed before if you think this is a=20 > bug. Do you any suggestion? > > Compile this test program. > $ gcc -g `pkg-config --cflags glib-2.0` gobject_value_arrays.c -o=20 > gobject -lgobject-2.0 > > I reviewed the origin patch from Martin Jansa and found that my patch is=20 different from him. My patch is to remove the check of qsort_r and the=20 origin had still this check. I also update my patch because it have two file:// in SRC_URI. Thank Martin Jansa. > > _______________________________________________ > Openembedded-core mailing list > Openembedded-core@lists.openembedded.org > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core --------------000105040209030609000502 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-MIME-Autoconverted: from 8bit to quoted-printable by mail.windriver.com id q169G7t1029567 On 2012=E5=B9=B402=E6=9C=8806=E6=97=A5 15:37, Xiaofeng Yan wrote:
On 2012=E5=B9=B402=E6=9C=8806=E6=97=A5 15:05, Martin Jansa wrote:
On Mon, Feb 06, 2012 at 02:41:09PM +0800, Xiaofeng Yan wrote:
From: Xiaofeng Yan<xiaofeng.yan@windriver.com>

The function "g_qsort_with_data" has two kinds of realization. One calls qsort_r
from libglib, the other realize itself. the realization method is controlled by macro "HAVE_QSORT_R".
qsort_r from libglib cause sort error. Add no-check-qsort_r.patch for fixing this problem.
This patch is to fix bug 1959. After this patch is merged into OE-core I will open a new bug to
track the reason why qsort_r report normally. For finishing lsb test in YOCTO 1.2 M3 this patch
is applied temporarily.

Pull URL: git://git.pokylinux.org/poky-contrib.git
=C2=A0=C2=A0 Branch: xiaofeng/1959
=C2=A0=C2=A0 Browse: http://git.pokylinux.o= rg/cgit.cgi/poky-contrib/log/?h=3Dxiaofeng/1959

Thanks,
=C2=A0=C2=A0=C2=A0=C2=A0 Xiaofeng Yan<xiaofeng.yan@wind= river.com>
---
no-check-qsort_r.patch is the same as remove.test.for.qsort_r.patch from
ht= tp://git.openembedded.org/openembedded-core/commit/?id=3D74f43e3b269c01db= a7cd5215a8e825229bd5ecff

which was later removed from glib-2.0-native in
ht= tp://git.openembedded.org/openembedded-core/commit/?id=3D37579d7d74d127c9= 0c1e078d05c5bf4ba0b3f755

and later removed completely in
ht= tp://git.openembedded.org/openembedded-core/commit/?id=3D812342e44e9e361a= 0fa93cfcfe65cf4dd6f2fb57

Are you sure this patch is right fix? And no need to split adding patch
itself and adding it to SRC_URI to 2 patches.

Cheers,

Hi Martin Jansa,
I reviewed the above link, no-check-qsort_r.patch is=C2=A0 the same= as remove.test.for.qsort_r.patch really. Please check attachment for the test program from LSB 4.1. I think we need this patch because it is an error really. LSB report the next information:

After sorting array, the 0-th element of array should be 1, but is 2
After sorting array, the 1-th element of array should be 2, but is 1
After sorting array, the 2-th element of array should be 3, but is 5
After sorting array, the 4-th element of array should be 5, but is 3


I can reuse this patch which was removed before if you think this is a bug. Do you any suggestion?

Compile this test program.
$ gcc -g `pkg-config --cflags glib-2.0` gobject_value_arrays.c -o gobject -lgobject-2.0


I reviewed the origin patch from Martin Jansa and found that my patch is different from him.=C2=A0 My patch is to remove the check of qsort_r and the origin had still this check. I also update my patch because
it have two file:= // in SRC_URI.
Thank Martin Jansa.


_______________________________________________
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.linuxtogo.org/cgi-b=
in/mailman/listinfo/openembedded-core

--------------000105040209030609000502--