From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail1.windriver.com ([147.11.146.13]) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1TbotO-00053D-SP for openembedded-core@lists.openembedded.org; Fri, 23 Nov 2012 09:45:47 +0100 Received: from ALA-HCA.corp.ad.wrs.com (ala-hca.corp.ad.wrs.com [147.11.189.40]) by mail1.windriver.com (8.14.5/8.14.3) with ESMTP id qAN8Vd5b024205 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 23 Nov 2012 00:31:39 -0800 (PST) Received: from [128.224.163.154] (128.224.163.154) by ALA-HCA.corp.ad.wrs.com (147.11.189.50) with Microsoft SMTP Server (TLS) id 14.2.318.4; Fri, 23 Nov 2012 00:31:36 -0800 Message-ID: <50AF3476.1000401@windriver.com> Date: Fri, 23 Nov 2012 16:31:50 +0800 From: ChenQi User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:11.0) Gecko/20120410 Thunderbird/11.0.1 MIME-Version: 1.0 To: Saul Wold References: <50AD6F8E.50807@linux.intel.com> In-Reply-To: <50AD6F8E.50807@linux.intel.com> X-Originating-IP: [128.224.163.154] Cc: Zhenfeng.Zhao@windriver.com, openembedded-core@lists.openembedded.org Subject: Re: [PATCH 1/1] autotools.bbclass: use relative path for acpaths whenever possible X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 23 Nov 2012 08:45:47 -0000 Content-Type: multipart/alternative; boundary="------------030405090600040306010904" --------------030405090600040306010904 Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit On 11/22/2012 08:19 AM, Saul Wold wrote: > On 11/08/2012 09:47 PM, Qi.Chen@windriver.com wrote: >> From: Chen Qi >> >> When the TMPDIR is very long, say, 410 characters, aclocal would >> fail because the argument list is too long. This patch is an effort >> to use relative path for acpaths whenever possible, aiming at >> making the build system work correctly when the sanity check says OK. >> >> With the current implementation of autoreconf, it's impossible to >> thoroughly replace absolute path with relative path. Therefore, we >> use relative path when there's on subdirectory to configure; otherwise, >> we use absolute path. >> > Have you done a world build long path names with this patch on 1.4 and > does it cause any failures with autoreconf and subdirs? > > Sau! > Hi Saul, Thank you for your suggestion. I did do the test before I sent out the patch, but at that time I only used 'bitbake core-image-sato'. However when I re-tested it yesterday with 'bitbake world', qt4-native configure failed. (Sorry for my carelessness.) The failure is not caused by the changes in autotools.bbclass. It failed to configure because *the path is too long to be fitted into the static char array defined in its qconfig.cpp file*. Below are some details: src/corelib/global/qconfig.cpp:11:67: *error: initializer-string for array of chars is too long* [-fpermissive] The statements in qconfig.cpp that caused this failure are: static const char qt_configure_prefix_path_str [*256* + 12] = "qt_prfxpath=/buildarea/qchen1/poky/build-autotools-test/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs\ /subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/s\ ubs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs0/sysroots/x86_64-linux/usr"; static const char qt_configure_documentation_path_str[*256* + 12] = "qt_docspath=/buildarea/qchen1/poky/build-autotools-test/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs\ /subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/s\ ubs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs0/sysroots/x86_64-linux/usr/doc"; ... Anyway it should be fixed as part of bug#2766 (long TMPDIR problem). Thus, I'll rework on bug#2766 and try to fix it. However, statements in qconfig.cpp lead me to doubt about the length limit of TMPDIR (currently 410). As we see in qconfig.cpp, */if the TMPDIR has chars more than 256, it would fail for certain./* *I think we should re-consider the length limit for TMPDIR.* Thanks, Chen Qi > >> [YOCTO #2766] >> >> Signed-off-by: Chen Qi >> --- >> meta/classes/autotools.bbclass | 14 +++++++++----- >> 1 file changed, 9 insertions(+), 5 deletions(-) >> >> diff --git a/meta/classes/autotools.bbclass >> b/meta/classes/autotools.bbclass >> index b97d74b..86d8939 100644 >> --- a/meta/classes/autotools.bbclass >> +++ b/meta/classes/autotools.bbclass >> @@ -123,8 +123,11 @@ autotools_do_configure() { >> rm -f `dirname $ac`/configure >> done ) >> if [ -e ${S}/configure.in -o -e ${S}/configure.ac ]; then >> + [ -e configure.in ] && CONFIGURE_AC=configure.in || >> CONFIGURE_AC=configure.ac >> olddir=`pwd` >> cd ${S} >> + # Determine whether there's subdirs to configure >> + grep -q -m 1 AC_CONFIG_SUBDIRS $CONFIGURE_AC && sub_cfg=1 || >> sub_cfg=0 >> # Remove any previous copy of the m4 macros >> rm -rf ${B}/aclocal-copy/ >> ACLOCAL="aclocal --system-acdir=${B}/aclocal-copy/" >> @@ -132,6 +135,11 @@ autotools_do_configure() { >> acpaths= >> for i in `find ${S} -maxdepth 2 -name \*.m4|grep -v >> 'aclocal.m4'| \ >> grep -v 'acinclude.m4' | grep -v 'aclocal-copy' | >> sed -e 's,\(.*/\).*$,\1,'|sort -u`; do >> + # If no subdirs to configure, we use relative path >> + # This is used for supporting long TMPDIR in Yocto >> + if [ $sub_cfg == 0 ]; then >> + i=`echo $i | sed -e 's#${S}#\.#'` >> + fi >> acpaths="$acpaths -I $i" >> done >> else >> @@ -161,11 +169,7 @@ autotools_do_configure() { >> if ! echo ${EXTRA_AUTORECONF} | grep -q "aclocal"; then >> rm -f aclocal.m4 >> fi >> - if [ -e configure.in ]; then >> - CONFIGURE_AC=configure.in >> - else >> - CONFIGURE_AC=configure.ac >> - fi >> + >> if grep "^[[:space:]]*AM_GLIB_GNU_GETTEXT" $CONFIGURE_AC >> >/dev/null; then >> if grep "sed.*POTFILES" $CONFIGURE_AC >/dev/null; then >> : do nothing -- we still have an old unmodified >> configure.ac >> > --------------030405090600040306010904 Content-Type: text/html; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit On 11/22/2012 08:19 AM, Saul Wold wrote:
On 11/08/2012 09:47 PM, Qi.Chen@windriver.com wrote:
From: Chen Qi <Qi.Chen@windriver.com>

When the TMPDIR is very long, say, 410 characters, aclocal would
fail because the argument list is too long. This patch is an effort
to use relative path for acpaths whenever possible, aiming at
making the build system work correctly when the sanity check says OK.

With the current implementation of autoreconf, it's impossible to
thoroughly replace absolute path with relative path. Therefore, we
use relative path when there's on subdirectory to configure; otherwise,
we use absolute path.

Have you done a world build long path names with this patch on 1.4 and does it cause any failures with autoreconf and subdirs?

Sau!

Hi Saul,

Thank you for your suggestion.

I did do the test before I sent out the patch, but at that time I only used 'bitbake core-image-sato'.
However when I re-tested it yesterday with 'bitbake world', qt4-native configure failed.
(Sorry for my carelessness.)

The failure is not caused by the changes in autotools.bbclass.
It failed to configure because the path is too long to be fitted into the static char array defined in its qconfig.cpp file. 

Below are some details:
src/corelib/global/qconfig.cpp:11:67: error: initializer-string for array of chars is too long [-fpermissive]
The statements in qconfig.cpp that caused this failure are:
static const char qt_configure_prefix_path_str       [256 + 12] = "qt_prfxpath=/buildarea/qchen1/poky/build-autotools-test/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs\
/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/s\
ubs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs0/sysroots/x86_64-linux/usr";
static const char qt_configure_documentation_path_str[256 + 12] = "qt_docspath=/buildarea/qchen1/poky/build-autotools-test/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs\
/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/s\
ubs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs/subs0/sysroots/x86_64-linux/usr/doc";
...


Anyway it should be fixed as part of bug#2766 (long TMPDIR problem).
Thus, I'll rework on bug#2766 and try to fix it.

However, statements in qconfig.cpp lead me to doubt about the length limit of TMPDIR (currently 410).
As we see in qconfig.cpp, if the TMPDIR has chars more than 256, it would fail for certain.

I think we should re-consider the length limit for TMPDIR.


Thanks,
Chen Qi




[YOCTO #2766]

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
---
  meta/classes/autotools.bbclass |   14 +++++++++-----
  1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/meta/classes/autotools.bbclass b/meta/classes/autotools.bbclass
index b97d74b..86d8939 100644
--- a/meta/classes/autotools.bbclass
+++ b/meta/classes/autotools.bbclass
@@ -123,8 +123,11 @@ autotools_do_configure() {
          rm -f `dirname $ac`/configure
          done )
      if [ -e ${S}/configure.in -o -e ${S}/configure.ac ]; then
+               [ -e configure.in ] && CONFIGURE_AC=configure.in || CONFIGURE_AC=configure.ac
          olddir=`pwd`
          cd ${S}
+        # Determine whether there's subdirs to configure
+        grep -q -m 1 AC_CONFIG_SUBDIRS $CONFIGURE_AC && sub_cfg=1 || sub_cfg=0
          # Remove any previous copy of the m4 macros
          rm -rf ${B}/aclocal-copy/
          ACLOCAL="aclocal --system-acdir=${B}/aclocal-copy/"
@@ -132,6 +135,11 @@ autotools_do_configure() {
              acpaths=
              for i in `find ${S} -maxdepth 2 -name \*.m4|grep -v 'aclocal.m4'| \
                  grep -v 'acinclude.m4' | grep -v 'aclocal-copy' | sed -e 's,\(.*/\).*$,\1,'|sort -u`; do
+                    # If no subdirs to configure, we use relative path
+                    # This is used for supporting long TMPDIR in Yocto
+                    if [ $sub_cfg == 0 ]; then
+                        i=`echo $i | sed -e 's#${S}#\.#'`
+                fi
                  acpaths="$acpaths -I $i"
              done
          else
@@ -161,11 +169,7 @@ autotools_do_configure() {
          if ! echo ${EXTRA_AUTORECONF} | grep -q "aclocal"; then
              rm -f aclocal.m4
          fi
-        if [ -e configure.in ]; then
-            CONFIGURE_AC=configure.in
-        else
-            CONFIGURE_AC=configure.ac
-        fi
+
          if grep "^[[:space:]]*AM_GLIB_GNU_GETTEXT" $CONFIGURE_AC >/dev/null; then
              if grep "sed.*POTFILES" $CONFIGURE_AC >/dev/null; then
                  : do nothing -- we still have an old unmodified configure.ac




--------------030405090600040306010904--