Nabble - T2

Re: gnome install path

2009-11-11T05:11:50Z

Dear Gerardo,

Gerardo Di Iorio wrote:
> Hi,
> is possible define the gnome install path in the "/usr" and the
> sysconfdir in "/etc"?
> Can send an example ?
>
> regards
> Gerardo Di Iorio
>
Yes, a target can control this fine grained, for every package,
theoretically every package's file.

For the case of Gnome (and other major "optional" software pieces, that
we, by default, install into /opt, such as KDE, Xfce, E17, Mozilla/*,
OpenOffice.org, etc.) you can simply change their "group" prefix in the
Config, "Install prefix of ...", some quick grep over the config.in files:

package/e17/imlib2/config.in: SDECFG_PKG_E17_CORE_PREFIX
"opt/e17"
package/emulators/vmware/config.in:
SDECFG_PKG_VMWARE_PREFIX 'opt/vmware'
package/gnome14/glib12/config.in:
SDECFG_PKG_GNOME14_CORE_PREFIX "opt/gnome"
package/gnome2/glib/config.in:
SDECFG_PKG_GNOME2_CORE_PREFIX "opt/gnome2"
package/gnustep/gnustep-make/config.in:
SDECFG_PKG_GNUSTEP_CORE_PREFIX ""
package/kde/kdelibs/config.in: SDECFG_PKG_KDE_CORE_PREFIX
"opt/kde4"
package/kde/qt/config.in: text 'Install prefix for Qt'
SDECFG_PKG_QT_PREFIX "opt/qt3"
package/kde/qt4/config.in: text 'Install prefix for Qt4'
SDECFG_PKG_QT4_PREFIX "opt/qt4"
package/lxde/lxde-common/config.in:
SDECFG_PKG_LXDE_CORE_PREFIX "opt/lxde"
package/network/apache/config.in: SDECFG_PKG_APACHE_PREFIX
"opt/apache"
package/network/bacula/config.in: SDECFG_PKG_BACULA_PREFIX
"opt/bacula"
package/network/honeyd/config.in: SDECFG_PKG_HONEYD_PREFIX
"opt/honeyd"
package/network/ldapdns/config.in: SDECFG_PKG_LDAPDNS_PREFIX
"opt/ldapdns"
package/network/nagios/config.in: SDECFG_PKG_NAGIOS_PREFIX
"opt/nagios"
package/network/pound/config.in: SDECFG_PKG_POUND_PREFIX
"opt/pound"
package/network/unreal-ircd/config.in:
SDECFG_PKG_UNREAL_PREFIX "/opt/unrealircd"
package/network/vnuml/config.in: SDECFG_PKG_VNUML_PREFIX
"opt/vnuml"
package/perl/perl/config.in: text 'Site prefix'
SDECFG_PKG_PERL5_SITE_PREFIX "usr"
package/perl/perl/config.in: text 'Vendor prefix'
SDECFG_PKG_PERL5_VENDOR_PREFIX "opt"
package/security/freeradius/config.in:
SDECFG_PKG_FREERADIUS_PREFIX "opt/freeradius"
package/security/libprelude/config.in:
SDECFG_PKG_PRELUDE_PREFIX "opt/prelude"
package/xfce4/libxfce4util/config.in:
SDECFG_PKG_XFCE4_CORE_PREFIX "opt/xfce4"
package/zope/zope/config.in: SDECFG_PKG_ZOPE_PREFIX "opt/zope"
package/zope/zope/config.in: SDECFG_PKG_ZOPE_PRODDIR
"$SDECFG_PKG_ZOPE_PREFIX/products"

Greetings,

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Qt4 compile fills partition

2009-11-11T05:07:33Z

Hi,

Barry Kauler wrote:
> My apologies for wasting your time.
> I found that I had target/ ..... /pkg_qt4.conf
> that was causing the problem.
>

No problem, you're welcome! Good to know it wasn't directly our fault :-)

René

> Regards,
> Barry Kauler
>
>
> On 11/6/09, René Rebe <rene@...> wrote:
>
>> Hi,
>>
>> Barry Kauler wrote:
>>
>>> I got T2 out of svn about 5 - 6 days ago.
>>>
>>> When it got to Qt4, I had 7.1GB free space in the partition. After a
>>> very long time, a few hours, the free space had dropped to under 1GB,
>>> so I killed the compile.
>>>
>>> build/ ... /var/adm/logs/5-qt4.out is the culprit, it just keeps growing .
>>>
>>> I found that Qt asks a question, but it seems that T2 gives the wrong
>>> answer, so Qt asks the same question again, and again...
>>>
>>> "Which edition of Qt do you want to use ?
>>>
>>> Type 'c' if you want to use the Commercial Edition.
>>> Type 'o' if you want to use the Open Source Edition."
>>>
>>> Regards,
>>> Barry Kauler
>>>
>>>
>> Yeah, that annoyed me once, years(?) ago, when they changed their
>> questioning, too. However, it currently builds for me:
>>
>> -rw-r--r-- 1 root root 9358063 Oct 2 10:58
>> build/reference-8.0-trunk-reference-x86-pentiumpro-linux/var/adm/logs/5-qt4.log
>> -rw-r--r-- 1 root root 8677388 Nov 4 01:22
>> build/reference64-8.0-trunk-reference-x86-64-linux/var/adm/logs/5-qt4.log
>>
>> The code in package/kde/qt4/qt4.conf:
>>
>> ...
>> # accept the GPL
>> echo "o
>> yes" | ./configure $confopt
>> ...
>>
>> Maybe it asks a second time on your system, after re-generating
>> something, or so?
>>
>> René
>>
>> --
>> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
>> http://exactcode.de | http://t2-project.org | http://rene.rebe.name
>>

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Qt4 compile fills partition

2009-11-10T03:05:17Z

My apologies for wasting your time.
I found that I had target/ ..... /pkg_qt4.conf
that was causing the problem.

Regards,
Barry Kauler

On 11/6/09, René Rebe <rene@...> wrote:

> Hi,
>
> Barry Kauler wrote:
>> I got T2 out of svn about 5 - 6 days ago.
>>
>> When it got to Qt4, I had 7.1GB free space in the partition. After a
>> very long time, a few hours, the free space had dropped to under 1GB,
>> so I killed the compile.
>>
>> build/ ... /var/adm/logs/5-qt4.out is the culprit, it just keeps growing .
>>
>> I found that Qt asks a question, but it seems that T2 gives the wrong
>> answer, so Qt asks the same question again, and again...
>>
>> "Which edition of Qt do you want to use ?
>>
>> Type 'c' if you want to use the Commercial Edition.
>> Type 'o' if you want to use the Open Source Edition."
>>
>> Regards,
>> Barry Kauler
>>
>
> Yeah, that annoyed me once, years(?) ago, when they changed their
> questioning, too. However, it currently builds for me:
>
> -rw-r--r-- 1 root root 9358063 Oct 2 10:58
> build/reference-8.0-trunk-reference-x86-pentiumpro-linux/var/adm/logs/5-qt4.log
> -rw-r--r-- 1 root root 8677388 Nov 4 01:22
> build/reference64-8.0-trunk-reference-x86-64-linux/var/adm/logs/5-qt4.log
>
> The code in package/kde/qt4/qt4.conf:
>
> ...
> # accept the GPL
> echo "o
> yes" | ./configure $confopt
> ...
>
> Maybe it asks a second time on your system, after re-generating
> something, or so?
>
> René
>
> --
> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
> http://exactcode.de | http://t2-project.org | http://rene.rebe.name
>
>

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

gnome install path

2009-11-09T23:01:29Z

Hi,
is possible define the gnome install path in the "/usr" and the
sysconfdir in "/etc"?
Can send an example ?

regards
Gerardo Di Iorio
--
http://www.gerardodiiorio.com

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: epdfview patch

2009-11-09T01:00:59Z

Hi,

On 08.11.2009, at 11:33, Gerardo Di Iorio wrote:

> Hi,
> this is the patch for update epdfview at last version

Committed revision 34844.

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
http://exactcode.com | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

epdfview patch

2009-11-08T02:33:15Z

Hi,
this is the patch for update epdfview at last version
--
http://www.gerardodiiorio.com

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

epdfview.patch (1K) Download Attachment

Re: new package wicd

2009-11-07T07:27:39Z

Hello again,

Mika Pesu wrote:
> Hi all,
>
> Here's the files for new package wicd (http://wicd.sourceforge.net).

I added the .cache from my test build, as well as your name to the
Maintainer tag (:-) and converted from custom (deprecated, we really
should remove it :-) to hooks:

-wicd_cm() {
- python ./setup.py configure
- python ./setup.py install --root=$root
-}
- custmain="wicd_cm"

+makeopt=
+makeinstopt=
+hook_add premake 5 "python ./setup.py configure"
+hook_add inmake 5 "python ./setup.py install --root=$root"

I was asked about those "spurious":

"The use of custmain is deprecated, please use hooks instead."

build warnings the other week in IRC, and well, my answer was: we still
got some packages left using it, including the linux kernel packages. I
just grep'ed how big the backlog is, and ouhm well, what shall I say,
it's still there:

# grep custmain= package/*/*/*.conf | wc -l
101

Patches for migrations welcome, in many cases it should be very trivial
janitorial work, ... :-)

Have a nice weekend,

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Mercurial support for Download and E17 update

2009-11-07T07:14:16Z

Hi Mika,

Mika Pesu wrote:
> Hi all,
>
> Here's some patches that where lingering on my package tree.
>
> download_hg.patch => Adds support for mercurial version system, for
> now use clone method, rather than archive (=svn export). On my test
> package, it didnt find revisions on archive method.
Committed revision 34825.

> e17_update.patch => Few enlightenment package updates. Some packages
> doesnt build, but i think they didnt build on current svn:head
> versions either. Some of the packages has moved to e17 svn old tree,
> meaning they are not maintained anymore. At somepoint these could be
> removed, in favor of something else, there where some new replacing
> packages in proto stage but i didnt package them at this point. This
> is still work in progress because of the broken packages. I will check
> them once more when i get fresh build started. There should be proper
> release from e17 in december.
Committed revision 34826.

Thanks,
René

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: new package wicd

2009-11-07T06:33:35Z

Hi Mika,

Mika Pesu wrote:
> Hi all,
>
> Here's the files for new package wicd (http://wicd.sourceforge.net).

It's so nice to see you back and your patches. I'll apply both mails
straight away, however, your year old SVN account should still be in
effect, so you could just commit bits :-) In case you need new auth data
let me know :-)

I'll re-reply to both mail when I committed them.

René

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

new package wicd

2009-11-07T06:13:33Z

Hi all,

Here's the files for new package wicd (http://wicd.sourceforge.net).

-Mika

# --- T2-COPYRIGHT-NOTE-BEGIN ---
# This copyright note is auto-generated by ./scripts/Create-CopyPatch.
#
# T2 SDE: package/.../wicd/wicd.conf
# Copyright (C) 2009 The T2 SDE Project
#
# More information can be found in the files COPYING and README.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; version 2 of the License. A copy of the
# GNU General Public License can be found in the file COPYING.
# --- T2-COPYRIGHT-NOTE-END ---

wicd_cm() {
python ./setup.py configure
python ./setup.py install --root=$root
}

custmain="wicd_cm"

[COPY] --- T2-COPYRIGHT-NOTE-BEGIN ---
[COPY] This copyright note is auto-generated by ./scripts/Create-CopyPatch.
[COPY]
[COPY] T2 SDE: package/.../wicd/wicd.desc
[COPY] Copyright (C) 2009 The T2 SDE Project
[COPY]
[COPY] More information can be found in the files COPYING and README.
[COPY]
[COPY] This program is free software; you can redistribute it and/or modify
[COPY] it under the terms of the GNU General Public License as published by
[COPY] the Free Software Foundation; version 2 of the License. A copy of the
[COPY] GNU General Public License can be found in the file COPYING.
[COPY] --- T2-COPYRIGHT-NOTE-END ---

[I] Wired and wireless network manager

[T] Open source wired and wireless network manager
[T] with simple interface with wide variety of settings

[U] http://wicd.sourceforge.net/
[A] wicd team <http://wicd.sourceforge.net>
[M] The T2 Project <t2@...>

[C] extra/network

[L] GPL
[S] Stable
[V] 1.6.2.2
[P] X -----5---9 184.600

[D] 3571585324 wicd-1.6.2.2.tar.bz2 http://dl.sourceforge.net/sourceforge/wicd/

#!/bin/bash

main_begin

block_begin(start, `Starting wicd')
mkdir /var/run/wicd 2>/dev/null
check(`D_prefix/usr/sbin/wicd')
sleep 1
chmod 777 /var/run/wicd{,/*socket}
block_end

block_begin(stop, `Stopping wicd.')
check(`killall -15 D_prefix/usr/sbin/wicd')
block_end

main_end

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Mercurial support for Download and E17 update

2009-11-07T03:44:30Z

Hi all,

Here's some patches that where lingering on my package tree.

download_hg.patch => Adds support for mercurial version system, for now
use clone method, rather than archive (=svn export). On my test package,
it didnt find revisions on archive method.

e17_update.patch => Few enlightenment package updates. Some packages
doesnt build, but i think they didnt build on current svn:head versions
either. Some of the packages has moved to e17 svn old tree, meaning they
are not maintained anymore. At somepoint these could be removed, in
favor of something else, there where some new replacing packages in
proto stage but i didnt package them at this point. This is still work
in progress because of the broken packages. I will check them once more
when i get fresh build started. There should be proper release from e17
in december.

-Mika

Index: scripts/Download
===================================================================
--- scripts/Download (revision 34823)
+++ scripts/Download (working copy)
@@ -689,6 +689,14 @@
options="${options#* }"
[ -n $options ] && cmdpp="(cd $module; git checkout $options)"
;;
+ hg|hg\+*) # allow any hg+ other transport and strip the hg+ part off
+ url="${protocol#hg+}://$url"
+
+ module="${url##*/}"
+ cmdline="hg clone $url $module"
+ options="${options#* }"
+ [ -n $options ] && cmdpp="(cd $module; hg clone $options)"
+ ;;
*)
echo "$cmdclient unrecognized!"
return 1
@@ -735,7 +743,7 @@
tarname="`basename $bzfile`"
echo "Preparing files for final tarball ..."
[ -n "$cmdpp" ] && eval "$cmdpp"
- find -type d $ -name CVS -o -name .svn -o -name .git $ | xargs rm -rf
+ find -type d $ -name CVS -o -name .svn -o -name .git -o -name .hg$ | xargs rm -rf

if [ `find -type f | wc -l` -gt 4 ]; then
find `basename $module` | xargs touch -t 200001010000

Index: package/e17/enna/enna.desc
===================================================================
--- package/e17/enna/enna.desc (revision 34823)
+++ package/e17/enna/enna.desc (working copy)
@@ -27,9 +27,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r1983
[P] X -----5---9 170.200

[O] . $base/package/*/*/e17-conf.in

-[D] X enna-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/MISC/enna -r41533
+[D] X enna-r1983.tar.bz2 hg+http://hg.geexbox.org/enna -r 1983
Index: package/e17/estickies/estickies.desc
===================================================================
--- package/e17/estickies/estickies.desc (revision 34823)
+++ package/e17/estickies/estickies.desc (working copy)
@@ -27,9 +27,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] X -----5---9 490.300

[O] . $base/package/*/*/e17-conf.in

-[D] X estickies-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/estickies -r41533
+[D] X estickies-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/estickies -r43316
Index: package/e17/etk/etk.desc
===================================================================
--- package/e17/etk/etk.desc (revision 34823)
+++ package/e17/etk/etk.desc (working copy)
@@ -27,9 +27,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r43316
[P] X -----5---9 138.900

[O] . $base/package/*/*/e17-conf.in

-[D] X etk-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/etk -r41533
+[D] X etk-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/OLD/etk -r43316
Index: package/e17/ethumb/ethumb.desc
===================================================================
--- package/e17/ethumb/ethumb.desc (revision 34823)
+++ package/e17/ethumb/ethumb.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] X -----5---9 149.300

[O] . $base/package/*/*/e17-conf.in

-[D] X ethumb-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/PROTO/ethumb -r41533
+[D] X ethumb-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/ethumb -r43316
Index: package/e17/esmart/esmart.desc
===================================================================
--- package/e17/esmart/esmart.desc (revision 34823)
+++ package/e17/esmart/esmart.desc (working copy)
@@ -27,9 +27,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r43316
[P] X -----5---9 138.100

[O] . $base/package/*/*/e17-conf.in

-[D] X esmart-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/esmart -r41533
+[D] X esmart-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/esmart -r43316
Index: package/e17/embryo/embryo.desc
===================================================================
--- package/e17/embryo/embryo.desc (revision 34823)
+++ package/e17/embryo/embryo.desc (working copy)
@@ -33,9 +33,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] 2009-07-29
[P] X -----5---9 137.900

[O] . $base/package/*/*/e17-conf.in

-[D] X embryo-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/embryo -r41533
+[D] 224058634 embryo-0.9.9.062.tar.bz2 http://download.enlightenment.org/snapshots/2009-07-29/
Index: package/e17/elementary/elementary.desc
===================================================================
--- package/e17/elementary/elementary.desc (revision 34823)
+++ package/e17/elementary/elementary.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] 2009-07-29
[P] X -----5---9 170.200

[O] . $base/package/*/*/e17-conf.in

-[D] X elementary-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/TMP/st/elementary -r41533
+[D] 1455585772 elementary-0.5.1.0.tar.bz2 http://download.enlightenment.org/snapshots/2009-07-29/
Index: package/e17/exalt/exalt.desc
===================================================================
--- package/e17/exalt/exalt.desc (revision 34823)
+++ package/e17/exalt/exalt.desc (working copy)
@@ -26,7 +26,7 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] X -----5---9 149.300

[E] webkit
@@ -34,4 +34,4 @@
[O] . $base/package/*/*/e17-conf.in
[O] var_append confopt ' ' '--disable-shave'

-[D] X exalt-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/PROTO/exalt -r41533
+[D] X exalt-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/exalt -r43316
Index: package/e17/emprint/emprint.desc
===================================================================
--- package/e17/emprint/emprint.desc (revision 34823)
+++ package/e17/emprint/emprint.desc (working copy)
@@ -27,9 +27,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r43316
[P] X -----5---9 170.200

[O] . $base/package/*/*/e17-conf.in

-[D] X emprint-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/emprint -r41533
+[D] X emprint-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/emprint -r43316
Index: package/e17/empower/empower.desc
===================================================================
--- package/e17/empower/empower.desc (revision 34823)
+++ package/e17/empower/empower.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] X -----5---9 490.400

[O] . $base/package/*/*/e17-conf.in

-[D] X empower-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/empower -r41533
+[D] X empower-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/empower -r43316
Index: package/e17/e/e.desc
===================================================================
--- package/e17/e/e.desc (revision 34823)
+++ package/e17/e/e.desc (working copy)
@@ -28,9 +28,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] 2009-07-29
[P] X -----5---9 490.300

[O] . $base/package/*/*/e17-conf.in

-[D] X e-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/e -r41533
+[D] 400424540 enlightenment-0.16.999.062.tar.bz2 http://download.enlightenment.org/snapshots/2009-07-29/enlightenment-0.16.999.062.tar.bz2
Index: package/e17/ewl/ewl.desc
===================================================================
--- package/e17/ewl/ewl.desc (revision 34823)
+++ package/e17/ewl/ewl.desc (working copy)
@@ -28,9 +28,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r43316
[P] X -----5---9 149.200

[O] . $base/package/*/*/e17-conf.in

-[D] X ewl-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/ewl -r41533
+[D] X ewl-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/ewl -r43316
Index: package/e17/express/express.desc
===================================================================
--- package/e17/express/express.desc (revision 34823)
+++ package/e17/express/express.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] O -----5---9 490.300

[O] . $base/package/*/*/e17-conf.in

-[D] X express-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/BROKEN express -r41533
+[D] X express-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/BROKEN express/express -r43316
Index: package/e17/enterminus/enterminus.desc
===================================================================
--- package/e17/enterminus/enterminus.desc (revision 34823)
+++ package/e17/enterminus/enterminus.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r43316
[P] X -----5---9 138.900

[O] . $base/package/*/*/e17-conf.in

-[D] X enterminus-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/PROTO/enterminus -r41533
+[D] X enterminus-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/PROTO/enterminus -r43316
Index: package/e17/ecore/ecore.conf
===================================================================
--- package/e17/ecore/ecore.conf (revision 34823)
+++ package/e17/ecore/ecore.conf (working copy)
@@ -23,7 +23,7 @@

# SDL - ecore_evas is b0rken currently if sdl enabled - 080402
#pkginstalled libsdl && var_append confopt " " "--enable-ecore-sdl"
-var_append confopt " " "--disable-ecore-evas-sdl"
+#var_append confopt " " "--disable-ecore-evas-sdl"

# enable framebuffer
var_append confopt " " "--enable-ecore-fb --enable-ecore-evas-fb"
Index: package/e17/ecore/ecore.desc
===================================================================
--- package/e17/ecore/ecore.desc (revision 34823)
+++ package/e17/ecore/ecore.desc (working copy)
@@ -31,9 +31,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] 2009-07-29
[P] X -----5---9 137.400

[O] . $base/package/*/*/e17-conf.in

-[D] X ecore-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/ecore -r41533
+[D] 1244364489 ecore-0.9.9.062.tar.bz2 http://download.enlightenment.org/snapshots/2009-07-29/
Index: package/e17/evas/evas.desc
===================================================================
--- package/e17/evas/evas.desc (revision 34823)
+++ package/e17/evas/evas.desc (working copy)
@@ -31,9 +31,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] X -----5---9 137.000

[O] . $base/package/*/*/e17-conf.in

-[D] X evas-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/evas -r41533
+[D] X evas-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/evas -r43316
Index: package/e17/ephoto/ephoto.desc
===================================================================
--- package/e17/ephoto/ephoto.desc (revision 34823)
+++ package/e17/ephoto/ephoto.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] X -----5---9 490.400

[O] . $base/package/*/*/e17-conf.in

-[D] X ephoto-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/ephoto -r41533
+[D] X ephoto-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/ephoto -r43316
Index: package/e17/eclair/eclair.desc
===================================================================
--- package/e17/eclair/eclair.desc (revision 34823)
+++ package/e17/eclair/eclair.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] X -----5---9 490.300

[O] . $base/package/*/*/e17-conf.in

-[D] X eclair-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/OLD/eclair -r41533
+[D] X eclair-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/OLD/eclair -r43316
Index: package/e17/edbus/edbus.desc
===================================================================
--- package/e17/edbus/edbus.desc (revision 34823)
+++ package/e17/edbus/edbus.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] 2009-07-29
[P] X -----5---9 170.200

[O] . $base/package/*/*/e17-conf.in

-[D] X edbus-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/e_dbus -r41533
+[D] 323078611 e_dbus-0.5.0.062.tar.bz2 http://download.enlightenment.org/snapshots/2009-07-29/
Index: package/e17/etk-extra/etk-extra.desc
===================================================================
--- package/e17/etk-extra/etk-extra.desc (revision 34823)
+++ package/e17/etk-extra/etk-extra.desc (working copy)
@@ -27,9 +27,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r43316
[P] X -----5---9 170.300

[O] . $base/package/*/*/e17-conf.in

-[D] X etk-extra-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/etk_extra -r41533
+[D] X etk-extra-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/OLD/etk_extra -r43316
Index: package/e17/exml/exml.desc
===================================================================
--- package/e17/exml/exml.desc (revision 34823)
+++ package/e17/exml/exml.desc (working copy)
@@ -28,9 +28,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r43316
[P] X -----5---9 149.250

[O] . $base/package/*/*/e17-conf.in

-[D] X exml-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/exml -r41533
+[D] X exml-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/exml -r43316
Index: package/e17/edje/edje.desc
===================================================================
--- package/e17/edje/edje.desc (revision 34823)
+++ package/e17/edje/edje.desc (working copy)
@@ -33,9 +33,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] 2009-07-29
[P] X -----5---9 138.000

[O] . $base/package/*/*/e17-conf.in

-[D] X edje-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/edje -r41533
+[D] 3637152148 edje-0.9.92.062.tar.bz2 http://download.enlightenment.org/snapshots/2009-07-29/
Index: package/e17/efreet/efreet.desc
===================================================================
--- package/e17/efreet/efreet.desc (revision 34823)
+++ package/e17/efreet/efreet.desc (working copy)
@@ -30,9 +30,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] 2009-07-29
[P] X -----5---9 138.800

[O] . $base/package/*/*/e17-conf.in

-[D] X efreet-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/efreet -r41533
+[D] 2709873108 efreet-0.5.0.062.tar.bz2 http://download.enlightenment.org/snapshots/2009-07-29/
Index: package/e17/erss/erss.desc
===================================================================
--- package/e17/erss/erss.desc (revision 34823)
+++ package/e17/erss/erss.desc (working copy)
@@ -26,9 +26,9 @@

[L] GPL
[S] Alpha
-[V] x35932
+[V] r43316
[P] X -----5---9 170.200

[O] . $base/package/*/*/e17-conf.in

-[D] X erss-r35932.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/MISC/erss -r35932
+[D] X erss-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/OLD/MISC/erss -r43316
Index: package/e17/eina/eina.desc
===================================================================
--- package/e17/eina/eina.desc (revision 34823)
+++ package/e17/eina/eina.desc (working copy)
@@ -27,9 +27,9 @@

[L] GPL
[S] Beta
-[V] r41533
+[V] r43316
[P] X -----5---9 136.050

[O] . $base/package/*/*/e17-conf.in

-[D] X eina-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/eina -r41533
+[D] X eina-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/eina -r43316
Index: package/e17/rage/rage.desc
===================================================================
--- package/e17/rage/rage.desc (revision 34823)
+++ package/e17/rage/rage.desc (working copy)
@@ -27,9 +27,9 @@

[L] GPL
[S] Alpha
-[V] r41533
+[V] r43316
[P] X -----5---9 170.200

[O] . $base/package/*/*/e17-conf.in

-[D] X rage-r41533.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/rage -r41533
+[D] X rage-r43316.tar.bz2 svn+http://svn.enlightenment.org/svn/e/trunk/rage -r43316

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Qt4 compile fills partition

2009-11-05T08:24:57Z

Hi,

Barry Kauler wrote:

> I got T2 out of svn about 5 - 6 days ago.
>
> When it got to Qt4, I had 7.1GB free space in the partition. After a
> very long time, a few hours, the free space had dropped to under 1GB,
> so I killed the compile.
>
> build/ ... /var/adm/logs/5-qt4.out is the culprit, it just keeps growing.
>
> I found that Qt asks a question, but it seems that T2 gives the wrong
> answer, so Qt asks the same question again, and again...
>
> "Which edition of Qt do you want to use ?
>
> Type 'c' if you want to use the Commercial Edition.
> Type 'o' if you want to use the Open Source Edition."
>
> Regards,
> Barry Kauler
>

Yeah, that annoyed me once, years(?) ago, when they changed their
questioning, too. However, it currently builds for me:

-rw-r--r-- 1 root root 9358063 Oct 2 10:58
build/reference-8.0-trunk-reference-x86-pentiumpro-linux/var/adm/logs/5-qt4.log
-rw-r--r-- 1 root root 8677388 Nov 4 01:22
build/reference64-8.0-trunk-reference-x86-64-linux/var/adm/logs/5-qt4.log

The code in package/kde/qt4/qt4.conf:

...
# accept the GPL
echo "o
yes" | ./configure $confopt
...

Maybe it asks a second time on your system, after re-generating
something, or so?

René

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Sqlite 3.6.19 needs tcl

2009-11-05T08:19:07Z

Hi Barry,

Barry Kauler wrote:

> Using T2 from SVN 5 - 6 days ago.
>
> Puppy Linux is a cutdown distro and we avoid unnecessary dependencies.
> The Sqlite package version 3.6.19 needs Tcl to compile. This is
> despite '--disable-tcl'.
>
> I rolled back to Sqlite 3.6.12, that I had used in T2 build early this
> year, and yep, that is fine, does not need Tcl.
>
> I joined the Sqlite mail list and asked about this. I was informed
> that I should use the "amalgamation" source package. I see too that
> they have now released 3.6.20:
>
> http://www.sqlite.org/download.html
>

Thanks for the email. I updated sqlite this morning as part of our daily
auto-updates. Their different packaging must be quite new, I'll check
them and make sure the T2 package builds without TCL.

René

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

[HOWTO] encrypted swap and home partitions

2009-11-05T07:27:14Z

Hey,

This is supported with r34790 on 3rd of November (As I understand this should be included in 8.0-rc2 release). You'll have to recompile sysvinit if you want that on your production system. It will create file /etc/init.d/system.new so you'll have to rename it.

TODO: there is no rootfs support!!!

!!!BACKUP ALL YOUR DATA AND DO IT ON YOUR OWN RISK!!!
(it's safe if you won't mix up device names)

/dev/sda6 will be used for /home/baldzius
/dev/sda8 will be used for swap

Note: I'll be doing this on my production system so I'll have different partitions created for that, please change device names accordingly to your system.
(Haven't tested but LVM should work fine)

Install cryptsetup:

./scripts/Emerge-Pkg cryptsetup

First we will start from swap. Encrypting swap is very easy and straight forward, all you need to do is add a line in /etc/crypttab
(don't worry if this file is missing in your system):

swap /dev/sda8 defaults

Next comment out swap line in /etc/fstab and add a new one (you can change existing one if you know what are you doing):

/dev/mapper/swap swap swap defaults 0 0

That's it. Time to reboot and check the system

# cat /proc/swaps
Filename Type Size Used Priority
/dev/mapper/swap partition 3903752 0 -1

# cryptsetup status swap
/dev/mapper/swap is active:
cipher: aes-cbc-essiv:sha256

keysize: 256 bits
device: /dev/sda8
offset: 0 sectors
size: 7807527 sectors
mode: read/write

WARNING: system script won't do anything if you have swap created as LUKS device, you'll have to take care of it by yourself and use other than swap name.

As you can see by default it uses aes-cbc-essiv cipher with 256bits keysize. You can make changes in /etc/crypttab if you want to customize it:

swap /dev/sda8 /dev/urandom -c aes-xts-plain -s 512

options field is treated as string so don't worry.

NOTE: you can google about differences between /dev/random and /dev/urandom in case you wonder why I've used the later one.

WARNING: right now /etc/init.d/system init scripts supports aes , sha256 and sha512 by loading those modules, make sure you've added required modules if you want to use something else.

Reboot

Checking if changes worked:

# cryptsetup status swap
/dev/mapper/swap is active:
cipher: aes-xts-plain
keysize: 512 bits
device: /dev/sda8
offset: 0 sectors
size: 7807527 sectors

mode: read/write

So far so good.

Basically I would recommend the default value as it looks cleaner:

swap /dev/sda8 defaults

NOTE: I didn't implement crypt support for /tmp . I saw no point as we mount it to shared memmory. If you still want to encrypt tmp you can encrypt it as LUKS device just like /home below.

Next step I am going to encrypt /home dir. As I am doing this on production system and already have /home I will do it for specific user (baldzius) with spare partition created.

I will use /dev/sda6

Fill partition with random data (this will take a while depending on the size of your partition):

I would recommend skiping this step, unless you're very paranoic :)

# dd if=/dev/urandom of=/dev/sda6

NOTE: the step below is needed only in case if you skiped swap steps and went straight to encrypting home directory. You don't need to load modules if you have already swap encrypted as system init script have them already loaded.

Load crypt modules:

# modprobe dm-mod
# modprobe aes
# modprobe sha256

Initialize the partition (use any cipher you want):

# cryptsetup -c aes-cbc-essiv:sha256 -y -s 256 luksFormat /dev/sda6

WARNING!
========
This will overwrite data on /dev/sda6 irrevocably.

Are you sure? (Type uppercase yes): YES
Enter LUKS passphrase:
Verify passphrase:

Type the password of your choise.

Create device mapping:

# cryptsetup luksOpen /dev/sda6 home
Enter passphrase for /dev/sda6:
Key slot 0 unlocked.

You can check the device:

# ls -la /dev/mapper/
total 0K
drwxr-xr-x 2 root root 100 Nov 1 13:59 .

drwxr-xr-x 15 root root 12620 Nov 1 13:59 ..
crw------- 1 root root 10, 59 Nov 1 12:25 control
brw------- 1 root root 253, 0 Nov 1 13:59 home
brw------- 1 root root 253, 1 Nov 1 12:25 swap

# dmsetup ls
home (253, 0)
swap (253, 1)

The list will be bigger if you are using LVM

Create file system, I will use ext4 (pick up any you like):

# mkfs.ext4 /dev/mapper/home

Now you can do testing to see if it works by mounting it and doing some file operations, unmount it when you are done.

Also don't forget to remove the mapping:

# cryptsetup luksClose home

Final step is to mount it after each reboot, add to /etc/fstab :

/dev/mapper/home /home/baldzius ext4 relatime,errors=remount-ro 0 2

That's it, time for reboot.

On system startup you'll be asked to type for passphrase. Volume won't be mounted if you fail for 3x times.

Check the system:

# df -h
...
/dev/mapper/home 9.2G 149M 8.6G 2% /home/baldzius

# cryptsetup status home
/dev/mapper/home is active:
cipher: aes-cbc-essiv:sha256
keysize: 256 bits
device: /dev/sda6
offset: 2056 sectors
size: 19532921 sectors
mode: read/write

Looks good. Now you're encrypted, enjoy! :)

BR,
Aldas

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Sqlite 3.6.19 needs tcl

2009-11-04T14:24:03Z

Using T2 from SVN 5 - 6 days ago.

Puppy Linux is a cutdown distro and we avoid unnecessary dependencies.
The Sqlite package version 3.6.19 needs Tcl to compile. This is
despite '--disable-tcl'.

I rolled back to Sqlite 3.6.12, that I had used in T2 build early this
year, and yep, that is fine, does not need Tcl.

I joined the Sqlite mail list and asked about this. I was informed
that I should use the "amalgamation" source package. I see too that
they have now released 3.6.20:

http://www.sqlite.org/download.html

Regards,
Barry Kauler

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Qt4 compile fills partition

2009-11-04T14:11:54Z

I got T2 out of svn about 5 - 6 days ago.

When it got to Qt4, I had 7.1GB free space in the partition. After a
very long time, a few hours, the free space had dropped to under 1GB,
so I killed the compile.

build/ ... /var/adm/logs/5-qt4.out is the culprit, it just keeps growing.

I found that Qt asks a question, but it seems that T2 gives the wrong
answer, so Qt asks the same question again, and again...

"Which edition of Qt do you want to use ?

Type 'c' if you want to use the Commercial Edition.
Type 'o' if you want to use the Open Source Edition."

Regards,
Barry Kauler

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Recent binutils update

2009-11-03T17:53:49Z

OK, have t2 installed as a virtual client on Fedora12-Beta

Building t2-trunk now.

Building glibc takes about 10 minutes - much better than 4 hrs.

Parallel build using 2 jobs

On Sun, 2009-11-01 at 10:07 -0600, Bob Gustafson wrote:

> Yes, I noticed that my 2nd patch file did not work after I reloaded
> glibc and did svn update. Inspecting the configure file showed that it
> already had the changes and the patch file could not find the old
> patterns.
>
> My sister's old laptop takes about 4 hrs 2 min 3 sec to build glibc..
>
> Masochistic. I need to get my virtual box configured.
>
> Bob G
>
> On Sun, 2009-11-01 at 09:17 +0200, Mika Pesu wrote:
> > Hi,
> >
> > Thanks for testing and modifying the upstream patch. From the svn log i
> > noticed that René commited some binutils patch for glibc:
> > Commit by *rene* :: r*34767* /trunk/package/base/glibc/
> > (any-cc-ok.patch.cross binutils-22.patch):
> >
> > * fixed glibc/configure for the new binutils version
> >
> > Look's like it's fixed now at svn.
> >
> > -Mika
> >
> >
> > Bob Gustafson kirjoitti:
> > > As one who has a little free time, I did some cut and paste from the git
> > > page you referenced. And then discovered that the git listing seems to
> > > add an additional line return and two leading spaces (or something like
> > > that) to each line listed.
> > >
> > > I did a manual edit of the two files and then successfully compiled
> > > glibc. This does not mean that my keystrokes are correct.. Looking at
> > > the upstream patch, I think I like your patch files better. The
> > > selection regex looks a little more consistent than the upstream patch.
> > >
> > > If anyone is interested, I have attached the two patch files. They could
> > > be dropped into directory package/base/glibc
> > >
> > > Bob G
> > >
> > > On Fri, 2009-10-30 at 20:21 +0200, Mika Pesu wrote:
> > >
> > >> Hi list,
> > >>
> > >> The recent update of binutils broke glibc, because glibc fails to
> > >> recognize ld --version that comes from binutils.
> > >>
> > >> With some git browsing i found the upstream commit to fix it, here's the
> > >> link:
> > >> http://sourceware.org/git/?p=glibc.git;a=commitdiff;h=0549ce8ec1f66a41c59815e50dd48ffd74fabb52;hp=6dfeb5245bca9f2ef94815ed9a7b979a9e0be744
> > >>
> > >> I havent yet tryed the upstream patch, but if someone with free time
> > >> could test it.
> > >>
> > >> Attached is my local patch for glibc, but i guess upstream fix could be
> > >> better one.
> > >>
> > >> -Mika
> > >> plain text document attachment (binutils_recognize_for_glibc.patch)
> > >> diff --git a/configure b/configure
> > >> index 7555891..fdfbd2d 100755
> > >> --- a/configure
> > >> +++ b/configure
> > >> @@ -4839,7 +4839,7 @@ $as_echo_n "checking version of $AS... " >&6; }
> > >> ac_prog_version=`$AS --version 2>&1 | sed -n 's/^.*GNU assembler.* $[0-9]*\.[0-9.]*$.*$/\1/p'`
> > >> case $ac_prog_version in
> > >> '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
> > >> - 2.1[3-9]*)
> > >> + 2.1[3-9]*|[2-9].[2-9]*)
> > >> ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
> > >> *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;
> > >>
> > >> @@ -4902,7 +4902,7 @@ $as_echo_n "checking version of $LD... " >&6; }
> > >> ac_prog_version=`$LD --version 2>&1 | sed -n 's/^.*GNU ld.* $[0-9][0-9]*\.[0-9.]*$.*$/\1/p'`
> > >> case $ac_prog_version in
> > >> '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
> > >> - 2.1[3-9]*)
> > >> + 2.1[3-9]*|[2-9].[2-9]*)
> > >> ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
> > >> *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;
> > >>
> > >> diff --git a/configure.in b/configure.in
> > >> index d54e952..05c396a 100644
> > >> --- a/configure.in
> > >> +++ b/configure.in
> > >> @@ -897,10 +897,10 @@ AC_SUBST(MIG)dnl Needed by sysdeps/mach/configure.in
> > >> # Accept binutils 2.13 or newer.
> > >> AC_CHECK_PROG_VER(AS, $AS, --version,
> > >> [GNU assembler.* $[0-9]*\.[0-9.]*$],
> > >> - [2.1[3-9]*], AS=: critic_missing="$critic_missing as")
> > >> + [2.1[3-9]*|[2-9].[2-9]*], AS=: critic_missing="$critic_missing as")
> > >> AC_CHECK_PROG_VER(LD, $LD, --version,
> > >> [GNU ld.* $[0-9][0-9]*\.[0-9.]*$],
> > >> - [2.1[3-9]*], LD=: critic_missing="$critic_missing ld")
> > >> + [2.1[3-9]*|[2-9].[2-9]*], LD=: critic_missing="$critic_missing ld")
> > >>
> > >> # We need the physical current working directory. We cannot use the
> > >> # "pwd -P" shell builtin since that's not portable. Instead we try to
> > >>
> > >>
> >
>
>
>
> -----------------------------------------------------------
> If you wish to unsubscribe from this mailing, send mail to
> lists@... with a subject of: unsubscribe t2

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Build-Pkg mine is OK, but fails under Build-Target]

2009-11-02T01:33:32Z

Hi all,

dialog part committed:

Committed revision 34772.

Have fun,

Mika Pesu wrote:

> Hi
>
> I attached util-linux and dialog patches. Dont really remember why i
> have also changed util-linux build priority, but the attached dialog
> patch priority should fix the mine package build.
>
> -Mika
>
>
> Bob Gustafson kirjoitti:
>> Hi Mika
>>
>> If you can just post your changes to the list, I'm sure that they will
>> find their way into the t2-trunk.
>>
>> The dialog package is listed as a dependency to mine, but it isn't being
>> built. Your comments seem right on.
>>
>> Bob G
>>
>> On Mon, 2009-10-26 at 19:45 +0200, Mika Pesu wrote:
>>
>>> Hi,
>>>
>>> I noticed this my self too, the key is to include dialog package as
>>> well and modify it's build stages and priority. Dont remember
>>> offhand but i think dialog package was missing stage 1 build and
>>> it's build priority was after mine package. I think this was the
>>> cause of mine build fail, that it is, if you get the same error on
>>> stage 1 mine build.
>>>
>>> I can't commit the changes my self to svn at the moment, but if
>>> someone with svn write access could do it and verify dialog build
>>> options that those are right.
>>>
>>> -Mika
>>>
>>>
>>> Bob Gustafson kirjoitti:
>>>
>>>> I am just following the Documentation and am trying to build the T2
>>>> system itself. My build system was a small old laptop which seemed to
>>>> trundle on ok, except when the Build-Target couldn't find dialog.h
>>>>
>>>> I now have a new system which has the Intel virtual bit and lots of
>>>> horsepower. Just brought it up a couple of hours ago running
>>>> Fedora-12-Beta. F12 seems to install and run pretty well, even with
>>>> hardware RAID, which was a problem with Fedora-11
>>>>
>>>> My intention is to bring up T2 as a virtual client with Fedora-12
>>>> as the
>>>> base. Lots of moving parts at the moment though.
>>>>
>>>> I would appreciate if you would take a look at the Build-Target of the
>>>> current t2-trunk as it is in svn. My build broke in mine where it
>>>> could
>>>> not find the dialog package. I sent several emails to the list
>>>> detailing
>>>> my progress and problem.
>>>>
>>>> Thanks for your clues. I will take them into account when I get my
>>>> virtual T2 client installed.
>>>>
>>>> Bob G
>>>>
>>>>
>>>> On Sun, 2009-10-25 at 10:22 +0100, Rene Rebe wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> On 22.10.2009, at 20:41, Bob Gustafson wrote:
>>>>>
>>>>>
>>>>>> Reading some of the handbook, it seems that Build-Pkg does not check
>>>>>> dependencies, but that Build-Target does (??) Perhaps this is the
>>>>>> problem.
>>>>>>
>>>>>>
>>>>> Neither does. Emerge-Pkg does, but it is only meant for in-system
>>>>> package installation and updates.
>>>>>
>>>>>
>>>>>> mine fails looking for dialog.h
>>>>>>
>>>>>> There are lots of dialog.h files sprinkled around in the t2-trunk
>>>>>> directory, and there is a [DEP] dialog in the mine.cache file.
>>>>>>
>>>>>> ./scripts/Build-Pkg mine works fine:
>>>>>>
>>>>>> == 13:28:04 =[9]=> Building archiver/mine [0.23 8.0-trunk].
>>>>>> -> Reading package config: package/archiver/mine/mine.conf
>>>>>> -> Preparing build in src.mine.default.20091022.132759.329.susie
>>>>>> -> Building. Writing output to $root/var/adm/logs/9-mine.out
>>>>>> -> $root/var/adm/logs/9-mine.out -> 9-mine.log
>>>>>> == 10/22/09 13:30:02 =[9]=> Finished building package mine
>>>>>>
>>>>>>
>>>>> Build-Pkg without any other option installs into your running system
>>>>> (like Emerge-Pkg).
>>>>>
>>>>>
>>>>>> The local var/adm/logs directory shows an err file which happened
>>>>>> about
>>>>>> ten minutes ago during the Build-Target run and a much older
>>>>>> 0-mine.log file
>>>>>>
>>>>>> Note that the 9-mine.log file from the above Build-Pkg run appears
>>>>>> in
>>>>>> the system root directory /var/adm/logs
>>>>>>
>>>>>>
>>>>> That's the default of Build-Pkg. You would need to specify some more
>>>>> options like -chroot and -root etc to exactly mimic what happens
>>>>> during the target build.
>>>>>
>>>>>
>>>>>> These things are a puzzle to me. Why are these files scattered
>>>>>> around
>>>>>> and more importantly, why is mine failing to build under
>>>>>> Build-Target?
>>>>>>
>>>>>>
>>>>> Because then it builds in the cross compile / sandbox, most probably
>>>>> some dependency or so is missing. I also recently changed mine to
>>>>> cross compile with the gasgui fronted, because I needed that. Maybe a
>>>>> typo slipped in.
>>>>>
>>>>>
>>>>> What kind of target and configuration do you build?
>>>>>
>>>>> --
>>>>> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
>>>>> http://exactcode.com | http://t2-project.org |
>>>>> http://rene.rebe.name
>>>>>
>>>>>
>>>>>
>>>>
>>>> ----------------------------------------------------------- If you
>>>> wish to unsubscribe from this mailing, send mail to
>>>> lists@... with a subject of: unsubscribe t2
>>>>
>>>>
>>>>
>>
>>
>>
>

Re: Recent binutils update

2009-11-01T08:07:40Z

Yes, I noticed that my 2nd patch file did not work after I reloaded
glibc and did svn update. Inspecting the configure file showed that it
already had the changes and the patch file could not find the old
patterns.

My sister's old laptop takes about 4 hrs 2 min 3 sec to build glibc..

Masochistic. I need to get my virtual box configured.

Bob G

On Sun, 2009-11-01 at 09:17 +0200, Mika Pesu wrote:

> Hi,
>
> Thanks for testing and modifying the upstream patch. From the svn log i
> noticed that René commited some binutils patch for glibc:
> Commit by *rene* :: r*34767* /trunk/package/base/glibc/
> (any-cc-ok.patch.cross binutils-22.patch):
>
> * fixed glibc/configure for the new binutils version
>
> Look's like it's fixed now at svn.
>
> -Mika
>
>
> Bob Gustafson kirjoitti:
> > As one who has a little free time, I did some cut and paste from the git
> > page you referenced. And then discovered that the git listing seems to
> > add an additional line return and two leading spaces (or something like
> > that) to each line listed.
> >
> > I did a manual edit of the two files and then successfully compiled
> > glibc. This does not mean that my keystrokes are correct.. Looking at
> > the upstream patch, I think I like your patch files better. The
> > selection regex looks a little more consistent than the upstream patch.
> >
> > If anyone is interested, I have attached the two patch files. They could
> > be dropped into directory package/base/glibc
> >
> > Bob G
> >
> > On Fri, 2009-10-30 at 20:21 +0200, Mika Pesu wrote:
> >
> >> Hi list,
> >>
> >> The recent update of binutils broke glibc, because glibc fails to
> >> recognize ld --version that comes from binutils.
> >>
> >> With some git browsing i found the upstream commit to fix it, here's the
> >> link:
> >> http://sourceware.org/git/?p=glibc.git;a=commitdiff;h=0549ce8ec1f66a41c59815e50dd48ffd74fabb52;hp=6dfeb5245bca9f2ef94815ed9a7b979a9e0be744
> >>
> >> I havent yet tryed the upstream patch, but if someone with free time
> >> could test it.
> >>
> >> Attached is my local patch for glibc, but i guess upstream fix could be
> >> better one.
> >>
> >> -Mika
> >> plain text document attachment (binutils_recognize_for_glibc.patch)
> >> diff --git a/configure b/configure
> >> index 7555891..fdfbd2d 100755
> >> --- a/configure
> >> +++ b/configure
> >> @@ -4839,7 +4839,7 @@ $as_echo_n "checking version of $AS... " >&6; }
> >> ac_prog_version=`$AS --version 2>&1 | sed -n 's/^.*GNU assembler.* $[0-9]*\.[0-9.]*$.*$/\1/p'`
> >> case $ac_prog_version in
> >> '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
> >> - 2.1[3-9]*)
> >> + 2.1[3-9]*|[2-9].[2-9]*)
> >> ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
> >> *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;
> >>
> >> @@ -4902,7 +4902,7 @@ $as_echo_n "checking version of $LD... " >&6; }
> >> ac_prog_version=`$LD --version 2>&1 | sed -n 's/^.*GNU ld.* $[0-9][0-9]*\.[0-9.]*$.*$/\1/p'`
> >> case $ac_prog_version in
> >> '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
> >> - 2.1[3-9]*)
> >> + 2.1[3-9]*|[2-9].[2-9]*)
> >> ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
> >> *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;
> >>
> >> diff --git a/configure.in b/configure.in
> >> index d54e952..05c396a 100644
> >> --- a/configure.in
> >> +++ b/configure.in
> >> @@ -897,10 +897,10 @@ AC_SUBST(MIG)dnl Needed by sysdeps/mach/configure.in
> >> # Accept binutils 2.13 or newer.
> >> AC_CHECK_PROG_VER(AS, $AS, --version,
> >> [GNU assembler.* $[0-9]*\.[0-9.]*$],
> >> - [2.1[3-9]*], AS=: critic_missing="$critic_missing as")
> >> + [2.1[3-9]*|[2-9].[2-9]*], AS=: critic_missing="$critic_missing as")
> >> AC_CHECK_PROG_VER(LD, $LD, --version,
> >> [GNU ld.* $[0-9][0-9]*\.[0-9.]*$],
> >> - [2.1[3-9]*], LD=: critic_missing="$critic_missing ld")
> >> + [2.1[3-9]*|[2-9].[2-9]*], LD=: critic_missing="$critic_missing ld")
> >>
> >> # We need the physical current working directory. We cannot use the
> >> # "pwd -P" shell builtin since that's not portable. Instead we try to
> >>
> >>
>

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Recent binutils update

2009-11-01T00:17:46Z

Hi,

Thanks for testing and modifying the upstream patch. From the svn log i
noticed that René commited some binutils patch for glibc:
Commit by *rene* :: r*34767* /trunk/package/base/glibc/
(any-cc-ok.patch.cross binutils-22.patch):

* fixed glibc/configure for the new binutils version

Look's like it's fixed now at svn.

-Mika

Bob Gustafson kirjoitti:

> As one who has a little free time, I did some cut and paste from the git
> page you referenced. And then discovered that the git listing seems to
> add an additional line return and two leading spaces (or something like
> that) to each line listed.
>
> I did a manual edit of the two files and then successfully compiled
> glibc. This does not mean that my keystrokes are correct.. Looking at
> the upstream patch, I think I like your patch files better. The
> selection regex looks a little more consistent than the upstream patch.
>
> If anyone is interested, I have attached the two patch files. They could
> be dropped into directory package/base/glibc
>
> Bob G
>
> On Fri, 2009-10-30 at 20:21 +0200, Mika Pesu wrote:
>
>> Hi list,
>>
>> The recent update of binutils broke glibc, because glibc fails to
>> recognize ld --version that comes from binutils.
>>
>> With some git browsing i found the upstream commit to fix it, here's the
>> link:
>> http://sourceware.org/git/?p=glibc.git;a=commitdiff;h=0549ce8ec1f66a41c59815e50dd48ffd74fabb52;hp=6dfeb5245bca9f2ef94815ed9a7b979a9e0be744
>>
>> I havent yet tryed the upstream patch, but if someone with free time
>> could test it.
>>
>> Attached is my local patch for glibc, but i guess upstream fix could be
>> better one.
>>
>> -Mika
>> plain text document attachment (binutils_recognize_for_glibc.patch)
>> diff --git a/configure b/configure
>> index 7555891..fdfbd2d 100755
>> --- a/configure
>> +++ b/configure
>> @@ -4839,7 +4839,7 @@ $as_echo_n "checking version of $AS... " >&6; }
>> ac_prog_version=`$AS --version 2>&1 | sed -n 's/^.*GNU assembler.* $[0-9]*\.[0-9.]*$.*$/\1/p'`
>> case $ac_prog_version in
>> '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
>> - 2.1[3-9]*)
>> + 2.1[3-9]*|[2-9].[2-9]*)
>> ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
>> *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;
>>
>> @@ -4902,7 +4902,7 @@ $as_echo_n "checking version of $LD... " >&6; }
>> ac_prog_version=`$LD --version 2>&1 | sed -n 's/^.*GNU ld.* $[0-9][0-9]*\.[0-9.]*$.*$/\1/p'`
>> case $ac_prog_version in
>> '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
>> - 2.1[3-9]*)
>> + 2.1[3-9]*|[2-9].[2-9]*)
>> ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
>> *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;
>>
>> diff --git a/configure.in b/configure.in
>> index d54e952..05c396a 100644
>> --- a/configure.in
>> +++ b/configure.in
>> @@ -897,10 +897,10 @@ AC_SUBST(MIG)dnl Needed by sysdeps/mach/configure.in
>> # Accept binutils 2.13 or newer.
>> AC_CHECK_PROG_VER(AS, $AS, --version,
>> [GNU assembler.* $[0-9]*\.[0-9.]*$],
>> - [2.1[3-9]*], AS=: critic_missing="$critic_missing as")
>> + [2.1[3-9]*|[2-9].[2-9]*], AS=: critic_missing="$critic_missing as")
>> AC_CHECK_PROG_VER(LD, $LD, --version,
>> [GNU ld.* $[0-9][0-9]*\.[0-9.]*$],
>> - [2.1[3-9]*], LD=: critic_missing="$critic_missing ld")
>> + [2.1[3-9]*|[2-9].[2-9]*], LD=: critic_missing="$critic_missing ld")
>>
>> # We need the physical current working directory. We cannot use the
>> # "pwd -P" shell builtin since that's not portable. Instead we try to
>>
>>

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Recent binutils update

2009-10-31T20:44:52Z

As one who has a little free time, I did some cut and paste from the git
page you referenced. And then discovered that the git listing seems to
add an additional line return and two leading spaces (or something like
that) to each line listed.

I did a manual edit of the two files and then successfully compiled
glibc. This does not mean that my keystrokes are correct.. Looking at
the upstream patch, I think I like your patch files better. The
selection regex looks a little more consistent than the upstream patch.

If anyone is interested, I have attached the two patch files. They could
be dropped into directory package/base/glibc

Bob G

On Fri, 2009-10-30 at 20:21 +0200, Mika Pesu wrote:

> Hi list,
>
> The recent update of binutils broke glibc, because glibc fails to
> recognize ld --version that comes from binutils.
>
> With some git browsing i found the upstream commit to fix it, here's the
> link:
> http://sourceware.org/git/?p=glibc.git;a=commitdiff;h=0549ce8ec1f66a41c59815e50dd48ffd74fabb52;hp=6dfeb5245bca9f2ef94815ed9a7b979a9e0be744
>
> I havent yet tryed the upstream patch, but if someone with free time
> could test it.
>
> Attached is my local patch for glibc, but i guess upstream fix could be
> better one.
>
> -Mika
> plain text document attachment (binutils_recognize_for_glibc.patch)
> diff --git a/configure b/configure
> index 7555891..fdfbd2d 100755
> --- a/configure
> +++ b/configure
> @@ -4839,7 +4839,7 @@ $as_echo_n "checking version of $AS... " >&6; }
> ac_prog_version=`$AS --version 2>&1 | sed -n 's/^.*GNU assembler.* $[0-9]*\.[0-9.]*$.*$/\1/p'`
> case $ac_prog_version in
> '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
> - 2.1[3-9]*)
> + 2.1[3-9]*|[2-9].[2-9]*)
> ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
> *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;
>
> @@ -4902,7 +4902,7 @@ $as_echo_n "checking version of $LD... " >&6; }
> ac_prog_version=`$LD --version 2>&1 | sed -n 's/^.*GNU ld.* $[0-9][0-9]*\.[0-9.]*$.*$/\1/p'`
> case $ac_prog_version in
> '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
> - 2.1[3-9]*)
> + 2.1[3-9]*|[2-9].[2-9]*)
> ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
> *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;
>
> diff --git a/configure.in b/configure.in
> index d54e952..05c396a 100644
> --- a/configure.in
> +++ b/configure.in
> @@ -897,10 +897,10 @@ AC_SUBST(MIG)dnl Needed by sysdeps/mach/configure.in
> # Accept binutils 2.13 or newer.
> AC_CHECK_PROG_VER(AS, $AS, --version,
> [GNU assembler.* $[0-9]*\.[0-9.]*$],
> - [2.1[3-9]*], AS=: critic_missing="$critic_missing as")
> + [2.1[3-9]*|[2-9].[2-9]*], AS=: critic_missing="$critic_missing as")
> AC_CHECK_PROG_VER(LD, $LD, --version,
> [GNU ld.* $[0-9][0-9]*\.[0-9.]*$],
> - [2.1[3-9]*], LD=: critic_missing="$critic_missing ld")
> + [2.1[3-9]*|[2-9].[2-9]*], LD=: critic_missing="$critic_missing ld")
>
> # We need the physical current working directory. We cannot use the
> # "pwd -P" shell builtin since that's not portable. Instead we try to
>

[configure.patch]

--- a/configure +++ b/configure @@ -4839,7 +4839,7 @@ ac_prog_version=`$AS --version 2>&1 | sed -n 's/^.*GNU assembler.* $[0-9]*\.[0-9.]*$.*$/\1/p'` case $ac_prog_version in '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;; - 2.1[3-9]*) + 2.1[3-9]*|2.1[0-9][0-9]*|2.[2-9]*|[3-9].*) ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;; *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;; @@ -4902,7 +4902,7 @@ ac_prog_version=`$LD --version 2>&1 | sed -n 's/^.*GNU ld.* $[0-9][0-9]*\.[0-9.]*$.*$/\1/p'` case $ac_prog_version in '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;; - 2.1[3-9]*) + 2.1[3-9]*|2.1[0-9][0-9]*|2.[2-9]*|3-9.*) ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;; *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;

[configure.in.patch]

--- a/configure.in +++ b/configure.in @@ -897,10 +897,10 @@ # Accept binutils 2.13 or newer. AC_CHECK_PROG_VER(AS, $AS, --version, [GNU assembler.* $[0-9]*\.[0-9.]*$], - [2.1[3-9]*], AS=: critic_missing="$critic_missing as") + [2.1[3-9]*|2.1[0-9][0-9]*|2.[2-9]*|[3-9].*], AS=: critic_missing="$critic_missing as") AC_CHECK_PROG_VER(LD, $LD, --version, [GNU ld.* $[0-9][0-9]*\.[0-9.]*$], - [2.1[3-9]*], LD=: critic_missing="$critic_missing ld") + [2.1[3-9]*|2.1[0-9][0-9]*|2.[2-9]*]|[3-9].*, LD=: critic_missing="$critic_missing ld") # We need the physical current working directory. We cannot use the # "pwd -P" shell builtin since that's not portable. Instead we try to

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

bash parallel build

2009-10-31T01:01:00Z

Hi,

I get shell.h:41:23: error: pathnames.h: No such file or directory error
on my quad-core build. This can be solved by adding NOPARALLEL flag to
bash or with attached patch i found from somewhere.....

-Mika

add missing dep

patch by Peter Alfredsen

--- bash-4.0/Makefile.in
+++ bash-4.0/Makefile.in
@@ -1022,10 +1022,11 @@
subst.o: flags.h jobs.h siglist.h execute_cmd.h ${BASHINCDIR}/filecntl.h trap.h pathexp.h
subst.o: mailcheck.h input.h $(DEFSRC)/getopt.h $(DEFSRC)/common.h
subst.o: bashline.h bashhist.h ${GLOB_LIBSRC}/strmatch.h
subst.o: ${BASHINCDIR}/chartypes.h
subst.o: ${BASHINCDIR}/shmbutil.h
+subst.o: ${DEFDIR}/builtext.h
test.o: bashtypes.h ${BASHINCDIR}/posixstat.h ${BASHINCDIR}/filecntl.h
test.o: shell.h syntax.h config.h bashjmp.h ${BASHINCDIR}/posixjmp.h command.h ${BASHINCDIR}/stdc.h error.h
test.o: general.h xmalloc.h bashtypes.h variables.h arrayfunc.h conftypes.h array.h hashlib.h
test.o: quit.h ${BASHINCDIR}/maxpath.h unwind_prot.h dispose_cmd.h
test.o: make_cmd.h subst.h sig.h pathnames.h externs.h test.h

--- bash-4.0/lib/sh/Makefile.in
+++ bash-4.0/lib/sh/Makefile.in
@@ -136,6 +136,9 @@
${BUILD_DIR}/version.h: ${BUILD_DIR}/config.h ${BUILD_DIR}/Makefile Makefile
-( cd ${BUILD_DIR} && ${MAKE} ${MFLAGS} version.h )

+${BUILD_DIR}/pathnames.h: ${BUILD_DIR}/config.h ${BUILD_DIR}/Makefile Makefile
+ -( cd ${BUILD_DIR} && ${MAKE} ${MFLAGS} pathnames.h )
+
# rules for losing makes, like SunOS
casemod.o: casemod.c
clktck.o: clktck.c

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Recent binutils update

2009-10-30T11:21:29Z

Hi list,

The recent update of binutils broke glibc, because glibc fails to
recognize ld --version that comes from binutils.

With some git browsing i found the upstream commit to fix it, here's the
link:
http://sourceware.org/git/?p=glibc.git;a=commitdiff;h=0549ce8ec1f66a41c59815e50dd48ffd74fabb52;hp=6dfeb5245bca9f2ef94815ed9a7b979a9e0be744

I havent yet tryed the upstream patch, but if someone with free time
could test it.

Attached is my local patch for glibc, but i guess upstream fix could be
better one.

-Mika

diff --git a/configure b/configure
index 7555891..fdfbd2d 100755
--- a/configure
+++ b/configure
@@ -4839,7 +4839,7 @@ $as_echo_n "checking version of $AS... " >&6; }
ac_prog_version=`$AS --version 2>&1 | sed -n 's/^.*GNU assembler.* $[0-9]*\.[0-9.]*$.*$/\1/p'`
case $ac_prog_version in
'') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
- 2.1[3-9]*)
+ 2.1[3-9]*|[2-9].[2-9]*)
ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
*) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;

@@ -4902,7 +4902,7 @@ $as_echo_n "checking version of $LD... " >&6; }
ac_prog_version=`$LD --version 2>&1 | sed -n 's/^.*GNU ld.* $[0-9][0-9]*\.[0-9.]*$.*$/\1/p'`
case $ac_prog_version in
'') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;;
- 2.1[3-9]*)
+ 2.1[3-9]*|[2-9].[2-9]*)
ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;;
*) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;;

diff --git a/configure.in b/configure.in
index d54e952..05c396a 100644
--- a/configure.in
+++ b/configure.in
@@ -897,10 +897,10 @@ AC_SUBST(MIG)dnl Needed by sysdeps/mach/configure.in
# Accept binutils 2.13 or newer.
AC_CHECK_PROG_VER(AS, $AS, --version,
[GNU assembler.* $[0-9]*\.[0-9.]*$],
- [2.1[3-9]*], AS=: critic_missing="$critic_missing as")
+ [2.1[3-9]*|[2-9].[2-9]*], AS=: critic_missing="$critic_missing as")
AC_CHECK_PROG_VER(LD, $LD, --version,
[GNU ld.* $[0-9][0-9]*\.[0-9.]*$],
- [2.1[3-9]*], LD=: critic_missing="$critic_missing ld")
+ [2.1[3-9]*|[2-9].[2-9]*], LD=: critic_missing="$critic_missing ld")

# We need the physical current working directory. We cannot use the
# "pwd -P" shell builtin since that's not portable. Instead we try to

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Package split

2009-10-29T01:42:47Z

Hi All,

What is the way to

a) split a single package (for example: bzip2) in more than one binary
file (bzip2-1.0.5.tar.bz2; bzip2-dev-1.0.5.tar.bz2;
bzip2-doc-1.0.5.tar.bz2; etc.)
(postinstall system hook?)

b) make enter all these files in packages.db

The purpose is to have the option to install smaller target.

all the best,

helasz

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Reminder: T2 hardware pages

2009-10-28T11:17:06Z

Hi all,

so, I finally got one of those "netbooks":

http://t2-project.org/hardware/portable/Nokia/Booklet-3G/

(I only check out all the other ones listed there in local stores, ... :-)

I'd like to remind that we highly appreciate to get hardware listing
contributions, especially when they have a little personal touch and
special hardware configuration details. They influence quite a bit how
new people can find T2, at all!

Have fun,

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

OpenDNS by default r34746

2009-10-28T08:43:16Z

Hi all,

it always annoyed me -when hacking on "something new"- to lookup some
name server (I used to memorize the one from my former university,
however they stopped to accept incoming requests, so I quickly learned
to forget about it, ...). I thus added the OpenDNS into the default
/etc/resolv.conf.

I know this is a debatable choice. I do not receive any cent from them
doing so, and we could even run our own open nameserver and get the
google adsense for the unmatched wildcard requests (but I'm not feeling
like setting one up right away, I may, however, in the near future:-).

This only affects unconfigured networks, and without DHCP anyway. The
default /etc/resolv.conf usually has a short life and is overwritten
quickly after the installation anyway. So this should only affect
low-level hackery, who happens to deal with chroot environments during
testing, and new PCB deployment, testing, in labs, without DHCP, anyway.

Have fun,

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: LXDE on t2

2009-10-28T08:36:45Z

Hey,

Aldas Nabazas wrote:

> Hi,
>
> I've added LXDE support couple of days ago so I would like to share
> some information with you. If someone wants to try it make sure you
> are at least at r34717
>
> ./scripts/Emerge-Pkg -deps=none openbox lxde-common lxmenu-data
> menu-cache lxsession lxpanel should be enough for minimal experience.
>
> I've tried myself for couple of minutes and it looks lovely even
> though I will stay on fluxbox.
>
> There are still some programs (like lxinput, lxlauncher,
> lxsession-edit and so on) left out which can be packaged. It should be
> pretty easy and straight forward by using one of existing lxde
> packages as template like lxterminal.
>
> Patches & feedback are welcome.
>
> Enjoy,
> Aldas

Thanks!

--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10785 Berlin
http://exactcode.de | http://t2-project.org | http://rene.rebe.name

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

LXDE on t2

2009-10-28T06:37:51Z

Hi,

I've added LXDE support couple of days ago so I would like to share some information with you. If someone wants to try it make sure you are at least at r34717

./scripts/Emerge-Pkg -deps=none openbox lxde-common lxmenu-data menu-cache lxsession lxpanel should be enough for minimal experience.

I've tried myself for couple of minutes and it looks lovely even though I will stay on fluxbox.

There are still some programs (like lxinput, lxlauncher, lxsession-edit and so on) left out which can be packaged. It should be pretty easy and straight forward by using one of existing lxde packages as template like lxterminal.

Patches & feedback are welcome.

Enjoy,
Aldas

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

[PATCH] support usb devices for install/init script

2009-10-28T05:01:42Z

Hi Rene,

I am attaching the patch so it can be reviewed. I wrote it long time ago and it was working for me fine as I could install T2 using usb key on host or as guest for Virtual Box.

Cheers,
Aldas

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

add_usb4_install_init.patch (2K) Download Attachment

Re: log4cxx and gcc 4.3.4

2009-10-28T04:52:05Z

Hi Marian,

Your patch worked but it wasn't enough , had to make another very small fix, please test and let me know if it's fine.

r34741 and r34742

Cheers,
Aldas

On Thu, Oct 22, 2009 at 8:30 AM, Marian Aldenhövel <marian.aldenhoevel@...> wrote:

Hi,

The version 0.9.7 of log4cxx that is in T2 apparently does not build with GCC 4.3.4. It is complaining about memcpy, strtok, strlen etc not being definied.

I managed to find out that there was a header-cleanup at some stage so that the offending source files now need to #include <cstring> to fix those problems.

I tried my (very unexperienced) hand at creating a patch for that problem and have attached it.

There is a version 0.10.x of log4cxx out, but it adds a dependency on the apache portable runtime which I did not want to put into my build just yet. So patching 0.9.7 is valid for me, and maybe for others as well.

Ciao, MM

--- log4cxx-0.9.7/src/socketoutputstream.cpp.vanilla 2009-10-22 09:45:32.000000000 +0200
+++ log4cxx-0.9.7/src/socketoutputstream.cpp 2009-10-22 09:45:54.000000000 +0200
@@ -13,6 +13,8 @@
* See the License for the specific language governing permissions and
* limitations under the License.
*/
+
+#include <cstring>

#include <log4cxx/helpers/socketoutputstream.h>
#include <log4cxx/helpers/socket.h>
--- log4cxx-0.9.7/src/stringtokenizer.cpp.vanilla 2009-10-22 10:03:12.000000000 +0200
+++ log4cxx-0.9.7/src/stringtokenizer.cpp 2009-10-22 10:03:45.000000000 +0200
@@ -13,6 +13,8 @@
* See the License for the specific language governing permissions and
* limitations under the License.
*/
+
+#include <cstring>

#include <log4cxx/helpers/stringtokenizer.h>

--- log4cxx-0.9.7/src/socketinputstream.cpp.vanilla 2009-10-22 10:04:37.000000000 +0200
+++ log4cxx-0.9.7/src/socketinputstream.cpp 2009-10-22 10:04:46.000000000 +0200
@@ -13,6 +13,8 @@
* See the License for the specific language governing permissions and
* limitations under the License.
*/
+
+#include <cstring>

#include <log4cxx/helpers/socketinputstream.h>
#include <log4cxx/helpers/socket.h>
--- log4cxx-0.9.7/src/xmlsocketappender.cpp.vanilla 2009-10-22 10:12:43.000000000 +0200
+++ log4cxx-0.9.7/src/xmlsocketappender.cpp 2009-10-22 10:12:50.000000000 +0200
@@ -14,6 +14,8 @@
* limitations under the License.
*/

+#include <cstring>
+
#include <log4cxx/net/xmlsocketappender.h>
#include <log4cxx/helpers/loglog.h>
#include <log4cxx/helpers/socketoutputstream.h>
--- log4cxx-0.9.7/src/timezone.cpp.vanilla 2009-10-22 10:12:12.000000000 +0200
+++ log4cxx-0.9.7/src/timezone.cpp 2009-10-22 10:12:19.000000000 +0200
@@ -13,6 +13,8 @@
* See the License for the specific language governing permissions and
* limitations under the License.
*/
+
+#include <cstring>

#include <log4cxx/helpers/timezone.h>
#include <locale>
--- log4cxx-0.9.7/tests/src/drfatestcase.cpp.vanilla 2009-10-22 10:13:07.000000000 +0200
+++ log4cxx-0.9.7/tests/src/drfatestcase.cpp 2009-10-22 10:13:13.000000000 +0200
@@ -14,6 +14,8 @@
* limitations under the License.
*/

+#include <cstring>
+
#include <cppunit/TestFixture.h>
#include <cppunit/extensions/HelperMacros.h>

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Build-Pkg mine is OK, but fails under Build-Target]

2009-10-28T03:54:56Z

Super. Your one character changes to each of two files did the trick.

It is now compiling Build-Target and is well past the mine (field) after
recompiling base/util-linux and /base/dialog

Still building on the old laptop so it will be awhile before I know if
it goes to the end. I am working with svn update of t2-trunk as of a few
minutes ago.

Thanks again for your patch.

Bob G

On Tue, 2009-10-27 at 16:04 +0200, Mika Pesu wrote:

> Hi
>
> I attached util-linux and dialog patches. Dont really remember why i
> have also changed util-linux build priority, but the attached dialog
> patch priority should fix the mine package build.
>
> -Mika
>
>
> Bob Gustafson kirjoitti:
> > Hi Mika
> >
> > If you can just post your changes to the list, I'm sure that they will
> > find their way into the t2-trunk.
> >
> > The dialog package is listed as a dependency to mine, but it isn't being
> > built. Your comments seem right on.
> >
> > Bob G
> >
> > On Mon, 2009-10-26 at 19:45 +0200, Mika Pesu wrote:
> >
> >> Hi,
> >>
> >> I noticed this my self too, the key is to include dialog package as well
> >> and modify it's build stages and priority. Dont remember offhand but i
> >> think dialog package was missing stage 1 build and it's build priority
> >> was after mine package. I think this was the cause of mine build fail,
> >> that it is, if you get the same error on stage 1 mine build.
> >>
> >> I can't commit the changes my self to svn at the moment, but if someone
> >> with svn write access could do it and verify dialog build options that
> >> those are right.
> >>
> >> -Mika
> >>
> >>
> >> Bob Gustafson kirjoitti:
> >>
> >>> I am just following the Documentation and am trying to build the T2
> >>> system itself. My build system was a small old laptop which seemed to
> >>> trundle on ok, except when the Build-Target couldn't find dialog.h
> >>>
> >>> I now have a new system which has the Intel virtual bit and lots of
> >>> horsepower. Just brought it up a couple of hours ago running
> >>> Fedora-12-Beta. F12 seems to install and run pretty well, even with
> >>> hardware RAID, which was a problem with Fedora-11
> >>>
> >>> My intention is to bring up T2 as a virtual client with Fedora-12 as the
> >>> base. Lots of moving parts at the moment though.
> >>>
> >>> I would appreciate if you would take a look at the Build-Target of the
> >>> current t2-trunk as it is in svn. My build broke in mine where it could
> >>> not find the dialog package. I sent several emails to the list detailing
> >>> my progress and problem.
> >>>
> >>> Thanks for your clues. I will take them into account when I get my
> >>> virtual T2 client installed.
> >>>
> >>> Bob G
> >>>
> >>>
> >>> On Sun, 2009-10-25 at 10:22 +0100, Rene Rebe wrote:
> >>>
> >>>
> >>>> Hi,
> >>>>
> >>>> On 22.10.2009, at 20:41, Bob Gustafson wrote:
> >>>>
> >>>>
> >>>>
> >>>>> Reading some of the handbook, it seems that Build-Pkg does not check
> >>>>> dependencies, but that Build-Target does (??) Perhaps this is the
> >>>>> problem.
> >>>>>
> >>>>>
> >>>>>
> >>>> Neither does. Emerge-Pkg does, but it is only meant for in-system
> >>>> package installation and updates.
> >>>>
> >>>>
> >>>>
> >>>>> mine fails looking for dialog.h
> >>>>>
> >>>>> There are lots of dialog.h files sprinkled around in the t2-trunk
> >>>>> directory, and there is a [DEP] dialog in the mine.cache file.
> >>>>>
> >>>>> ./scripts/Build-Pkg mine works fine:
> >>>>>
> >>>>> == 13:28:04 =[9]=> Building archiver/mine [0.23 8.0-trunk].
> >>>>> -> Reading package config: package/archiver/mine/mine.conf
> >>>>> -> Preparing build in src.mine.default.20091022.132759.329.susie
> >>>>> -> Building. Writing output to $root/var/adm/logs/9-mine.out
> >>>>> -> $root/var/adm/logs/9-mine.out -> 9-mine.log
> >>>>> == 10/22/09 13:30:02 =[9]=> Finished building package mine
> >>>>>
> >>>>>
> >>>>>
> >>>> Build-Pkg without any other option installs into your running system
> >>>> (like Emerge-Pkg).
> >>>>
> >>>>
> >>>>
> >>>>> The local var/adm/logs directory shows an err file which happened
> >>>>> about
> >>>>> ten minutes ago during the Build-Target run and a much older
> >>>>> 0-mine.log file
> >>>>>
> >>>>> Note that the 9-mine.log file from the above Build-Pkg run appears
> >>>>> in
> >>>>> the system root directory /var/adm/logs
> >>>>>
> >>>>>
> >>>>>
> >>>> That's the default of Build-Pkg. You would need to specify some more
> >>>> options like -chroot and -root etc to exactly mimic what happens
> >>>> during the target build.
> >>>>
> >>>>
> >>>>
> >>>>> These things are a puzzle to me. Why are these files scattered
> >>>>> around
> >>>>> and more importantly, why is mine failing to build under
> >>>>> Build-Target?
> >>>>>
> >>>>>
> >>>>>
> >>>> Because then it builds in the cross compile / sandbox, most probably
> >>>> some dependency or so is missing. I also recently changed mine to
> >>>> cross compile with the gasgui fronted, because I needed that. Maybe a
> >>>> typo slipped in.
> >>>>
> >>>>
> >>>> What kind of target and configuration do you build?
> >>>>
> >>>> --
> >>>> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
> >>>> http://exactcode.com | http://t2-project.org | http://rene.rebe.name
> >>>>
> >>>>
> >>>>
> >>>>
> >>>
> >>> -----------------------------------------------------------
> >>> If you wish to unsubscribe from this mailing, send mail to
> >>> lists@... with a subject of: unsubscribe t2
> >>>
> >>>
> >>>
> >>>
> >
> >
> >
>
> plain text document attachment (dialog.patch)
> Index: package/base/dialog/dialog.desc
> ===================================================================
> --- package/base/dialog/dialog.desc (revision 34726)
> +++ package/base/dialog/dialog.desc (working copy)
> @@ -36,6 +36,6 @@
> [L] GPL
> [S] Stable
> [V] 1.0-20050911
> -[P] X -----5---9 113.900
> +[P] X -1---5---9 113.900
>
> [D] 1535550711 dialog-1.0-20050911.tgz ftp://invisible-island.net/dialog/
> plain text document attachment (util-linux.patch)
> Index: package/base/util-linux/util-linux.desc
> ===================================================================
> --- package/base/util-linux/util-linux.desc (revision 34726)
> +++ package/base/util-linux/util-linux.desc (working copy)
> @@ -29,6 +29,6 @@
> [L] GPL
> [S] Stable
> [V] 2.16.1
> -[P] X -1-3-----9 115.000
> +[P] X -1-3-----9 112.000
>
> [D] 1920311447 util-linux-ng-2.16.1.tar.bz2 ftp://ftp.kernel.org/pub/linux/utils/util-linux-ng/v2.16/

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Build-Pkg mine is OK, but fails under Build-Target]

2009-10-27T07:04:54Z

Hi

I attached util-linux and dialog patches. Dont really remember why i
have also changed util-linux build priority, but the attached dialog
patch priority should fix the mine package build.

-Mika

Bob Gustafson kirjoitti:

> Hi Mika
>
> If you can just post your changes to the list, I'm sure that they will
> find their way into the t2-trunk.
>
> The dialog package is listed as a dependency to mine, but it isn't being
> built. Your comments seem right on.
>
> Bob G
>
> On Mon, 2009-10-26 at 19:45 +0200, Mika Pesu wrote:
>
>> Hi,
>>
>> I noticed this my self too, the key is to include dialog package as well
>> and modify it's build stages and priority. Dont remember offhand but i
>> think dialog package was missing stage 1 build and it's build priority
>> was after mine package. I think this was the cause of mine build fail,
>> that it is, if you get the same error on stage 1 mine build.
>>
>> I can't commit the changes my self to svn at the moment, but if someone
>> with svn write access could do it and verify dialog build options that
>> those are right.
>>
>> -Mika
>>
>>
>> Bob Gustafson kirjoitti:
>>
>>> I am just following the Documentation and am trying to build the T2
>>> system itself. My build system was a small old laptop which seemed to
>>> trundle on ok, except when the Build-Target couldn't find dialog.h
>>>
>>> I now have a new system which has the Intel virtual bit and lots of
>>> horsepower. Just brought it up a couple of hours ago running
>>> Fedora-12-Beta. F12 seems to install and run pretty well, even with
>>> hardware RAID, which was a problem with Fedora-11
>>>
>>> My intention is to bring up T2 as a virtual client with Fedora-12 as the
>>> base. Lots of moving parts at the moment though.
>>>
>>> I would appreciate if you would take a look at the Build-Target of the
>>> current t2-trunk as it is in svn. My build broke in mine where it could
>>> not find the dialog package. I sent several emails to the list detailing
>>> my progress and problem.
>>>
>>> Thanks for your clues. I will take them into account when I get my
>>> virtual T2 client installed.
>>>
>>> Bob G
>>>
>>>
>>> On Sun, 2009-10-25 at 10:22 +0100, Rene Rebe wrote:
>>>
>>>
>>>> Hi,
>>>>
>>>> On 22.10.2009, at 20:41, Bob Gustafson wrote:
>>>>
>>>>
>>>>
>>>>> Reading some of the handbook, it seems that Build-Pkg does not check
>>>>> dependencies, but that Build-Target does (??) Perhaps this is the
>>>>> problem.
>>>>>
>>>>>
>>>>>
>>>> Neither does. Emerge-Pkg does, but it is only meant for in-system
>>>> package installation and updates.
>>>>
>>>>
>>>>
>>>>> mine fails looking for dialog.h
>>>>>
>>>>> There are lots of dialog.h files sprinkled around in the t2-trunk
>>>>> directory, and there is a [DEP] dialog in the mine.cache file.
>>>>>
>>>>> ./scripts/Build-Pkg mine works fine:
>>>>>
>>>>> == 13:28:04 =[9]=> Building archiver/mine [0.23 8.0-trunk].
>>>>> -> Reading package config: package/archiver/mine/mine.conf
>>>>> -> Preparing build in src.mine.default.20091022.132759.329.susie
>>>>> -> Building. Writing output to $root/var/adm/logs/9-mine.out
>>>>> -> $root/var/adm/logs/9-mine.out -> 9-mine.log
>>>>> == 10/22/09 13:30:02 =[9]=> Finished building package mine
>>>>>
>>>>>
>>>>>
>>>> Build-Pkg without any other option installs into your running system
>>>> (like Emerge-Pkg).
>>>>
>>>>
>>>>
>>>>> The local var/adm/logs directory shows an err file which happened
>>>>> about
>>>>> ten minutes ago during the Build-Target run and a much older
>>>>> 0-mine.log file
>>>>>
>>>>> Note that the 9-mine.log file from the above Build-Pkg run appears
>>>>> in
>>>>> the system root directory /var/adm/logs
>>>>>
>>>>>
>>>>>
>>>> That's the default of Build-Pkg. You would need to specify some more
>>>> options like -chroot and -root etc to exactly mimic what happens
>>>> during the target build.
>>>>
>>>>
>>>>
>>>>> These things are a puzzle to me. Why are these files scattered
>>>>> around
>>>>> and more importantly, why is mine failing to build under
>>>>> Build-Target?
>>>>>
>>>>>
>>>>>
>>>> Because then it builds in the cross compile / sandbox, most probably
>>>> some dependency or so is missing. I also recently changed mine to
>>>> cross compile with the gasgui fronted, because I needed that. Maybe a
>>>> typo slipped in.
>>>>
>>>>
>>>> What kind of target and configuration do you build?
>>>>
>>>> --
>>>> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
>>>> http://exactcode.com | http://t2-project.org | http://rene.rebe.name
>>>>
>>>>
>>>>
>>>>
>>>
>>> -----------------------------------------------------------
>>> If you wish to unsubscribe from this mailing, send mail to
>>> lists@... with a subject of: unsubscribe t2
>>>
>>>
>>>
>>>
>
>
>

Index: package/base/dialog/dialog.desc
===================================================================
--- package/base/dialog/dialog.desc (revision 34726)
+++ package/base/dialog/dialog.desc (working copy)
@@ -36,6 +36,6 @@
[L] GPL
[S] Stable
[V] 1.0-20050911
-[P] X -----5---9 113.900
+[P] X -1---5---9 113.900

[D] 1535550711 dialog-1.0-20050911.tgz ftp://invisible-island.net/dialog/

Index: package/base/util-linux/util-linux.desc
===================================================================
--- package/base/util-linux/util-linux.desc (revision 34726)
+++ package/base/util-linux/util-linux.desc (working copy)
@@ -29,6 +29,6 @@
[L] GPL
[S] Stable
[V] 2.16.1
-[P] X -1-3-----9 115.000
+[P] X -1-3-----9 112.000

[D] 1920311447 util-linux-ng-2.16.1.tar.bz2 ftp://ftp.kernel.org/pub/linux/utils/util-linux-ng/v2.16/

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: compiling a single kernel module package

2009-10-27T01:47:49Z

Hi.

Thanks for your help. I used

hook_add premake 5 abort

to get a working copy of the kernels directory. After that I can reconfigure the environment as you showed and compile my kernel module.

CU,

Jens

-------- Original-Nachricht --------
> Datum: Fri, 23 Oct 2009 09:48:59 +0200
> Von: Rene Rebe <rene@...>
> An: Toan Pham <tpham3783@...>
> CC: "Jens Möller" <komischername@...>, T2 developers mailing list <t2@...>
> Betreff: Re: [t2] compiling a single kernel module package

> It should break at all, but maybe exit is not enough, as it may not
> return a non-zero exit code. exit 1 might help. Better yet is just
> using the T2 function "abort":
>
> hook_add postmake 5 abort
>
> On 23.10.2009, at 06:50, Toan Pham wrote:
>
> > Rene,
> >
> > That's exactly what i've been doing. To enter the sandbox i added
> > hook in the premake in the file /target/<name>/pkg_linux26.conf, ie.
> >
> > hook_add premake 5 "exit"
> >
> > it works fine; however, i notice that when i try to add the same
> > hook for
> > postmake, postflist or finish, the build process does not break.
> > Instead
> > it finishes building as usual. I am not sure if that is normal.
> >
> >
> > Thank you.
> >
> > Toan
> >
> >
> > On Thu, Oct 22, 2009 at 4:46 AM, Rene Rebe <rene@...> wrote:
> >> Hi,
> >> On 21.10.2009, at 13:11, Jens Möller wrote:
> >>
> >> Hi @all!
> >>
> >> I would like to modify a linux kernel module. Unfortunately, the
> >> module
> >> package has a postlinux.conf that does not seem to be executed when
> >> I start
> >> a compile for that single package. Recompiling the whole kernel is
> >> not much
> >> fun when developing a kernel module.
> >>
> >> Is there a way to compile that single kernel module package?
> >>
> >>
> >> In the past it was once possible to only build the third party
> >> modules by
> >> setting: SDEDEBUG_LINUX_SUBMODULES_ONLY, but due to changes in the
> >> kernel
> >> build system I think this does not work correctly anymore. But it's
> >> certainly fixable:
> >> SDEDEBUG_LINUX_SUBMODULES_ONLY scripts/Build-Target -job 5-linux26
> >> # etc.
> >> ...
> >> However, you can also let the kernel build, and then enter the
> >> sandbox to
> >> continue developing / fixing where the build stopped in your module:
> >> cd src.linuux26*
> >> ./debuig.sh
> >> and then also include the sub-module defines, like:
> >> . debug.buildenv.<name of module>
> >> and continue building it with something like eval $MAKE $makeopt #
> >> etc. as
> >> appropriated for the module
> >> Regards,
> >> René
> >> --
> >> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
> >> http://exactcode.com | http://t2-project.org | http://
> >> rene.rebe.name
> >>
> >> -----------------------------------------------------------
> >> If you wish to unsubscribe from this mailing, send mail to
> >> lists@... with a subject of: unsubscribe t2
> >>
> >>
>
> --
> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
> http://exactcode.com | http://t2-project.org | http://rene.rebe.name

--
Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/atbrowser

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Build-Pkg mine is OK, but fails under Build-Target

2009-10-26T11:06:17Z

Hi Mika

If you can just post your changes to the list, I'm sure that they will
find their way into the t2-trunk.

The dialog package is listed as a dependency to mine, but it isn't being
built. Your comments seem right on.

Bob G

On Mon, 2009-10-26 at 19:45 +0200, Mika Pesu wrote:

> Hi,
>
> I noticed this my self too, the key is to include dialog package as well
> and modify it's build stages and priority. Dont remember offhand but i
> think dialog package was missing stage 1 build and it's build priority
> was after mine package. I think this was the cause of mine build fail,
> that it is, if you get the same error on stage 1 mine build.
>
> I can't commit the changes my self to svn at the moment, but if someone
> with svn write access could do it and verify dialog build options that
> those are right.
>
> -Mika
>
>
> Bob Gustafson kirjoitti:
> > I am just following the Documentation and am trying to build the T2
> > system itself. My build system was a small old laptop which seemed to
> > trundle on ok, except when the Build-Target couldn't find dialog.h
> >
> > I now have a new system which has the Intel virtual bit and lots of
> > horsepower. Just brought it up a couple of hours ago running
> > Fedora-12-Beta. F12 seems to install and run pretty well, even with
> > hardware RAID, which was a problem with Fedora-11
> >
> > My intention is to bring up T2 as a virtual client with Fedora-12 as the
> > base. Lots of moving parts at the moment though.
> >
> > I would appreciate if you would take a look at the Build-Target of the
> > current t2-trunk as it is in svn. My build broke in mine where it could
> > not find the dialog package. I sent several emails to the list detailing
> > my progress and problem.
> >
> > Thanks for your clues. I will take them into account when I get my
> > virtual T2 client installed.
> >
> > Bob G
> >
> >
> > On Sun, 2009-10-25 at 10:22 +0100, Rene Rebe wrote:
> >
> >> Hi,
> >>
> >> On 22.10.2009, at 20:41, Bob Gustafson wrote:
> >>
> >>
> >>> Reading some of the handbook, it seems that Build-Pkg does not check
> >>> dependencies, but that Build-Target does (??) Perhaps this is the
> >>> problem.
> >>>
> >>>
> >> Neither does. Emerge-Pkg does, but it is only meant for in-system
> >> package installation and updates.
> >>
> >>
> >>> mine fails looking for dialog.h
> >>>
> >>> There are lots of dialog.h files sprinkled around in the t2-trunk
> >>> directory, and there is a [DEP] dialog in the mine.cache file.
> >>>
> >>> ./scripts/Build-Pkg mine works fine:
> >>>
> >>> == 13:28:04 =[9]=> Building archiver/mine [0.23 8.0-trunk].
> >>> -> Reading package config: package/archiver/mine/mine.conf
> >>> -> Preparing build in src.mine.default.20091022.132759.329.susie
> >>> -> Building. Writing output to $root/var/adm/logs/9-mine.out
> >>> -> $root/var/adm/logs/9-mine.out -> 9-mine.log
> >>> == 10/22/09 13:30:02 =[9]=> Finished building package mine
> >>>
> >>>
> >> Build-Pkg without any other option installs into your running system
> >> (like Emerge-Pkg).
> >>
> >>
> >>> The local var/adm/logs directory shows an err file which happened
> >>> about
> >>> ten minutes ago during the Build-Target run and a much older
> >>> 0-mine.log file
> >>>
> >>> Note that the 9-mine.log file from the above Build-Pkg run appears
> >>> in
> >>> the system root directory /var/adm/logs
> >>>
> >>>
> >> That's the default of Build-Pkg. You would need to specify some more
> >> options like -chroot and -root etc to exactly mimic what happens
> >> during the target build.
> >>
> >>
> >>> These things are a puzzle to me. Why are these files scattered
> >>> around
> >>> and more importantly, why is mine failing to build under
> >>> Build-Target?
> >>>
> >>>
> >> Because then it builds in the cross compile / sandbox, most probably
> >> some dependency or so is missing. I also recently changed mine to
> >> cross compile with the gasgui fronted, because I needed that. Maybe a
> >> typo slipped in.
> >>
> >>
> >> What kind of target and configuration do you build?
> >>
> >> --
> >> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
> >> http://exactcode.com | http://t2-project.org | http://rene.rebe.name
> >>
> >>
> >>
> >
> >
> >
> > -----------------------------------------------------------
> > If you wish to unsubscribe from this mailing, send mail to
> > lists@... with a subject of: unsubscribe t2
> >
> >
> >
>

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

Re: Build-Pkg mine is OK, but fails under Build-Target

2009-10-26T10:45:38Z

Hi,

I noticed this my self too, the key is to include dialog package as well
and modify it's build stages and priority. Dont remember offhand but i
think dialog package was missing stage 1 build and it's build priority
was after mine package. I think this was the cause of mine build fail,
that it is, if you get the same error on stage 1 mine build.

I can't commit the changes my self to svn at the moment, but if someone
with svn write access could do it and verify dialog build options that
those are right.

-Mika

Bob Gustafson kirjoitti:

> I am just following the Documentation and am trying to build the T2
> system itself. My build system was a small old laptop which seemed to
> trundle on ok, except when the Build-Target couldn't find dialog.h
>
> I now have a new system which has the Intel virtual bit and lots of
> horsepower. Just brought it up a couple of hours ago running
> Fedora-12-Beta. F12 seems to install and run pretty well, even with
> hardware RAID, which was a problem with Fedora-11
>
> My intention is to bring up T2 as a virtual client with Fedora-12 as the
> base. Lots of moving parts at the moment though.
>
> I would appreciate if you would take a look at the Build-Target of the
> current t2-trunk as it is in svn. My build broke in mine where it could
> not find the dialog package. I sent several emails to the list detailing
> my progress and problem.
>
> Thanks for your clues. I will take them into account when I get my
> virtual T2 client installed.
>
> Bob G
>
>
> On Sun, 2009-10-25 at 10:22 +0100, Rene Rebe wrote:
>
>> Hi,
>>
>> On 22.10.2009, at 20:41, Bob Gustafson wrote:
>>
>>
>>> Reading some of the handbook, it seems that Build-Pkg does not check
>>> dependencies, but that Build-Target does (??) Perhaps this is the
>>> problem.
>>>
>>>
>> Neither does. Emerge-Pkg does, but it is only meant for in-system
>> package installation and updates.
>>
>>
>>> mine fails looking for dialog.h
>>>
>>> There are lots of dialog.h files sprinkled around in the t2-trunk
>>> directory, and there is a [DEP] dialog in the mine.cache file.
>>>
>>> ./scripts/Build-Pkg mine works fine:
>>>
>>> == 13:28:04 =[9]=> Building archiver/mine [0.23 8.0-trunk].
>>> -> Reading package config: package/archiver/mine/mine.conf
>>> -> Preparing build in src.mine.default.20091022.132759.329.susie
>>> -> Building. Writing output to $root/var/adm/logs/9-mine.out
>>> -> $root/var/adm/logs/9-mine.out -> 9-mine.log
>>> == 10/22/09 13:30:02 =[9]=> Finished building package mine
>>>
>>>
>> Build-Pkg without any other option installs into your running system
>> (like Emerge-Pkg).
>>
>>
>>> The local var/adm/logs directory shows an err file which happened
>>> about
>>> ten minutes ago during the Build-Target run and a much older
>>> 0-mine.log file
>>>
>>> Note that the 9-mine.log file from the above Build-Pkg run appears
>>> in
>>> the system root directory /var/adm/logs
>>>
>>>
>> That's the default of Build-Pkg. You would need to specify some more
>> options like -chroot and -root etc to exactly mimic what happens
>> during the target build.
>>
>>
>>> These things are a puzzle to me. Why are these files scattered
>>> around
>>> and more importantly, why is mine failing to build under
>>> Build-Target?
>>>
>>>
>> Because then it builds in the cross compile / sandbox, most probably
>> some dependency or so is missing. I also recently changed mine to
>> cross compile with the gasgui fronted, because I needed that. Maybe a
>> typo slipped in.
>>
>>
>> What kind of target and configuration do you build?
>>
>> --
>> René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
>> http://exactcode.com | http://t2-project.org | http://rene.rebe.name
>>
>>
>>
>
>
>
> -----------------------------------------------------------
> If you wish to unsubscribe from this mailing, send mail to
> lists@... with a subject of: unsubscribe t2
>
>
>

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2