| View previous topic :: View next topic |
| Author |
Message |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
Posted: Wed Feb 03, 2010 3:40 pm Post subject: |
|
|
bump. ^^
I remerged OOo, removed my OOo 3 config and the formula text is back to normal, though the termination on double-click after restart and load persists.
time to build a debug version ...
edit #1
disabled graphite
- may it ever happen that someone has a use for it ping me. thanks.
edit #2
backtrace of the termination of double-click on formula.
http://pastebin.com/d7dca5742
edit #3
the only experimental option set I can think of ( o, ha ) is the glib GIO code. let's see ... _________________ LibreOffice|split-boost|trans-follow xcb port|instruction set analyzer
_________________
sanity . . . is not with me |
|
| Back to top |
|
 |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5314 Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
 |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
Posted: Mon Feb 08, 2010 8:31 pm Post subject: |
|
|
humm, need to fix graphite depend. somehow skipped DEPEND.
good to know they differ. still, have the same name and use flag now.
debian uses system silgraphite, so I guess it is safe.
anyone may test if any use flags of silgraphite are required.
edit #1
ehhh, bump? ..!  _________________ LibreOffice|split-boost|trans-follow xcb port|instruction set analyzer
_________________
sanity . . . is not with me |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5314 Location: Vienna, Austria; Germany; hello world :)
|
Posted: Tue Feb 09, 2010 2:27 pm Post subject: |
|
|
| geki wrote: | humm, need to fix graphite depend. somehow skipped DEPEND.
good to know they differ. still, have the same name and use flag now.
debian uses system silgraphite, so I guess it is safe.
anyone may test if any use flags of silgraphite are required.
edit #1
ehhh, bump? ..!  |
I'll be using silgraphite all use-flags enabled (since those are also enabled in my /etc/make.conf)
thanks for fixing the dependency & bumping
I'm on it ...
emerge will start soon _________________ Unofficial minimal livecd x86/amd64 w/reiser4+truecrypt (by Neo2)
2.6.37.2_plus_v1: BFS, CFS,THP,compaction, zcache or TOI
Hardcore Linux user since 2004  |
|
| Back to top |
|
 |
albright Veteran


Joined: 16 Nov 2003 Posts: 1705 Location: Near Toronto
|
Posted: Tue Feb 09, 2010 4:30 pm Post subject: |
|
|
this one is not working for me:
| Code: | Failed to emerge app-office/openoffice-3.2.0.12_pre20100208
patching file connectivity/prj/build.lst
Reversed (or previously applied) patch detected! Assume -R? [n]
Apply anyway? [n] y
Hunk #1 FAILED at 28.
1 out of 1 hunk FAILED -- saving rejects to file connectivity/prj/build.lst.rej
Testing patch connectivity-build-fix-mac.diff failed. at /mnt/sdb1/tmp/portage/app-office/openoffice-3.2.0.12_pre20100208/work/ooo-build-045191cbbde92ec54bf24c5261dddc56c274cf7c/patches/apply.pl line 98.
make: *** [stamp/patch.apply] Error 1 |
It makes no difference whether I "apply anyway" or not. _________________ .... there is nothing - absolutely nothing - half so much worth
doing as simply messing about with Linux ...
(apologies to Kenneth Graeme) |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5314 Location: Vienna, Austria; Germany; hello world :)
|
Posted: Tue Feb 09, 2010 4:45 pm Post subject: |
|
|
| albright wrote: | this one is not working for me:
| Code: | Failed to emerge app-office/openoffice-3.2.0.12_pre20100208
patching file connectivity/prj/build.lst
Reversed (or previously applied) patch detected! Assume -R? [n]
Apply anyway? [n] y
Hunk #1 FAILED at 28.
1 out of 1 hunk FAILED -- saving rejects to file connectivity/prj/build.lst.rej
Testing patch connectivity-build-fix-mac.diff failed. at /mnt/sdb1/tmp/portage/app-office/openoffice-3.2.0.12_pre20100208/work/ooo-build-045191cbbde92ec54bf24c5261dddc56c274cf7c/patches/apply.pl line 98.
make: *** [stamp/patch.apply] Error 1 |
It makes no difference whether I "apply anyway" or not. |
++
| Quote: | Reversed (or previously applied) patch detected! Assume -R? [n] y
< /var/tmp/portage/app-office/openoffice-3.2.0.12_pre20100208-r1/work/ooo-build-045191cbbde92ec54bf24c5261dddc56c274cf7c/patches/dev300/./connectivity-build-fix-mac.diff patch -l -p0 -d /var/tmp/portage/app-office/openoffice-3.2.0.12_pre20100208-r1/work/ooo-build-045191cbbde92ec54bf24c5261dddc56c274cf7c/build/OOO320_m12 > /dev/null
Failed to patch connectivity-build-fix-mac.diff
- this is often caused by duplicate sections in a patch.
you will need to manually reverse this patch
make: *** [stamp/patch.apply] Error 1
* ERROR: app-office/openoffice-3.2.0.12_pre20100208-r1 failed:
* make failed
*
* Call stack:
* ebuild.sh, line 48: Called src_compile
* environment, line 13659: Called die
* The specific snippet of code:
* make || die "make failed"
|
_________________ Unofficial minimal livecd x86/amd64 w/reiser4+truecrypt (by Neo2)
2.6.37.2_plus_v1: BFS, CFS,THP,compaction, zcache or TOI
Hardcore Linux user since 2004  |
|
| Back to top |
|
 |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
|
| Back to top |
|
 |
albright Veteran


Joined: 16 Nov 2003 Posts: 1705 Location: Near Toronto
|
Posted: Tue Feb 09, 2010 10:08 pm Post subject: |
|
|
| Quote: | | removed from apply. |
thank you kindly - working now _________________ .... there is nothing - absolutely nothing - half so much worth
doing as simply messing about with Linux ...
(apologies to Kenneth Graeme) |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5314 Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
 |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
|
| Back to top |
|
 |
sian Apprentice

Joined: 26 Jan 2004 Posts: 288
|
Posted: Sat Feb 13, 2010 6:28 pm Post subject: |
|
|
Just one question, what is the difference between the 20100210 build and the official 3.2 version ? _________________ Kernel : 2.6.32 - E17 - OpenRC-0.61 - gcc-4.4.3 |
|
| Back to top |
|
 |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5314 Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
 |
cruzki123 Apprentice

Joined: 16 May 2008 Posts: 193
|
Posted: Sat Feb 13, 2010 9:04 pm Post subject: |
|
|
| It is possible to make the build of the jmalloc in a ebuild? I have troubles in compiling this library. Exist actualized documentation on that? |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5314 Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
 |
cruzki123 Apprentice

Joined: 16 May 2008 Posts: 193
|
Posted: Sun Feb 14, 2010 12:27 am Post subject: |
|
|
That isn't the point. You shouldn't do this, portage is there for a reason.
If not, why not use LFS for build everything? Imagine you uninstall openoffice for wherever reason (or disable the jemalloc use flag) then you will have a library orphaned in your system that portage did not know anything. In this case, is only a single file, but if you do this often you will finish with a lot of cruft in your system.
Anyway, I will try this night the jemalloc use flag. But now openoffice start a lot quicker.
PS: are we compilling openoffice with custom-useflags or are filtered? |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5314 Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
 |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
Posted: Sun Feb 14, 2010 2:09 pm Post subject: |
|
|
| cruzki123 wrote: | | That isn't the point. You shouldn't do this, portage is there for a reason. | we have a volunteer to create a jemalloc ebuild, not jmalloc.
sorry, I could not resist.  _________________ LibreOffice|split-boost|trans-follow xcb port|instruction set analyzer
_________________
sanity . . . is not with me |
|
| Back to top |
|
 |
cruzki123 Apprentice

Joined: 16 May 2008 Posts: 193
|
Posted: Sun Feb 14, 2010 6:08 pm Post subject: |
|
|
Well I will try to make an ebuild but this would be my first time
On other hand, with jemalloc use flag enable, compiling hangs without error in this point.
| Code: |
Compiling: idlc/source/preproc/tokens.c
Compiling: idlc/source/preproc/unix.c
Compiling: idlc/source/wrap_parser.cxx
Making: idlcpp.lib
------------------------------
Making: idlcpp
unx
Compiling: idlc/source/idlcmain.cxx
Compiling: idlc/source/idlc.cxx
Compiling: idlc/source/idlccompile.cxx
Compiling: idlc/source/idlcproduce.cxx
Compiling: idlc/source/errorhandler.cxx
Compiling: idlc/source/options.cxx
Compiling: idlc/source/fehelper.cxx
Compiling: idlc/source/astdeclaration.cxx
Compiling: idlc/source/astscope.cxx
Compiling: idlc/source/aststack.cxx
Compiling: idlc/source/astdump.cxx
Compiling: idlc/source/astinterface.cxx
Compiling: idlc/source/aststruct.cxx
Compiling: idlc/source/aststructinstance.cxx
Compiling: idlc/source/astoperation.cxx
Compiling: idlc/source/astconstant.cxx
Compiling: idlc/source/astenum.cxx
Compiling: idlc/source/astarray.cxx
Compiling: idlc/source/astunion.cxx
Compiling: idlc/source/astexpression.cxx
Compiling: idlc/source/astservice.cxx
------------------------------
Making: idlc
unx
Module 'idlc' delivered successfully. 3 files copied, 4 files unchanged
=============
Building module udkapi
Entering /home/portage/portage/app-office/openoffice-3.2.0.12.20100210/work/ooo-build-0280fcc31445f7a443717caeb61b1dbe91ae2608/build/OOO320_m12/udkapi/com/sun/star/uno
Entering /home/portage/portage/app-office/openoffice-3.2.0.12.20100210/work/ooo-build-0280fcc31445f7a443717caeb61b1dbe91ae2608/build/OOO320_m12/udkapi/com/sun/star/bridge
mkout -- version: 1.8
mkout -- version: 1.8
|
There was no error in dmesg nor the log. It was hang there and do nothing more. |
|
| Back to top |
|
 |
kernelOfTruth Watchman


Joined: 20 Dec 2005 Posts: 5314 Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
 |
cruzki123 Apprentice

Joined: 16 May 2008 Posts: 193
|
Posted: Sun Feb 14, 2010 7:47 pm Post subject: |
|
|
well If I quit jemalloc use flag, it compiles fine so...
No, there isn't any massege in dmesg nor in the log.
EDIT:
Ussing FEATURES="-sandbox -usersandbox" I obtain:
| Code: |
Entering /home/portage/portage/app-office/openoffice-3.2.0.12.20100210/work/ooo-build-0280fcc31445f7a443717caeb61b1dbe91ae2608/build/OOO320_m12/udkapi/com/sun/star/uno
Entering /home/portage/portage/app-office/openoffice-3.2.0.12.20100210/work/ooo-build-0280fcc31445f7a443717caeb61b1dbe91ae2608/build/OOO320_m12/udkapi/com/sun/star/bridge
mkout -- version: 1.8
mkout -- version: 1.8
/bin/bash: line 1: 16908 Segmentation fault LD_LIBRARY_PATH=${LD_LIBRARY_PATH+${LD_LIBRARY_PATH}:}/home/portage/portage/app-office/openoffice-3.2.0.12.20100210/work/ooo-build-0280fcc31445f7a443717caeb61b1dbe91ae2608/build/OOO320_m12/solver/320/unxlngx6.pro/lib /home/portage/portage/app-office/openoffice-3.2.0.12.20100210/work/ooo-build-0280fcc31445f7a443717caeb61b1dbe91ae2608/build/OOO320_m12/solver/320/unxlngx6.pro/bin/idlc @/home/portage/portage/app-office/openoffice-3.2.0.12.20100210/temp/mkbFHbrV
dmake: Error code 139, while making '../../../../unxlngx6.pro/ucr/cssbridge.db'
/bin/bash: line 1: 16910 Segmentation fault LD_LIBRARY_PATH=${LD_LIBRARY_PATH+${LD_LIBRARY_PATH}:}/home/portage/portage/app-office/openoffice-3.2.0.12.20100210/work/ooo-build-0280fcc31445f7a443717caeb61b1dbe91ae2608/build/OOO320_m12/solver/320/unxlngx6.pro/lib /home/portage/portage/app-office/openoffice-3.2.0.12.20100210/work/ooo-build-0280fcc31445f7a443717caeb61b1dbe91ae2608/build/OOO320_m12/solver/320/unxlngx6.pro/bin/idlc @/home/portage/portage/app-office/openoffice-3.2.0.12.20100210/temp/mkneCRcU
dmake: Error code 139, while making '../../../../unxlngx6.pro/ucr/cssuno.db'
|
On other things, I have done the jemalloc ebuild
| Code: |
# Distributed under the terms of the GNU General Public License v2
# $Header: $
EAPI="2"
DESCRIPTION="Alternaive malloc from *BSD"
HOMEPAGE="http://mxr.mozilla.org/mozilla-central/source/memory/jemalloc/"
LICENSE="bsd"
SLOT="0"
KEYWORDS="~amd64"
IUSE=""
DEPEND=""
RDEPEND="${DEPEND}"
RESTRICT="fetch"
pkg_nofetch(){
einfo "Please download:"
einfo " - jemalloc.c"
einfo " - jemalloc.h"
einfo " - rb.h"
einfo "from ${HOMEPAGE} and place them in ${DISTDIR}"
}
src_unpack() {
mkdir ${S}
cp /home/portage/distfiles/{jemalloc.c,jemalloc.h,rb.h} ${S}
}
src_compile() {
cd ${S}
gcc -shared -DMOZ_MEMORY -DMOZ_MEMORY_LINUX ${CFLAGS} -fPIC -pthread ${LDFLAGS} jemalloc.c -o libjemalloc.so || die
}
src_install() {
mkdir -p "${D}/usr/lib64/"
cp "${S}/libjemalloc.so" "${D}/usr/lib64/libjemalloc.so" || die "Install failed!"
}
pkg_postinst(){
strip -s /usr/lib64/libjemalloc.so
}
|
I don't like the the restrict fetch, but in some sense this is a "svn" build. Also, I don't like the /usr/lib64/ line in src_install. I suspect I can put an if and depending of the arch put lib or lib64.
Fell free to use or put in the overlay  |
|
| Back to top |
|
 |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
Posted: Sun Feb 14, 2010 11:06 pm Post subject: |
|
|
okay, improved it a bit. dev-libs/jemalloc be welcomed. old one without segfaults hopefully.
| Quote: | | >>> Installing (1 of 1) dev-libs/jemalloc-0_p20091102 | anyone may send patches to improve, thanks!
edit #1
and down to dev-libs/jemalloc-0_p20080620. this one works.
check ebuild. quite easy to test new mercurial revisions.
take care, at some point q?.h files were added!
o, I added '-DMALLOC_DECOMMIT' to test.
have fun!  _________________ LibreOffice|split-boost|trans-follow xcb port|instruction set analyzer
_________________
sanity . . . is not with me |
|
| Back to top |
|
 |
geki Advocate


Joined: 13 May 2004 Posts: 2057 Location: Germania
|
|
| Back to top |
|
 |
albright Veteran


Joined: 16 Nov 2003 Posts: 1705 Location: Near Toronto
|
Posted: Fri Feb 19, 2010 5:47 pm Post subject: |
|
|
what is, and what are the advantages of, this jemalloc thing?
tia, _________________ .... there is nothing - absolutely nothing - half so much worth
doing as simply messing about with Linux ...
(apologies to Kenneth Graeme) |
|
| Back to top |
|
 |
|