KDE4 Overlay (Part 5)

Message

Berniyh · Post by **Berniyh** » Sat Apr 26, 2008 9:49 am

tranquilcool wrote:inking CXX executable testsh
/usr/kde/svn/lib/libkio.so.5: undefined reference to `Strigi::AnalysisResult::AnalysisResult(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, long, Strigi::IndexWriter&, Strigi::StreamAnalyzer&, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'
collect2: ld returned 1 exit status
make[2]: *** [ksmserver/tests/testsh] Error 1
make[1]: *** [ksmserver/tests/CMakeFiles/testsh.dir/all] Error 2
make: *** [all] Error 2

i have never been able to install ksmserver because of the above errors.
i have tried all strigi versions but the problem remains the same.
i don't know if anybody has had this same problem and any
kind of solution.

Did you recompile in the correct order? (soprano --> kdelibs --> kdepimlibs --> kdebase-stuff --> rest)
If in doubt, try

Code: Select all

paludis -i1 --dl-reinstall-targets always kdebase-scm

wrc1944 · Post by **wrc1944** » Sat Apr 26, 2008 2:13 pm

On a "paludis -i sys-apps/paludis" (part of a paludis -i world update) I got this failure after about a 2 hr. compile of sys-apps/paludis. Previously in the compile, i noticed no other errors- it appeared to go fine up until the final install. Any ideas/fixes, or can I just continue using the version I currently have installed (0.26.0_pre4)? BTW, why does this package take so long to compile, considering it's small size (1.4Mb tarball)- all the tests it runs?

Code: Select all

/usr/bin/make  check-TESTS                                                                                    
make[4]: Entering directory `/var/tmp/paludis/sys-apps-paludis-0.26.0/work/paludis-0.26.0/hooks'              
>>> setup for test ./eselect_env_update_TEST                                                                  
Test program ./eselect_env_update_TEST:                                                                       
* eselect_env_update: .. OK                                                                                   
>>> cleanup for test ./eselect_env_update_TEST                                                                
PASS: eselect_env_update_TEST                                                                                 
>>> setup for test ./news_TEST                                                                                
Test program ./news_TEST:                                                                                     
* news_none: .. OK                                                                                            
* news_one:                                                                                                   
 * No unread news items found                                                                                 
.!{retcode: 1} FAIL                                                                                           
>>> cleanup for test ./news_TEST                                                                              
FAIL: news_TEST                                                                                               
===================                                                                                           
1 of 2 tests failed                                                                                           
===================                                                                                           
make[4]: *** [check-TESTS] Error 1                                                                            
make[4]: Leaving directory `/var/tmp/paludis/sys-apps-paludis-0.26.0/work/paludis-0.26.0/hooks'               
make[3]: *** [check-am] Error 2                                                                               
make[3]: Leaving directory `/var/tmp/paludis/sys-apps-paludis-0.26.0/work/paludis-0.26.0/hooks'               
make[2]: *** [check-recursive] Error 1                                                                        
make[2]: Leaving directory `/var/tmp/paludis/sys-apps-paludis-0.26.0/work/paludis-0.26.0/hooks'               
make[1]: *** [check-recursive] Error 1                                                                        
make[1]: Leaving directory `/var/tmp/paludis/sys-apps-paludis-0.26.0/work/paludis-0.26.0'                     
make: *** [check] Error 2                                                                                     

!!! ERROR in sys-apps/paludis-0.26.0:
!!! In src_test at line 3727         
!!! Make check failed                

!!! Call stack:
!!!    * src_test (/var/tmp/paludis/sys-apps-paludis-0.26.0/temp/loadsaveenv:3727)
!!!    * ebuild_f_test (/usr/libexec/paludis/0/src_test.bash:69)                  
!!!    * ebuild_main (/usr/libexec/paludis/ebuild.bash:460)                       
!!!    * main (/usr/libexec/paludis/ebuild.bash:479)                              


!!! This package failed inside the test phase. You should read
!!!    http://paludis.pioto.org/faq/stricter.html#testfailures
!!! for more information on packages with test phase failures.


diefunc: making ebuild PID 15918 exit with error
die trap: exiting with error.

Install error:
  * In program paludis (--log-level warning) -i sys-apps/paludis:
  * When performing install action from command line:
  * When executing install task:
  * When installing 'sys-apps/paludis-0.26.0:0::paludis-overlay':
  * Install error: Install failed for 'sys-apps/paludis-0.26.0:0::paludis-overlay'



Summary of failures:

* sys-apps/paludis-0.26.0:0::paludis-overlay: failure

Total: 1 package, 0 successes, 0 skipped, 1 failure, 0 unreached

Resume command: paludis --log-level warning --install --add-to-world-spec 'sys-apps/paludis' --serialised 0.25 'package;=sys-apps/paludis-0.26.0:0::paludis-overlay;installed;has_all_deps;P'

 * You have 2 unread news items (use 'eselect news' to read)

 * Searching for configuration files requiring action...
 * Found 4 files in /usr/kde/svn/share/config
 * Found files in 1 directories
 * Your action is required

gentoo-audio rob #

Philantrop · Post by **Philantrop** » Sat Apr 26, 2008 2:50 pm

wrc1944 wrote:On a "paludis -i sys-apps/paludis" (part of a paludis -i world update) I got this failure after

Please file a bug for this at http://paludis.pioto.org/trac/. This thread is not likely to be read by the Paludis developers.

Of course, this is true for any Paludis bug you might encounter. We can't really help you with such stuff here.

wrc1944 · Post by **wrc1944** » Sat Apr 26, 2008 4:32 pm

Just saw on dirk/dashboard that kdepimlibs was building as of 801365 (assumed that meant the akonadi problem was fixed), so I went ahead and started a complete kde-scm rebuild. Lots of plasma/panel etc. stuff had many updates- so I thought things might be good again.

However, after getting through soprano and kdelibs 801400 fine, kdepimlibs 801426 and 801428 still fails with

Code: Select all

 CMake Error: Could not find Akonadi includes.                                                                 
-- Configuring done                                                                                           

!!! ERROR in kde-base/kdepimlibs-scm:
!!! In cmake-utils_src_configureout at line 1058
!!! Cmake failed                                

!!! Call stack:
!!!    * cmake-utils_src_configureout (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:1058)
!!!    * kde4-base_src_configure (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3052)     
!!!    * kde4-base_src_compile (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3032)       
!!!    * src_compile (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3904)                 
!!!    * ebuild_f_compile (/usr/libexec/paludis/1/src_compile.bash:49)                                
!!!    * ebuild_main (/usr/libexec/paludis/ebuild.bash:460)                                           
!!!    * main (/usr/libexec/paludis/ebuild.bash:479)                                                  

diefunc: making ebuild PID 10440 exit with error
die trap: exiting with error.

Is it recommended to try and build with kdepimlibs r801365, or just wait on another fix to appear? Or, has anyone got through this yet, and have a fix I can use locally?

Berniyh · Post by **Berniyh** » Sat Apr 26, 2008 4:53 pm

wrc1944 wrote:Just saw on dirk/dashboard that kdepimlibs was building as of 801365 (assumed that meant the akonadi problem was fixed), so I went ahead and started a complete kde-scm rebuild. Lots of plasma/panel etc. stuff had many updates- so I thought things might be good again.

However, after getting through soprano and kdelibs 801400 fine, kdepimlibs 801426 and 801428 still fails with
Code: Select all
 CMake Error: Could not find Akonadi includes.                                                                 
-- Configuring done                                                                                           

!!! ERROR in kde-base/kdepimlibs-scm:
!!! In cmake-utils_src_configureout at line 1058
!!! Cmake failed                                

!!! Call stack:
!!!    * cmake-utils_src_configureout (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:1058)
!!!    * kde4-base_src_configure (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3052)     
!!!    * kde4-base_src_compile (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3032)       
!!!    * src_compile (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3904)                 
!!!    * ebuild_f_compile (/usr/libexec/paludis/1/src_compile.bash:49)                                
!!!    * ebuild_main (/usr/libexec/paludis/ebuild.bash:460)                                           
!!!    * main (/usr/libexec/paludis/ebuild.bash:479)                                                  

diefunc: making ebuild PID 10440 exit with error
die trap: exiting with error.
Is it recommended to try and build with kdepimlibs r801365, or just wait on another fix to appear? Or, has anyone got through this yet, and have a fix I can use locally?

You'll have to wait until kdesupport/akonadi has been packaged.

wrc1944 · Post by **wrc1944** » Sat Apr 26, 2008 5:13 pm

Thanks Berniyh,
I just tried 801365 with export ESVN_REVISION=801365 and got the same error- I'll keep checking and try again when it's packaged.

phoenix_me · Post by **phoenix_me** » Sat Apr 26, 2008 6:40 pm

wrc1944 wrote:On a "paludis -i sys-apps/paludis" (part of a paludis -i world update) I got this failure after about a 2 hr. compile of sys-apps/paludis.

I don't know why it takes 2hr mine need less than 10 minutes to compile paludis.

Code: Select all

time paludis -ip paludis
[...cut...]
Updating world file

* adding sys-apps/paludis

 * Checking whether the GNU info directory needs updating...
 * No updates needed

 * No unread news items found

 * Searching for configuration files requiring action...
 * No configuration file updates required

real 587.38
user 819.68
sys 132.97

and some cpu info:

Code: Select all

cat /proc/cpuinfo 
processor       : 0          
vendor_id       : GenuineIntel
cpu family      : 6           
model           : 15          
model name      : Intel(R) Core(TM)2 CPU          6300  @ 1.86GHz
stepping        : 6                                              
cpu MHz         : 3010.248                                       
cache size      : 2048 KB                                        
physical id     : 0                                              
siblings        : 2                                              
core id         : 0                                              
cpu cores       : 2                                              
fdiv_bug        : no                                             
hlt_bug         : no                                             
f00f_bug        : no                                             
coma_bug        : no                                             
fpu             : yes                                            
fpu_exception   : yes                                            
cpuid level     : 10                                             
wp              : yes                                            
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc arch_perfmon pebs bts pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr lahf_lm                                                                                              
bogomips        : 6023.39                                                                                    
clflush size    : 64                                                                                         

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6           
model           : 15          
model name      : Intel(R) Core(TM)2 CPU          6300  @ 1.86GHz
stepping        : 6
cpu MHz         : 3010.248
cache size      : 2048 KB
physical id     : 0
siblings        : 2
core id         : 1
cpu cores       : 2
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc arch_perfmon pebs bts pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr lahf_lm
bogomips        : 6020.29
clflush size    : 64

wrc1944 · Post by **wrc1944** » Sat Apr 26, 2008 7:06 pm

phoenix_me wrote:

I don't know why it takes 2hr mine need less than 10 minutes to compile paludis.

I know your Intel Core 2 6300 is considerable faster than an athlon-xp 3000, but 12 times as fast (120 min. vs 10min.)? That's hard to believe, and I know quite a lot about processors, etc., as I've built many computers, doing much research on componets.

When you built paludis, did you notice if there were lots of tests running all along during the compile? I had many periods of what seemed like hundreds and hundreds of endless non-stop testing routines.

creidiki · Post by **creidiki** » Sat Apr 26, 2008 7:14 pm

use --checks none

wrc1944 · Post by **wrc1944** » Sat Apr 26, 2008 9:22 pm

creidiki,
Sounds like that would solve it, but didn't I read somewhere in the docs or wiki that disabling checks was not a good idea?

phoenix_me · Post by **phoenix_me** » Sat Apr 26, 2008 11:21 pm

wrc1944 wrote:phoenix_me wrote:
I don't know why it takes 2hr mine need less than 10 minutes to compile paludis.
I know your Intel Core 2 6300 is considerable faster than an athlon-xp 3000, but 12 times as fast (120 min. vs 10min.)? That's hard to believe, and I know quite a lot about processors, etc., as I've built many computers, doing much research on componets.

When you built paludis, did you notice if there were lots of tests running all along during the compile? I had many periods of what seemed like hundreds and hundreds of endless non-stop testing routines.

I was measure compilation time using time command, the result of it is in my last post [real 587.38 sec].

I can paste the build log if you want to check for testing routines. (I don't know if there was many of those as I was in kitchen while compilation)

What's the clock of your athlon? About 2GHz? mine is 1,8@3Ghz per core = 6Ghz but this is only 3 times faster. Let's look for something more... my GCC-4.3.0 which is first that have -march=core2, what else - sometimes I am using a little trick to speed-up compilation time. I am moving my build directory to RAM.

Code: Select all

none                    /var/tmp/paludis tmpfs          nr_inodes=1M,size=2000M

but I don't know what's the impact in time for paludis compilation (will try tommorow)
The USE flags can be reponsible for extended compilation time mine for paludis are like below:

Code: Select all

paludis -ip paludis
Building target list...
Building dependency list...

These packages will be installed:

* sys-apps/paludis [R 0.26.0] <target>
    Reasons: app-admin/python-updater-0.2:0::installed
    bash-completion -doc glsa inquisitio -pink -portage python -qa -ruby -vim-syntax visibility -zsh-completion

Total: 1 package (1 rebuild)

 * No unread news items found

My compilator FLAGS are:

Code: Select all

cat /etc/paludis/bashrc 
# bashrc                           
# This file created by portage2paludis.bash
CHOST="i686-pc-linux-gnu"                  

CFLAGS="-march=core2 -O2 -fomit-frame-pointer -DNDEBUG"

CXXFLAGS="${CFLAGS}"
LDFLAGS="-Wl,-O1"
MAKEOPTS="-j3"

if [[ "${CATEGORY}/${PN}" == "app-crypt/truecrypt" ]] ; then
    CFLAGS="-march=native -O2 -fomit-frame-pointer -DNDEBUG"
    CXXFLAGS="${CFLAGS}"
fi

All this together can be your ally or enemy regarding compilation time.

wrc1944 · Post by **wrc1944** » Sun Apr 27, 2008 12:48 am

phoenix_me,
I really appreciate your detailed replies- thanks much- really good info

I was mistaken about the cpu- I forgot I wasn't on my main box, but on a testing box, where the cpu is only an athlon-xp 1700 which runs at about 1.5Gb. It's noticably slower than my other athlon-xp 3000 box.

I think having 2 cores with a more modern architecture and that MAKEOPTS="-j3" option makes a huge difference in compiling time. Not to mention having the build directory in RAM- I knew about that trick but have never tried it, as I didn't think I've ever had enough installed to manage it.

I'm curious about that -DNDEBUG cflag. I looked in man:gcc (4.2.3 on the testing box) and couldn't find a reference. Will investigate further.

Don't bother about pasting your log file- maybe just notice if you have those testing routines, or not.
BTW, did you get the new version 0.26.0 successfully installed- mine failed (above) at the end. If so, I must have something wrong on my paludis or system setup.

tranquilcool · Post by **tranquilcool** » Sun Apr 27, 2008 6:00 am

Berniyh wrote:
tranquilcool wrote:inking CXX executable testsh
/usr/kde/svn/lib/libkio.so.5: undefined reference to `Strigi::AnalysisResult::AnalysisResult(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, long, Strigi::IndexWriter&, Strigi::StreamAnalyzer&, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'
collect2: ld returned 1 exit status
make[2]: *** [ksmserver/tests/testsh] Error 1
make[1]: *** [ksmserver/tests/CMakeFiles/testsh.dir/all] Error 2
make: *** [all] Error 2

i have never been able to install ksmserver because of the above errors.
i have tried all strigi versions but the problem remains the same.
i don't know if anybody has had this same problem and any
kind of solution.
Did you recompile in the correct order? (soprano --> kdelibs --> kdepimlibs --> kdebase-stuff --> rest)
If in doubt, try
Code: Select all
paludis -i1 --dl-reinstall-targets always kdebase-scm

yes i have tried in that order but all the same.
obviously if i use --checks none it installs but then i have series of other
problems with the strigi thing.

titoucha · Post by **titoucha** » Sun Apr 27, 2008 10:50 am

Hi, i have one problem with my panel, it's not possible to save the config.
I start kde, the panel it's too small, i configure it and restart kde, never change it's small.
I don't find wher is the file for change the config.

Sorry for my poor english.

Berniyh · Post by **Berniyh** » Sun Apr 27, 2008 11:40 am

titoucha wrote:Hi, i have one problem with my panel, it's not possible to save the config.
I start kde, the panel it's too small, i configure it and restart kde, never change it's small.
I don't find wher is the file for change the config.

Sorry for my poor english.

You'll find the config file here:

Code: Select all

~/.kdesvn/share/config/plasma-appletsrc

That plasma doesn't save the config file is an upstream problem. You'll have to wait until they fix that.

titoucha · Post by **titoucha** » Sun Apr 27, 2008 12:13 pm

Thanks a lot

Boblenain · Post by **Boblenain** » Sun Apr 27, 2008 1:41 pm

Is there a thread of an rss, or anything in order to know in quite real-time what is broken in trunk ?

beso_1717 · Post by **beso_1717** » Sun Apr 27, 2008 1:57 pm

Berniyh wrote:
wrc1944 wrote:Just saw on dirk/dashboard that kdepimlibs was building as of 801365 (assumed that meant the akonadi problem was fixed), so I went ahead and started a complete kde-scm rebuild. Lots of plasma/panel etc. stuff had many updates- so I thought things might be good again.

However, after getting through soprano and kdelibs 801400 fine, kdepimlibs 801426 and 801428 still fails with
Code: Select all
 CMake Error: Could not find Akonadi includes.                                                                 
-- Configuring done                                                                                           

!!! ERROR in kde-base/kdepimlibs-scm:
!!! In cmake-utils_src_configureout at line 1058
!!! Cmake failed                                

!!! Call stack:
!!!    * cmake-utils_src_configureout (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:1058)
!!!    * kde4-base_src_configure (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3052)     
!!!    * kde4-base_src_compile (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3032)       
!!!    * src_compile (/var/tmp/paludis/kde-base-kdepimlibs-scm/temp/loadsaveenv:3904)                 
!!!    * ebuild_f_compile (/usr/libexec/paludis/1/src_compile.bash:49)                                
!!!    * ebuild_main (/usr/libexec/paludis/ebuild.bash:460)                                           
!!!    * main (/usr/libexec/paludis/ebuild.bash:479)                                                  

diefunc: making ebuild PID 10440 exit with error
die trap: exiting with error.
Is it recommended to try and build with kdepimlibs r801365, or just wait on another fix to appear? Or, has anyone got through this yet, and have a fix I can use locally?
You'll have to wait until kdesupport/akonadi has been packaged.

app-office/akonadi has hit the overlay and now kdepim seems to be able to find out akonadi includes.
i'm only wondering why does qt-phonon need the gstreamer lib and not the xine-lib?!

Berniyh · Post by **Berniyh** » Sun Apr 27, 2008 3:23 pm

beso_1717 wrote:app-office/akonadi has hit the overlay and now kdepim seems to be able to find out akonadi includes.
i'm only wondering why does qt-phonon need the gstreamer lib and not the xine-lib?!

Because qt-phonon atm only contains the backends for gstreamer, quicktime and directsound.
The xine backend still is part of kdelibs/kdebase.

tranquilcool · Post by **tranquilcool** » Sun Apr 27, 2008 4:40 pm

gone panel gone!

no panel at all.
if i create one and try
to add a widget it just crashes.

any help to get back the panel?

creidiki · Post by **creidiki** » Sun Apr 27, 2008 4:44 pm

wrc1944 wrote:creidiki,
Sounds like that would solve it, but didn't I read somewhere in the docs or wiki that disabling checks was not a good idea?

The bad idea is having it on, in gentoo.

beso_1717 · Post by **beso_1717** » Sun Apr 27, 2008 9:29 pm

i don't know if it's me that i'm broken but i seem unable to have plasma workspace to load. i've tried with the 4.0.3 config and plasma loads, but i cannot see the pannel and whenever i try to do something inside plasma it breaks with an error in /lib64/libc.so.6 and with the trace saying: this trace is useless.
now, after the qt-split force upgrade i cannot also use the old kde-4.0.3.
also plasma (801820) cannot be compiled now for the following error:

[ 61%] Built target plasma_engine_powermanagement
[ 61%] Generating plasma_engine_time_automoc.cpp
Generating timeengine.moc
Scanning dependencies of target plasma_engine_time
CMakeFiles/plasma_containment_desktop.dir/backgrounddialog.o: In function `findAllBackgrounds(BackgroundContainer const*, QString const&, float)':
backgrounddialog.cpp:(.text+0x1f97): undefined reference to `Plasma::Package::listInstalled(QString const&)'
collect2: ld returned 1 exit status
make[2]: *** [lib/plasma_containment_desktop.so] Error 1
make[1]: *** [plasma/containments/desktop/CMakeFiles/plasma_containment_desktop.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
[ 61%] [ 62%] Building CXX object plasma/dataengines/time/CMakeFiles/plasma_engine_time.dir/plasma_engine_time_automoc.o
[ 62%] Built target plasma_engine_tasks
Building CXX object plasma/dataengines/time/CMakeFiles/plasma_engine_time.dir/timeengine.o
Linking CXX shared module ../../../lib/plasma_engine_time.so
[ 62%] Built target plasma_engine_places
[ 62%] Built target plasma_engine_time
make: *** [all] Error 2

!!! ERROR in kde-base/plasma-scm:
!!! In cmake-utils_src_make at line 1197
!!! Make failed!

!!! Call stack:
!!! * cmake-utils_src_make (/var/tmp/paludis/kde-base-plasma-scm/temp/loadsaveenv:1197)
!!! * kde4-base_src_make (/var/tmp/paludis/kde-base-plasma-scm/temp/loadsaveenv:3178)
!!! * kde4-meta_src_make (/var/tmp/paludis/kde-base-plasma-scm/temp/loadsaveenv:3449)
!!! * kde4-meta_src_compile (/var/tmp/paludis/kde-base-plasma-scm/temp/loadsaveenv:3403)
!!! * src_compile (/var/tmp/paludis/kde-base-plasma-scm/temp/loadsaveenv:4285)
!!! * ebuild_f_compile (/usr/libexec/paludis/1/src_compile.bash:49)
!!! * ebuild_main (/usr/libexec/paludis/ebuild.bash:460)
!!! * main (/usr/libexec/paludis/ebuild.bash:479)

wrc1944 · Post by **wrc1944** » Sun Apr 27, 2008 10:51 pm

Well, I was all set to rebuild and see some progress, but no such luck.

The rebuild went great- out of 84 packages, only marble failed (complete log at http://pastebin.com/m1e09d705 ). I restarted X, and kde started like before (no panel, taskmanager, etc.), and now had a 3/16" grey and white checkerboard outline completely around the screen.

I then logged into a console, and removed .kdesvn, and rebooted.
Worse.

Now, all I can get is 2-3 of false starts with a checkerboard panel briefly appearing, and an immediate plasma crash dialog on top of a completely light grey screen. If I close the crash dialog, all I have left is a locked grey screen.

This is with r801788-801845 packages. Anyone else have better results, or know what's going on? Do I need a total qt rebuild first, and then again rebuild kde-scm?

Had to boot with gnome to post.....

EDIT: Recompiled libplasma and plasma with r801861, and panel is back, and I can add app launcher icons! Task manager works! Things are looking MUCH better

Berniyh · Post by **Berniyh** » Mon Apr 28, 2008 8:04 am

beso_1717 wrote:i don't know if it's me that i'm broken but i seem unable to have plasma workspace to load. i've tried with the 4.0.3 config and plasma loads, but i cannot see the pannel and whenever i try to do something inside plasma it breaks with an error in /lib64/libc.so.6 and with the trace saying: this trace is useless.
now, after the qt-split force upgrade i cannot also use the old kde-4.0.3.
also plasma (801820) cannot be compiled now for the following error::-(

I guess, that you didn't at least recompile libplasma before you recompiled plasma.
If that doesn't help try kdelibs -> libplasma -> plasma.
(And if you are at it, recompile the rest of kdebase, too

)

Berniyh · Post by **Berniyh** » Mon Apr 28, 2008 8:07 am

wrc1944 wrote:The rebuild went great- out of 84 packages, only marble failed (complete log at http://pastebin.com/m1e09d705 ). I restarted X, and kde started like before (no panel, taskmanager, etc.), and now had a 3/16" grey and white checkerboard outline completely around the screen.

This is the error, looks like some stuff upstream:

Code: Select all

/var/tmp/paludis/kde-base-marble-scm/work/marble/marble/src/lib/TileCreator.cpp:389:27: error: TileCreator.moc: No such file or directory