Calligra just became buildable again. There's just a handful of patches needed when building with Solaris Studio 12.3 and most of them are already applied in upstream, thanks to helpful Calligra developers.
The updated spec is available in the kde4-solaris repo.
Friday, December 23, 2011
Tuesday, November 1, 2011
Qt autotests - SFE vs KDE4
Qt comes with an extensive number of self tests and it's interesting to run them from time to time. In the FOSSqt.spec file of the kde4-solaris project, they can be turned on by defining run_autotests to 1, similarly in the SFEqt-gpp.spec of our SFE brethren.
A great number of them is failing on Solaris for various reasons - be it compiler issues, code bugs, or anything else, there's simply too many failures to demotivate anyone from trying to fix it.
So I thought it might be more interesting to compare the results from our build and SFE's - that way the difference would be more or less based on compiler differences (SFE uses gcc) and not anything else.
So here comes the comparison of the results for Qt 4.7.4:
Our Qt ran 491 tests, SFE's 489. There were 77 differences.
17 of the differences were irrelevant (difference in pointer address, test duration, slightly different messages, etc.). This includes tst_bic, tst_gestures, tst_maketestselftest, tst_modeltest, tst_q_func_info, tst_qcontiguouscache, tst_qdbusperformance, tst_qelapsedtimer, tst_qgraphicsitem, tst_qmainwindow, tst_qmenu, tst_qmutex, tst_qobjectperformance, tst_qscriptjstestsuite, tst_qscriptv8testsuite,
36 cases where we fail and SFE passes: tst_collections,
tst_compiler (mere prejudice, we should try to override the stupid #if defined(Q_CC_SUN) there),
tst_mediaobject (something's wrong with our phonon),
tst_q3sqlcursor, tst_q3sqlselectcursor, tst_qitemmodel, tst_qsql, tst_qsqldatabase, tst_qsqldriver, tst_qsqlquery, tst_qsqlquerymodel, tst_qsqlrelationaltablemodel, tst_qsqltablemodel, tst_qsqlthread, (hm, sqlite not set up properly),
tst_qaccessibility (a big fail here),
tst_qapplication (may be caused just by some difference in how the tests were run),
tst_qdatetime, tst_qdatetimeedit (caused by different environment when running the test - need to run it in the CET timezone),
tst_qdom, tst_qtextcodec, tst_qxmlsimplereader (mmm, asian encodings!),
tst_qgraphicsscene (this one is weird),
tst_qicoimageformat, tst_qicon, tst_qimage, tst_qimagereader, tst_qimagewriter, tst_qpixmap (deserves investigation, image format support),
tst_qmake (weird, need to have a look - perhaps bad mkspec used?)
tst_qmovie
tst_qnetworkconfiguration, tst_qnetworkconfigurationmanager, tst_qnetworksession (!)
tst_qwebpage
tst_uic, tst_uic3 (likely QTBUG-21523)
2 tests that SFE builds but we don't: tst_qdbusmarshall (QTBUG-22106), tst_qtconcurrentmap (QTBUG-12987)
4 tests that we build but SFE doesn't: tst_guiapplauncher, tst_qglthreads, tst_qwidget_window, tst_qwidget
20 tests where SFE fails while we pass:
tst_qabstractslider
tst_qcolumnview
tst_qcompleter
tst_qdialog
tst_qeventloop
tst_qfilesystemmodel
tst_qgl
tst_qglbuffer
tst_qgraphicsproxywidget
tst_qgraphicstransform
tst_qgraphicsview (not really a pass, but a few more tests are run in our build)
tst_qmdiarea
tst_qmessagebox (X Error: BadWindow)
tst_qprocess
tst_qreadwritelock
tst_qscriptengine
tst_qstring (no wide character support in the g++ build? Strange ... )
tst_qthread
tst_qtimeline
tst_xmlpatternsdiagnosticsts
It looks like the tests that included some performance measurements (tst_qcontiguouscache, tst_qdbusperformance, tst_qmutex, tst_qprocess) show better numbers for our bits, but I'd be far from interpreting it in that way. Both tests were run in a VirtualBox environment and the load on the host was quite different on the two occasions.
Webrev of the results (to save space only the udiff diffs were preserved)
The results bzipped
The tests source for reference
A great number of them is failing on Solaris for various reasons - be it compiler issues, code bugs, or anything else, there's simply too many failures to demotivate anyone from trying to fix it.
So I thought it might be more interesting to compare the results from our build and SFE's - that way the difference would be more or less based on compiler differences (SFE uses gcc) and not anything else.
So here comes the comparison of the results for Qt 4.7.4:
Our Qt ran 491 tests, SFE's 489. There were 77 differences.
17 of the differences were irrelevant (difference in pointer address, test duration, slightly different messages, etc.). This includes tst_bic, tst_gestures, tst_maketestselftest, tst_modeltest, tst_q_func_info, tst_qcontiguouscache, tst_qdbusperformance, tst_qelapsedtimer, tst_qgraphicsitem, tst_qmainwindow, tst_qmenu, tst_qmutex, tst_qobjectperformance, tst_qscriptjstestsuite, tst_qscriptv8testsuite,
36 cases where we fail and SFE passes: tst_collections,
tst_compiler (mere prejudice, we should try to override the stupid #if defined(Q_CC_SUN) there),
tst_mediaobject (something's wrong with our phonon),
tst_q3sqlcursor, tst_q3sqlselectcursor, tst_qitemmodel, tst_qsql, tst_qsqldatabase, tst_qsqldriver, tst_qsqlquery, tst_qsqlquerymodel, tst_qsqlrelationaltablemodel, tst_qsqltablemodel, tst_qsqlthread, (hm, sqlite not set up properly),
tst_qaccessibility (a big fail here),
tst_qapplication (may be caused just by some difference in how the tests were run),
tst_qdatetime, tst_qdatetimeedit (caused by different environment when running the test - need to run it in the CET timezone),
tst_qdom, tst_qtextcodec, tst_qxmlsimplereader (mmm, asian encodings!),
tst_qgraphicsscene (this one is weird),
tst_qicoimageformat, tst_qicon, tst_qimage, tst_qimagereader, tst_qimagewriter, tst_qpixmap (deserves investigation, image format support),
tst_qmake (weird, need to have a look - perhaps bad mkspec used?)
tst_qmovie
tst_qnetworkconfiguration, tst_qnetworkconfigurationmanager, tst_qnetworksession (!)
tst_qwebpage
tst_uic, tst_uic3 (likely QTBUG-21523)
2 tests that SFE builds but we don't: tst_qdbusmarshall (QTBUG-22106), tst_qtconcurrentmap (QTBUG-12987)
4 tests that we build but SFE doesn't: tst_guiapplauncher, tst_qglthreads, tst_qwidget_window, tst_qwidget
20 tests where SFE fails while we pass:
tst_qabstractslider
tst_qcolumnview
tst_qcompleter
tst_qdialog
tst_qeventloop
tst_qfilesystemmodel
tst_qgl
tst_qglbuffer
tst_qgraphicsproxywidget
tst_qgraphicstransform
tst_qgraphicsview (not really a pass, but a few more tests are run in our build)
tst_qmdiarea
tst_qmessagebox (X Error: BadWindow)
tst_qprocess
tst_qreadwritelock
tst_qscriptengine
tst_qstring (no wide character support in the g++ build? Strange ... )
tst_qthread
tst_qtimeline
tst_xmlpatternsdiagnosticsts
It looks like the tests that included some performance measurements (tst_qcontiguouscache, tst_qdbusperformance, tst_qmutex, tst_qprocess) show better numbers for our bits, but I'd be far from interpreting it in that way. Both tests were run in a VirtualBox environment and the load on the host was quite different on the two occasions.
Webrev of the results (to save space only the udiff diffs were preserved)
The results bzipped
The tests source for reference
Thursday, September 1, 2011
KDE 4.7.0 with Solaris Studio 12.3 beta
KDEpim, the last package of the KDE 4.7.0 suite I care for (bindings never really worked and I've given up hope there) just built using the Solaris Studio 12.3 beta compiler.
Unfortunately, just as with Studio 12.1, one cannot login into the KDE session, it freezes somewhere in the splash screen, but the apps seem to be running fine when launched from GNOME.
Unfortunately, just as with Studio 12.1, one cannot login into the KDE session, it freezes somewhere in the splash screen, but the apps seem to be running fine when launched from GNOME.
Tuesday, August 9, 2011
4W LED bulb from DX
I got my hands at the 4W LED bulb from DealExtreme.
I put it to a directional reading lamp that normally hosts a 60W incandescent and the result was not really satisfying.
While the reviews at DX (most of them apparently written by the seller) say the light output is the same as a 40W incandescent, this is simply not true. Unfortunately I don't have any 25W nor 40W bulbs to compare, but I don't doubt a 25W one would still outshine it.
With the LED bulb the lamp can really be only used for reading and nothing more, but it's good enough at that (if you're close enough). Another plus is that it fits well in the lamp - a fluorescent compact bulb would be unacceptable here and there's no heat and no risk of broken glass in case of an accident.
Below are pictures of a white wall taken with Canon A85 in manual mode, using the same white balance, aperture and exposure values. (The spot in the top right is a defect that developed in the past year or so, I just don't have the A590 at hand ATM)
As a ghetto scientific method of light output measurement I also took a look at the histograms that GIMP kindly provides. It looks like the LED bulb is very weak in the blue part of the spectrum and the median value 2.5x higher (likely even more as part of the red channel is burned out in the 60W one) for the incandescent supports my estimate that the light output is at best comparable to a 25W incandescent.
I put it to a directional reading lamp that normally hosts a 60W incandescent and the result was not really satisfying.
While the reviews at DX (most of them apparently written by the seller) say the light output is the same as a 40W incandescent, this is simply not true. Unfortunately I don't have any 25W nor 40W bulbs to compare, but I don't doubt a 25W one would still outshine it.
With the LED bulb the lamp can really be only used for reading and nothing more, but it's good enough at that (if you're close enough). Another plus is that it fits well in the lamp - a fluorescent compact bulb would be unacceptable here and there's no heat and no risk of broken glass in case of an accident.
Below are pictures of a white wall taken with Canon A85 in manual mode, using the same white balance, aperture and exposure values. (The spot in the top right is a defect that developed in the past year or so, I just don't have the A590 at hand ATM)
As a ghetto scientific method of light output measurement I also took a look at the histograms that GIMP kindly provides. It looks like the LED bulb is very weak in the blue part of the spectrum and the median value 2.5x higher (likely even more as part of the red channel is burned out in the 60W one) for the incandescent supports my estimate that the light output is at best comparable to a 25W incandescent.
4W LED |
60W incancescent |
4W histogram |
60W histogram |
Saturday, July 16, 2011
Sun Java on Fedora 15
For the next time I need it: https://www.martineve.com/2011/06/08/six-lines-to-get-sun-java-running-on-fedora-15-x64/
because the upstream instructions are hopelessly useless...
because the upstream instructions are hopelessly useless...
Saturday, July 9, 2011
KDE 4.7 RC1
Some screenshots from KDE 4.7 RC1.
I don't have konsole yet as it was split into another package and so far only the bare minimum required for KDE gdm-integration is ready.
I don't have konsole yet as it was split into another package and so far only the bare minimum required for KDE gdm-integration is ready.
Sunday, July 3, 2011
OS upgrade
Over the weekend I've migrated to Fedora 15. I've briefly considered going back to Mandriva, but since Mageia forked off it, bot distros are weaker than the original one. Of course Fedora is not a miracle either, it's nowhere near the level of comfort I was used to from Mandriva and I've already been hit by a bug in the kernel, but at least I can expect every possible piece of software to be available with decent stability.
Thursday, June 23, 2011
KDE 4.6.4 on Solaris
It took slightly longer than I expected (thank you pkg team), but finally, as of today KDE 4.6.4 is buildable from the kde4-solaris project repository.
Upstream lists a number of bugfixes, but on the Solaris side I'm not as happy.
As some people already let me know, you can encounter a plasmatic crash right at startup which prevents the desktop from loading.A workaround is to
# mv /opt/kde4/lib/libkholidays.so.4.6.0 /opt/kde4/lib/libkholidays.so.4.6.0.bak
but you'll lose the clock/calendar applet :-(
Another inconvenience is a number of warning popups shown when connecting to secured websites in Konqueror. They also start a knotify process that gives out annoying bell and eats a lot of CPU.
Last bad news is the loss of KDEaccessibility. Upstream has introduced hard dependency on speechd, so we can't compile until someone creates a spec for it.
The rest seems to work as usual :-)
Upstream lists a number of bugfixes, but on the Solaris side I'm not as happy.
As some people already let me know, you can encounter a plasmatic crash right at startup which prevents the desktop from loading.A workaround is to
# mv /opt/kde4/lib/libkholidays.so.4.6.0 /opt/kde4/lib/libkholidays.so.4.6.0.bak
but you'll lose the clock/calendar applet :-(
Another inconvenience is a number of warning popups shown when connecting to secured websites in Konqueror. They also start a knotify process that gives out annoying bell and eats a lot of CPU.
Last bad news is the loss of KDEaccessibility. Upstream has introduced hard dependency on speechd, so we can't compile until someone creates a spec for it.
The rest seems to work as usual :-)
Thursday, June 9, 2011
3G
I'm considering getting a Kindle so I was reading some info about it. At the page explaining the difference between 3G and WiFi version I found a link to a map of 3G coverage across the world.
Apart from missing 3G coverage in the Baltic states, which is apparently an error, who do you think is the EU's worst covered country?
Could it be the Czech Republic, where the office regulating telcos spends taxpayer's millions in badly camouflaged telcos' PR campaigns, instead of fighting the cartel that's formed in the market ?
(taken from http://client0.cellmaps.com/viewer.html?cov=1, license unknown)
Apart from missing 3G coverage in the Baltic states, which is apparently an error, who do you think is the EU's worst covered country?
Could it be the Czech Republic, where the office regulating telcos spends taxpayer's millions in badly camouflaged telcos' PR campaigns, instead of fighting the cartel that's formed in the market ?
(taken from http://client0.cellmaps.com/viewer.html?cov=1, license unknown)
Tuesday, May 31, 2011
digikam
One more step closer to usable desktop - I've got Digikam 2.0.0 beta5 running on my laptop.
It works quite well actually, apart from the usual lag associated with anything Qt4 on Solaris. It hasn't crashed during the 1 hour I played with it, so I consider it stable :-).
The only thing I found crashing is showfoto - the crapfest called semantic desktop is showing its ugly face there again, with misbehaving strigi analyzer ruining the show. It's good to know we're not relying on some unmaintained pile of rubbish there.
It works quite well actually, apart from the usual lag associated with anything Qt4 on Solaris. It hasn't crashed during the 1 hour I played with it, so I consider it stable :-).
The only thing I found crashing is showfoto - the crapfest called semantic desktop is showing its ugly face there again, with misbehaving strigi analyzer ruining the show. It's good to know we're not relying on some unmaintained pile of rubbish there.
Sunday, May 22, 2011
Hugin on Solaris
I've just built hugin and made some panorama's!
Finally, I'm a bit closer to the mutimedia richness of Linux (!) on my Solaris 11 desktop :-)
Panoramic backyard of a gym in Hrusice:
The simple build approach I took worked quite well, I just had to workaround the stupid cairo bug in S11 regarding libpng and add a few simple patches here and there.
If anyone's interested, a snapshot of the specs is at bionicmutton. I've yet to decide how to proceed with this project, whether to blend it with kde-solaris or with SFE.
Finally, I'm a bit closer to the mutimedia richness of Linux (!) on my Solaris 11 desktop :-)
Panoramic backyard of a gym in Hrusice:
The simple build approach I took worked quite well, I just had to workaround the stupid cairo bug in S11 regarding libpng and add a few simple patches here and there.
If anyone's interested, a snapshot of the specs is at bionicmutton. I've yet to decide how to proceed with this project, whether to blend it with kde-solaris or with SFE.
Saturday, May 21, 2011
"Security" by Pearson VUE
The whole concept of "Security questions", usually used to recover an online account in case of lost credentials is quite flawed by definition, but what Pearson VUE, the company managing tests and certification (mostly) in IT is doing is beyond my comprehension.
Have a look at the below screenshot.
And their Security tips do not make it any better - they limit the length of the answer and make it caps insensitive. And asking me not to tell my spouse's nick name to anyone ???
Have a look at the below screenshot.
Pearson VUE is serious about account safety. That’s why we’re requiring all web users to create security questions. Select two questions below, one from each drop-down menu. Then type your answers in the provided fields and click Next.Doesn't sound THAT bad, does it? But check the preselected questions:
We’ve designed questions that would be hard to guess.
Who is your favorite actor, musician, or artist?
What was your favorite place to visit as a child?
What is your favorite song?
What is the title of your favorite book?
What was your most memorable gift as a child?
What was the name of your first toy animal?
Who is your favorite athlete?
What is the first and last name of the best man at your wedding?Sounds like exactly the type of questions that have answers on one's Facebook or blog page, doesn't it?
What is the name of the first company for which you worked?
What is your dream occupation?
What is your oldest sibling's (brother or sister) nickname?
What is your favorite food?
In what city did you first meet your spouse or significant other?
What is your spouse or significant other's nickname?
And their Security tips do not make it any better - they limit the length of the answer and make it caps insensitive. And asking me not to tell my spouse's nick name to anyone ???
* Answers may contain no more than 50 characters.Combine this with the way they manage their site's SSL certificates ... :
* Answers are NOT case sensitive (caps or no caps are allowed).
* Make sure the answer is not related to your username or password in any way.
* Do not tell anyone this information.
* Do not send this information by email.
* Change this information periodically in Change Sign In under My Account.
Sunday, May 15, 2011
building hugin: error: expected class-name before '{' token
I've got the below error when trying to build hugin. As it turned out my wxWidgets were compiled without Drag&Drop support, even when I supplier --enable-dnd to the configure script.
It's because --with-x11 does not support it.
To find out if dnd is really enabled use
$ grep wxUSE_DRAG_AND_DROP /pat/to/build/dir/lib/wx/include/*/wx/setup.h
#define wxUSE_DRAG_AND_DROP 1
[ 64%] Building CXX object src/hugin1/icpfind/CMakeFiles/icpfindlib.dir/AutoCtrlPointCreator.cpp.o
cd /home/test/packages/BUILD/hugin-2010.4.0/build/src/hugin1/icpfind && /usr/gcc/4.5/bin/g++ -Dicpfindlib_EXPORTS -D_FILE_OFFSET_BITS=64 -D_LARGE_FILES -D__WXUNIVERSAL__ -D__WXX11__ -D_REENTRANT -pthreads -O2 -g -fPIC -I/home/test/packages/BUILD/hugin-2010.4.0/build/src -I/home/test/packages/BUILD/hugin-2010.4.0/src/hugin_base -I/home/test/packages/BUILD/hugin-2010.4.0/src/foreign -I/home/test/packages/BUILD/hugin-2010.4.0/src/foreign/vigra -I/home/test/packages/BUILD/hugin-2010.4.0/src/celeste -I/home/test/packages/BUILD/hugin-2010.4.0/build/src/celeste -I/home/test/packages/BUILD/hugin-2010.4.0/src -I/opt/kde4/include/OpenEXR -I/opt/kde4/include -isystem /opt/kde4/lib/wx/include/x11univ-unicode-release-2.8 -isystem /opt/kde4/include/wx-2.8 -I/home/test/packages/BUILD/hugin-2010.4.0/src/hugin1 -o CMakeFiles/icpfindlib.dir/AutoCtrlPointCreator.cpp.o -c /home/test/packages/BUILD/hugin-2010.4.0/src/hugin1/icpfind/AutoCtrlPointCreator.cpp
In file included from /home/test/packages/BUILD/hugin-2010.4.0/src/hugin1/hugin/huginApp.h:30:0,
from /home/test/packages/BUILD/hugin-2010.4.0/src/hugin1/icpfind/AutoCtrlPointCreator.cpp:38:
/home/test/packages/BUILD/hugin-2010.4.0/src/hugin1/hugin/MainFrame.h:64:1: error: expected class-name before '{' token
Sunday, May 1, 2011
KDE 4.6.2 on Solaris 11, with gcc
Since some interesting KDE/Qt applications are not Sun/Oracle Studio friendly I've been pondering the use of gcc for some time already.
And here it is:
The lack of icons is due to Solaris bug (an older libpng version used stupidly by cairo) that I did not try to workaround.
I used SFEgcc from pkg.opensolaris.cz.
I ditched the idea of reusing the kde-solaris workspace and for the moment built all packages with mere
./configure --prefix=/opt/kde4 && gmake && gmake install
or
PKG_CONFIG_PATH=/opt/kde4/lib/pkgconfig:/usr/lib/pkgconfig cmake .. -DCMAKE_INSTALL_PREFIX=/opt/kde4 && gmake && gmake install
No crazy mesh of *FLAGS, no nothing, just the defaults.
Only about 5 patches were needed for to have gdm-integration complete.
Unfortunately the session feels about as sluggish as when built with Studio.
And here it is:
The lack of icons is due to Solaris bug (an older libpng version used stupidly by cairo) that I did not try to workaround.
I used SFEgcc from pkg.opensolaris.cz.
I ditched the idea of reusing the kde-solaris workspace and for the moment built all packages with mere
./configure --prefix=/opt/kde4 && gmake && gmake install
or
PKG_CONFIG_PATH=/opt/kde4/lib/pkgconfig:/usr/lib/pkgconfig cmake .. -DCMAKE_INSTALL_PREFIX=/opt/kde4 && gmake && gmake install
No crazy mesh of *FLAGS, no nothing, just the defaults.
Only about 5 patches were needed for to have gdm-integration complete.
Unfortunately the session feels about as sluggish as when built with Studio.
Elephant droppings
You can buy elephant droppings in the Prague ZOO, a bucket costs about 2€, plus deposit. It looks like a nice marketing stunt by the new director Bobek :-) (bobek = dropping)
Name 'em & shame 'em
Somebody in Říčany got really sick of their neighbor's behaviour.
And there's no wonder. The garbage appears there too often.
Feeding Google with rough translation: This garbage and her sanitary towels are dumped by Mrs. Eva Plíhalová, Sadová 1015 Říčany. Other pigs will be named here the next time.
Sunday, April 17, 2011
Thursday, April 14, 2011
KDE 4.6.2 on Solaris 11
KDE 4.6.2 is here and now you can have it on your Solaris/OpenIndiana box as well.
Apart from the upstream fixes we refreshed boost to v. 1.46.1, phonon to 4.5.0 and Qt to 4.7.2.
More patches got upstreamed, especially on the Webkit front, thanks to Ben. Ben is also working hard on Solaris 10 support.
Since we lost the build machine, publishing the binary packages may lag a bit, so if you want to give KDE a try, create a fresh zone and follow the effortless guide to build them. It takes about half a day on a mid-range laptop and it's almost fully automated so it's nothing to be scared of.
Apart from the upstream fixes we refreshed boost to v. 1.46.1, phonon to 4.5.0 and Qt to 4.7.2.
More patches got upstreamed, especially on the Webkit front, thanks to Ben. Ben is also working hard on Solaris 10 support.
Since we lost the build machine, publishing the binary packages may lag a bit, so if you want to give KDE a try, create a fresh zone and follow the effortless guide to build them. It takes about half a day on a mid-range laptop and it's almost fully automated so it's nothing to be scared of.
Tuesday, April 12, 2011
Czech University of Life Sciences Prague
I just learned that couple of years ago my alma mater changed its english name from 'The Czech University of Agriculture Prague' to 'Czech University of Life Sciences Prague'.
I guess it sounds much cooler on the CVs and particularly the graduates of the IT or Economical Faculty must've really appreciated that :-)
Personally I find it ridiculous as they kept the Czech name intact ...
Excuse me for now, I'll update my Linked-in profile ...
I guess it sounds much cooler on the CVs and particularly the graduates of the IT or Economical Faculty must've really appreciated that :-)
Personally I find it ridiculous as they kept the Czech name intact ...
Excuse me for now, I'll update my Linked-in profile ...
Wednesday, April 6, 2011
Thursday, March 17, 2011
bone marrow
Reading about the guys trying to cool down Fukushima and the alert sent to the European Bone Marrow centers reminded me that my contact info in the bone marrow donors register is way outdated. I have moved six times since I registered and the phone number I put there ceased to exist ten years ago. To my pleasant surprise I could use an online form to update my contact details, so in the unlikely event of someone matching my HLA being in need I can do my bit in making the world a better place.
Friday, February 25, 2011
failing beadm
All of a sudden I was unable to create a new environment.
Huh. After a bit of googling a learned about the BE_PRINT_ERR variable and that helped me to find the culprit:
hajma@briza:~$ pfexec beadm create test Unable to create test. Doesn't exist.
Huh. After a bit of googling a learned about the BE_PRINT_ERR variable and that helped me to find the culprit:
hajma@briza:~$ BE_PRINT_ERR=true pfexec beadm create test get_mountpoint_from_vfstab: failed to open vfstab (/export/testzone/root/etc/vfstab) _update_vfstab: failed to open vfstab (/tmp/.be.TTayGg/etc/vfstab): No such file or directory be_copy_zones: failed to update new BE's vfstab (zbe-1) be_copy: failed to process zones be_copy: destroying partially created boot environment Unable to create test. Doesn't exist.It looks like the testzone was to blame. Since I did not have anything of a value in that zone, I just deleted it and everything started working again.
Thursday, February 24, 2011
Wednesday, February 16, 2011
AM200g again
A year and half ago I said I'd come back to the AM200g firmware when I have the time. While WRT the time it's not really the case, I managed to install RedHat 9 (yup, that's the requirement!) to VirtualBox and run the
To my surprise the firmware built absolutely smoothly, without a single glitch.
I'm going to test the firmware now so if you don't see me online you know what happened :-)
I don't have any plans for tuning the firmware, but if the router is still alive when the provider starts offering IPv6, it may come handy.
./consumer_install
script that Asus' engineers provided.To my surprise the firmware built absolutely smoothly, without a single glitch.
I'm going to test the firmware now so if you don't see me online you know what happened :-)
I don't have any plans for tuning the firmware, but if the router is still alive when the provider starts offering IPv6, it may come handy.
Thursday, February 10, 2011
Calligra on Solaris
Calligra, the fork of KOffice, is now buildable on Solaris. Some parts are missing, however - Krita and Tables are not present because they require the Eigen linear algebra library. Eigen is basically a bunch of C++ templates and if there's one field where Sun/Oracle Studio is losing breath, it's templates. Due to a bug in Studio, anything that touches Eigen is unbuildable :-(
Also there are some nasty crashes, like in Plan and some modules fail to load due to broken linkage so you only get limited feature in Kformula, Flow and others, but this can well improve as soon as we run some ldd checks on the binaries.
Here're some screenshots:
Also there are some nasty crashes, like in Plan and some modules fail to load due to broken linkage so you only get limited feature in Kformula, Flow and others, but this can well improve as soon as we run some ldd checks on the binaries.
Here're some screenshots:
Plan WorkPackage
Flow
Karbon
Kexi
KThesaurus
Words
Braindump
Labels:
Calligra,
KDE,
kde-solaris,
KOffice,
OpenSolaris,
Solaris,
Studio
Monday, January 31, 2011
EvalBot at home
The Stellaris development kit I ordered back in October has finally arrived today. Kudos to TI for keeping their promise and delivering it for the discounted price. I'll do the assembly when the boy comes home from the kindergarten so we can have some fun together :-)
Thursday, January 27, 2011
KDE 4.6.0 for Solaris
Binary KDE 4.6.0 packages for Solaris 11 are available at the usual place. The only parts missing are kdebindings, kdeplasma-addons and kdepim. Unfortunately these are not buildable at the moment. On the other hand 53 localization packages are present.
Other packages worth mentioning are Amarok 2.4.0, QtCreator 2.0.1 and Boost 1.45
There is an interesting powerdevil related crash when one logs into the session, but apart from that it works quite nicely.
The packages were built on b151a so OpenIndiana users might prefer to build them themselves.
Other packages worth mentioning are Amarok 2.4.0, QtCreator 2.0.1 and Boost 1.45
There is an interesting powerdevil related crash when one logs into the session, but apart from that it works quite nicely.
The packages were built on b151a so OpenIndiana users might prefer to build them themselves.
Wednesday, January 26, 2011
Friday, January 14, 2011
KDE 4.5.5 for Solaris 11
$SUBJ is out in the wild, you can get it from the repository at bionicmutton.
There's not many changes upstream, but concerning Solaris support, we're workarounding a stupid cairo bug in S11e that caused issues with anything png-related.
enjoy
There's not many changes upstream, but concerning Solaris support, we're workarounding a stupid cairo bug in S11e that caused issues with anything png-related.
enjoy
Wednesday, January 5, 2011
No spam
A few weeks ago we put a label saying 'No advertisements' on our mailbox.
Of course the frequency of the warm fuzzy feelings that accompany the act of throwing a big box of paper waste into the blue paper bins has drastically decreased but something's telling me the nature appreciates it more this way :-)
It also gives us quite a unique feeling as in our town only 275 other people do the same. Wow :-(
Of course the frequency of the warm fuzzy feelings that accompany the act of throwing a big box of paper waste into the blue paper bins has drastically decreased but something's telling me the nature appreciates it more this way :-)
It also gives us quite a unique feeling as in our town only 275 other people do the same. Wow :-(
Subscribe to:
Posts (Atom)