Today I deleted the whole objects directory and did an svn update. $./configure $makehdimage /Haiku That gave something that booted properly to the console. Then I deleted the objects dir again (just to make sure), changed makehdimage to: #if [ x$RUN_WITHOUT_APP_SERVER = x ]; then # define only, if not already set export RUN_WITHOUT_APP_SERVER=0 #fi export TARGET_PLATFORM=haiku compiled everything again and then changed the bootscript to: [snip] launch $SERVERS/app_server picasso # launch app_server launch $SERVERS/registrar _roster_thread_ # launch registrar launch $SERVERS/input_server launch $SERVERS/debug_server # launch debug_server [snip] launch beos/apps/MiniTerminal This just seems to hang on the boot screen. I've tried various launch lines (removing the picasso, adding sleeps) but they don't appear to help. Pressing escape on boot shows the last line as entering kernel at... and then nothing else. I have a nvidia card, I know Michael uses a Radeon, so that could be the issue. My main concern before was the default build without the app server not booting either, but that seems fine when I tried it today, so don't worry too much. Simon > "Simon Taylor" <simontaylor1@xxxxxxxxxxxx> wrote: > > I've just grabbed the whole SVN repository (took over 6 hours with > > dialup!), and done a makehdimage onto a blank (initialised BFS) > > Haiku > > partition. > > > > The boot screen appears and there is some hard disk activity for a > > few > > seconds, but then nothing. I was trying to get miniterminal working > > as > > mmlr described at http://forum.mlotz.ch/viewtopic.php?p=354, but I > > put > > makehdimage back to export RUN_WITHOUT_APP_SERVER=1 and left > > everything > > else as it was still to no avail. > > You've tried to load the app_server? > Anyway, if your basic configuration is correct (ie. you haven't > messed > with the build), try to remove the object files for Application.cpp > and > Bitmap.cpp in objects/x86.R1/kits/, and call makehdimage again. > If you switch to use the app_server, you have to do the same thing > again, but don't export RUN_WITHOUT_APP_SERVER=1 in this case :-) > > AFAICT the current release works without any problems. > > Bye, > Axel. > >