The collected works of sgefant

it crashes here when trying to load the default .B.blend file. Maybe this backtrace from gdb helps:



The same happens when selecting "Load Factory Settings" from the File menu.

greetings,
sgefant

PS: the link in your first post points to blender-2.48-irix-b.tar.gz, while the file on the server is named blender-2.28-irix-b.tar.gz

That crash probably is not related to python. Do you have a .B.blend file in your home directory? if not, try creating one by running a older version of blender and pressing ctrl+u. If blender 2.48 works then, you've hit the same bug as I did.

fzalfa: I just tested blender on an R5k O2 and it works (as long as there is a ~/.B.blend). What is the exact error you are getting?

greetings,
sgefant

hamei wrote: edit :
upped Python to 2.52
Irix 6.5.30
Fuel

Code: Select all

fewel 5% ./blender
guessing './blender' == '/usr/people/wacko/irix6/./blender'
Compiled with Python version 2.5.2.
Could not find platform independent libraries <prefix>
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
'import site' failed; use -v for traceback
Checking for installed Python... No installed Python found.
Only built-in modules are available.  Some scripts may not run.
Continuing happily.
Bus error


No .B.blend and I don't have one lying around. A placebo .B.blend makes no difference

You can copy any .blend file to ~/.B.blend. Alternatively you could download an older version of blender, run it and press CTRL+U to save a new .B.blend.

Unfortunately blender crashes on Irix when trying to load files saved on 64 bit systems (lile the default .B.blend). I haven't been able yet to track down this bug. However, current svn should be compilable with gcc (using gmake), so any brave soul willing to have a look at this issue is more than welcome.

To use gcc, one has to add "export IRIX_USE_GCC=true" to their user-def.mk file. Here's mine for reference (using python 2.4 from nekoware):



edit: BTW, the python error messages are completely harmless and not related to the segfault. They just mean that some python scripts won't run.

greetings,
sgefant

I've already mentioned this in another thread , but apparently nobody noticed.

Blender 2.49a got released more than a month ago. It should work on multi-cpu systems, so please test & use it a lot to justify building blender on IRIX. You can get it from the usual place: http://www.blender.org/download/get-blender/

greetings,
sgefant

PS: Upcoming Blender 2.5 will be using Python 3.1 (and depend on it, unlike previous version). It would be nice if someone could create a nekoware package of python 3.1.

It's been built with MipsPRO. However, gcc builds don't seem to be much slower.
I must admit that I don't quite get the point of the question.

nekonoko wrote:

sgefant wrote: PS: Upcoming Blender 2.5 will be using Python 3.1 (and depend on it, unlike previous version). It would be nice if someone could create a nekoware package of python 3.1.

I tried a quick attempt at a compile and it died with:

Code: Select all

c99 -c -OPT:Olimit=0 -DNDEBUG -O  -I. -IInclude -I./Include -I/usr/nekoware/include  -DPy_BUILD_CORE -o Python/pythonrun.o Python/pythonrun.c
cc-1020 c99: ERROR File = Python/pythonrun.c, Line = 662
The identifier "PATH_MAX" is undefined.

static wchar_t env_home[PATH_MAX+1];
^

1 error detected in the compilation of "Python/pythonrun.c".
gmake: *** [Python/pythonrun.o] Error 2

If you already have patches for it, let me know and I'll put everything together for Nekoware. Otherwise we'll have to wait until someone can port it.

Sure I will, but I won't be able to take a look at what needs to be done to Python for a couple of days.