[Octave-bug-tracker] [bug #47183] Problems with instructions on http://wiki.octave.org/MXE - and suggested improvements

[Michael Maloney]

Follow-up Comment #3, bug #47183 (project octave):

Re problems with crashes: I started things off in each case with clean sweep
of everything, before restarting any make steps, so I don't think this was the
problem. I only used restarts for a specific configuration - wiping things
completely if I decided to run configure again. At this point I'm not using
the 64 bit addressing.

The reason I'm not using the ppas to install a binary is that the default
quantum-depth used with graphicsmagick is 8bit and I need a version of Octave
with built with a 16 bits version  graphicsMagick. 

I do want a stable version - and in each case that is what I have built,
either via the standard configure and make approach - or via the mxe
approach.

I have done this using the method of downloading the source and then running
configure etc to build and install Octave - and get a 16 bit version this way
(after downloading and building GraphicsMagick as a 16 bits version).

The reason I looked at the MXE procedure was that I also needed a 16 bit image
version of 64bit Octave for Windows, and was able to do this via the
mxe-Octave method for Windows. 

I thought then it might be useful to get familiar with the mxe  procedure for
a linux build.

I'm not an expert on this stuff and ended up spending a fair amount of time to
get things working with the mxe-Octave Linux version. By posting the details
here I hope at least to make things easier for others until such time as any
changes do get made to the procedure.

Thank you for looking at this.

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?47183>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/