Hi, On Sun, 22 Oct 2006, Rob Browning wrote:
> > I looked into this problem and with the attached patch I can > > successfully compile the package. The most important part is the > > SHORT_ALIGN define, as m68k has different alignment rules. > > I noticed that you increased the stack limit, but at least from an > initial investigation, I'm not sure that SHORT_ALIGN should actually > affect the stack size. > > Is there any chance that it doesn't, and that you just noticed that > Guile's default stack limit might be too low in general? The SHORT_ALIGN define causes SCM_STACKITEM to be a short and thus it influences SCM_STACK_OVERFLOW_P in stackchk.h, so it's only half of what it was before. I had some debug option enabled, which might cause a little more stack usage, but only with the increased stack I then could complete the tests. bye, Roman -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
