Re: Analysis of remaining xm-host headers

["Zack Weinberg" <zackw at Stanford dot EDU>]

On Wed, Mar 14, 2001 at 03:33:35AM -0600, Sam TH wrote:
> On Wed, Mar 14, 2001 at 12:43:54AM -0800, Zack Weinberg wrote:
> > i386/xm-beos.h, rs6000/xm-beos.h: A shocking amount of crud which
> > makes sense only if beos can't run configure.  I'm 99% sure this is
> > not the case.  The only necessary bits appear to be the
> > include-directory adjustments, and an unconditional #define
> > USE_C_ALLOCA.  On the latter, aomment explains that beos stacks are
> > permanently limited to 64KB, but it says "may change after DR9" so it
> > may be obsolete.
> 
> BeOS R 4.5 can definitely run configure on both of these platforms
> (and lots more, to boot).  I don't know about earlier versions, but
> 4.5 is something like 18 months old, and since upgrades are free, and
> BeOS is only a desktop system, I think that assuming that configure
> runs on BeOS is quite safe.  

This is good to know.  If I feed you a patch, can you test it?

> What would be a good way to test the max stack size?

Something like

#include <stddef.h>
#include <string.h>

void func(size_t n)
{
  char *p = __builtin_alloca(n);
  memset (p, 0x5A, n);
}

int main(void)
{
  size_t i;
  for (i = 1024; i < 8 * 1024 * 1024; i <<= 1)
    func(i);
  return 0;
}

run that under the debugger and see what i is when you crash for lack
of stack space.  8MB is the default stack limit on Linux, which means
that if GCC commonly wanted more than that, people would be screaming.

zw