Real purpose for failure in "Too restrictive sanity check"

[Jeffrey A Law]

  In message <20000421142202.F8511@atrey.karlin.mff.cuni.cz>you write:
  > >   In message <20000420134203.D9657@atrey.karlin.mff.cuni.cz>you write:
  > >   > 
  > >   > Wed Apr 19 12:54:31 MET DST 2000  Jan Hubicka  <jh@suse.cz>
  > >   > 	* calls.c (expand_call): Call compute_argument_block_size right
  > >   > 	before allocating the block; update comment; don't do alignment
  > >   > 	sanity checking for sibbling call; use args_size instead of
  > >   > 	unadjusted_args_size before args_size is adjusted.
  > > As I mentioned in my previous message.  I can't test this since it does n
  > ot
  > > apply to the current sources.
  > > 
  > Jeff, here is the even more updated patch for you.  The previous one collid
  > ed with my
  > another patch I checked in yesterday evening.
  > 
  > I am sorry for the breakage my loop.c patch caused. Hope that this one will
  > work for you.  I am just running bootstrap on ia32 (end of stage1 right now
  > ). I
  > don't expect problem (since the patch is equivallent to previous one), but 
  > I
  > will let you know if something goes wrong.
  > 
  > Honza
  > 
  > 
  > Wed Apr 19 12:54:31 MET DST 2000  Jan Hubicka  <jh@suse.cz>
  > 	* calls.c (expand_call): Call compute_argument_block_size right
  > 	before allocating the block; update comment; don't do alignment
  > 	sanity checking for sibbling call; use args_size instead of
  > 	unadjusted_args_size before args_size is adjusted.
I tried to bootstrap the PA64 port, but it's still failing the test in
calls.c:

#ifdef PREFERRED_STACK_BOUNDARY
      /* Stack must be properly aligned now.  */
      if (pass && stack_pointer_delta % preferred_unit_stack_boundary)
        abort ();
#endif

I tried to apply the patch in this message, but every hunk failed.

This is getting annoying.

jeff