ACCUMULATE_OUTGOING_ARGS as runtime argument take 2

Jan Hubicka hubicka@atrey.karlin.mff.cuni.cz
Wed Mar 29 01:49:00 GMT 2000


> On Mon, Mar 27, 2000 at 09:48:11PM +0200, Jan Hubicka wrote:
> > Here is updated version of my patch to turn ACCUMULATE_OUTGOING_ARGS
> > to an runtime option.  It now contains all necesarry changes to backends.
> > It bootstraps on i386 in all three modes, passes regression tests
> > and bootstraps on m68k (with some unreleated tweaking of current CVS tree).
> 
> Ok.
Thanks for the really exceptional patience while reading this patch.
> 
> Some grammar:
Uff, thats realy embrasling. I will have to subscribe some english language
lessons again.
> 
> ! #ifndef STACK_GROWS_DOWNWARD
> ! #ifdef ARGS_GROW_DOWNWARD
> !   if (!ACCUMULATE_OUTGOING_ARGS)
>   #else
> !   if (0)
>   #endif
> + #else
> +   if (1)
> + #endif
> 
> Ug.  I realize this isn't much worse than the original, but can
> we get this cleaned up next?
Yes. In fact I originally wanted to include this in the original patch, but then
I god completely anoyed by it, so I decided to do this later.
(I've prepared some ground for this by avoiding ifdefs on ARGS_REVERSED)

I was thinking about killing STACK_GWORS_DOWNWARD and ARGS_* macros and
do instad STACK_GROWTH (or STACK_DIRECTION?) macro, that can be 1 for stack
going uppward and -1 otherwise.

Then we can often use it as multiplier for the values to get everything working
right (like in our combine_stack_adjustments pass BTW it is even safe now on
!STACK_GROWS_DOWNWARD machines?).  What do you think?

Is there some so weird port needing to change this definitions at runtime?
I remember seeing some note about this on HP-PA with ARGS_* macro.

Honza


More information about the Gcc-patches mailing list