[PATCH] scheduler register pressure excessive stack space

[law@redhat.com]

 In message <200205291733.NAA24442@makai.watson.ibm.com>, David Edelsohn 
writes:
 > 	Yes, it is yet another latent bug which I had not yet tracked
 > down.  The patch itself is correct, but I did not want to check in a patch
 > exposing latent bugs I could not fix.  All of the intervening patches have
 > been for latent bugs leading to this patch, but apparently there still are
 > more.
 > 
 > 	The following link is my reply to Richard Henderson's original
 > approval of the patch where I withdrew it due to failures I discovered:
 > 
 > http://gcc.gnu.org/ml/gcc-patches/2002-04/msg00094.html
Is it just the libcall stuff that's causing problems?  Or the no-unit part,
or the fix for SCHED_GROUP_P?  Your message isn't specific about what part
of the patch was causing failures.  I can revert everything, but I'd prefer
not to do that.

jeff