Re: other/7114: ICE building strcoll.op from glibc-2.2.5

[Geoff Keating <geoffk@geoffk.org>]

The following reply was made to PR other/7114; it has been noted by GNATS.

From: Geoff Keating <geoffk@geoffk.org>
To: dje@watson.ibm.com
Cc: amodra@bigpond.net.au, d.mueller@elsoft.ch, gcc-gnats@gcc.gnu.org,
   gcc-patches@gcc.gnu.org
Subject: Re: other/7114: ICE building strcoll.op from glibc-2.2.5
Date: Wed, 17 Jul 2002 12:10:12 -0700

 > cc: amodra@bigpond.net.au, d.mueller@elsoft.ch, gcc-gnats@gcc.gnu.org,
 >    gcc-patches@gcc.gnu.org
 > Date: Wed, 17 Jul 2002 11:42:02 -0400
 > From: David Edelsohn <dje@watson.ibm.com>
 > 
 > >>>>> Geoff Keating writes:
 > 
 > Geoff> So, thanks for testing it!  We now know this doesn't work. :-)
 > 
 > 	The patch was tested on SVR4 by Franz and showed no new failures:
 > 
 > http://gcc.gnu.org/ml/gcc-patches/1999-03n/msg00584.html
 
 The testsuite probably doesn't check that r30 is not clobbered in a
 nested function when profiling is switched on.  Actually, Alan, could
 you write a test for that?  (That's something I missed while reviewing
 your patch.)
 
 > PUSH/POP cannot work on PowerPC.  On AIX PUSH/POP were corrupting the
 > stack.
 
 Because they were implemented wrongly?
 
 Clearly, push/pop can work, because procedures push and pop call
 frames all the time.  It's just necessary to do it the right way. 
 
 For AIX, of course, push/pop is unnecessary, and Alan's patch didn't
 add it.
 
 -- 
 - Geoffrey Keating <geoffk@geoffk.org> <geoffk@redhat.com>