Re: egcs-1.0.1-prerelease powerpc-ibm-aix4.1.4.0 results (g++ regression)

["Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>]

 > From: David Edelsohn <dje@watson.ibm.com>
 > 
 > >>>>> Andrey Slepuhin writes:
 > 
 > Andrey> I already reported the same regressions with egcs-1.0 on
 > Andrey> powerpc-ibm-aix4.2.1.0.
 > Andrey> They are really G++ regressions - compiler is "bailing out" instead
 > Andrey> of producing more error messages. I beleive that these regressions
 > Andrey> are fixed in egcs-971215 (at least on my machine all is Ok).
 > 
 > 	If you already reported these for egcs-1.0 then why is Kaveh
 > referring to them as new regressions in egcs-1.0.1?
 > David

	I apologize.  I was mistaken when I reported that the
regressions occurred between egcs-1.0 -> egcs-1.0.1-prerelease.  In
fact the regressions occurred somewhere between egcs-971127 ->
egcs-1.0, so the bug is in the released egcs and is still in 1.0.1 but
as you mentioned, its fixed in egcs-971215.

	The gist is that we have this regression in the release branch
since 1.0.  But it doesn't happen in the snapshots just prior and just
after the releases.  If you don't think its serious enough to be worth
worrying about, then I'm okay with it. 

		--Kaveh
--
Kaveh R. Ghazi			Project Manager / Custom Development
ghazi@caip.rutgers.edu		ICon CMT Corp.