From: pavenis AT lanet DOT lv Message-ID: To: DJ Delorie , djgpp-workers AT delorie DOT com Date: Thu, 9 Sep 1999 17:31:53 +0300 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Re: gcc-2.95 In-reply-to: <199909091416.KAA22042@envy.delorie.com> References: (pavenis AT lanet DOT lv) X-mailer: Pegasus Mail for Win32 (v3.12a) Reply-To: djgpp-workers AT delorie DOT com X-Mailing-List: djgpp-workers AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk On 9 Sep 99, at 10:16, DJ Delorie wrote: > > Is there a -m option that disables this? If so, would it make sense > these days to make 486+ instructions the default? > -m386 Bad example which got SIGILL on 386 when compiled with -march=pentium and -march=-i486 works Ok on 486 (also with -march=pentium). However we cannot be sure there is no code which will crash in that case. Also current Slackware beta (glibc-2.1.1 now) is compiled for i386 (but sometime earlier there were a mix) Perhaps I don't have other choise as to rebuild gcc-2.95.1 for i386-pc-msdosdjgpp. Perhaps no need to do that with gcc-2.95 as I think we could delete it when all will be Ok with 2.95.1. Only question is what to do with port of egcs-1.1.2 as that's the only version which contains recent enough version of GPC. Also according to mails in gcc mailing list there may be 2.95.2 next month. Andris Andris