Re: [Mjpeg-users] jpeg-mmx/jdapimin.c with gcc 3.3

2003-05-31 Thread Wolfgang Fritz
Steven M. Schultz wrote: > Hi - > > >>From: Wolfgang Fritz <[EMAIL PROTECTED]> >> >>>gcc -O6 -I. -c -o jdapimin.o jdapimin.c >>>/tmp/cc7IfzBn.s: Assembler messages: >>>/tmp/cc7IfzBn.s:619: Error: symbol `NOT_SUPPORTED' is already defined >>>gmake: *** [jdapimin.o] Error 1 >>> >> >>The reason f

Re: [Mjpeg-users] jpeg-mmx/jdapimin.c with gcc 3.3

2003-05-31 Thread Steven M. Schultz
Hi - > From: Wolfgang Fritz <[EMAIL PROTECTED]> > > gcc -O6 -I. -c -o jdapimin.o jdapimin.c > > /tmp/cc7IfzBn.s: Assembler messages: > > /tmp/cc7IfzBn.s:619: Error: symbol `NOT_SUPPORTED' is already defined > > gmake: *** [jdapimin.o] Error 1 > > > > The reason for this it that gcc inlines th

Re: [Mjpeg-users] jpeg-mmx/jdapimin.c with gcc 3.3

2003-05-31 Thread Wolfgang Fritz
Steven M. Schultz wrote: > Hi - > > If you're using gcc 3.3 (as SuSE 8.2 does) and the building > of the jpeg-mmx library fails with: > > gcc -O6 -I. -c -o jdapimin.o jdapimin.c > /tmp/cc7IfzBn.s: Assembler messages: > /tmp/cc7IfzBn.s:619: Error: symbol `NOT_SUPPORTED' is already d

Re: [Mjpeg-users] jpeg-mmx/jdapimin.c with gcc 3.3

2003-05-31 Thread Al Bogner
On Friday 30 May 2003 17:14, Steven M. Schultz wrote: > > From: Ronald Bultje <[EMAIL PROTECTED]> > > > > > Here is a patch to fix the problem - it seems that -O6 is > > > too high and causes the module to be miscompiled somehow. > > > Changing that to -O2 fixes the problem: > > > > Hm, even t

Re: [Mjpeg-users] jpeg-mmx/jdapimin.c with gcc 3.3

2003-05-31 Thread Steven M. Schultz
Hi ! > From: Ronald Bultje <[EMAIL PROTECTED]> > > Here is a patch to fix the problem - it seems that -O6 is too high > > and causes the module to be miscompiled somehow. Changing that to > > -O2 fixes the problem: > > Hm, even though -O>=4 doesn't guarantee to give proper compilati

Re: [Mjpeg-users] jpeg-mmx/jdapimin.c with gcc 3.3

2003-05-30 Thread Ronald Bultje
Hey Steven & all, On Fri, 2003-05-30 at 07:03, Steven M. Schultz wrote: > Here is a patch to fix the problem - it seems that -O6 is too high > and causes the module to be miscompiled somehow. Changing that to > -O2 fixes the problem: Hm, even though -O>=4 doesn't guarantee to