This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: bogus .quad in i386/att.h


David> and it broke cygwin bootstrap 
David> http://gcc.gnu.org/ml/gcc-bugs/2001-10/msg00084.html

Good point.

Jan> I am not quite sure if it is cleaner to ifdef around each ASM_QUAD
Jan> use, or if we can just define ASM_QUAD to ".quad" everywhere...

Well, ".quad" is definitely not part of the standard AT&T assembler
syntax.   Elsewhere in the compiler, it already knows how to do this 
anyway.   Isn't that what UNALGINED_DOUBLE_INT_ASM_OP is about?   
Why invent something new here?

RJL


> 
> -----Original Message-----
> From: Robert Lipe [mailto:robertlipe@usa.net]
> Sent: Tuesday, 9 October 2001 11:38 
> To: gcc@gcc.gnu.org
> Cc: jh@suze.cz
> Subject: bogus .quad in i386/att.h
> 
> 
> Jan, your commit that starts: 
> 
> Tue Oct  2 12:46:01 CEST 2001  Bo Thorsen     <bo@suse.co.uk>,
>                                Andreas Jaeger <aj@suse.de>,
>                                Jan Hubicka    <jh@suse.cz>
> 
>         * doc/invoke.texi (i386 Options): Document x86-64 options.
>         (i386 and x86-64 Options): Rename i386 options section.
> 
> 
> added an entry for ASM_QUAD in att.h.   But the ChangeLog does not
> call out any change to att.h.
> 
> ..quad is definitely not something supported by the AT&T-derived
> assemblers.  Maybe it's a GAS thing and should be in gas.h.  Maybe it's
> an x86-64 thing and should be in x86-64.h.  But att.h sure seems like
> the wrong place for it.  This does show up as testsuite failures.
> 
> Can you please move this?
> 
> Thank you,
> RJL


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]