|
From: | Bernardo Innocenti |
Subject: | Re: patch for longlong.h] |
Date: | Tue, 05 Aug 2003 21:14:40 +0200 |
User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030703 |
Torbjorn Granlund wrote:
Bernardo Innocenti <address@hidden> writes: Hello,this patch should probably be applied to your copy of longlong.has well as GCC's, otherwise things will break when GCC 3.4 comes out:http://gcc.gnu.org/ml/gcc-patches/2003-08/msg00178.htmlIf gcc 3.4 incorrectly defines __mc68020__ for cpu32, isn't gcc really what needs fixing?
That's not incorrect. CPU32 actually has a 68020 core. It has always been handled as a 68020 internally (see m68k.h/m68k.c). The only difference is that CPU32 doesn't have bit-field instructions, and there are really very few places where they are used in inline asm. It's much easier to take care of those special exceptions instead of adding a test for __mcpu32__ along with __m68020__ in all other places. -- // Bernardo Innocenti - Develer S.r.l., R&D dept. \X/ http://www.develer.com/ Please don't send Word attachments - http://www.gnu.org/philosophy/no-word-attachments.html
[Prev in Thread] | Current Thread | [Next in Thread] |