This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [m68k] Convert CPP_SPEC to TARGET_CPU_CPP_BUILTINS
- From: Joel Sherrill <joel dot sherrill at OARcorp dot com>
- To: Gunther Nikl <gni at gecko dot de>
- Cc: Bernardo Innocenti <bernie at develer dot com>, gcc-patches at gcc dot gnu dot org
- Date: Tue, 16 Sep 2003 09:31:37 -0500
- Subject: Re: [m68k] Convert CPP_SPEC to TARGET_CPU_CPP_BUILTINS
- Organization: OAR Corporation
- References: <3F64F6A2.5030204@develer.com> <jer82ipjtt.fsf@sykes.suse.de> <3F65BB28.3060004@develer.com> <20030915144423.GA30190@lorien.int.gecko.de> <3F6616A9.8020601@develer.com> <3F6706CD.1000108@OARcorp.com> <20030916135117.GA46624@lorien.int.gecko.de>
Gunther Nikl wrote:
On Tue, Sep 16, 2003 at 07:49:17AM -0500, Joel Sherrill wrote:
mc68030: We use this to distinguish 68030 from the 68020+FPU and
68040. There are some subtle differences that matter.
So -m68030 was really used and the build depends on mc68030 beeing
defined?
As a disclaimer, I do not have the 680x0 docs in front of me and
this code is old.
The RTEMS cache management code has unique code for the mc68030.
According to the comment, the mc68020 does not have data cache
while the 68030 does. Further down, there is another block of
code for the mc68040 and mc68060 which use the same code.
Also, I recall that the FPU in the 68040 is not functionally
equivalent to the 68881/2 or 68030 FPU and some instructions
have to be emulated.
AFAIK GCC doesn't distinguish the 68030 from the 68020+FPU but
there are reasons the application code would need to know this
difference.
Gunther
--
Joel Sherrill, Ph.D. Director of Research & Development
joel@OARcorp.com On-Line Applications Research
Ask me about RTEMS: a free RTOS Huntsville AL 35805
Support Available (256) 722-9985