[gnu.utils.bug] gcc & M68881 fp processor

lethin@WHEATIES.AI.MIT.EDU (Richard Lethin) (10/25/88)

gcc has a bug in that it compiles assuming that a M68881 floating
point coprocessor is available, when I compile it on cream-of-wheat.
I cannot run it on kaboom (a 3/50).  Additionally, I cannot run gcc
itself on a 3/50 because it's compiled assuming a M68881 fp
coprocessor!

The default should be the equivalent of the Sun cc -fsoft switch.

or a switch sould be provided to get -fsoft behavior and a manual
page should be made available to document it.

-------

tmb@WHEATIES.AI.MIT.EDU (10/25/88)

	gcc has a bug in that it compiles assuming that a M68881 floating
	point coprocessor is available, when I compile it on cream-of-wheat.
	I cannot run it on kaboom (a 3/50).  Additionally, I cannot run gcc
	itself on a 3/50 because it's compiled assuming a M68881 fp
	coprocessor!
	
	The default should be the equivalent of the Sun cc -fsoft switch.
	
	or a switch sould be provided to get -fsoft behavior and a manual
	page should be made available to document it.

No bug. This is quite well documented in "Internals of GNU CC" (which
is the one and only user's manual):

   These are the '-m' options defined in the 68000 machine description:
   
   ...
   
   '-m68881'	Generate output containing 68881 instructions for floating
   		point. This is the default if you use the unmodified sources.
   
   '-msoft-float'	Generate output containing library calls for floating point.
   
   ...
   
   
The manual page doesn't list most of the switches.

If there is a bug, it's with the Sun kernel. It should provide co-processor
emulation.