egcs-19980502 powerpc-ibm-aix4.1.4.0 falling off end of non-void func

[Kaveh R. Ghazi]

	From rs6000.md, movdf_hardfloat32, movdf_softfloat32, movdi_32
and one other place generate code for functions in insn-output.c which
look like this:

 > static char *
 > output_418 (operands, insn)
 >      rtx *operands ATTRIBUTE_UNUSED;
 >      rtx insn ATTRIBUTE_UNUSED;
 > {
 >  
 > {
 >   switch (which_alternative)
 >     {
 >     case 0:
 >       /* We normally copy the low-numbered register first.  However, if
 >          the first register operand 0 is the same as the second register of
 >          operand 1, we must copy in the opposite order.  */
 >       if (REGNO (operands[0]) == REGNO (operands[1]) + 1)
 >         return "mr %L0,%L1\n\tmr %0,%1";
 >       else
 >         return "mr %0,%1\n\tmr %L0,%L1";
 >     case 1:
 >       /* If the low-address word is used in the address, we must load it
 >          last.  Otherwise, load it first.  Note that we cannot have
 >          auto-increment in that case since the address register is known to be
 >          dead.  */
 >       if (refers_to_regno_p (REGNO (operands[0]), REGNO (operands[0]) + 1,
 >                              operands [1], 0))
 >         return "ld %L0,%L1\n\tld %0,%1";
 >       else
 >         return "ld%U1 %0,%1\n\tld %L0,%L1";
 >     case 2:
 >       return "std%U0 %1,%0\n\tstd %L1,%L0";
 >     }
 > }
 > }

	Gcc warns about this:

 > insn-output.c: In function `output_418':
 > insn-output.c:1509: warning: control reaches end of non-void function


	So I think a default: case for the switch should be added to
either abort or return a valid default value.

		--Kaveh
--
Kaveh R. Ghazi			Project Manager / Custom Development
ghazi@caip.rutgers.edu		Icon CMT Corp.