Use simplify_replace_rtx rather than wrap_constant

Richard Sandiford rdsandiford@googlemail.com
Wed Oct 21 22:17:00 GMT 2009


Richard Sandiford <rdsandiford@googlemail.com> writes:
> Index: gcc/config/m32r/m32r.c
> ===================================================================
> --- gcc/config/m32r/m32r.c	2009-10-21 20:59:55.000000000 +0100
> +++ gcc/config/m32r/m32r.c	2009-10-21 21:00:01.000000000 +0100
> @@ -1212,6 +1212,7 @@ m32r_setup_incoming_varargs (CUMULATIVE_
>  m32r_is_insn (rtx insn)
>  {
>    return (NONDEBUG_INSN_P (insn)
> +	  && !DEBUG_INSN_P (insn)
>  	  && GET_CODE (PATTERN (insn)) != USE
>  	  && GET_CODE (PATTERN (insn)) != CLOBBER
>  	  && GET_CODE (PATTERN (insn)) != ADDR_VEC);
> Index: gcc/config/mips/mips.c
> ===================================================================
> --- gcc/config/mips/mips.c	2009-10-21 20:59:55.000000000 +0100
> +++ gcc/config/mips/mips.c	2009-10-21 21:00:01.000000000 +0100
> @@ -9533,7 +9533,10 @@ mips_restore_gp_from_cprestore_slot (rtx
>    gcc_assert (TARGET_ABICALLS && TARGET_OLDABI && epilogue_completed);
>  
>    if (!cfun->machine->must_restore_gp_when_clobbered_p)
> -    return;
> +    {
> +      emit_note (NOTE_INSN_DELETED);
> +      return;
> +    }
>  
>    if (TARGET_MIPS16)
>      {
> Index: gcc/testsuite/lib/target-supports.exp
> ===================================================================
> --- gcc/testsuite/lib/target-supports.exp	2009-10-21 20:59:55.000000000 +0100
> +++ gcc/testsuite/lib/target-supports.exp	2009-10-21 21:00:01.000000000 +0100
> @@ -1568,6 +1568,7 @@ proc check_effective_target_arm_thumb2_o
>  # otherwise.  Cache the result.
>  
>  proc check_effective_target_arm_neon_hw { } {
> +    return 0
>      return [check_runtime arm_neon_hw_available {
>  	int
>  	main (void)

And as you will have guessed, this lot wasn't supposed to be there.
Tonight really hasn't been my night.

Richard



More information about the Gcc-patches mailing list