[PATCH] Fix signed bitfield BIT_NOT_EXPR expansion (PR middle-end/52209)

Richard Guenther richard.guenther@gmail.com
Mon Feb 13 10:48:00 GMT 2012


On Sat, Feb 11, 2012 at 12:49 PM, Jakub Jelinek <jakub@redhat.com> wrote:
> Hi!
>
> In July Richard changed reduce_bit_field BIT_NOT_EXPR expansion from
> NOT unop to XOR with all the bits in the bitfield's precision set.
> Unfortunately that is correct for unsigned bitfields only, for signed
> bitfields, where op0 is already sign-extended to its mode before this,
> expanding this as NOT is the right thing.
>
> Fixed thusly, bootstrapped/regtested on x86_64-linux and i686-linux, ok for
> trunk?

Ok.

Thanks,
Richard.

> 2012-02-11  Jakub Jelinek  <jakub@redhat.com>
>
>        PR middle-end/52209
>        * expr.c (expand_expr_real_2) <case BIT_NOT_EXPR>: Only expand using
>        XOR for reduce_bit_field if type is unsigned.
>
>        * gcc.c-torture/execute/pr52209.c: New test.
>
> --- gcc/expr.c.jj       2012-02-07 16:05:51.000000000 +0100
> +++ gcc/expr.c  2012-02-11 10:08:44.162924423 +0100
> @@ -8582,8 +8582,9 @@ expand_expr_real_2 (sepops ops, rtx targ
>       if (modifier == EXPAND_STACK_PARM)
>        target = 0;
>       /* In case we have to reduce the result to bitfield precision
> -        expand this as XOR with a proper constant instead.  */
> -      if (reduce_bit_field)
> +        for unsigned bitfield expand this as XOR with a proper constant
> +        instead.  */
> +      if (reduce_bit_field && TYPE_UNSIGNED (type))
>        temp = expand_binop (mode, xor_optab, op0,
>                             immed_double_int_const
>                               (double_int_mask (TYPE_PRECISION (type)), mode),
> --- gcc/testsuite/gcc.c-torture/execute/pr52209.c.jj    2012-02-11 10:09:46.080571803 +0100
> +++ gcc/testsuite/gcc.c-torture/execute/pr52209.c       2012-02-11 10:09:28.000000000 +0100
> @@ -0,0 +1,14 @@
> +/* PR middle-end/52209 */
> +
> +extern void abort (void);
> +struct S0 { int f2 : 1; } c;
> +int b;
> +
> +int
> +main ()
> +{
> +  b = -1 ^ c.f2;
> +  if (b != -1)
> +    abort ();
> +  return 0;
> +}
>
>        Jakub



More information about the Gcc-patches mailing list