On 27/11/2016 00:28, Marc Glisse wrote:
> On Sat, 26 Nov 2016, Paolo Bonzini wrote:
>
>> --- match.pd (revision 242742)
>> +++ match.pd (working copy)
>> @@ -2554,6 +2554,19 @@
>> (cmp (bit_and@2 @0 integer_pow2p@1) @1)
>> (icmp @2 { build_zero_cst (TREE_TYPE (@0)); })))
>>
>> +/* If we have (A & C) != 0 ? D : 0 where C and D are powers of 2,
>> + convert this into a shift of (A & C). */
>> +(simplify
>> + (cond
>> + (ne (bit_and@2 @0 integer_pow2p@1) integer_zerop)
>> + integer_pow2p@3 integer_zerop)
>> + (with {
>> + int shift = wi::exact_log2 (@3) - wi::exact_log2 (@1);
>> + }
>> + (if (shift > 0)
>> + (lshift (convert @2) { build_int_cst (integer_type_node, shift); })
>> + (convert (rshift @2 { build_int_cst (integer_type_node, -shift);
>> })))))
>
> What happens if @1 is the sign bit, in a signed type? Do we get an
> arithmetic shift right?
It shouldn't happen because the canonical form of a sign bit test is A <
0 (that's the pattern immediately after). However I can add an "if" if
preferred, or change the pattern to do the AND after the shift.
(bit_and
(if (shift > 0)
(lshift (convert @0) { build_int_cst (integer_type_node, shift); })
(convert (rshift @0 { build_int_cst (integer_type_node, -shift); })))
@3)
What do you think?
Paolo
