Re: Fold more boolean expressions

[Marc Glisse <marc.glisse@inria.fr>]

On Thu, 20 Sep 2018, Richard Biener wrote:

> On Sat, Sep 15, 2018 at 8:01 AM MCC CS <mcccs@gmx.com> wrote:
>>
>> Sorry for doing the same mistake twice. Is this OK, and do
>> I need to test it again after the first version of this
>> patch?
>>
>> 2018-09-15 MCC CS <deswurstes@users.noreply.github.com>
>>
>>         gcc/
>>         PR tree-optimization/87261
>>         * match.pd: Add boolean optimizations,
>>         fix whitespace.
>>
>> 2018-09-15 MCC CS <deswurstes@users.noreply.github.com>
>>
>>         gcc/testsuite/
>>         PR tree-optimization/87261
>>         * gcc.dg/pr87261.c: New test.
>>
>> Index: gcc/match.pd
>> ===================================================================
>> --- gcc/match.pd        (revision 264170)
>> +++ gcc/match.pd        (working copy)
>> @@ -92,7 +92,7 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
>>    IFN_FMA IFN_FMS IFN_FNMA IFN_FNMS)
>>  (define_operator_list COND_TERNARY
>>    IFN_COND_FMA IFN_COND_FMS IFN_COND_FNMA IFN_COND_FNMS)
>> -
>> +
>>  /* As opposed to convert?, this still creates a single pattern, so
>>     it is not a suitable replacement for convert? in all cases.  */
>>  (match (nop_convert @0)
>> @@ -106,7 +106,7 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
>>        && tree_nop_conversion_p (TREE_TYPE (type), TREE_TYPE (TREE_TYPE (@0))))))
>>  /* This one has to be last, or it shadows the others.  */
>>  (match (nop_convert @0)
>> - @0)
>> + @0)
>>
>>  /* Transform likes of (char) ABS_EXPR <(int) x> into (char) ABSU_EXPR <x>
>>     ABSU_EXPR returns unsigned absolute value of the operand and the operand
>> @@ -285,7 +285,7 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
>>       And not for _Fract types where we can't build 1.  */
>>    (if (!integer_zerop (@0) && !ALL_FRACT_MODE_P (TYPE_MODE (type)))
>>     { build_one_cst (type); }))
>> - /* X / abs (X) is X < 0 ? -1 : 1.  */
>> + /* X / abs (X) is X < 0 ? -1 : 1.  */
>>   (simplify
>>     (div:C @0 (abs @0))
>>     (if (INTEGRAL_TYPE_P (type)
>> @@ -921,6 +921,31 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
>>    (bitop:c @0 (bit_not (bitop:cs @0 @1)))
>>    (bitop @0 (bit_not @1))))
>>
>> +/* (~x & y) | ~(x | y) -> ~x */
>> +(simplify
>> + (bit_ior:c (bit_and:c (bit_not@2 @0) @1) (bit_not (bit_ior:c @0 @1)))
>> + @2)
>> +
>> +/* (x | y) ^ (x | ~y) -> ~x */
>> +(simplify
>> + (bit_xor:c (bit_ior:c @0 @1) (bit_ior:c @0 (bit_not @1)))
>> + (bit_not @0))
>> +
>> +/* (x & y) | ~(x | y) -> ~(x ^ y) */
>> +(simplify
>> + (bit_ior:c (bit_and @0 @1) (bit_not:s (bit_ior:s @0 @1)))
>
> I think this misses :cs on the bit_and.

For :c, shouldn't canonicalization make the order of @0 and @1 consistent 
for bit_and and bit_ior?

>> + (bit_not (bit_xor @0 @1)))
>> +
>> +/* (~x | y) ^ (x ^ y) -> x | ~y */
>> +(simplify
>> + (bit_xor:c (bit_ior:cs (bit_not @0) @1) (bit_xor:c @0 @1))
>> + (bit_ior @0 (bit_not @1)))
>
> :s on the bit_xor
>
>> +/* (x ^ y) | ~(x | y) -> ~(x & y) */
>> +(simplify
>> + (bit_ior:c (bit_xor @0 @1) (bit_not:s (bit_ior @0 @1)))
>> + (bit_not (bit_and @0 @1)))
>
> :cs on the bit_xor, :s on the second bit_ior
>
> Otherwise looks OK to me.

-- 
Marc Glisse