Re: [PATCH v2] match.pd: rewrite select to branchless expression

[Michael Collison <collison@rivosinc.com>]

Richard,

Can you submit this patch for me while I sort out git write access?

On 11/18/22 07:57, Richard Biener wrote:
> On Fri, Nov 11, 2022 at 3:28 AM Michael Collison <collison@rivosinc.com> wrote:
>> This patches transforms ((x & 0x1) == 0) ? y : z <op> y -into
>> (-(typeof(y))(x & 0x1) & z) <op> y, where op is a '^' or a '|'. It also
>> transforms (cond (and (x , 0x1) != 0), (z op y), y ) into (-(and (x ,
>> 0x1)) & z ) op y.
>>
>> Matching this patterns allows GCC to generate branchless code for one of
>> the functions in coremark.
>>
>> Bootstrapped and tested on x86 and RISC-V. Okay?
> OK.
>
> Thanks,
> Richard.
>
>> Michael.
>>
>> 2022-11-10  Michael Collison  <collison@rivosinc.com>
>>
>>       * match.pd ((x & 0x1) == 0) ? y : z <op> y
>>       -> (-(typeof(y))(x & 0x1) & z) <op> y.
>>
>> 2022-11-10  Michael Collison <collison@rivosinc.com>
>>
>>       * gcc.dg/tree-ssa/branchless-cond.c: New test.
>>
>> ---
>>
>> Changes in v2:
>>
>> - Rewrite comment to use C syntax
>>
>> - Guard against 1-bit types
>>
>> - Simplify pattern by using zero_one_valued_p
>>
>>    gcc/match.pd                                  | 24 +++++++++++++++++
>>    .../gcc.dg/tree-ssa/branchless-cond.c         | 26 +++++++++++++++++++
>>    2 files changed, 50 insertions(+)
>>    create mode 100644 gcc/testsuite/gcc.dg/tree-ssa/branchless-cond.c
>>
>> diff --git a/gcc/match.pd b/gcc/match.pd
>> index 194ba8f5188..258531e9046 100644
>> --- a/gcc/match.pd
>> +++ b/gcc/match.pd
>> @@ -3486,6 +3486,30 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
>>      (cond (le @0 integer_zerop@1) (negate@2 @0) integer_zerop@1)
>>      (max @2 @1))
>>
>> +/* ((x & 0x1) == 0) ? y : z <op> y -> (-(typeof(y))(x & 0x1) & z) <op> y */
>> +(for op (bit_xor bit_ior)
>> + (simplify
>> +  (cond (eq zero_one_valued_p@0
>> +            integer_zerop)
>> +        @1
>> +        (op:c @2 @1))
>> +  (if (INTEGRAL_TYPE_P (type)
>> +       && TYPE_PRECISION (type) > 1
>> +       && (INTEGRAL_TYPE_P (TREE_TYPE (@0))))
>> +       (op (bit_and (negate (convert:type @0)) @2) @1))))
>> +
>> +/* ((x & 0x1) == 0) ? z <op> y : y -> (-(typeof(y))(x & 0x1) & z) <op> y */
>> +(for op (bit_xor bit_ior)
>> + (simplify
>> +  (cond (ne zero_one_valued_p@0
>> +            integer_zerop)
>> +       (op:c @2 @1)
>> +        @1)
>> +  (if (INTEGRAL_TYPE_P (type)
>> +       && TYPE_PRECISION (type) > 1
>> +       && (INTEGRAL_TYPE_P (TREE_TYPE (@0))))
>> +       (op (bit_and (negate (convert:type @0)) @2) @1))))
>> +
>>    /* Simplifications of shift and rotates.  */
>>
>>    (for rotate (lrotate rrotate)
>> diff --git a/gcc/testsuite/gcc.dg/tree-ssa/branchless-cond.c b/gcc/testsuite/gcc.dg/tree-ssa/branchless-cond.c
>> new file mode 100644
>> index 00000000000..68087ae6568
>> --- /dev/null
>> +++ b/gcc/testsuite/gcc.dg/tree-ssa/branchless-cond.c
>> @@ -0,0 +1,26 @@
>> +/* { dg-do compile } */
>> +/* { dg-options "-O2 -fdump-tree-optimized" } */
>> +
>> +int f1(unsigned int x, unsigned int y, unsigned int z)
>> +{
>> +  return ((x & 1) == 0) ? y : z ^ y;
>> +}
>> +
>> +int f2(unsigned int x, unsigned int y, unsigned int z)
>> +{
>> +  return ((x & 1) != 0) ? z ^ y : y;
>> +}
>> +
>> +int f3(unsigned int x, unsigned int y, unsigned int z)
>> +{
>> +  return ((x & 1) == 0) ? y : z | y;
>> +}
>> +
>> +int f4(unsigned int x, unsigned int y, unsigned int z)
>> +{
>> +  return ((x & 1) != 0) ? z | y : y;
>> +}
>> +
>> +/* { dg-final { scan-tree-dump-times " -" 4 "optimized" } } */
>> +/* { dg-final { scan-tree-dump-times " & " 8 "optimized" } } */
>> +/* { dg-final { scan-tree-dump-not "if" "optimized" } } */
>> --
>> 2.34.1
>>