[PATCH] i386: Add peephole for lea and zero extend [PR 101716]

[Uros Bizjak]

On Fri, Aug 13, 2021 at 2:48 AM Hongyu Wang <hongyu.wang@intel.com> wrote:
>
> Hi,
>
> For lea + zero_extendsidi insns, if dest of lea and src of zext are the
> same, combine them with single leal under 64bit target since 32bit
> register will be automatically zero-extended.
>
> Bootstrapped and regtested on x86_64-linux-gnu{-m32,}.
> Ok for master?
>
> gcc/ChangeLog:
>
>         PR target/101716
>         * config/i386/i386.md (*lea<mode>_zext): New define_insn.
>         (define_peephole2): New peephole2 to combine zero_extend
>         with lea.
>
> gcc/testsuite/ChangeLog:
>
>         PR target/101716
>         * gcc.target/i386/pr101716.c: New test.

This form should be covered by ix86_decompose_address via
address_no_seg_operand predicate. Combine creates:

Trying 6 -> 7:
   6: {r86:DI=r87:DI<<0x1;clobber flags:CC;}
     REG_DEAD r87:DI
     REG_UNUSED flags:CC
   7: r85:DI=zero_extend(r86:DI#0)
     REG_DEAD r86:DI
Failed to match this instruction:
(set (reg:DI 85)
   (and:DI (ashift:DI (reg:DI 87)
           (const_int 1 [0x1]))
       (const_int 4294967294 [0xfffffffe])))

which does not fit:

      else if (GET_CODE (addr) == AND
           && const_32bit_mask (XEXP (addr, 1), DImode))

After reload, we lose SUBREG, so REE does not trigger on:

(insn 17 3 7 2 (set (reg:DI 0 ax [86])
       (mult:DI (reg:DI 5 di [87])
           (const_int 2 [0x2]))) "pr101716.c":4:13 204 {*leadi}
    (nil))
(insn 7 17 13 2 (set (reg:DI 0 ax [85])
       (zero_extend:DI (reg:SI 0 ax [86]))) "pr101716.c":4:19 136
{*zero_extendsidi2}
    (nil))

So, the question is if the combine pass really needs to zero-extend
with 0xfffffffe, the left shift << 1 guarantees zero in the LSB, so
0xffffffff should be better and in line with canonical zero-extension
RTX.

> ---
>  gcc/config/i386/i386.md                  | 20 ++++++++++++++++++++
>  gcc/testsuite/gcc.target/i386/pr101716.c | 11 +++++++++++
>  2 files changed, 31 insertions(+)
>  create mode 100644 gcc/testsuite/gcc.target/i386/pr101716.c
>
> diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
> index 4a8e8fea290..6739dbd799b 100644
> --- a/gcc/config/i386/i386.md
> +++ b/gcc/config/i386/i386.md
> @@ -5187,6 +5187,26 @@
>         (const_string "SI")
>         (const_string "<MODE>")))])
>
> +;; combine zero_extendsidi with lea to use leal.
> +(define_insn "*lea<mode>_zext"
> +  [(set (match_operand:DI 0 "register_operand" "=r")
> +       (zero_extend:DI
> +           (match_operand:SWI48 1 "address_no_seg_operand" "Ts")))]
> +  "TARGET_64BIT"
> +  "lea{l}\t{%E1, %k0|%k0,%E1}")

The above can lead to invalid RTX: (zero_extend:DI (... DImode RTX)).

Uros.

> +
> +(define_peephole2
> +  [(set (match_operand:SWI48 0 "general_reg_operand")
> +       (match_operand:SWI48 1 "address_no_seg_operand"))
> +   (set (match_operand:DI 2 "general_reg_operand")
> +       (zero_extend:DI (match_operand:SI 3 "general_reg_operand")))]
> +  "TARGET_64BIT && ix86_hardreg_mov_ok (operands[2], operands[1])
> +   && REGNO (operands[0]) == REGNO (operands[3])
> +   && (REGNO (operands[2]) == REGNO (operands[3])
> +      || peep2_reg_dead_p (2, operands[3]))"
> +  [(set (match_dup 2)
> +       (zero_extend:DI (match_dup 1)))])
> +
>  (define_peephole2
>    [(set (match_operand:SWI48 0 "register_operand")
>         (match_operand:SWI48 1 "address_no_seg_operand"))]
> diff --git a/gcc/testsuite/gcc.target/i386/pr101716.c b/gcc/testsuite/gcc.target/i386/pr101716.c
> new file mode 100644
> index 00000000000..0b684755c2f
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/pr101716.c
> @@ -0,0 +1,11 @@
> +/* PR target/101716 */
> +/* { dg-do compile { target { ! ia32 } } } */
> +/* { dg-options "-O2" } */
> +
> +/* { dg-final { scan-assembler "leal\[\\t \]\*eax" } } */
> +/* { dg-final { scan-assembler-not "movl\[\\t \]\*eax" } } */
> +
> +unsigned long long sample1(unsigned long long m) {
> +    unsigned int t = -1;
> +    return (m << 1) & t;
> +}
> --
> 2.18.1
>