https://gcc.gnu.org/g:cdb5b1eb1115600dabcc2ba4ffa639eef3e2a7b1

commit r15-4252-gcdb5b1eb1115600dabcc2ba4ffa639eef3e2a7b1
Author: Pan Li <pan2...@intel.com>
Date:   Wed Oct 9 22:33:10 2024 +0800

    Match: Support form 3 for scalar signed integer SAT_TRUNC
    
    This patch would like to support the form 3 of the scalar signed
    integer SAT_TRUNC.  Aka below example:
    
    Form 3:
      #define DEF_SAT_S_TRUNC_FMT_3(NT, WT, NT_MIN, NT_MAX) \
      NT __attribute__((noinline))                          \
      sat_s_trunc_##WT##_to_##NT##_fmt_3 (WT x)             \
      {                                                     \
        NT trunc = (NT)x;                                   \
        return (WT)NT_MIN < x && x <= (WT)NT_MAX            \
          ? trunc                                           \
          : x < 0 ? NT_MIN : NT_MAX;                        \
      }
    
    DEF_SAT_S_TRUNC_FMT_3(int8_t, int16_t, INT8_MIN, INT8_MAX)
    
    Before this patch:
       4   │ __attribute__((noinline))
       5   │ int8_t sat_s_sub_int8_t_fmt_3 (int8_t x, int8_t y)
       6   │ {
       7   │   signed char _1;
       8   │   signed char _2;
       9   │   int8_t _3;
      10   │   __complex__ signed char _6;
      11   │   _Bool _8;
      12   │   signed char _9;
      13   │   signed char _10;
      14   │   signed char _11;
      15   │
      16   │ ;;   basic block 2, loop depth 0
      17   │ ;;    pred:       ENTRY
      18   │   _6 = .SUB_OVERFLOW (x_4(D), y_5(D));
      19   │   _2 = IMAGPART_EXPR <_6>;
      20   │   if (_2 != 0)
      21   │     goto <bb 4>; [50.00%]
      22   │   else
      23   │     goto <bb 3>; [50.00%]
      24   │ ;;    succ:       4
      25   │ ;;                3
      26   │
      27   │ ;;   basic block 3, loop depth 0
      28   │ ;;    pred:       2
      29   │   _1 = REALPART_EXPR <_6>;
      30   │   goto <bb 5>; [100.00%]
      31   │ ;;    succ:       5
      32   │
      33   │ ;;   basic block 4, loop depth 0
      34   │ ;;    pred:       2
      35   │   _8 = x_4(D) < 0;
      36   │   _9 = (signed char) _8;
      37   │   _10 = -_9;
      38   │   _11 = _10 ^ 127;
      39   │ ;;    succ:       5
      40   │
      41   │ ;;   basic block 5, loop depth 0
      42   │ ;;    pred:       3
      43   │ ;;                4
      44   │   # _3 = PHI <_1(3), _11(4)>
      45   │   return _3;
      46   │ ;;    succ:       EXIT
      47   │
      48   │ }
    
    After this patch:
       4   │ __attribute__((noinline))
       5   │ int8_t sat_s_trunc_int16_t_to_int8_t_fmt_3 (int16_t x)
       6   │ {
       7   │   int8_t _3;
       8   │
       9   │ ;;   basic block 2, loop depth 0
      10   │ ;;    pred:       ENTRY
      11   │   _3 = .SAT_TRUNC (x_4(D)); [tail call]
      12   │   return _3;
      13   │ ;;    succ:       EXIT
      14   │
      15   │ }
    
    The below test suites are passed for this patch.
    * The rv64gcv fully regression test.
    * The x86 bootstrap test.
    * The x86 fully regression test.
    
    gcc/ChangeLog:
    
            * match.pd: Add case 3 matching pattern for signed SAT_TRUNC.
    
    Signed-off-by: Pan Li <pan2...@intel.com>

Diff:
---
 gcc/match.pd | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/gcc/match.pd b/gcc/match.pd
index 754c1f39d8b5..33886bd58d7e 100644
--- a/gcc/match.pd
+++ b/gcc/match.pd
@@ -3478,6 +3478,8 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
    wide_int limit_0 = wi::mask (otype_prec, false, itype_prec); // Aka 255
    wide_int limit_1 = wi::uhwi ((HOST_WIDE_INT_1U << otype_prec) - 3,
                                itype_prec); // Aka 253
+   wide_int limit_2 = wi::uhwi ((HOST_WIDE_INT_1U << otype_prec) - 2,
+                               itype_prec); // Aka 254
    wide_int otype_max = wi::mask (otype_prec - 1, false, otype_prec);
    wide_int itype_max = wi::mask (otype_prec - 1, false, itype_prec);
    wide_int int_cst_1 = wi::to_wide (@1);
@@ -3485,6 +3487,7 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
    wide_int int_cst_3 = wi::to_wide (@3);
   }
   (if (((wi::eq_p (int_cst_1, offset) && wi::eq_p (int_cst_2, limit_0))
+        || (wi::eq_p (int_cst_1, itype_max) && wi::eq_p (int_cst_2, limit_2))
         || (wi::eq_p (int_cst_1, itype_max) && wi::eq_p (int_cst_2, limit_1)))
        && wi::eq_p (int_cst_3, otype_max))))))

Reply via email to