diff mbox series

rs6000: optimize li+rldicr+cmpd==>rotldi+cmpldi for 16bits

Message ID 20220317104740.4131405-1-guojiufu@linux.ibm.com
State New
Headers show
Series rs6000: optimize li+rldicr+cmpd==>rotldi+cmpldi for 16bits | expand

Commit Message

Jiufu Guo March 17, 2022, 10:47 a.m. UTC
When checking eq/neq with a constant which has only 16bits, then it can
be optimized to check the rotated data.  By this, the constant building
is optimized.

As the example in PR103743:
For "in == 0x8000000000000000LL", this patch generates:
	rotldi %r3,%r3,16
        cmpldi %cr0,%r3,32768
instead:
        li %r9,-1
        rldicr %r9,%r9,0,0
        cmpd %cr0,%r3,%r9

This patch pass bootstrap and regtest on ppc64 and ppc64le.
Ok for trunk?  Thanks!

BR,
Jiufu


	PR target/103743

gcc/ChangeLog:

	* config/rs6000/rs6000.cc (rot_to_16bits): New function.
	(rs6000_generate_compare): Compare rotated const for eq.

gcc/testsuite/ChangeLog:

	* gcc.target/powerpc/pr103743.c: New test.
	* gcc.target/powerpc/pr103743_1.c: New test.

---
 gcc/config/rs6000/rs6000.cc                   | 62 +++++++++++++
 gcc/testsuite/gcc.target/powerpc/pr103743.c   | 48 ++++++++++
 gcc/testsuite/gcc.target/powerpc/pr103743_1.c | 87 +++++++++++++++++++
 3 files changed, 197 insertions(+)
 create mode 100644 gcc/testsuite/gcc.target/powerpc/pr103743.c
 create mode 100644 gcc/testsuite/gcc.target/powerpc/pr103743_1.c
diff mbox series

Patch

diff --git a/gcc/config/rs6000/rs6000.cc b/gcc/config/rs6000/rs6000.cc
index 3afe78f5d04..4aa1d0ca4ab 100644
--- a/gcc/config/rs6000/rs6000.cc
+++ b/gcc/config/rs6000/rs6000.cc
@@ -14860,6 +14860,42 @@  rs6000_reverse_condition (machine_mode mode, enum rtx_code code)
     return reverse_condition (code);
 }
 
+/* If SRC is a constant with only 16bits, return the number by which
+   the constant can be rotated to lowest 16bits.
+   Return 0, if SRC does not meet requirements.
+   Set *sgn to 1 if the rotated cst is negative, otherwise set it to 0.
+   Set *res_cst to the rotated constant.  */
+
+static int
+rot_to_16bits (rtx src, machine_mode mode, bool *sgn, rtx *res_cst)
+{
+  if (!(src && CONST_INT_P (src)))
+    return 0;
+
+  unsigned HOST_WIDE_INT C = INTVAL (src);
+
+  /* If all 0 except low 16bits.  */
+  int leadz = clz_hwi (C) + 16;
+  rtx cst = simplify_gen_binary (ROTATE, mode, src, GEN_INT (leadz));
+  if (satisfies_constraint_K (cst))
+    {
+      *res_cst = cst;
+      return leadz;
+    }
+
+  /* If all 1 except low 15bits.  */
+  leadz = clz_hwi (~C) + 15;
+  cst = simplify_gen_binary (ROTATE, mode, src, GEN_INT (leadz));
+  if (satisfies_constraint_I (cst))
+    {
+      *sgn = true;
+      *res_cst = cst;
+      return leadz;
+    }
+
+  return 0;
+}
+
 /* Generate a compare for CODE.  Return a brand-new rtx that
    represents the result of the compare.  */
 
@@ -14889,6 +14925,32 @@  rs6000_generate_compare (rtx cmp, machine_mode mode)
   else
     comp_mode = CCmode;
 
+  /* "i == C" ==> "rotl(i,N) == rotl(C,N)" if rotl(C,N) only low 16bits.  */
+  if ((code == NE || code == EQ) && mode == DImode)
+    {
+      /* The constant would already been set to a reg in the last insn.  */
+      rtx_insn *last = get_last_insn_anywhere ();
+      rtx set = last ? single_set (last) : NULL_RTX;
+      rtx src = (set && SET_DEST (set) == op1) ? SET_SRC (set) : NULL_RTX;
+
+      /* It constant may be in constant pool. */
+      if (src && MEM_P (src))
+	src = avoid_constant_pool_reference (src);
+
+      rtx cst = NULL_RTX;
+      bool sgn = false;
+      int n = rot_to_16bits (src, mode, &sgn, &cst);
+      if (n >= 16 && n <= 48)
+	{
+	  rtx dest = gen_reg_rtx (mode);
+	  emit_insn (
+	    gen_rtx_SET (dest, gen_rtx_ROTATE (mode, op0, GEN_INT (n))));
+	  op0 = dest;
+	  op1 = cst;
+	  comp_mode = sgn ? CCmode : CCUNSmode;
+	}
+    }
+
   /* If we have an unsigned compare, make sure we don't have a signed value as
      an immediate.  */
   if (comp_mode == CCUNSmode && CONST_INT_P (op1)
diff --git a/gcc/testsuite/gcc.target/powerpc/pr103743.c b/gcc/testsuite/gcc.target/powerpc/pr103743.c
new file mode 100644
index 00000000000..606e5e51c4f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/pr103743.c
@@ -0,0 +1,48 @@ 
+/* { dg-do compile } */
+/* { dg-require-effective-target lp64 } */
+/* { dg-options "-O2" } */
+/* { dg-final { scan-assembler-times "cmpldi" 8  } } */
+/* { dg-final { scan-assembler-times "cmpdi" 4  } } */
+/* { dg-final { scan-assembler-times "rotldi" 8  } } */
+
+int foo (int a);
+
+int __attribute__ ((noinline)) udi_fun (unsigned long long in)
+{
+  if (in == (0x8642000000000000ULL))
+    return foo (1);
+  if (in == (0x7642000000000000ULL))
+    return foo (12);
+  if (in == (0x8000000000000000ULL))
+    return foo (32);
+  if (in == (0x8642FFFFFFFFFFFFULL))
+    return foo (46);
+  if (in == (0x7642FFFFFFFFFFFFULL))
+    return foo (51);
+  if (in == (0x7567000000ULL))
+    return foo (9);
+  if (in == (0xFFF8567FFFFFFFFFULL))
+    return foo (19);
+
+  return 0;
+}
+
+int __attribute__ ((noinline)) di_fun (long long in)
+{
+  if (in == (0x8642000000000000LL))
+    return foo (1);
+  if (in == (0x7642000000000000LL))
+    return foo (12);
+  if (in == (0x8000000000000000LL))
+    return foo (32);
+  if (in == (0x8642FFFFFFFFFFFFLL))
+    return foo (46);
+  if (in == (0x7642FFFFFFFFFFFFLL))
+    return foo (51);
+  if (in == (0x7567000000LL))
+    return foo (9);
+  if (in == (0xFFF8567FFFFFFFFFLL))
+    return foo (19);
+
+  return 0;
+}
diff --git a/gcc/testsuite/gcc.target/powerpc/pr103743_1.c b/gcc/testsuite/gcc.target/powerpc/pr103743_1.c
new file mode 100644
index 00000000000..56fbbe4d4f1
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/pr103743_1.c
@@ -0,0 +1,87 @@ 
+/* { dg-do run } */
+/* { dg-options "-O2 -std=c99" } */
+
+int
+foo (int a)
+{
+  return a + 6;
+}
+
+int __attribute__ ((noinline)) udi_fun (unsigned long long in)
+{
+  if (in == (0x8642000000000000ULL))
+    return foo (1);
+  if (in == (0x7642000000000000ULL))
+    return foo (12);
+  if (in == (0x8000000000000000ULL))
+    return foo (32);
+  if (in == (0x8642FFFFFFFFFFFFULL))
+    return foo (46);
+  if (in == (0x7642FFFFFFFFFFFFULL))
+    return foo (51);
+  if (in == (0x7567000000ULL))
+    return foo (9);
+  if (in == (0xFFF8567FFFFFFFFFULL))
+    return foo (19);
+  
+  return 0;
+}
+
+int __attribute__ ((noinline)) di_fun (long long in)
+{
+  if (in == (0x8642000000000000LL))
+    return foo (1);
+  if (in == (0x7642000000000000LL))
+    return foo (12);
+  if (in == (0x8000000000000000LL))
+    return foo (32);
+  if (in == (0x8642FFFFFFFFFFFFLL))
+    return foo (46);
+  if (in == (0x7642FFFFFFFFFFFFLL))
+    return foo (51);
+  return 0;
+}
+
+int
+main ()
+{
+  int e = 0;
+  if (udi_fun (6) != 0)
+    e++;
+  if (udi_fun (0x8642000000000000ULL) != foo (1))
+    e++;
+  if (udi_fun (0x7642000000000000ULL) != foo (12))
+    e++;
+  if (udi_fun (0x8000000000000000ULL) != foo (32))
+    e++;
+  if (udi_fun (0x8642FFFFFFFFFFFFULL) != foo (46))
+    e++;
+  if (udi_fun (0x7642FFFFFFFFFFFFULL) != foo (51))
+    e++;
+  if (udi_fun (0x7567000000ULL) != foo (9))
+    e++;
+  if (udi_fun (0xFFF8567FFFFFFFFFULL) != foo (19))
+    e++;
+
+  if (di_fun (6) != 0)
+    e++;
+  if (di_fun (0x8642000000000000LL) != foo (1))
+    e++;
+  if (di_fun (0x7642000000000000LL) != foo (12))
+    e++;
+  if (di_fun (0x8000000000000000LL) != foo (32))
+    e++;
+  if (di_fun (0x8642FFFFFFFFFFFFLL) != foo (46))
+    e++;
+  if (di_fun (0x7642FFFFFFFFFFFFLL) != foo (51))
+    e++;
+  if (udi_fun (0x7567000000LL) != foo (9))
+    e++;
+  if (udi_fun (0xFFF8567FFFFFFFFFLL) != foo (19))
+    e++;
+
+  if (e)
+    __builtin_abort ();
+  return 0;
+}
+