i386: Add new insvti_lowpart_1 and insvdi_lowpart_1 patterns.

This patch implements another of Uros' suggestions, to investigate a
insvti_lowpart_1 pattern to improve TImode parameter passing on x86_64.
In PR 88873, the RTL the middle-end expands for passing V2DF in TImode
is subtly different from what it does for V2DI in TImode, sufficiently so
that my explanations for why insvti_lowpart_1 isn't required don't apply
in this case.

This patch adds an insvti_lowpart_1 pattern, complementing the existing
insvti_highpart_1 pattern, and also a 32-bit variant, insvdi_lowpart_1.
Because the middle-end represents 128-bit constants using CONST_WIDE_INT
and 64-bit constants using CONST_INT, it's easiest to treat these as
different patterns, rather than attempt <dwi> parameterization.

This patch also includes a peephole2 (actually a pair) to transform
xchg instructions into mov instructions, when one of the destinations
is unused.  This optimization is required to produce the optimal code
sequences below.

For the 64-bit case:

__int128 foo(__int128 x, unsigned long long y)
{
  __int128 m = ~((__int128)~0ull);
  __int128 t = x & m;
  __int128 r = t | y;
  return r;
}

Before:
        xchgq   %rdi, %rsi
        movq    %rdx, %rax
        xorl    %esi, %esi
        xorl    %edx, %edx
        orq     %rsi, %rax
        orq     %rdi, %rdx
        ret

After:
        movq    %rdx, %rax
        movq    %rsi, %rdx
        ret

For the 32-bit case:

long long bar(long long x, int y)
{
  long long mask = ~0ull << 32;
  long long t = x & mask;
  long long r = t | (unsigned int)y;
  return r;
}

Before:
        pushl   %ebx
        movl    12(%esp), %edx
        xorl    %ebx, %ebx
        xorl    %eax, %eax
        movl    16(%esp), %ecx
        orl     %ebx, %edx
        popl    %ebx
        orl     %ecx, %eax
        ret

After:
        movl    12(%esp), %eax
        movl    8(%esp), %edx
        ret

2023-07-10  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog
	* config/i386/i386.md (peephole2): Transform xchg insn with a
	REG_UNUSED note to a (simple) move.
	(*insvti_lowpart_1): New define_insn_and_split.
	(*insvdi_lowpart_1): Likewise.

gcc/testsuite/ChangeLog
	* gcc.target/i386/insvdi_lowpart-1.c: New test case.
	* gcc.target/i386/insvti_lowpart-1.c: Likewise.

3 files changed, 92 insertions, 0 deletions

diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
index 621cdd9..844deea 100644
--- a/gcc/config/i386/i386.md
+++ b/gcc/config/i386/i386.md
@@ -3243,6 +3243,30 @@
   [(parallel [(set (match_dup 1) (match_dup 2))
 	      (set (match_dup 2) (match_dup 1))])])
 
+;; Convert xchg with a REG_UNUSED note to a mov (variant #1).
+(define_peephole2
+  [(parallel [(set (match_operand:SWI 0 "general_reg_operand")
+		   (match_operand:SWI 1 "general_reg_operand"))
+	      (set (match_dup 1) (match_dup 0))])]
+  "((REGNO (operands[0]) != AX_REG
+     && REGNO (operands[1]) != AX_REG)
+    || optimize_size < 2
+    || !optimize_insn_for_size_p ())
+   && peep2_reg_dead_p (1, operands[0])"
+  [(set (match_dup 1) (match_dup 0))])
+
+;; Convert xchg with a REG_UNUSED note to a mov (variant #2).
+(define_peephole2
+  [(parallel [(set (match_operand:SWI 0 "general_reg_operand")
+		   (match_operand:SWI 1 "general_reg_operand"))
+	      (set (match_dup 1) (match_dup 0))])]
+  "((REGNO (operands[0]) != AX_REG
+     && REGNO (operands[1]) != AX_REG)
+    || optimize_size < 2
+    || !optimize_insn_for_size_p ())
+   && peep2_reg_dead_p (1, operands[1])"
+  [(set (match_dup 0) (match_dup 1))])
+
 ;; Convert moves to/from AX_REG into xchg with -Oz.
 (define_peephole2
   [(set (match_operand:SWI48 0 "general_reg_operand")
@@ -3573,6 +3597,48 @@
   split_double_concat (TImode, operands[0], operands[4], operands[2]);
   DONE;
 })
+
+(define_insn_and_split "*insvti_lowpart_1"
+  [(set (match_operand:TI 0 "nonimmediate_operand" "=ro,r,r,&r")
+	(any_or_plus:TI
+	  (and:TI
+	    (match_operand:TI 1 "nonimmediate_operand" "r,m,r,m")
+	    (match_operand:TI 3 "const_scalar_int_operand" "n,n,n,n"))
+	  (zero_extend:TI
+	    (match_operand:DI 2 "nonimmediate_operand" "r,r,m,m"))))]
+  "TARGET_64BIT
+   && CONST_WIDE_INT_P (operands[3])
+   && CONST_WIDE_INT_NUNITS (operands[3]) == 2
+   && CONST_WIDE_INT_ELT (operands[3], 0) == 0
+   && CONST_WIDE_INT_ELT (operands[3], 1) == -1"
+  "#"
+  "&& reload_completed"
+  [(const_int 0)]
+{
+  operands[4] = gen_highpart (DImode, operands[1]);
+  split_double_concat (TImode, operands[0], operands[2], operands[4]);
+  DONE;
+})
+
+(define_insn_and_split "*insvdi_lowpart_1"
+  [(set (match_operand:DI 0 "nonimmediate_operand" "=ro,r,r,&r")
+	(any_or_plus:DI
+	  (and:DI
+	    (match_operand:DI 1 "nonimmediate_operand" "r,m,r,m")
+	    (match_operand:DI 3 "const_int_operand" "n,n,n,n"))
+	  (zero_extend:DI
+	    (match_operand:SI 2 "nonimmediate_operand" "r,r,m,m"))))]
+  "!TARGET_64BIT
+   && CONST_INT_P (operands[3])
+   && UINTVAL (operands[3]) == 0xffffffff00000000ll"
+  "#"
+  "&& reload_completed"
+  [(const_int 0)]
+{
+  operands[4] = gen_highpart (SImode, operands[1]);
+  split_double_concat (DImode, operands[0], operands[2], operands[4]);
+  DONE;
+})
 
 ;; Floating point push instructions.
 
diff --git a/gcc/testsuite/gcc.target/i386/insvdi_lowpart-1.c b/gcc/testsuite/gcc.target/i386/insvdi_lowpart-1.c
new file mode 100644
index 0000000..4d94fec
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/insvdi_lowpart-1.c
@@ -0,0 +1,13 @@
+/* { dg-do compile { target ia32 } } */
+/* { dg-options "-O2" } */
+
+long long foo(long long x, int y)
+{
+  long long mask = ~0ull << 32;
+  long long t = x & mask;
+  long long r = t | (unsigned int)y;
+  return r;
+}
+
+/* { dg-final { scan-assembler-not "xorl" } } */
+/* { dg-final { scan-assembler-not "orq" } } */
diff --git a/gcc/testsuite/gcc.target/i386/insvti_lowpart-1.c b/gcc/testsuite/gcc.target/i386/insvti_lowpart-1.c
new file mode 100644
index 0000000..4e1fbbb
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/insvti_lowpart-1.c
@@ -0,0 +1,13 @@
+/* { dg-do compile { target int128 } } */
+/* { dg-options "-O2" } */
+
+__int128 foo(__int128 x, unsigned long long y)
+{
+  __int128 m = ~((__int128)~0ull);
+  __int128 t = x & m;
+  __int128 r = t | y;
+  return r;
+}
+
+/* { dg-final { scan-assembler-not "xorl" } } */
+/* { dg-final { scan-assembler-not "orq" } } */