i386: Add cbranchti4 pattern to i386.md (for -m32 compare_by_pieces).

This patch fixes some very odd (unanticipated) code generation by
compare_by_pieces with -m32 -mavx, since the recent addition of the
cbranchoi4 pattern.  The issue is that cbranchoi4 is available with
TARGET_AVX, but cbranchti4 is currently conditional on TARGET_64BIT
which results in the odd behaviour (thanks to OPTAB_WIDEN) that with
-m32 -mavx, compare_by_pieces ends up (inefficiently) widening 128-bit
comparisons to 256-bits before performing PTEST.

This patch fixes this by providing a cbranchti4 pattern that's available
with either TARGET_64BIT or TARGET_SSE4_1.

For the test case below (again from PR 104610):

int foo(char *a)
{
    static const char t[] = "0123456789012345678901234567890";
    return __builtin_memcmp(a, &t[0], sizeof(t)) == 0;
}

GCC with -m32 -O2 -mavx currently produces the bonkers:

foo:    pushl   %ebp
        movl    %esp, %ebp
        andl    $-32, %esp
        subl    $64, %esp
        movl    8(%ebp), %eax
        vmovdqa .LC0, %xmm4
        movl    $0, 48(%esp)
        vmovdqu (%eax), %xmm2
        movl    $0, 52(%esp)
        movl    $0, 56(%esp)
        movl    $0, 60(%esp)
        movl    $0, 16(%esp)
        movl    $0, 20(%esp)
        movl    $0, 24(%esp)
        movl    $0, 28(%esp)
        vmovdqa %xmm2, 32(%esp)
        vmovdqa %xmm4, (%esp)
        vmovdqa (%esp), %ymm5
        vpxor   32(%esp), %ymm5, %ymm0
        vptest  %ymm0, %ymm0
        jne     .L2
        vmovdqu 16(%eax), %xmm7
        movl    $0, 48(%esp)
        movl    $0, 52(%esp)
        vmovdqa %xmm7, 32(%esp)
        vmovdqa .LC1, %xmm7
        movl    $0, 56(%esp)
        movl    $0, 60(%esp)
        movl    $0, 16(%esp)
        movl    $0, 20(%esp)
        movl    $0, 24(%esp)
        movl    $0, 28(%esp)
        vmovdqa %xmm7, (%esp)
        vmovdqa (%esp), %ymm1
        vpxor   32(%esp), %ymm1, %ymm0
        vptest  %ymm0, %ymm0
        je      .L6
.L2:	movl    $1, %eax
        xorl    $1, %eax
        vzeroupper
        leave
        ret
.L6:	xorl    %eax, %eax
        xorl    $1, %eax
        vzeroupper
        leave
        ret

with this patch, we now generate the (slightly) more sensible:

foo:	vmovdqa .LC0, %xmm0
        movl    4(%esp), %eax
        vpxor   (%eax), %xmm0, %xmm0
        vptest  %xmm0, %xmm0
        jne     .L2
        vmovdqa .LC1, %xmm0
        vpxor   16(%eax), %xmm0, %xmm0
        vptest  %xmm0, %xmm0
        je      .L5
.L2:	movl    $1, %eax
        xorl    $1, %eax
        ret
.L5:	xorl    %eax, %eax
        xorl    $1, %eax
        ret

2023-06-28  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog
	* config/i386/i386-expand.cc (ix86_expand_branch): Also use ptest
	for TImode comparisons on 32-bit architectures.
	* config/i386/i386.md (cbranch<mode>4): Change from SDWIM to
	SWIM1248x to exclude/avoid TImode being conditional on -m64.
	(cbranchti4): New define_expand for TImode on both TARGET_64BIT
	and/or with TARGET_SSE4_1.
	* config/i386/predicates.md (ix86_timode_comparison_operator):
	New predicate that depends upon TARGET_64BIT.
	(ix86_timode_comparison_operand): Likewise.

gcc/testsuite/ChangeLog
	* gcc.target/i386/pieces-memcmp-2.c: New test case.

4 files changed, 45 insertions, 3 deletions

diff --git a/gcc/config/i386/i386-expand.cc b/gcc/config/i386/i386-expand.cc
index 9a8d244..567248d 100644
--- a/gcc/config/i386/i386-expand.cc
+++ b/gcc/config/i386/i386-expand.cc
@@ -2365,6 +2365,7 @@ ix86_expand_branch (enum rtx_code code, rtx op0, rtx op1, rtx label)
   /* Handle special case - vector comparsion with boolean result, transform
      it using ptest instruction.  */
   if (GET_MODE_CLASS (mode) == MODE_VECTOR_INT
+      || (mode == TImode && !TARGET_64BIT)
       || mode == OImode)
     {
       rtx flag = gen_rtx_REG (CCZmode, FLAGS_REG);
@@ -2372,7 +2373,7 @@ ix86_expand_branch (enum rtx_code code, rtx op0, rtx op1, rtx label)
 
       gcc_assert (code == EQ || code == NE);
 
-      if (mode == OImode)
+      if (GET_MODE_CLASS (mode) != MODE_VECTOR_INT)
 	{
 	  op0 = lowpart_subreg (p_mode, force_reg (mode, op0), mode);
 	  op1 = lowpart_subreg (p_mode, force_reg (mode, op1), mode);
diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
index 15c0310..a82cc35 100644
--- a/gcc/config/i386/i386.md
+++ b/gcc/config/i386/i386.md
@@ -1352,8 +1352,8 @@
 
 (define_expand "cbranch<mode>4"
   [(set (reg:CC FLAGS_REG)
-	(compare:CC (match_operand:SDWIM 1 "nonimmediate_operand")
-		    (match_operand:SDWIM 2 "<general_operand>")))
+	(compare:CC (match_operand:SWIM1248x 1 "nonimmediate_operand")
+		    (match_operand:SWIM1248x 2 "<general_operand>")))
    (set (pc) (if_then_else
 	       (match_operator 0 "ordered_comparison_operator"
 		[(reg:CC FLAGS_REG) (const_int 0)])
@@ -1368,6 +1368,22 @@
   DONE;
 })
 
+(define_expand "cbranchti4"
+  [(set (reg:CC FLAGS_REG)
+	(compare:CC (match_operand:TI 1 "nonimmediate_operand")
+		    (match_operand:TI 2 "ix86_timode_comparison_operand")))
+   (set (pc) (if_then_else
+	       (match_operator 0 "ix86_timode_comparison_operator"
+		[(reg:CC FLAGS_REG) (const_int 0)])
+	       (label_ref (match_operand 3))
+	       (pc)))]
+  "TARGET_64BIT || TARGET_SSE4_1"
+{
+  ix86_expand_branch (GET_CODE (operands[0]),
+		      operands[1], operands[2], operands[3]);
+  DONE;
+})
+
 (define_expand "cbranchoi4"
   [(set (reg:CC FLAGS_REG)
 	(compare:CC (match_operand:OI 1 "nonimmediate_operand")
diff --git a/gcc/config/i386/predicates.md b/gcc/config/i386/predicates.md
index fb07707..7ddbe01 100644
--- a/gcc/config/i386/predicates.md
+++ b/gcc/config/i386/predicates.md
@@ -1641,6 +1641,18 @@
                (match_operand 0 "comparison_operator")
                (match_operand 0 "ix86_trivial_fp_comparison_operator")))
 
+;; Return true if we can perform this comparison on TImode operands.
+(define_predicate "ix86_timode_comparison_operator"
+  (if_then_else (match_test "TARGET_64BIT")
+		(match_operand 0 "ordered_comparison_operator")
+		(match_operand 0 "bt_comparison_operator")))
+
+;; Return true if this is a valid second operand for a TImode comparison.
+(define_predicate "ix86_timode_comparison_operand"
+  (if_then_else (match_test "TARGET_64BIT")
+		(match_operand 0 "x86_64_general_operand")
+		(match_operand 0 "nonimmediate_operand")))
+
 ;; Nearly general operand, but accept any const_double, since we wish
 ;; to be able to drop them into memory rather than have them get pulled
 ;; into registers.
diff --git a/gcc/testsuite/gcc.target/i386/pieces-memcmp-2.c b/gcc/testsuite/gcc.target/i386/pieces-memcmp-2.c
new file mode 100644
index 0000000..6f996fa
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pieces-memcmp-2.c
@@ -0,0 +1,13 @@
+/* { dg-do compile { target ia32 } } */
+/* { dg-options "-O2 -mavx2" } */
+
+int foo(char *a)
+{
+    static const char t[] = "0123456789012345678901234567890";
+    return __builtin_memcmp(a, &t[0], sizeof(t)) == 0;
+}
+
+/* { dg-final { scan-assembler-not "movl\[ \\t]*\\\$0," } } */
+/* { dg-final { scan-assembler-not "vptest\[ \\t]*%ymm" } } */
+/* { dg-final { scan-assembler-times "vptest\[ \\t]*%xmm" 2 } } */
+