From 8968e92cc65ef4cfe9dbba86b7873490139100b0 Mon Sep 17 00:00:00 2001
From: Kyrylo Tkachov <kyrylo.tkachov@arm.com>
Date: Mon, 19 Jun 2023 10:56:37 +0100
Subject: [PATCH] simplify-rtx: Simplify VEC_CONCAT of SUBREG and VEC_CONCAT
 from same vector

In the testcase for this patch we try to vec_concat the lowpart and highpart of a vector, but the lowpart is expressed as a subreg.
simplify-rtx.cc does not recognise this and combine ends up trying to match:
Trying 7 -> 8:
    7: r93:V2SI=vec_select(r95:V4SI,parallel)
    8: r97:V4SI=vec_concat(r95:V4SI#0,r93:V2SI)
      REG_DEAD r95:V4SI
      REG_DEAD r93:V2SI
Failed to match this instruction:
(set (reg:V4SI 97)
    (vec_concat:V4SI (subreg:V2SI (reg/v:V4SI 95 [ a ]) 0)
        (vec_select:V2SI (reg/v:V4SI 95 [ a ])
            (parallel:V4SI [
                    (const_int 2 [0x2])
                    (const_int 3 [0x3])
                ]))))

This should be just (set (reg:V4SI 97) (reg:V4SI 95)). This patch adds such a simplification.
The testcase is a bit artificial, but I do have other aarch64-specific patterns that I want to optimise later
that rely on this simplification happening.

Without this patch for the testcase we generate:
foo:
        dup     d31, v0.d[1]
        ins     v0.d[1], v31.d[0]
        ret

whereas we should just not generate anything as the operation is ultimately a no-op.

Bootstrapped and tested on aarch64-none-linux-gnu and aarch64_be-none-elf.

gcc/ChangeLog:

	* simplify-rtx.cc (simplify_context::simplify_binary_operation_1):
	Simplify vec_concat of lowpart subreg and high part vec_select.

gcc/testsuite/ChangeLog:

	* gcc.target/aarch64/simd/low-high-combine_1.c: New test.
---
 gcc/simplify-rtx.cc                           | 11 ++++++++
 .../aarch64/simd/low-high-combine_1.c         | 28 +++++++++++++++++++
 2 files changed, 39 insertions(+)
 create mode 100644 gcc/testsuite/gcc.target/aarch64/simd/low-high-combine_1.c

diff --git a/gcc/simplify-rtx.cc b/gcc/simplify-rtx.cc
index 21b7eb484d0..9c68d360672 100644
--- a/gcc/simplify-rtx.cc
+++ b/gcc/simplify-rtx.cc
@@ -4860,6 +4860,17 @@ simplify_ashift:
 	    return simplify_gen_binary (VEC_SELECT, mode, XEXP (trueop0, 0),
 					gen_rtx_PARALLEL (VOIDmode, vec));
 	  }
+	/* (vec_concat:
+	     (subreg_lowpart:N OP)
+	     (vec_select:N OP P))  -->  OP when P selects the high half
+	    of the OP.  */
+	if (GET_CODE (trueop0) == SUBREG
+	    && subreg_lowpart_p (trueop0)
+	    && GET_CODE (trueop1) == VEC_SELECT
+	    && SUBREG_REG (trueop0) == XEXP (trueop1, 0)
+	    && !side_effects_p (XEXP (trueop1, 0))
+	    && vec_series_highpart_p (op1_mode, mode, XEXP (trueop1, 1)))
+	  return XEXP (trueop1, 0);
       }
       return 0;
 
diff --git a/gcc/testsuite/gcc.target/aarch64/simd/low-high-combine_1.c b/gcc/testsuite/gcc.target/aarch64/simd/low-high-combine_1.c
new file mode 100644
index 00000000000..0b502d593f2
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/simd/low-high-combine_1.c
@@ -0,0 +1,28 @@
+/* { dg-do compile } */
+/* { dg-additional-options "-O" } */
+/* { dg-final { check-function-bodies "**" "" "" } } */
+
+#include <arm_neon.h>
+
+/*
+** foo_le:	{ target aarch64_little_endian }
+**	ret
+*/
+
+int32x4_t
+foo_le (int32x4_t a)
+{
+  return vcombine_s32 (vget_low_s32 (a), vget_high_s32 (a));
+}
+
+/*
+** foo_be:	{ target aarch64_big_endian }
+**	ret
+*/
+
+int32x4_t
+foo_be (int32x4_t a)
+{
+  return vcombine_s32 (vget_high_s32 (a), vget_low_s32 (a));
+}
+