tree-optimization/121405 - missed VN with aggregate copy

The following handles value-numbering of a BIT_FIELD_REF of
a register that's defined by a load by looking up a subset
load similar to how we handle bit-and masked loads.  This
allows the testcase to be simplified by two FRE passes,
the first one will create the BIT_FIELD_REF.

	PR tree-optimization/121405
	* tree-ssa-sccvn.cc (visit_nary_op): Handle BIT_FIELD_REF
	with reference def by looking up a combination of both.

	* gcc.dg/tree-ssa/ssa-fre-107.c: New testcase.
	* gcc.target/i386/pr90579.c: Adjust.

1 files changed, 29 insertions, 0 deletions

diff --git a/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-107.c b/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-107.c
new file mode 100644
index 0000000..f80baf3
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-107.c
@@ -0,0 +1,29 @@
+/* { dg-do compile } */
+/* { dg-options "-O -fdump-tree-optimized" } */
+
+struct vec_char_16
+{
+  unsigned char raw[2];
+};
+
+static inline struct vec_char_16
+Dup128VecFromValues(unsigned char t0, unsigned char t1)
+{
+  struct vec_char_16 result;
+  result.raw[0] = t0;
+  result.raw[1] = t1;
+  return result;
+}
+
+int f(unsigned char t0, unsigned char t1)
+{
+  struct vec_char_16 a = Dup128VecFromValues(t0, t1);
+  struct vec_char_16 b;
+  __builtin_memcpy(&b, &a, sizeof(a));
+  return b.raw[0] + b.raw[1];
+}
+
+/* Ideally we'd optimize this at FRE1 time but we only replace
+   the loads from b.raw[] with BIT_FIELD_REFs which get optimized
+   only later in the next FRE.  */
+/* { dg-final { scan-tree-dump-not "MEM" "optimized" } } */