forwprop: Recongize a store of integral zero for optimize_aggr_zeroprop.

While looking into the gimple level after optimization of the highway code
from google, I noticed in .optimized we still have:
```
  MEM <vector(8) short int> [(short int *)&a] = { 0, 0, 0, 0, 0, 0, 0, 0 };
  D.4398 = a;
  a ={v} {CLOBBER(eos)};
  D.4389 = D.4398;
  D.4390 = D.4389;
  D.4361 = D.4390;
  D.4195 = D.4361;
  return D.4195;
```
Note this is with SRA disabled since I noticed there is better code generation with
SRA disabled but that is a different story and I will get to that later on.

Which could be just optimized to a single store of `{}` .

The reason why the optimize_agr_copyprop does not handle the above is there was clobbers
inbetween the store in the last forwprop pass and currently don't copy after the first use.
While optimize_aggr_zeroprop does handle copying over clobbers just fine.

So this allows the recognization of the store to a to be like a memset to optimize_aggr_zeroprop
and then the result just falls through.

Bootstrapped and tested on x86_64-linux-gnu.

gcc/ChangeLog:

	* tree-ssa-forwprop.cc (optimize_aggr_zeroprop): Recognize stores
	of integer_zerop as memset of 0.

gcc/testsuite/ChangeLog:

	* gcc.dg/torture/copy-prop-aggr-zero-1.c: New test.
	* gcc.dg/torture/copy-prop-aggr-zero-2.c: New test.
	* gcc.dg/tree-ssa/copy-prop-aggregate-zero-1.c: New test.
	* gcc.dg/tree-ssa/copy-prop-aggregate-zero-2.c: New test.
	* gcc.dg/tree-ssa/copy-prop-aggregate-zero-3.c: New test.

Signed-off-by: Andrew Pinski <quic_apinski@quicinc.com>

diff --git a/gcc/testsuite/gcc.dg/torture/copy-prop-aggr-zero-1.c b/gcc/testsuite/gcc.dg/torture/copy-prop-aggr-zero-1.c
new file mode 100644 (file)
index 0000000..5c457b9
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/torture/copy-prop-aggr-zero-1.c
@@ -0,0 +1,28 @@
+/* { dg-do run } */
+
+/* Make sure a bit-field store of 0 cause the whole assignment become 0. */
+
+struct s1
+{
+  unsigned char c:1;
+  unsigned char d:7;
+};
+
+__attribute__((noinline))
+struct s1 f(struct s1 a)
+{
+  a.c = 0;
+  struct s1 t = a;
+  return t;
+}
+
+int main()
+{
+  struct s1 a = {1, 2};
+  struct s1 b = f(a);
+  if (b.c != 0)
+    __builtin_abort();
+  if (b.d != 2)
+    __builtin_abort();
+  return 0;
+}

diff --git a/gcc/testsuite/gcc.dg/torture/copy-prop-aggr-zero-2.c b/gcc/testsuite/gcc.dg/torture/copy-prop-aggr-zero-2.c
new file mode 100644 (file)
index 0000000..f1da161
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/torture/copy-prop-aggr-zero-2.c
@@ -0,0 +1,28 @@
+/* { dg-do run } */
+
+/* Make sure a bit-field store of 0 cause the whole assignment become 0. */
+
+struct s1
+{
+  unsigned char d:7;
+  unsigned char c:1;
+};
+
+__attribute__((noinline))
+struct s1 f(struct s1 a)
+{
+  a.c = 0;
+  struct s1 t = a;
+  return t;
+}
+
+int main()
+{
+  struct s1 a = {2, 1};
+  struct s1 b = f(a);
+  if (b.c != 0)
+    __builtin_abort();
+  if (b.d != 2)
+    __builtin_abort();
+  return 0;
+}

diff --git a/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-1.c b/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-1.c
new file mode 100644 (file)
index 0000000..577a5b5
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-1.c
@@ -0,0 +1,28 @@
+/* { dg-do compile } */ 
+/* { dg-options "-O1 -fno-tree-sra -fdump-tree-optimized -fdump-tree-forwprop1-details" } */
+
+extern void link_error (void);
+
+/* Check for copyprop on structs with zeroing.  */
+#define vector16 __attribute__((vector_size(64)))
+
+struct g
+{
+  vector16 unsigned char t;
+};
+
+struct g f(void)
+{
+  struct g temp_struct1 ;
+  temp_struct1.t = (vector16 unsigned char){};
+  struct g temp_struct2 = temp_struct1;
+  struct g temp_struct3 = temp_struct2;
+  struct g temp_struct4 = temp_struct3;
+  return temp_struct4;
+}
+
+/* There should be no references to any of "temp_struct*"
+   temporaries.  */
+/* { dg-final { scan-tree-dump-times "temp_struct" 0 "optimized" } } */
+/* Also check that forwprop pass did the copy prop. */
+/* { dg-final { scan-tree-dump-times "after previous" 4 "forwprop1" } } */

diff --git a/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-2.c b/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-2.c
new file mode 100644 (file)
index 0000000..ce3c612
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-2.c
@@ -0,0 +1,25 @@
+/* { dg-do compile } */ 
+/* { dg-options "-O1 -fno-tree-sra -fdump-tree-optimized -fdump-tree-forwprop1-details" } */
+
+extern void link_error (void);
+
+struct g
+{
+  unsigned int t;
+};
+
+struct g f(void)
+{
+  struct g temp_struct1 ;
+  temp_struct1.t = 0;
+  struct g temp_struct2 = temp_struct1;
+  struct g temp_struct3 = temp_struct2;
+  struct g temp_struct4 = temp_struct3;
+  return temp_struct4;
+}
+
+/* There should be no references to any of "temp_struct*"
+   temporaries.  */
+/* { dg-final { scan-tree-dump-times "temp_struct" 0 "optimized" } } */
+/* Also check that forwprop pass did the copy prop. */
+/* { dg-final { scan-tree-dump-times "after previous" 4 "forwprop1" } } */

diff --git a/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-3.c b/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-3.c
new file mode 100644 (file)
index 0000000..94ce965
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/tree-ssa/copy-prop-aggregate-zero-3.c
@@ -0,0 +1,25 @@
+/* { dg-do compile } */ 
+/* { dg-options "-O1 -fno-tree-sra -fdump-tree-optimized -fdump-tree-forwprop1-details" } */
+
+extern void link_error (void);
+
+struct g
+{
+  _Complex unsigned int t;
+};
+
+struct g f(void)
+{
+  struct g temp_struct1 ;
+  temp_struct1.t = 0;
+  struct g temp_struct2 = temp_struct1;
+  struct g temp_struct3 = temp_struct2;
+  struct g temp_struct4 = temp_struct3;
+  return temp_struct4;
+}
+
+/* There should be no references to any of "temp_struct*"
+   temporaries.  */
+/* { dg-final { scan-tree-dump-times "temp_struct" 0 "optimized" } } */
+/* Also check that forwprop pass did the copy prop. */
+/* { dg-final { scan-tree-dump-times "after previous" 4 "forwprop1" } } */

diff --git a/gcc/tree-ssa-forwprop.cc b/gcc/tree-ssa-forwprop.cc
index 3d38d88844bff1a9a65394cfa793e6ad593fc65f..1cde5f85150e040999b01f5103fb479d480e006f 100644 (file)
--- a/gcc/tree-ssa-forwprop.cc
+++ b/gcc/tree-ssa-forwprop.cc
@@ -1340,6 +1340,21 @@ optimize_aggr_zeroprop (gimple_stmt_iterator *gsip)
            }
        }
     }
+  /* A store of integer (scalar, vector or complex) zeros is
+     a zero store. */
+  else if (gimple_store_p (stmt)
+          && gimple_assign_single_p (stmt)
+          && integer_zerop (gimple_assign_rhs1 (stmt)))
+    {
+      tree rhs = gimple_assign_rhs1 (stmt);
+      tree type = TREE_TYPE (rhs);
+      dest = gimple_assign_lhs (stmt);
+      ao_ref_init (&read, dest);
+      /* For integral types, the type precision needs to be a multiply of BITS_PER_UNIT. */
+      if (INTEGRAL_TYPE_P (type)
+         && (TYPE_PRECISION (type) % BITS_PER_UNIT) != 0)
+       dest = NULL_TREE;
+    }
   else if (gimple_store_p (stmt)
           && gimple_assign_single_p (stmt)
           && TREE_CODE (gimple_assign_rhs1 (stmt)) == CONSTRUCTOR