gcc/address-calculation-optimization-within-loop.patch

diff -Nurp a/gcc/testsuite/gcc.dg/pr94269.c b/gcc/testsuite/gcc.dg/pr94269.c
--- a/gcc/testsuite/gcc.dg/pr94269.c	1970-01-01 08:00:00.000000000 +0800
+++ b/gcc/testsuite/gcc.dg/pr94269.c	2020-04-17 17:04:50.608000000 +0800
@@ -0,0 +1,26 @@
+/* { dg-do compile { target aarch64*-*-* } } */
+/* { dg-options "-O2 -ftree-loop-vectorize -funsafe-math-optimizations -march=armv8.2-a+sve -msve-vector-bits=256" } */
+
+float
+foo(long n, float *x, int inc_x,
+            float *y, int inc_y)
+{
+  float dot = 0.0;
+  int ix = 0, iy = 0;
+
+  if (n < 0) {
+    return dot;
+  }
+
+  int i = 0;
+  while (i < n) {
+    dot += y[iy] * x[ix];
+    ix  += inc_x;
+    iy  += inc_y;
+    i++;
+  }
+
+  return dot;
+}
+
+/* { dg-final { scan-assembler-not "smaddl" { target aarch64*-*-* } } } */
diff -Nurp a/gcc/tree-ssa-math-opts.c b/gcc/tree-ssa-math-opts.c
--- a/gcc/tree-ssa-math-opts.c	2020-04-17 16:43:59.540000000 +0800
+++ b/gcc/tree-ssa-math-opts.c	2020-04-17 16:48:34.072036000 +0800
@@ -2721,11 +2721,14 @@ convert_plusminus_to_widen (gimple_stmt_
      multiply-and-accumulate instructions.
 
      If the widened-multiplication result has more than one uses, it is
-     probably wiser not to do the conversion.  */
+     probably wiser not to do the conversion.  Also restrict this operation
+     to single basic block to avoid moving the multiply to a different block
+     with a higher execution frequency.  */
   if (code == PLUS_EXPR
       && (rhs1_code == MULT_EXPR || rhs1_code == WIDEN_MULT_EXPR))
     {
       if (!has_single_use (rhs1)
+	  || gimple_bb (rhs1_stmt) != gimple_bb (stmt)
 	  || !is_widening_mult_p (rhs1_stmt, &type1, &mult_rhs1,
 				  &type2, &mult_rhs2))
 	return false;
@@ -2735,6 +2738,7 @@ convert_plusminus_to_widen (gimple_stmt_
   else if (rhs2_code == MULT_EXPR || rhs2_code == WIDEN_MULT_EXPR)
     {
       if (!has_single_use (rhs2)
+	  || gimple_bb (rhs2_stmt) != gimple_bb (stmt)
 	  || !is_widening_mult_p (rhs2_stmt, &type1, &mult_rhs1,
 				  &type2, &mult_rhs2))
 	return false;
diff -Nurp a/gcc/testsuite/gcc.target/aarch64/sve/var_stride_1.c b/gcc/testsuite/gcc.target/aarch64/sve/var_stride_1.c
--- a/gcc/testsuite/gcc.target/aarch64/sve/var_stride_1.c	2020-03-31 09:51:36.000000000 +0800
+++ b/gcc/testsuite/gcc.target/aarch64/sve/var_stride_1.c	2020-04-29 10:55:44.937471475 +0800
@@ -17,7 +17,6 @@ f (TYPE *x, TYPE *y, unsigned short n, l
 /* { dg-final { scan-assembler {\tstr\tw[0-9]+} } } */
 /* Should multiply by (VF-1)*4 rather than (257-1)*4.  */
 /* { dg-final { scan-assembler-not {, 1024} } } */
-/* { dg-final { scan-assembler-not {\t.bfiz\t} } } */
 /* { dg-final { scan-assembler-not {lsl[^\n]*[, ]10} } } */
 /* { dg-final { scan-assembler-not {\tcmp\tx[0-9]+, 0} } } */
 /* { dg-final { scan-assembler-not {\tcmp\tw[0-9]+, 0} } } */