RISC-V: Support FP lround/lroundf auto vectorization

This patch would like to support the FP lround/lroundf auto vectorization.

* long lround (double) for rv64
* long lroundf (float) for rv32

Due to the limitation that only the same size of data type are allowed
in the vectorier, the standard name lroundmn2 only act on DF => DI for
rv64, and SF => SI for rv32.

Given we have code like:

void
test_lround (long *out, double *in, unsigned count)
{
  for (unsigned i = 0; i < count; i++)
    out[i] = __builtin_lround (in[i]);
}

Before this patch:
.L3:
  ...
  fld      fa5,0(a1)
  fcvt.l.d a5,fa5,rmm
  sd       a5,-8(a0)
  ...
  bne      a1,a4,.L3

After this patch:
  frrm     a6
  ...
  fsrmi    4 // RMM
.L3:
  ...
  vsetvli     a3,zero,e64,m1,ta,ma
  vfcvt.x.f.v v1,v1
  vsetvli     zero,a2,e64,m1,ta,ma
  vse32.v     v1,0(a0)
  ...
  bne         a2,zero,.L3
  ...
  fsrm     a6

The rest part like SF => DI/HF => DI/DF => SI/HF => SI will be covered
by TARGET_VECTORIZE_BUILTIN_VECTORIZED_FUNCTION.

gcc/ChangeLog:

	* config/riscv/autovec.md (lround<mode><v_i_l_ll_convert>2): New
	pattern for lround/lroundf.
	* config/riscv/riscv-protos.h (enum insn_type): New enum value.
	(expand_vec_lround): New func decl for expanding lround.
	* config/riscv/riscv-v.cc (expand_vec_lround): New func impl
	for expanding lround.

gcc/testsuite/ChangeLog:

	* gcc.target/riscv/rvv/autovec/unop/math-lround-0.c: New test.
	* gcc.target/riscv/rvv/autovec/unop/math-lround-1.c: New test.
	* gcc.target/riscv/rvv/autovec/unop/math-lround-run-0.c: New test.
	* gcc.target/riscv/rvv/autovec/unop/math-lround-run-1.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/math-lround-0.c: New test.
	* gcc.target/riscv/rvv/autovec/vls/math-lround-1.c: New test.

Signed-off-by: Pan Li <pan2.li@intel.com>

9 files changed, 264 insertions, 0 deletions

diff --git a/gcc/config/riscv/autovec.md b/gcc/config/riscv/autovec.md
index ebc51ea..33b1172 100644
--- a/gcc/config/riscv/autovec.md
+++ b/gcc/config/riscv/autovec.md
@@ -2321,3 +2321,13 @@
     DONE;
   }
 )
+
+(define_expand "lround<mode><v_i_l_ll_convert>2"
+  [(match_operand:<V_I_L_LL_CONVERT>    0 "register_operand")
+   (match_operand:V_VLS_FCONVERT_I_L_LL 1 "register_operand")]
+  "TARGET_VECTOR && !flag_trapping_math && !flag_rounding_math"
+  {
+    riscv_vector::expand_vec_lround (operands[0], operands[1], <MODE>mode, <V_I_L_LL_CONVERT>mode);
+    DONE;
+  }
+)
diff --git a/gcc/config/riscv/riscv-protos.h b/gcc/config/riscv/riscv-protos.h
index 8c9f7e0..b7eeeb8 100644
--- a/gcc/config/riscv/riscv-protos.h
+++ b/gcc/config/riscv/riscv-protos.h
@@ -302,6 +302,7 @@ enum insn_type : unsigned int
   UNARY_OP_TAMA = __MASK_OP_TAMA | UNARY_OP_P,
   UNARY_OP_TAMU = __MASK_OP_TAMU | UNARY_OP_P,
   UNARY_OP_FRM_DYN = UNARY_OP | FRM_DYN_P,
+  UNARY_OP_FRM_RMM = UNARY_OP | FRM_RMM_P,
   UNARY_OP_TAMU_FRM_DYN = UNARY_OP_TAMU | FRM_DYN_P,
   UNARY_OP_TAMU_FRM_RUP = UNARY_OP_TAMU | FRM_RUP_P,
   UNARY_OP_TAMU_FRM_RDN = UNARY_OP_TAMU | FRM_RDN_P,
@@ -475,6 +476,7 @@ void expand_vec_round (rtx, rtx, machine_mode, machine_mode);
 void expand_vec_trunc (rtx, rtx, machine_mode, machine_mode);
 void expand_vec_roundeven (rtx, rtx, machine_mode, machine_mode);
 void expand_vec_lrint (rtx, rtx, machine_mode, machine_mode);
+void expand_vec_lround (rtx, rtx, machine_mode, machine_mode);
 #endif
 bool sew64_scalar_helper (rtx *, rtx *, rtx, machine_mode,
 			  bool, void (*)(rtx *, rtx));
diff --git a/gcc/config/riscv/riscv-v.cc b/gcc/config/riscv/riscv-v.cc
index a75eb59..b61c745 100644
--- a/gcc/config/riscv/riscv-v.cc
+++ b/gcc/config/riscv/riscv-v.cc
@@ -4122,4 +4122,14 @@ expand_vec_lrint (rtx op_0, rtx op_1, machine_mode vec_fp_mode,
   emit_vec_cvt_x_f (op_0, op_1, UNARY_OP_FRM_DYN, vec_fp_mode);
 }
 
+void
+expand_vec_lround (rtx op_0, rtx op_1, machine_mode vec_fp_mode,
+		   machine_mode vec_long_mode)
+{
+  gcc_assert (known_eq (GET_MODE_SIZE (vec_fp_mode),
+			GET_MODE_SIZE (vec_long_mode)));
+
+  emit_vec_cvt_x_f (op_0, op_1, UNARY_OP_FRM_RMM, vec_fp_mode);
+}
+
 } // namespace riscv_vector
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-0.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-0.c
new file mode 100644
index 0000000..32b7348
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-0.c
@@ -0,0 +1,19 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv -mabi=lp64d -O3 -ftree-vectorize -fno-vect-cost-model -ffast-math -fno-schedule-insns -fno-schedule-insns2" } */
+/* { dg-final { check-function-bodies "**" "" } } */
+
+#include "test-math.h"
+
+/*
+** test_double_long___builtin_lround:
+**   frrm\s+[atx][0-9]+
+**   ...
+**   fsrmi\s+4
+**   ...
+**   vsetvli\s+[atx][0-9]+,\s*zero,\s*e64,\s*m1,\s*ta,\s*ma
+**   vfcvt\.x\.f\.v\s+v[0-9]+,\s*v[0-9]+
+**   ...
+**   fsrm\s+[atx][0-9]+
+**   ret
+*/
+TEST_UNARY_CALL_CVT (double, long, __builtin_lround)
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-1.c
new file mode 100644
index 0000000..a4d6fcf
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-1.c
@@ -0,0 +1,19 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv32gcv -mabi=ilp32f -O3 -ftree-vectorize -fno-vect-cost-model -ffast-math -fno-schedule-insns -fno-schedule-insns2" } */
+/* { dg-final { check-function-bodies "**" "" } } */
+
+#include "test-math.h"
+
+/*
+** test_float_long___builtin_lroundf:
+**   frrm\s+[atx][0-9]+
+**   ...
+**   fsrmi\s+4
+**   ...
+**   vsetvli\s+[atx][0-9]+,\s*zero,\s*e32,\s*m1,\s*ta,\s*ma
+**   vfcvt\.x\.f\.v\s+v[0-9]+,\s*v[0-9]+
+**   ...
+**   fsrm\s+[atx][0-9]+
+**   ret
+*/
+TEST_UNARY_CALL_CVT (float, long, __builtin_lroundf)
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-run-0.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-run-0.c
new file mode 100644
index 0000000..ec4d9f5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-run-0.c
@@ -0,0 +1,72 @@
+/* { dg-do run { target { riscv_v && rv64 } } } */
+/* { dg-additional-options "-std=c99 -O3 -ftree-vectorize -fno-vect-cost-model -ffast-math" } */
+
+#include "test-math.h"
+
+#define ARRAY_SIZE 128
+
+double in[ARRAY_SIZE];
+long out[ARRAY_SIZE];
+long ref[ARRAY_SIZE];
+
+TEST_UNARY_CALL_CVT (double, long, __builtin_lround)
+TEST_ASSERT (long)
+
+TEST_INIT_CVT (double, 1.2, long, __builtin_lround (1.2), 1)
+TEST_INIT_CVT (double, -1.2, long, __builtin_lround (-1.2), 2)
+TEST_INIT_CVT (double, 0.5, long, __builtin_lround (0.5), 3)
+TEST_INIT_CVT (double, -0.5, long, __builtin_lround (-0.5), 4)
+TEST_INIT_CVT (double, 0.1, long, __builtin_lround (0.1), 5)
+TEST_INIT_CVT (double, -0.1, long, __builtin_lround (-0.1), 6)
+TEST_INIT_CVT (double, 3.0, long, __builtin_lround (3.0), 7)
+TEST_INIT_CVT (double, -3.0, long, __builtin_lround (-3.0), 8)
+TEST_INIT_CVT (double, 4503599627370495.5, long, __builtin_lround (4503599627370495.5), 9)
+TEST_INIT_CVT (double, 4503599627370497.0, long, __builtin_lround (4503599627370497.0), 10)
+TEST_INIT_CVT (double, -4503599627370495.5, long, __builtin_lround (-4503599627370495.5), 11)
+TEST_INIT_CVT (double, -4503599627370496.0, long, __builtin_lround (-4503599627370496.0), 12)
+TEST_INIT_CVT (double, 0.0, long, __builtin_lround (-0.0), 13)
+TEST_INIT_CVT (double, -0.0, long, __builtin_lround (-0.0), 14)
+TEST_INIT_CVT (double, 9223372036854774784.0, long, __builtin_lround (9223372036854774784.0), 15)
+TEST_INIT_CVT (double, 9223372036854775808.0, long, 0x7fffffffffffffff, 16)
+TEST_INIT_CVT (double, -9223372036854775808.0, long, __builtin_lround (-9223372036854775808.0), 17)
+TEST_INIT_CVT (double, -9223372036854777856.0, long, 0x8000000000000000, 18)
+TEST_INIT_CVT (double, __builtin_inf (), long, __builtin_lround (__builtin_inf ()), 19)
+TEST_INIT_CVT (double, -__builtin_inf (), long, __builtin_lround (-__builtin_inf ()), 20)
+TEST_INIT_CVT (double, __builtin_nan (""), long, 0x7fffffffffffffff, 21)
+
+/* According to the manual as below.
+
+   If x is a NaN or an infinity, or the rounded value is too large to
+   be stored in a long (long long in the case of the ll* functions),
+   then a domain error occurs, and the return value is unspecified.
+
+   Some reference are hard-code instead of leveraging scalar __builtin_lround.
+*/
+
+int
+main ()
+{
+  RUN_TEST_CVT (double, long, 1, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 2, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 3, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 4, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 5, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 6, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 7, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 8, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 9, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 10, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 11, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 12, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 13, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 14, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 15, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 16, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 17, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 18, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 19, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 20, __builtin_lround, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (double, long, 21, __builtin_lround, in, out, ref, ARRAY_SIZE);
+
+  return 0;
+}
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-run-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-run-1.c
new file mode 100644
index 0000000..76e4e43
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/unop/math-lround-run-1.c
@@ -0,0 +1,72 @@
+/* { dg-do run { target { riscv_v && rv32 } } } */
+/* { dg-additional-options "-std=c99 -O3 -ftree-vectorize -fno-vect-cost-model -ffast-math" } */
+
+#include "test-math.h"
+
+#define ARRAY_SIZE 128
+
+float in[ARRAY_SIZE];
+long out[ARRAY_SIZE];
+long ref[ARRAY_SIZE];
+
+TEST_UNARY_CALL_CVT (float, long, __builtin_lroundf)
+TEST_ASSERT (long)
+
+TEST_INIT_CVT (float, 1.2, long, __builtin_lroundf (1.2), 1)
+TEST_INIT_CVT (float, -1.2, long, __builtin_lroundf (-1.2), 2)
+TEST_INIT_CVT (float, 0.5, long, __builtin_lroundf (0.5), 3)
+TEST_INIT_CVT (float, -0.5, long, __builtin_lroundf (-0.5), 4)
+TEST_INIT_CVT (float, 0.1, long, __builtin_lroundf (0.1), 5)
+TEST_INIT_CVT (float, -0.1, long, __builtin_lroundf (-0.1), 6)
+TEST_INIT_CVT (float, 3.0, long, __builtin_lroundf (3.0), 7)
+TEST_INIT_CVT (float, -3.0, long, __builtin_lroundf (-3.0), 8)
+TEST_INIT_CVT (float, 8388607.5, long, __builtin_lroundf (8388607.5), 9)
+TEST_INIT_CVT (float, 8388609.0, long, __builtin_lroundf (8388609.0), 10)
+TEST_INIT_CVT (float, -8388607.5, long, __builtin_lroundf (-8388607.5), 11)
+TEST_INIT_CVT (float, -8388609.0, long, __builtin_lroundf (-8388609.0), 12)
+TEST_INIT_CVT (float, 0.0, long, __builtin_lroundf (-0.0), 13)
+TEST_INIT_CVT (float, -0.0, long, __builtin_lroundf (-0.0), 14)
+TEST_INIT_CVT (float, 2147483520.0, long, __builtin_lroundf (2147483520.0), 15)
+TEST_INIT_CVT (float, 2147483648.0, long, 0x7fffffff, 16)
+TEST_INIT_CVT (float, -2147483648.0, long, __builtin_lroundf (-2147483648.0), 17)
+TEST_INIT_CVT (float, -2147483904.0, long, 0x80000000, 18)
+TEST_INIT_CVT (float, __builtin_inf (), long, __builtin_lroundf (__builtin_inff ()), 19)
+TEST_INIT_CVT (float, -__builtin_inf (), long, __builtin_lroundf (-__builtin_inff ()), 20)
+TEST_INIT_CVT (float, __builtin_nanf (""), long, 0x7fffffff, 21)
+
+/* According to the manual as below.
+
+   If x is a NaN or an infinity, or the rounded value is too large to
+   be stored in a long (long long in the case of the ll* functions),
+   then a domain error occurs, and the return value is unspecified.
+
+   Some reference are hard-code instead of leveraging scalar __builtin_lround.
+*/
+
+int
+main ()
+{
+  RUN_TEST_CVT (float, long, 1, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 2, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 3, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 4, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 5, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 6, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 7, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 8, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 9, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 10, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 11, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 12, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 13, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 14, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 15, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 16, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 17, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 18, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 19, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 20, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+  RUN_TEST_CVT (float, long, 21, __builtin_lroundf, in, out, ref, ARRAY_SIZE);
+
+  return 0;
+}
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/math-lround-0.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/math-lround-0.c
new file mode 100644
index 0000000..c2a9f6b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/math-lround-0.c
@@ -0,0 +1,30 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv_zvfh_zvl4096b -mabi=lp64d -O3 --param=riscv-autovec-lmul=m8 -ffast-math -fdump-tree-optimized" } */
+
+#include "def.h"
+
+DEF_OP_V_CVT (lround, 1, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 2, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 4, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 8, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 16, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 32, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 64, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 128, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 256, double, long, __builtin_lround)
+DEF_OP_V_CVT (lround, 512, double, long, __builtin_lround)
+
+/* { dg-final { scan-assembler-not {csrr} } } */
+/* { dg-final { scan-tree-dump-not "1,1" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "2,2" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "4,4" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "16,16" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "32,32" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "64,64" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "128,128" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "256,256" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "512,512" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "1024,1024" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "2048,2048" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "4096,4096" "optimized" } } */
+/* { dg-final { scan-assembler-times {vfcvt\.x\.f\.v\s+v[0-9]+,\s*v[0-9]+} 9 } } */
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/math-lround-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/math-lround-1.c
new file mode 100644
index 0000000..5a43133
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/vls/math-lround-1.c
@@ -0,0 +1,30 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv32gcv_zvfh_zvl4096b -mabi=ilp32f -O3 --param=riscv-autovec-lmul=m8 -ffast-math -fdump-tree-optimized" } */
+
+#include "def.h"
+
+DEF_OP_V_CVT (lroundf, 1, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 2, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 4, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 8, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 16, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 32, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 64, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 128, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 256, float, long, __builtin_lroundf)
+DEF_OP_V_CVT (lroundf, 512, float, long, __builtin_lroundf)
+
+/* { dg-final { scan-assembler-not {csrr} } } */
+/* { dg-final { scan-tree-dump-not "1,1" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "2,2" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "4,4" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "16,16" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "32,32" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "64,64" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "128,128" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "256,256" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "512,512" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "1024,1024" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "2048,2048" "optimized" } } */
+/* { dg-final { scan-tree-dump-not "4096,4096" "optimized" } } */
+/* { dg-final { scan-assembler-times {vfcvt\.x\.f\.v\s+v[0-9]+,\s*v[0-9]+} 9 } } */