[3/3] [AArch64][vect] vec_widen_lshift pattern

Add aarch64 vec_widen_lshift_lo/hi patterns and fix bug it triggers in
mid-end. This pattern takes one vector with N elements of size S, shifts
each element left by the element width and stores the results as N
elements of size 2*s (in 2 result vectors). The aarch64 backend
implements this with the shll,shll2 instruction pair.

gcc/ChangeLog:

	* config/aarch64/aarch64-simd.md: Add vec_widen_lshift_hi/lo<mode>
	patterns.
	* tree-vect-stmts.c (vectorizable_conversion): Fix for widen_lshift
	case.

gcc/testsuite/ChangeLog:

	* gcc.target/aarch64/vect-widen-lshift.c: New test.

3 files changed, 131 insertions, 2 deletions

diff --git a/gcc/config/aarch64/aarch64-simd.md b/gcc/config/aarch64/aarch64-simd.md
index e8c951f..68baf41 100644
--- a/gcc/config/aarch64/aarch64-simd.md
+++ b/gcc/config/aarch64/aarch64-simd.md
@@ -4664,8 +4664,74 @@
   [(set_attr "type" "neon_sat_shift_reg<q>")]
 )
 
+(define_expand "vec_widen_<sur>shiftl_lo_<mode>"
+  [(set (match_operand:<VWIDE> 0 "register_operand" "=w")
+	(unspec:<VWIDE> [(match_operand:VQW 1 "register_operand" "w")
+			 (match_operand:SI 2
+			   "aarch64_simd_shift_imm_bitsize_<ve_mode>" "i")]
+			 VSHLL))]
+  "TARGET_SIMD"
+  {
+    rtx p = aarch64_simd_vect_par_cnst_half (<MODE>mode, <nunits>, false);
+    emit_insn (gen_aarch64_<sur>shll<mode>_internal (operands[0], operands[1],
+						     p, operands[2]));
+    DONE;
+  }
+)
+
+(define_expand "vec_widen_<sur>shiftl_hi_<mode>"
+   [(set (match_operand:<VWIDE> 0 "register_operand")
+	(unspec:<VWIDE> [(match_operand:VQW 1 "register_operand" "w")
+			 (match_operand:SI 2
+			   "immediate_operand" "i")]
+			  VSHLL))]
+   "TARGET_SIMD"
+   {
+    rtx p = aarch64_simd_vect_par_cnst_half (<MODE>mode, <nunits>, true);
+    emit_insn (gen_aarch64_<sur>shll2<mode>_internal (operands[0], operands[1],
+						      p, operands[2]));
+    DONE;
+   }
+)
+
 ;; vshll_n
 
+(define_insn "aarch64_<sur>shll<mode>_internal"
+  [(set (match_operand:<VWIDE> 0 "register_operand" "=w")
+	(unspec:<VWIDE> [(vec_select:<VHALF>
+			    (match_operand:VQW 1 "register_operand" "w")
+			    (match_operand:VQW 2 "vect_par_cnst_lo_half" ""))
+			 (match_operand:SI 3
+			   "aarch64_simd_shift_imm_bitsize_<ve_mode>" "i")]
+			 VSHLL))]
+  "TARGET_SIMD"
+  {
+    if (INTVAL (operands[3]) == GET_MODE_UNIT_BITSIZE (<MODE>mode))
+      return "shll\\t%0.<Vwtype>, %1.<Vhalftype>, %3";
+    else
+      return "<sur>shll\\t%0.<Vwtype>, %1.<Vhalftype>, %3";
+  }
+  [(set_attr "type" "neon_shift_imm_long")]
+)
+
+(define_insn "aarch64_<sur>shll2<mode>_internal"
+  [(set (match_operand:<VWIDE> 0 "register_operand" "=w")
+	(unspec:<VWIDE> [(vec_select:<VHALF>
+			    (match_operand:VQW 1 "register_operand" "w")
+			    (match_operand:VQW 2 "vect_par_cnst_hi_half" ""))
+			 (match_operand:SI 3
+			   "aarch64_simd_shift_imm_bitsize_<ve_mode>" "i")]
+			 VSHLL))]
+  "TARGET_SIMD"
+  {
+    if (INTVAL (operands[3]) == GET_MODE_UNIT_BITSIZE (<MODE>mode))
+      return "shll2\\t%0.<Vwtype>, %1.<Vtype>, %3";
+    else
+      return "<sur>shll2\\t%0.<Vwtype>, %1.<Vtype>, %3";
+  }
+  [(set_attr "type" "neon_shift_imm_long")]
+)
+
 (define_insn "aarch64_<sur>shll_n<mode>"
   [(set (match_operand:<VWIDE> 0 "register_operand" "=w")
 	(unspec:<VWIDE> [(match_operand:VD_BHSI 1 "register_operand" "w")
diff --git a/gcc/testsuite/gcc.target/aarch64/vect-widen-lshift.c b/gcc/testsuite/gcc.target/aarch64/vect-widen-lshift.c
new file mode 100644
index 0000000..48a3719
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/vect-widen-lshift.c
@@ -0,0 +1,62 @@
+/* { dg-do run } */
+/* { dg-options "-O3 -save-temps" } */
+#include <stdint.h>
+#include <string.h>
+
+#pragma GCC target "+nosve"
+
+#define ARR_SIZE 1024
+
+/* Should produce an shll,shll2 pair*/
+void sshll_opt (int32_t *foo, int16_t *a, int16_t *b)
+{
+    for( int i = 0; i < ARR_SIZE - 3;i=i+4)
+    {
+        foo[i]   = a[i]   << 16;
+        foo[i+1] = a[i+1] << 16;
+        foo[i+2] = a[i+2] << 16;
+        foo[i+3] = a[i+3] << 16;
+    }
+}
+
+__attribute__((optimize (0)))
+void sshll_nonopt (int32_t *foo, int16_t *a, int16_t *b)
+{
+    for( int i = 0; i < ARR_SIZE - 3;i=i+4)
+    {
+        foo[i]   = a[i]   << 16;
+        foo[i+1] = a[i+1] << 16;
+        foo[i+2] = a[i+2] << 16;
+        foo[i+3] = a[i+3] << 16;
+    }
+}
+
+
+void __attribute__((optimize (0)))
+init(uint16_t *a, uint16_t *b)
+{
+    for( int i = 0; i < ARR_SIZE;i++)
+    {
+      a[i] = i;
+      b[i] = 2*i;
+    }
+}
+
+int __attribute__((optimize (0)))
+main()
+{
+    uint32_t foo_arr[ARR_SIZE];
+    uint32_t bar_arr[ARR_SIZE];
+    uint16_t a[ARR_SIZE];
+    uint16_t b[ARR_SIZE];
+
+    init(a, b);
+    sshll_opt(foo_arr, a, b);
+    sshll_nonopt(bar_arr, a, b);
+    if (memcmp(foo_arr, bar_arr, ARR_SIZE) != 0)
+      return 1;
+    return 0;
+}
+
+/* { dg-final { scan-assembler-times {\tshll\t} 1} } */
+/* { dg-final { scan-assembler-times {\tshll2\t} 1} } */
diff --git a/gcc/tree-vect-stmts.c b/gcc/tree-vect-stmts.c
index f88f07a..a4980a9 100644
--- a/gcc/tree-vect-stmts.c
+++ b/gcc/tree-vect-stmts.c
@@ -4935,8 +4935,9 @@ vectorizable_conversion (vec_info *vinfo,
 			 &vec_oprnds1);
       if (code == WIDEN_LSHIFT_EXPR)
 	{
-	  vec_oprnds1.create (ncopies * ninputs);
-	  for (i = 0; i < ncopies * ninputs; ++i)
+	  int oprnds_size = vec_oprnds0.length ();
+	  vec_oprnds1.create (oprnds_size);
+	  for (i = 0; i < oprnds_size; ++i)
 	    vec_oprnds1.quick_push (op1);
 	}
       /* Arguments are ready.  Create the new vector stmts.  */