aarch64: Add support for mfloat8x{8|16}_t types

gcc/ChangeLog:

	* config/aarch64/aarch64-builtins.cc
	(aarch64_init_simd_builtin_types): Initialise FP8 simd types.
	* config/aarch64/aarch64-builtins.h
	(enum aarch64_type_qualifiers): Add qualifier_modal_float bit.
	* config/aarch64/aarch64-simd-builtin-types.def:
	Add Mfloat8x{8|16}_t types.
	* config/aarch64/arm_neon.h: Add mfloat8x{8|16}_t typedefs.

gcc/testsuite/ChangeLog:

	* gcc.target/aarch64/movv16qi_2.c: Test mfloat as well.
	* gcc.target/aarch64/movv16qi_3.c: Ditto.
	* gcc.target/aarch64/movv2x16qi_1.c: Ditto.
	* gcc.target/aarch64/movv3x16qi_1.c: Ditto.
	* gcc.target/aarch64/movv4x16qi_1.c: Ditto.
	* gcc.target/aarch64/movv8qi_2.c: Ditto.
	* gcc.target/aarch64/movv8qi_3.c: Ditto.
	* gcc.target/aarch64/mfloat-init-1.c: New test.

gcc/config/aarch64/aarch64-builtins.cc

@@ -1220,6 +1220,10 @@ aarch64_init_simd_builtin_types (void)
   aarch64_simd_types[Bfloat16x4_t].eltype = bfloat16_type_node;
   aarch64_simd_types[Bfloat16x8_t].eltype = bfloat16_type_node;
 
+  /* Init FP8 element types.  */
+  aarch64_simd_types[Mfloat8x8_t].eltype = aarch64_mfp8_type_node;
+  aarch64_simd_types[Mfloat8x16_t].eltype = aarch64_mfp8_type_node;
+
   for (i = 0; i < nelts; i++)
     {
       tree eltype = aarch64_simd_types[i].eltype;

gcc/config/aarch64/aarch64-builtins.h

@@ -54,6 +54,8 @@ enum aarch64_type_qualifiers
   /* Lane indices selected in quadtuplets. - must be in range, and flipped for
      bigendian.  */
   qualifier_lane_quadtup_index = 0x1000,
+  /* Modal FP types.  */
+  qualifier_modal_float = 0x2000,
 };
 
 #define ENTRY(E, M, Q, G) E,

gcc/config/aarch64/aarch64-simd-builtin-types.def

@@ -52,3 +52,5 @@
   ENTRY (Float64x2_t, V2DF, none, 13)
   ENTRY (Bfloat16x4_t, V4BF, none, 14)
   ENTRY (Bfloat16x8_t, V8BF, none, 14)
+  ENTRY (Mfloat8x8_t, V8QI, modal_float, 13)
+  ENTRY (Mfloat8x16_t, V16QI, modal_float, 14)

gcc/config/aarch64/arm_neon.h

@@ -72,6 +72,9 @@ typedef __Poly16_t poly16_t;
 typedef __Poly64_t poly64_t;
 typedef __Poly128_t poly128_t;
 
+typedef __Mfloat8x8_t mfloat8x8_t;
+typedef __Mfloat8x16_t mfloat8x16_t;
+
 typedef __fp16 float16_t;
 typedef float float32_t;
 typedef double float64_t;

gcc/testsuite/gcc.target/aarch64/mfloat-init-1.c

@@ -0,0 +1,5 @@
+/* { dg-do assemble } */
+/* { dg-options "-O --save-temps" } */
+
+/* { dg-error "invalid conversion to type 'mfloat8_t" "" {target *-*-*} 0 } */
+__Mfloat8x8_t const_mf8x8 () { return (__Mfloat8x8_t) { 1, 1, 1, 1, 1, 1, 1, 1 }; }

gcc/testsuite/gcc.target/aarch64/movv16qi_2.c

@@ -17,6 +17,7 @@ TEST_GENERAL (__Bfloat16x8_t)
 TEST_GENERAL (__Float16x8_t)
 TEST_GENERAL (__Float32x4_t)
 TEST_GENERAL (__Float64x2_t)
+TEST_GENERAL (__Mfloat8x16_t)
 
 __Int8x16_t const_s8x8 () { return (__Int8x16_t) { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 }; }
 __Int16x8_t const_s16x4 () { return (__Int16x8_t) { 1, 0, 1, 0, 1, 0, 1, 0 }; }

gcc/testsuite/gcc.target/aarch64/movv16qi_3.c

@@ -22,6 +22,7 @@ TEST_VECTOR (__Bfloat16x8_t)
 TEST_VECTOR (__Float16x8_t)
 TEST_VECTOR (__Float32x4_t)
 TEST_VECTOR (__Float64x2_t)
+TEST_VECTOR (__Mfloat8x16_t)
 
 /*
 ** test___Int8x16_t:

gcc/testsuite/gcc.target/aarch64/movv2x16qi_1.c

@@ -19,6 +19,7 @@ TEST_VECTOR (float16x8x2_t)
 TEST_VECTOR (bfloat16x8x2_t)
 TEST_VECTOR (float32x4x2_t)
 TEST_VECTOR (float64x2x2_t)
+TEST_VECTOR (mfloat8x16x2_t)
 
 /*
 ** mov_int8x16x2_t:

gcc/testsuite/gcc.target/aarch64/movv3x16qi_1.c

@@ -19,6 +19,7 @@ TEST_VECTOR (float16x8x3_t)
 TEST_VECTOR (bfloat16x8x3_t)
 TEST_VECTOR (float32x4x3_t)
 TEST_VECTOR (float64x2x3_t)
+TEST_VECTOR (mfloat8x16x3_t)
 
 /*
 ** mov_int8x16x3_t:

gcc/testsuite/gcc.target/aarch64/movv4x16qi_1.c

@@ -19,6 +19,7 @@ TEST_VECTOR (float16x8x4_t)
 TEST_VECTOR (bfloat16x8x4_t)
 TEST_VECTOR (float32x4x4_t)
 TEST_VECTOR (float64x2x4_t)
+TEST_VECTOR (mfloat8x16x4_t)
 
 /*
 ** mov_int8x16x4_t:

gcc/testsuite/gcc.target/aarch64/movv8qi_2.c

@@ -17,6 +17,7 @@ TEST_GENERAL (__Bfloat16x4_t)
 TEST_GENERAL (__Float16x4_t)
 TEST_GENERAL (__Float32x2_t)
 TEST_GENERAL (__Float64x1_t)
+TEST_GENERAL (__Mfloat8x8_t)
 
 __Int8x8_t const_s8x8 () { return (__Int8x8_t) { 1, 1, 1, 1, 1, 1, 1, 1 }; }
 __Int16x4_t const_s16x4 () { return (__Int16x4_t) { 1, 0, 1, 0 }; }

gcc/testsuite/gcc.target/aarch64/movv8qi_3.c

@@ -22,6 +22,7 @@ TEST_VECTOR (__Bfloat16x4_t)
 TEST_VECTOR (__Float16x4_t)
 TEST_VECTOR (__Float32x2_t)
 TEST_VECTOR (__Float64x1_t)
+TEST_VECTOR (__Mfloat8x8_t)
 
 /*
 ** test___Int8x8_t: