@@ -14,6 +14,7 @@ VECT_VAR_DECL(expected_st2_0,uint,32,2) [] = { 0xfffffff0, 0xfffffff1 };
VECT_VAR_DECL(expected_st2_0,poly,8,8) [] = { 0xf0, 0xf1, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_0,poly,16,4) [] = { 0xfff0, 0xfff1, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st2_0,hfloat,16,4) [] = { 0xcc00, 0xcb80, 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_0,hfloat,32,2) [] = { 0xc1800000, 0xc1700000 };
VECT_VAR_DECL(expected_st2_0,int,16,8) [] = { 0xfff0, 0xfff1, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
@@ -24,6 +25,8 @@ VECT_VAR_DECL(expected_st2_0,uint,32,4) [] = { 0xfffffff0, 0xfffffff1,
0x0, 0x0 };
VECT_VAR_DECL(expected_st2_0,poly,16,8) [] = { 0xfff0, 0xfff1, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st2_0,hfloat,16,8) [] = { 0xcc00, 0xcb80, 0x0, 0x0,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_0,hfloat,32,4) [] = { 0xc1800000, 0xc1700000,
0x0, 0x0 };
@@ -39,6 +42,7 @@ VECT_VAR_DECL(expected_st2_1,uint,32,2) [] = { 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_1,poly,8,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_1,poly,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st2_1,hfloat,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_1,hfloat,32,2) [] = { 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_1,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
@@ -48,6 +52,8 @@ VECT_VAR_DECL(expected_st2_1,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
VECT_VAR_DECL(expected_st2_1,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_1,poly,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st2_1,hfloat,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st2_1,hfloat,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
/* Expected results for vst3, chunk 0. */
@@ -62,6 +68,7 @@ VECT_VAR_DECL(expected_st3_0,uint,32,2) [] = { 0xfffffff0, 0xfffffff1 };
VECT_VAR_DECL(expected_st3_0,poly,8,8) [] = { 0xf0, 0xf1, 0xf2, 0x0,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_0,poly,16,4) [] = { 0xfff0, 0xfff1, 0xfff2, 0x0 };
+VECT_VAR_DECL(expected_st3_0,hfloat,16,4) [] = { 0xcc00, 0xcb80, 0xcb00, 0x0 };
VECT_VAR_DECL(expected_st3_0,hfloat,32,2) [] = { 0xc1800000, 0xc1700000 };
VECT_VAR_DECL(expected_st3_0,int,16,8) [] = { 0xfff0, 0xfff1, 0xfff2, 0x0,
0x0, 0x0, 0x0, 0x0 };
@@ -73,6 +80,8 @@ VECT_VAR_DECL(expected_st3_0,uint,32,4) [] = { 0xfffffff0, 0xfffffff1,
0xfffffff2, 0x0 };
VECT_VAR_DECL(expected_st3_0,poly,16,8) [] = { 0xfff0, 0xfff1, 0xfff2, 0x0,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st3_0,hfloat,16,8) [] = { 0xcc00, 0xcb80, 0xcb00, 0x0,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_0,hfloat,32,4) [] = { 0xc1800000, 0xc1700000,
0xc1600000, 0x0 };
@@ -88,6 +97,7 @@ VECT_VAR_DECL(expected_st3_1,uint,32,2) [] = { 0xfffffff2, 0x0 };
VECT_VAR_DECL(expected_st3_1,poly,8,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_1,poly,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st3_1,hfloat,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_1,hfloat,32,2) [] = { 0xc1600000, 0x0 };
VECT_VAR_DECL(expected_st3_1,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
@@ -97,6 +107,8 @@ VECT_VAR_DECL(expected_st3_1,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
VECT_VAR_DECL(expected_st3_1,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_1,poly,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st3_1,hfloat,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_1,hfloat,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
/* Expected results for vst3, chunk 2. */
@@ -111,6 +123,7 @@ VECT_VAR_DECL(expected_st3_2,uint,32,2) [] = { 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_2,poly,8,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_2,poly,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st3_2,hfloat,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_2,hfloat,32,2) [] = { 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_2,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
@@ -120,6 +133,8 @@ VECT_VAR_DECL(expected_st3_2,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
VECT_VAR_DECL(expected_st3_2,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_2,poly,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st3_2,hfloat,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st3_2,hfloat,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
/* Expected results for vst4, chunk 0. */
@@ -134,6 +149,7 @@ VECT_VAR_DECL(expected_st4_0,uint,32,2) [] = { 0xfffffff0, 0xfffffff1 };
VECT_VAR_DECL(expected_st4_0,poly,8,8) [] = { 0xf0, 0xf1, 0xf2, 0xf3,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_0,poly,16,4) [] = { 0xfff0, 0xfff1, 0xfff2, 0xfff3 };
+VECT_VAR_DECL(expected_st4_0,hfloat,16,4) [] = { 0xcc00, 0xcb80, 0xcb00, 0xca80 };
VECT_VAR_DECL(expected_st4_0,hfloat,32,2) [] = { 0xc1800000, 0xc1700000 };
VECT_VAR_DECL(expected_st4_0,int,16,8) [] = { 0xfff0, 0xfff1, 0xfff2, 0xfff3,
0x0, 0x0, 0x0, 0x0 };
@@ -145,6 +161,8 @@ VECT_VAR_DECL(expected_st4_0,uint,32,4) [] = { 0xfffffff0, 0xfffffff1,
0xfffffff2, 0xfffffff3 };
VECT_VAR_DECL(expected_st4_0,poly,16,8) [] = { 0xfff0, 0xfff1, 0xfff2, 0xfff3,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st4_0,hfloat,16,8) [] = { 0xcc00, 0xcb80, 0xcb00, 0xca80,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_0,hfloat,32,4) [] = { 0xc1800000, 0xc1700000,
0xc1600000, 0xc1500000 };
@@ -160,6 +178,7 @@ VECT_VAR_DECL(expected_st4_1,uint,32,2) [] = { 0xfffffff2, 0xfffffff3 };
VECT_VAR_DECL(expected_st4_1,poly,8,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_1,poly,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st4_1,hfloat,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_1,hfloat,32,2) [] = { 0xc1600000, 0xc1500000 };
VECT_VAR_DECL(expected_st4_1,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
@@ -169,6 +188,8 @@ VECT_VAR_DECL(expected_st4_1,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
VECT_VAR_DECL(expected_st4_1,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_1,poly,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st4_1,hfloat,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_1,hfloat,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
/* Expected results for vst4, chunk 2. */
@@ -183,6 +204,7 @@ VECT_VAR_DECL(expected_st4_2,uint,32,2) [] = { 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_2,poly,8,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_2,poly,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st4_2,hfloat,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_2,hfloat,32,2) [] = { 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_2,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
@@ -192,6 +214,8 @@ VECT_VAR_DECL(expected_st4_2,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
VECT_VAR_DECL(expected_st4_2,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_2,poly,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st4_2,hfloat,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_2,hfloat,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
/* Expected results for vst4, chunk 3. */
@@ -206,6 +230,7 @@ VECT_VAR_DECL(expected_st4_3,uint,32,2) [] = { 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_3,poly,8,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_3,poly,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st4_3,hfloat,16,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_3,hfloat,32,2) [] = { 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_3,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
@@ -215,6 +240,8 @@ VECT_VAR_DECL(expected_st4_3,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
VECT_VAR_DECL(expected_st4_3,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_3,poly,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0 };
+VECT_VAR_DECL(expected_st4_3,hfloat,16,8) [] = { 0x0, 0x0, 0x0, 0x0,
+ 0x0, 0x0, 0x0, 0x0 };
VECT_VAR_DECL(expected_st4_3,hfloat,32,4) [] = { 0x0, 0x0, 0x0, 0x0 };
/* Declare additional input buffers as needed. */
@@ -229,6 +256,7 @@ VECT_VAR_DECL_INIT(buffer_vld2_lane, uint, 32, 2);
VECT_VAR_DECL_INIT(buffer_vld2_lane, uint, 64, 2);
VECT_VAR_DECL_INIT(buffer_vld2_lane, poly, 8, 2);
VECT_VAR_DECL_INIT(buffer_vld2_lane, poly, 16, 2);
+VECT_VAR_DECL_INIT(buffer_vld2_lane, float, 16, 2);
VECT_VAR_DECL_INIT(buffer_vld2_lane, float, 32, 2);
/* Input buffers for vld3_lane. */
@@ -242,6 +270,7 @@ VECT_VAR_DECL_INIT(buffer_vld3_lane, uint, 32, 3);
VECT_VAR_DECL_INIT(buffer_vld3_lane, uint, 64, 3);
VECT_VAR_DECL_INIT(buffer_vld3_lane, poly, 8, 3);
VECT_VAR_DECL_INIT(buffer_vld3_lane, poly, 16, 3);
+VECT_VAR_DECL_INIT(buffer_vld3_lane, float, 16, 3);
VECT_VAR_DECL_INIT(buffer_vld3_lane, float, 32, 3);
/* Input buffers for vld4_lane. */
@@ -255,6 +284,7 @@ VECT_VAR_DECL_INIT(buffer_vld4_lane, uint, 32, 4);
VECT_VAR_DECL_INIT(buffer_vld4_lane, uint, 64, 4);
VECT_VAR_DECL_INIT(buffer_vld4_lane, poly, 8, 4);
VECT_VAR_DECL_INIT(buffer_vld4_lane, poly, 16, 4);
+VECT_VAR_DECL_INIT(buffer_vld4_lane, float, 16, 4);
VECT_VAR_DECL_INIT(buffer_vld4_lane, float, 32, 4);
void exec_vstX_lane (void)
@@ -302,7 +332,7 @@ void exec_vstX_lane (void)
/* We need all variants in 64 bits, but there is no 64x2 variant,
nor 128 bits vectors of int8/uint8/poly8. */
-#define DECL_ALL_VSTX_LANE(X) \
+#define DECL_ALL_VSTX_LANE_NO_FP16(X) \
DECL_VSTX_LANE(int, 8, 8, X); \
DECL_VSTX_LANE(int, 16, 4, X); \
DECL_VSTX_LANE(int, 32, 2, X); \
@@ -319,11 +349,20 @@ void exec_vstX_lane (void)
DECL_VSTX_LANE(poly, 16, 8, X); \
DECL_VSTX_LANE(float, 32, 4, X)
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+#define DECL_ALL_VSTX_LANE(X) \
+ DECL_ALL_VSTX_LANE_NO_FP16(X); \
+ DECL_VSTX_LANE(float, 16, 4, X); \
+ DECL_VSTX_LANE(float, 16, 8, X)
+#else
+#define DECL_ALL_VSTX_LANE(X) DECL_ALL_VSTX_LANE_NO_FP16(X)
+#endif
+
#define DUMMY_ARRAY(V, T, W, N, L) VECT_VAR_DECL(V,T,W,N)[N*L]
/* Use the same lanes regardless of the size of the array (X), for
simplicity. */
-#define TEST_ALL_VSTX_LANE(X) \
+#define TEST_ALL_VSTX_LANE_NO_FP16(X) \
TEST_VSTX_LANE(, int, s, 8, 8, X, 7); \
TEST_VSTX_LANE(, int, s, 16, 4, X, 2); \
TEST_VSTX_LANE(, int, s, 32, 2, X, 0); \
@@ -340,7 +379,16 @@ void exec_vstX_lane (void)
TEST_VSTX_LANE(q, poly, p, 16, 8, X, 5); \
TEST_VSTX_LANE(q, float, f, 32, 4, X, 2)
-#define TEST_ALL_EXTRA_CHUNKS(X, Y) \
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+#define TEST_ALL_VSTX_LANE(X) \
+ TEST_ALL_VSTX_LANE_NO_FP16(X); \
+ TEST_VSTX_LANE(, float, f, 16, 4, X, 2); \
+ TEST_VSTX_LANE(q, float, f, 16, 8, X, 6)
+#else
+#define TEST_ALL_VSTX_LANE(X) TEST_ALL_VSTX_LANE_NO_FP16(X)
+#endif
+
+#define TEST_ALL_EXTRA_CHUNKS_NO_FP16(X, Y) \
TEST_EXTRA_CHUNK(int, 8, 8, X, Y); \
TEST_EXTRA_CHUNK(int, 16, 4, X, Y); \
TEST_EXTRA_CHUNK(int, 32, 2, X, Y); \
@@ -357,6 +405,15 @@ void exec_vstX_lane (void)
TEST_EXTRA_CHUNK(poly, 16, 8, X, Y); \
TEST_EXTRA_CHUNK(float, 32, 4, X, Y)
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+#define TEST_ALL_EXTRA_CHUNKS(X,Y) \
+ TEST_ALL_EXTRA_CHUNKS_NO_FP16(X, Y); \
+ TEST_EXTRA_CHUNK(float, 16, 4, X, Y); \
+ TEST_EXTRA_CHUNK(float, 16, 8, X, Y)
+#else
+#define TEST_ALL_EXTRA_CHUNKS(X,Y) TEST_ALL_EXTRA_CHUNKS_NO_FP16(X, Y)
+#endif
+
/* Declare the temporary buffers / variables. */
DECL_ALL_VSTX_LANE(2);
DECL_ALL_VSTX_LANE(3);
@@ -371,12 +428,18 @@ void exec_vstX_lane (void)
DUMMY_ARRAY(buffer_src, uint, 32, 2, 4);
DUMMY_ARRAY(buffer_src, poly, 8, 8, 4);
DUMMY_ARRAY(buffer_src, poly, 16, 4, 4);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ DUMMY_ARRAY(buffer_src, float, 16, 4, 4);
+#endif
DUMMY_ARRAY(buffer_src, float, 32, 2, 4);
DUMMY_ARRAY(buffer_src, int, 16, 8, 4);
DUMMY_ARRAY(buffer_src, int, 32, 4, 4);
DUMMY_ARRAY(buffer_src, uint, 16, 8, 4);
DUMMY_ARRAY(buffer_src, uint, 32, 4, 4);
DUMMY_ARRAY(buffer_src, poly, 16, 8, 4);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ DUMMY_ARRAY(buffer_src, float, 16, 8, 4);
+#endif
DUMMY_ARRAY(buffer_src, float, 32, 4, 4);
/* Check vst2_lane/vst2q_lane. */
@@ -400,6 +463,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st2_0, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st2_0, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st2_0, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st2_0, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st2_0, CMT);
+#endif
TEST_ALL_EXTRA_CHUNKS(2, 1);
#undef CMT
@@ -419,6 +486,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st2_1, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st2_1, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st2_1, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st2_1, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st2_1, CMT);
+#endif
/* Check vst3_lane/vst3q_lane. */
@@ -444,6 +515,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st3_0, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st3_0, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st3_0, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st3_0, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st3_0, CMT);
+#endif
TEST_ALL_EXTRA_CHUNKS(3, 1);
@@ -464,6 +539,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st3_1, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st3_1, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st3_1, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st3_1, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st3_1, CMT);
+#endif
TEST_ALL_EXTRA_CHUNKS(3, 2);
@@ -484,6 +563,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st3_2, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st3_2, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st3_2, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st3_2, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st3_2, CMT);
+#endif
/* Check vst4_lane/vst4q_lane. */
@@ -509,6 +592,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st4_0, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st4_0, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st4_0, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st4_0, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st4_0, CMT);
+#endif
TEST_ALL_EXTRA_CHUNKS(4, 1);
@@ -529,6 +616,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st4_1, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st4_1, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st4_1, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st4_1, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st4_1, CMT);
+#endif
TEST_ALL_EXTRA_CHUNKS(4, 2);
@@ -549,6 +640,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st4_2, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st4_2, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st4_2, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st4_2, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st4_2, CMT);
+#endif
TEST_ALL_EXTRA_CHUNKS(4, 3);
@@ -569,6 +664,10 @@ void exec_vstX_lane (void)
CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_st4_3, CMT);
CHECK(TEST_MSG, poly, 16, 8, PRIx16, expected_st4_3, CMT);
CHECK_FP(TEST_MSG, float, 32, 4, PRIx32, expected_st4_3, CMT);
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+ CHECK_FP(TEST_MSG, float, 16, 4, PRIx16, expected_st4_3, CMT);
+ CHECK_FP(TEST_MSG, float, 16, 8, PRIx16, expected_st4_3, CMT);
+#endif
}
int main (void)