aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorAndrea Corallo <andrea.corallo@arm.com>2020-10-29 15:11:37 +0100
committerAndrea Corallo <andrea.corallo@arm.com>2020-11-03 14:21:27 +0100
commit6170a793b7f5b4658751fc3a5ca51c10d18651f9 (patch)
treed69c3c48787b73b7fb3fa919aea0ded8eb1cbb55
parent890076673d4ade665470cfa8e5c92702bccfd2ee (diff)
downloadgcc-6170a793b7f5b4658751fc3a5ca51c10d18651f9.zip
gcc-6170a793b7f5b4658751fc3a5ca51c10d18651f9.tar.gz
gcc-6170a793b7f5b4658751fc3a5ca51c10d18651f9.tar.bz2
arm: Add vst1_bf16 + vst1q_bf16 intrinsics
gcc/ChangeLog 2020-10-29 Andrea Corallo <andrea.corallo@arm.com> * config/arm/arm_neon.h (vst1_bf16, vst1q_bf16): Add intrinsics. * config/arm/arm_neon_builtins.def : Touch for: __builtin_neon_vst1v4bf, __builtin_neon_vst1v8bf. gcc/testsuite/ChangeLog 2020-10-29 Andrea Corallo <andrea.corallo@arm.com> * gcc.target/arm/simd/vst1_bf16_1.c: New test.
-rw-r--r--gcc/config/arm/arm_neon.h14
-rw-r--r--gcc/config/arm/arm_neon_builtins.def5
-rw-r--r--gcc/testsuite/gcc.target/arm/simd/vst1_bf16_1.c29
3 files changed, 46 insertions, 2 deletions
diff --git a/gcc/config/arm/arm_neon.h b/gcc/config/arm/arm_neon.h
index b77175e..24aad33 100644
--- a/gcc/config/arm/arm_neon.h
+++ b/gcc/config/arm/arm_neon.h
@@ -19510,6 +19510,20 @@ vbfmlaltq_laneq_f32 (float32x4_t __r, bfloat16x8_t __a, bfloat16x8_t __b,
}
__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst1_bf16 (bfloat16_t * __a, bfloat16x4_t __b)
+{
+ __builtin_neon_vst1v4bf (__a, __b);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_bf16 (bfloat16_t * __a, bfloat16x8_t __b)
+{
+ __builtin_neon_vst1v8bf (__a, __b);
+}
+
+__extension__ extern __inline void
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vst2_bf16 (bfloat16_t * __ptr, bfloat16x4x2_t __val)
{
diff --git a/gcc/config/arm/arm_neon_builtins.def b/gcc/config/arm/arm_neon_builtins.def
index 07eda44..e3ab628 100644
--- a/gcc/config/arm/arm_neon_builtins.def
+++ b/gcc/config/arm/arm_neon_builtins.def
@@ -317,8 +317,9 @@ VAR12 (LOAD1LANE, vld1_lane,
v8qi, v4hi, v2si, v2sf, di, v16qi, v8hi, v4si, v4sf, v2di, v4bf, v8bf)
VAR10 (LOAD1, vld1_dup,
v8qi, v4hi, v2si, v2sf, di, v16qi, v8hi, v4si, v4sf, v2di)
-VAR12 (STORE1, vst1,
- v8qi, v4hi, v4hf, v2si, v2sf, di, v16qi, v8hi, v8hf, v4si, v4sf, v2di)
+VAR14 (STORE1, vst1,
+ v8qi, v4hi, v4hf, v2si, v2sf, di, v16qi, v8hi, v8hf, v4si, v4sf, v2di,
+ v4bf, v8bf)
VAR14 (STORE1LANE, vst1_lane,
v8qi, v4hi, v4hf, v2si, v2sf, di, v16qi, v8hi, v8hf, v4si, v4sf, v2di, v4bf, v8bf)
VAR13 (LOAD1, vld2,
diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1_bf16_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1_bf16_1.c
new file mode 100644
index 0000000..06fb58e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1_bf16_1.c
@@ -0,0 +1,29 @@
+/* { dg-do assemble } */
+/* { dg-require-effective-target arm_v8_2a_bf16_neon_ok } */
+/* { dg-add-options arm_v8_2a_bf16_neon } */
+/* { dg-additional-options "-save-temps -O2 -mfloat-abi=hard" } */
+/* { dg-final { check-function-bodies "**" "" } } */
+
+#include "arm_neon.h"
+
+/*
+**test_vst1_bf16:
+** vst1.16 {d0}, \[r0\]
+** bx lr
+*/
+void
+test_vst1_bf16 (bfloat16_t *a, bfloat16x4_t b)
+{
+ vst1_bf16 (a, b);
+}
+
+/*
+**test_vst1q_bf16:
+** vst1.16 {d0-d1}, \[r0\]
+** bx lr
+*/
+void
+test_vst1q_bf16 (bfloat16_t *a, bfloat16x8_t b)
+{
+ vst1q_bf16 (a, b);
+}