Implement vaddlvaq using the new MVE builtins framework.
2022-10-25 Christophe Lyon <christophe.l...@arm.com>
gcc/
* config/arm/arm-mve-builtins-base.cc (vaddlvaq): New.
* config/arm/arm-mve-builtins-base.def (vaddlvaq): New.
* config/arm/arm-mve-builtins-base.h (vaddlvaq): New.
* config/arm/arm_mve.h (vaddlvaq): Remove.
(vaddlvaq_p): Remove.
(vaddlvaq_u32): Remove.
(vaddlvaq_s32): Remove.
(vaddlvaq_p_s32): Remove.
(vaddlvaq_p_u32): Remove.
(__arm_vaddlvaq_u32): Remove.
(__arm_vaddlvaq_s32): Remove.
(__arm_vaddlvaq_p_s32): Remove.
(__arm_vaddlvaq_p_u32): Remove.
(__arm_vaddlvaq): Remove.
(__arm_vaddlvaq_p): Remove.
---
gcc/config/arm/arm-mve-builtins-base.cc | 1 +
gcc/config/arm/arm-mve-builtins-base.def | 1 +
gcc/config/arm/arm-mve-builtins-base.h | 1 +
gcc/config/arm/arm_mve.h | 74 ------------------------
4 files changed, 3 insertions(+), 74 deletions(-)
diff --git a/gcc/config/arm/arm-mve-builtins-base.cc
b/gcc/config/arm/arm-mve-builtins-base.cc
index 2dec15ac0b1..070a41c2d89 100644
--- a/gcc/config/arm/arm-mve-builtins-base.cc
+++ b/gcc/config/arm/arm-mve-builtins-base.cc
@@ -244,6 +244,7 @@ namespace arm_mve {
FUNCTION_WITHOUT_N (vabdq, VABDQ)
FUNCTION (vabsq, unspec_based_mve_function_exact_insn, (ABS, ABS, ABS, -1, -1,
-1, VABSQ_M_S, -1, VABSQ_M_F, -1, -1, -1))
FUNCTION_WITH_RTX_M_N (vaddq, PLUS, VADDQ)
+FUNCTION_PRED_P_S_U (vaddlvaq, VADDLVAQ)
FUNCTION_PRED_P_S_U (vaddlvq, VADDLVQ)
FUNCTION_PRED_P_S_U (vaddvq, VADDVQ)
FUNCTION_PRED_P_S_U (vaddvaq, VADDVAQ)
diff --git a/gcc/config/arm/arm-mve-builtins-base.def
b/gcc/config/arm/arm-mve-builtins-base.def
index b0de5af1013..62d2050b86d 100644
--- a/gcc/config/arm/arm-mve-builtins-base.def
+++ b/gcc/config/arm/arm-mve-builtins-base.def
@@ -20,6 +20,7 @@
#define REQUIRES_FLOAT false
DEF_MVE_FUNCTION (vabdq, binary, all_integer, mx_or_none)
DEF_MVE_FUNCTION (vabsq, unary, all_signed, mx_or_none)
+DEF_MVE_FUNCTION (vaddlvaq, unary_widen_acc, integer_32, p_or_none)
DEF_MVE_FUNCTION (vaddlvq, unary_acc, integer_32, p_or_none)
DEF_MVE_FUNCTION (vaddq, binary_opt_n, all_integer, mx_or_none)
DEF_MVE_FUNCTION (vaddvaq, unary_int32_acc, all_integer, p_or_none)
diff --git a/gcc/config/arm/arm-mve-builtins-base.h
b/gcc/config/arm/arm-mve-builtins-base.h
index fa2e97fd461..59754a03977 100644
--- a/gcc/config/arm/arm-mve-builtins-base.h
+++ b/gcc/config/arm/arm-mve-builtins-base.h
@@ -25,6 +25,7 @@ namespace functions {
extern const function_base *const vabdq;
extern const function_base *const vabsq;
+extern const function_base *const vaddlvaq;
extern const function_base *const vaddlvq;
extern const function_base *const vaddq;
extern const function_base *const vaddvaq;
diff --git a/gcc/config/arm/arm_mve.h b/gcc/config/arm/arm_mve.h
index c0891b7592a..8b61593c6b0 100644
--- a/gcc/config/arm/arm_mve.h
+++ b/gcc/config/arm/arm_mve.h
@@ -66,7 +66,6 @@
#define vmlsldavq(__a, __b) __arm_vmlsldavq(__a, __b)
#define vmlaldavxq(__a, __b) __arm_vmlaldavxq(__a, __b)
#define vrmlaldavhq(__a, __b) __arm_vrmlaldavhq(__a, __b)
-#define vaddlvaq(__a, __b) __arm_vaddlvaq(__a, __b)
#define vrmlsldavhxq(__a, __b) __arm_vrmlsldavhxq(__a, __b)
#define vrmlsldavhq(__a, __b) __arm_vrmlsldavhq(__a, __b)
#define vrmlaldavhxq(__a, __b) __arm_vrmlaldavhxq(__a, __b)
@@ -103,7 +102,6 @@
#define vrmlaldavhaxq(__a, __b, __c) __arm_vrmlaldavhaxq(__a, __b, __c)
#define vrmlsldavhaq(__a, __b, __c) __arm_vrmlsldavhaq(__a, __b, __c)
#define vrmlsldavhaxq(__a, __b, __c) __arm_vrmlsldavhaxq(__a, __b, __c)
-#define vaddlvaq_p(__a, __b, __p) __arm_vaddlvaq_p(__a, __b, __p)
#define vrmlaldavhq_p(__a, __b, __p) __arm_vrmlaldavhq_p(__a, __b, __p)
#define vrmlaldavhxq_p(__a, __b, __p) __arm_vrmlaldavhxq_p(__a, __b, __p)
#define vrmlsldavhq_p(__a, __b, __p) __arm_vrmlsldavhq_p(__a, __b, __p)
@@ -474,14 +472,12 @@
#define vctp64q_m(__a, __p) __arm_vctp64q_m(__a, __p)
#define vctp32q_m(__a, __p) __arm_vctp32q_m(__a, __p)
#define vctp16q_m(__a, __p) __arm_vctp16q_m(__a, __p)
-#define vaddlvaq_u32(__a, __b) __arm_vaddlvaq_u32(__a, __b)
#define vrmlsldavhxq_s32(__a, __b) __arm_vrmlsldavhxq_s32(__a, __b)
#define vrmlsldavhq_s32(__a, __b) __arm_vrmlsldavhq_s32(__a, __b)
#define vrmlaldavhxq_s32(__a, __b) __arm_vrmlaldavhxq_s32(__a, __b)
#define vrmlaldavhq_s32(__a, __b) __arm_vrmlaldavhq_s32(__a, __b)
#define vcvttq_f16_f32(__a, __b) __arm_vcvttq_f16_f32(__a, __b)
#define vcvtbq_f16_f32(__a, __b) __arm_vcvtbq_f16_f32(__a, __b)
-#define vaddlvaq_s32(__a, __b) __arm_vaddlvaq_s32(__a, __b)
#define vabavq_s8(__a, __b, __c) __arm_vabavq_s8(__a, __b, __c)
#define vabavq_s16(__a, __b, __c) __arm_vabavq_s16(__a, __b, __c)
#define vabavq_s32(__a, __b, __c) __arm_vabavq_s32(__a, __b, __c)
@@ -615,7 +611,6 @@
#define vrmlaldavhaxq_s32(__a, __b, __c) __arm_vrmlaldavhaxq_s32(__a, __b, __c)
#define vrmlsldavhaq_s32(__a, __b, __c) __arm_vrmlsldavhaq_s32(__a, __b, __c)
#define vrmlsldavhaxq_s32(__a, __b, __c) __arm_vrmlsldavhaxq_s32(__a, __b, __c)
-#define vaddlvaq_p_s32(__a, __b, __p) __arm_vaddlvaq_p_s32(__a, __b, __p)
#define vcvtbq_m_f16_f32(__a, __b, __p) __arm_vcvtbq_m_f16_f32(__a, __b, __p)
#define vcvtbq_m_f32_f16(__inactive, __a, __p)
__arm_vcvtbq_m_f32_f16(__inactive, __a, __p)
#define vcvttq_m_f16_f32(__a, __b, __p) __arm_vcvttq_m_f16_f32(__a, __b, __p)
@@ -624,7 +619,6 @@
#define vrmlaldavhxq_p_s32(__a, __b, __p) __arm_vrmlaldavhxq_p_s32(__a, __b,
__p)
#define vrmlsldavhq_p_s32(__a, __b, __p) __arm_vrmlsldavhq_p_s32(__a, __b, __p)
#define vrmlsldavhxq_p_s32(__a, __b, __p) __arm_vrmlsldavhxq_p_s32(__a, __b,
__p)
-#define vaddlvaq_p_u32(__a, __b, __p) __arm_vaddlvaq_p_u32(__a, __b, __p)
#define vrmlaldavhq_p_u32(__a, __b, __p) __arm_vrmlaldavhq_p_u32(__a, __b, __p)
#define vmvnq_m_n_s16(__inactive, __imm, __p) __arm_vmvnq_m_n_s16(__inactive,
__imm, __p)
#define vcmlaq_f16(__a, __b, __c) __arm_vcmlaq_f16(__a, __b, __c)
@@ -2253,13 +2247,6 @@ __arm_vctp16q_m (uint32_t __a, mve_pred16_t __p)
return __builtin_mve_vctp16q_mv8bi (__a, __p);
}
-__extension__ extern __inline uint64_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vaddlvaq_u32 (uint64_t __a, uint32x4_t __b)
-{
- return __builtin_mve_vaddlvaq_uv4si (__a, __b);
-}
-
__extension__ extern __inline int64_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vrmlsldavhxq_s32 (int32x4_t __a, int32x4_t __b)
@@ -2288,13 +2275,6 @@ __arm_vrmlaldavhq_s32 (int32x4_t __a, int32x4_t __b)
return __builtin_mve_vrmlaldavhq_sv4si (__a, __b);
}
-__extension__ extern __inline int64_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vaddlvaq_s32 (int64_t __a, int32x4_t __b)
-{
- return __builtin_mve_vaddlvaq_sv4si (__a, __b);
-}
-
__extension__ extern __inline uint32_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vabavq_s8 (uint32_t __a, int8x16_t __b, int8x16_t __c)
@@ -3182,13 +3162,6 @@ __arm_vrmlsldavhaxq_s32 (int64_t __a, int32x4_t __b,
int32x4_t __c)
return __builtin_mve_vrmlsldavhaxq_sv4si (__a, __b, __c);
}
-__extension__ extern __inline int64_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vaddlvaq_p_s32 (int64_t __a, int32x4_t __b, mve_pred16_t __p)
-{
- return __builtin_mve_vaddlvaq_p_sv4si (__a, __b, __p);
-}
-
__extension__ extern __inline int64_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vrmlaldavhq_p_s32 (int32x4_t __a, int32x4_t __b, mve_pred16_t __p)
@@ -3217,13 +3190,6 @@ __arm_vrmlsldavhxq_p_s32 (int32x4_t __a, int32x4_t __b,
mve_pred16_t __p)
return __builtin_mve_vrmlsldavhxq_p_sv4si (__a, __b, __p);
}
-__extension__ extern __inline uint64_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vaddlvaq_p_u32 (uint64_t __a, uint32x4_t __b, mve_pred16_t __p)
-{
- return __builtin_mve_vaddlvaq_p_uv4si (__a, __b, __p);
-}
-
__extension__ extern __inline uint64_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vrmlaldavhq_p_u32 (uint32x4_t __a, uint32x4_t __b, mve_pred16_t __p)
@@ -10047,13 +10013,6 @@ __arm_vrmlaldavhq (uint32x4_t __a, uint32x4_t __b)
return __arm_vrmlaldavhq_u32 (__a, __b);
}
-__extension__ extern __inline uint64_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vaddlvaq (uint64_t __a, uint32x4_t __b)
-{
- return __arm_vaddlvaq_u32 (__a, __b);
-}
-
__extension__ extern __inline int64_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vrmlsldavhxq (int32x4_t __a, int32x4_t __b)
@@ -10082,13 +10041,6 @@ __arm_vrmlaldavhq (int32x4_t __a, int32x4_t __b)
return __arm_vrmlaldavhq_s32 (__a, __b);
}
-__extension__ extern __inline int64_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vaddlvaq (int64_t __a, int32x4_t __b)
-{
- return __arm_vaddlvaq_s32 (__a, __b);
-}
-
__extension__ extern __inline uint32_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vabavq (uint32_t __a, int8x16_t __b, int8x16_t __c)
@@ -10964,13 +10916,6 @@ __arm_vrmlsldavhaxq (int64_t __a, int32x4_t __b,
int32x4_t __c)
return __arm_vrmlsldavhaxq_s32 (__a, __b, __c);
}
-__extension__ extern __inline int64_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vaddlvaq_p (int64_t __a, int32x4_t __b, mve_pred16_t __p)
-{
- return __arm_vaddlvaq_p_s32 (__a, __b, __p);
-}
-
__extension__ extern __inline int64_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vrmlaldavhq_p (int32x4_t __a, int32x4_t __b, mve_pred16_t __p)
@@ -10999,13 +10944,6 @@ __arm_vrmlsldavhxq_p (int32x4_t __a, int32x4_t __b,
mve_pred16_t __p)
return __arm_vrmlsldavhxq_p_s32 (__a, __b, __p);
}
-__extension__ extern __inline uint64_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vaddlvaq_p (uint64_t __a, uint32x4_t __b, mve_pred16_t __p)
-{
- return __arm_vaddlvaq_p_u32 (__a, __b, __p);
-}
-
__extension__ extern __inline uint64_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vrmlaldavhq_p (uint32x4_t __a, uint32x4_t __b, mve_pred16_t __p)
@@ -18698,18 +18636,6 @@ extern void *__ARM_undef;
int (*)[__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]:
__arm_vabavq_p_u16(__p0, __ARM_mve_coerce(__p1, uint16x8_t),
__ARM_mve_coerce(__p2, uint16x8_t), p3), \
int (*)[__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]:
__arm_vabavq_p_u32(__p0, __ARM_mve_coerce(__p1, uint32x4_t),
__ARM_mve_coerce(__p2, uint32x4_t), p3));})
-#define __arm_vaddlvaq(p0,p1) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)])0, \
- int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int32x4_t]: __arm_vaddlvaq_s32
(__ARM_mve_coerce3(p0, int), __ARM_mve_coerce(__p1, int32x4_t)), \
- int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint32x4_t]: __arm_vaddlvaq_u32
(__ARM_mve_coerce3(p0, int), __ARM_mve_coerce(__p1, uint32x4_t)));})
-
-#define __arm_vaddlvaq_p(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)])0, \
- int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int32x4_t]:
__arm_vaddlvaq_p_s32 (__ARM_mve_coerce3(p0, int), __ARM_mve_coerce(__p1,
int32x4_t), p2), \
- int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint32x4_t]:
__arm_vaddlvaq_p_u32 (__ARM_mve_coerce3(p0, int), __ARM_mve_coerce(__p1,
uint32x4_t), p2));})
-
#define __arm_vmladavaq(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
__typeof(p1) __p1 = (p1); \
__typeof(p2) __p2 = (p2); \
--
2.34.1
