From: Christophe Lyon <christophe.lyon@linaro.org>
To: gcc-patches@gcc.gnu.org, Kyrylo.Tkachov@arm.com,
richard.earnshaw@arm.com, richard.sandiford@arm.com
Cc: Christophe Lyon <christophe.lyon@linaro.org>
Subject: [PATCH 6/6] arm: [MVE intrinsics] rework vcmlaq
Date: Thu, 13 Jul 2023 10:22:24 +0000 [thread overview]
Message-ID: <20230713102224.1161596-6-christophe.lyon@linaro.org> (raw)
In-Reply-To: <20230713102224.1161596-1-christophe.lyon@linaro.org>
Implement vcmlaq using the new MVE builtins framework.
2023-07-13 Christophe Lyon <christophe.lyon@linaro.org>
gcc/
* config/arm/arm-mve-builtins-base.cc (vcmlaq, vcmlaq_rot90)
(vcmlaq_rot180, vcmlaq_rot270): New.
* config/arm/arm-mve-builtins-base.def (vcmlaq, vcmlaq_rot90)
(vcmlaq_rot180, vcmlaq_rot270): New.
* config/arm/arm-mve-builtins-base.h: (vcmlaq, vcmlaq_rot90)
(vcmlaq_rot180, vcmlaq_rot270): New.
* config/arm/arm-mve-builtins.cc
(function_instance::has_inactive_argument): Handle vcmlaq,
vcmlaq_rot90, vcmlaq_rot180, vcmlaq_rot270.
* config/arm/arm_mve.h (vcmlaq): Delete.
(vcmlaq_rot180): Delete.
(vcmlaq_rot270): Delete.
(vcmlaq_rot90): Delete.
(vcmlaq_m): Delete.
(vcmlaq_rot180_m): Delete.
(vcmlaq_rot270_m): Delete.
(vcmlaq_rot90_m): Delete.
(vcmlaq_f16): Delete.
(vcmlaq_rot180_f16): Delete.
(vcmlaq_rot270_f16): Delete.
(vcmlaq_rot90_f16): Delete.
(vcmlaq_f32): Delete.
(vcmlaq_rot180_f32): Delete.
(vcmlaq_rot270_f32): Delete.
(vcmlaq_rot90_f32): Delete.
(vcmlaq_m_f32): Delete.
(vcmlaq_m_f16): Delete.
(vcmlaq_rot180_m_f32): Delete.
(vcmlaq_rot180_m_f16): Delete.
(vcmlaq_rot270_m_f32): Delete.
(vcmlaq_rot270_m_f16): Delete.
(vcmlaq_rot90_m_f32): Delete.
(vcmlaq_rot90_m_f16): Delete.
(__arm_vcmlaq_f16): Delete.
(__arm_vcmlaq_rot180_f16): Delete.
(__arm_vcmlaq_rot270_f16): Delete.
(__arm_vcmlaq_rot90_f16): Delete.
(__arm_vcmlaq_f32): Delete.
(__arm_vcmlaq_rot180_f32): Delete.
(__arm_vcmlaq_rot270_f32): Delete.
(__arm_vcmlaq_rot90_f32): Delete.
(__arm_vcmlaq_m_f32): Delete.
(__arm_vcmlaq_m_f16): Delete.
(__arm_vcmlaq_rot180_m_f32): Delete.
(__arm_vcmlaq_rot180_m_f16): Delete.
(__arm_vcmlaq_rot270_m_f32): Delete.
(__arm_vcmlaq_rot270_m_f16): Delete.
(__arm_vcmlaq_rot90_m_f32): Delete.
(__arm_vcmlaq_rot90_m_f16): Delete.
(__arm_vcmlaq): Delete.
(__arm_vcmlaq_rot180): Delete.
(__arm_vcmlaq_rot270): Delete.
(__arm_vcmlaq_rot90): Delete.
(__arm_vcmlaq_m): Delete.
(__arm_vcmlaq_rot180_m): Delete.
(__arm_vcmlaq_rot270_m): Delete.
(__arm_vcmlaq_rot90_m): Delete.
---
gcc/config/arm/arm-mve-builtins-base.cc | 4 +
gcc/config/arm/arm-mve-builtins-base.def | 4 +
gcc/config/arm/arm-mve-builtins-base.h | 16 +-
gcc/config/arm/arm-mve-builtins.cc | 4 +
gcc/config/arm/arm_mve.h | 304 -----------------------
5 files changed, 22 insertions(+), 310 deletions(-)
diff --git a/gcc/config/arm/arm-mve-builtins-base.cc b/gcc/config/arm/arm-mve-builtins-base.cc
index 3ad8df304e8..e31095ae112 100644
--- a/gcc/config/arm/arm-mve-builtins-base.cc
+++ b/gcc/config/arm/arm-mve-builtins-base.cc
@@ -262,6 +262,10 @@ FUNCTION_WITH_RTX_M (vandq, AND, VANDQ)
FUNCTION_ONLY_N (vbrsrq, VBRSRQ)
FUNCTION (vcaddq_rot90, unspec_mve_function_exact_insn_rot, (UNSPEC_VCADD90, UNSPEC_VCADD90, UNSPEC_VCADD90, VCADDQ_ROT90_M_S, VCADDQ_ROT90_M_U, VCADDQ_ROT90_M_F))
FUNCTION (vcaddq_rot270, unspec_mve_function_exact_insn_rot, (UNSPEC_VCADD270, UNSPEC_VCADD270, UNSPEC_VCADD270, VCADDQ_ROT270_M_S, VCADDQ_ROT270_M_U, VCADDQ_ROT270_M_F))
+FUNCTION (vcmlaq, unspec_mve_function_exact_insn_rot, (-1, -1, UNSPEC_VCMLA, -1, -1, VCMLAQ_M_F))
+FUNCTION (vcmlaq_rot90, unspec_mve_function_exact_insn_rot, (-1, -1, UNSPEC_VCMLA90, -1, -1, VCMLAQ_ROT90_M_F))
+FUNCTION (vcmlaq_rot180, unspec_mve_function_exact_insn_rot, (-1, -1, UNSPEC_VCMLA180, -1, -1, VCMLAQ_ROT180_M_F))
+FUNCTION (vcmlaq_rot270, unspec_mve_function_exact_insn_rot, (-1, -1, UNSPEC_VCMLA270, -1, -1, VCMLAQ_ROT270_M_F))
FUNCTION (vcmulq, unspec_mve_function_exact_insn_rot, (-1, -1, UNSPEC_VCMUL, -1, -1, VCMULQ_M_F))
FUNCTION (vcmulq_rot90, unspec_mve_function_exact_insn_rot, (-1, -1, UNSPEC_VCMUL90, -1, -1, VCMULQ_ROT90_M_F))
FUNCTION (vcmulq_rot180, unspec_mve_function_exact_insn_rot, (-1, -1, UNSPEC_VCMUL180, -1, -1, VCMULQ_ROT180_M_F))
diff --git a/gcc/config/arm/arm-mve-builtins-base.def b/gcc/config/arm/arm-mve-builtins-base.def
index cbcf0d296cd..e7d466f2efd 100644
--- a/gcc/config/arm/arm-mve-builtins-base.def
+++ b/gcc/config/arm/arm-mve-builtins-base.def
@@ -158,6 +158,10 @@ DEF_MVE_FUNCTION (vandq, binary, all_float, mx_or_none)
DEF_MVE_FUNCTION (vbrsrq, binary_imm32, all_float, mx_or_none)
DEF_MVE_FUNCTION (vcaddq_rot90, binary, all_float, mx_or_none)
DEF_MVE_FUNCTION (vcaddq_rot270, binary, all_float, mx_or_none)
+DEF_MVE_FUNCTION (vcmlaq, ternary, all_float, m_or_none)
+DEF_MVE_FUNCTION (vcmlaq_rot90, ternary, all_float, m_or_none)
+DEF_MVE_FUNCTION (vcmlaq_rot180, ternary, all_float, m_or_none)
+DEF_MVE_FUNCTION (vcmlaq_rot270, ternary, all_float, m_or_none)
DEF_MVE_FUNCTION (vcmulq, binary, all_float, mx_or_none)
DEF_MVE_FUNCTION (vcmulq_rot90, binary, all_float, mx_or_none)
DEF_MVE_FUNCTION (vcmulq_rot180, binary, all_float, mx_or_none)
diff --git a/gcc/config/arm/arm-mve-builtins-base.h b/gcc/config/arm/arm-mve-builtins-base.h
index 875b333ebef..be3698b4f4c 100644
--- a/gcc/config/arm/arm-mve-builtins-base.h
+++ b/gcc/config/arm/arm-mve-builtins-base.h
@@ -33,14 +33,14 @@ extern const function_base *const vaddvaq;
extern const function_base *const vaddvq;
extern const function_base *const vandq;
extern const function_base *const vbrsrq;
-extern const function_base *const vcaddq_rot90;
extern const function_base *const vcaddq_rot270;
-extern const function_base *const vcmulq;
-extern const function_base *const vcmulq_rot90;
-extern const function_base *const vcmulq_rot180;
-extern const function_base *const vcmulq_rot270;
+extern const function_base *const vcaddq_rot90;
extern const function_base *const vclsq;
extern const function_base *const vclzq;
+extern const function_base *const vcmlaq;
+extern const function_base *const vcmlaq_rot180;
+extern const function_base *const vcmlaq_rot270;
+extern const function_base *const vcmlaq_rot90;
extern const function_base *const vcmpcsq;
extern const function_base *const vcmpeqq;
extern const function_base *const vcmpgeq;
@@ -49,6 +49,10 @@ extern const function_base *const vcmphiq;
extern const function_base *const vcmpleq;
extern const function_base *const vcmpltq;
extern const function_base *const vcmpneq;
+extern const function_base *const vcmulq;
+extern const function_base *const vcmulq_rot180;
+extern const function_base *const vcmulq_rot270;
+extern const function_base *const vcmulq_rot90;
extern const function_base *const vcreateq;
extern const function_base *const vdupq;
extern const function_base *const veorq;
@@ -56,8 +60,8 @@ extern const function_base *const vfmaq;
extern const function_base *const vfmasq;
extern const function_base *const vfmsq;
extern const function_base *const vhaddq;
-extern const function_base *const vhcaddq_rot90;
extern const function_base *const vhcaddq_rot270;
+extern const function_base *const vhcaddq_rot90;
extern const function_base *const vhsubq;
extern const function_base *const vmaxaq;
extern const function_base *const vmaxavq;
diff --git a/gcc/config/arm/arm-mve-builtins.cc b/gcc/config/arm/arm-mve-builtins.cc
index 7033e41a571..3272ece6326 100644
--- a/gcc/config/arm/arm-mve-builtins.cc
+++ b/gcc/config/arm/arm-mve-builtins.cc
@@ -670,6 +670,10 @@ function_instance::has_inactive_argument () const
return false;
if (mode_suffix_id == MODE_r
+ || base == functions::vcmlaq
+ || base == functions::vcmlaq_rot90
+ || base == functions::vcmlaq_rot180
+ || base == functions::vcmlaq_rot270
|| base == functions::vcmpeqq
|| base == functions::vcmpneq
|| base == functions::vcmpgeq
diff --git a/gcc/config/arm/arm_mve.h b/gcc/config/arm/arm_mve.h
index b9d3a876369..88b2e77ffd9 100644
--- a/gcc/config/arm/arm_mve.h
+++ b/gcc/config/arm/arm_mve.h
@@ -159,18 +159,10 @@
#define vcvtq_m(__inactive, __a, __p) __arm_vcvtq_m(__inactive, __a, __p)
#define vcvtbq_m(__a, __b, __p) __arm_vcvtbq_m(__a, __b, __p)
#define vcvttq_m(__a, __b, __p) __arm_vcvttq_m(__a, __b, __p)
-#define vcmlaq(__a, __b, __c) __arm_vcmlaq(__a, __b, __c)
-#define vcmlaq_rot180(__a, __b, __c) __arm_vcmlaq_rot180(__a, __b, __c)
-#define vcmlaq_rot270(__a, __b, __c) __arm_vcmlaq_rot270(__a, __b, __c)
-#define vcmlaq_rot90(__a, __b, __c) __arm_vcmlaq_rot90(__a, __b, __c)
#define vcvtmq_m(__inactive, __a, __p) __arm_vcvtmq_m(__inactive, __a, __p)
#define vcvtnq_m(__inactive, __a, __p) __arm_vcvtnq_m(__inactive, __a, __p)
#define vcvtpq_m(__inactive, __a, __p) __arm_vcvtpq_m(__inactive, __a, __p)
#define vcvtq_m_n(__inactive, __a, __imm6, __p) __arm_vcvtq_m_n(__inactive, __a, __imm6, __p)
-#define vcmlaq_m(__a, __b, __c, __p) __arm_vcmlaq_m(__a, __b, __c, __p)
-#define vcmlaq_rot180_m(__a, __b, __c, __p) __arm_vcmlaq_rot180_m(__a, __b, __c, __p)
-#define vcmlaq_rot270_m(__a, __b, __c, __p) __arm_vcmlaq_rot270_m(__a, __b, __c, __p)
-#define vcmlaq_rot90_m(__a, __b, __c, __p) __arm_vcmlaq_rot90_m(__a, __b, __c, __p)
#define vcvtq_x(__a, __p) __arm_vcvtq_x(__a, __p)
#define vcvtq_x_n(__a, __imm6, __p) __arm_vcvtq_x_n(__a, __imm6, __p)
@@ -286,10 +278,6 @@
#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)
#define vcvttq_m_f32_f16(__inactive, __a, __p) __arm_vcvttq_m_f32_f16(__inactive, __a, __p)
-#define vcmlaq_f16(__a, __b, __c) __arm_vcmlaq_f16(__a, __b, __c)
-#define vcmlaq_rot180_f16(__a, __b, __c) __arm_vcmlaq_rot180_f16(__a, __b, __c)
-#define vcmlaq_rot270_f16(__a, __b, __c) __arm_vcmlaq_rot270_f16(__a, __b, __c)
-#define vcmlaq_rot90_f16(__a, __b, __c) __arm_vcmlaq_rot90_f16(__a, __b, __c)
#define vcvtmq_m_s16_f16(__inactive, __a, __p) __arm_vcvtmq_m_s16_f16(__inactive, __a, __p)
#define vcvtnq_m_s16_f16(__inactive, __a, __p) __arm_vcvtnq_m_s16_f16(__inactive, __a, __p)
#define vcvtpq_m_s16_f16(__inactive, __a, __p) __arm_vcvtpq_m_s16_f16(__inactive, __a, __p)
@@ -298,10 +286,6 @@
#define vcvtnq_m_u16_f16(__inactive, __a, __p) __arm_vcvtnq_m_u16_f16(__inactive, __a, __p)
#define vcvtpq_m_u16_f16(__inactive, __a, __p) __arm_vcvtpq_m_u16_f16(__inactive, __a, __p)
#define vcvtq_m_u16_f16(__inactive, __a, __p) __arm_vcvtq_m_u16_f16(__inactive, __a, __p)
-#define vcmlaq_f32(__a, __b, __c) __arm_vcmlaq_f32(__a, __b, __c)
-#define vcmlaq_rot180_f32(__a, __b, __c) __arm_vcmlaq_rot180_f32(__a, __b, __c)
-#define vcmlaq_rot270_f32(__a, __b, __c) __arm_vcmlaq_rot270_f32(__a, __b, __c)
-#define vcmlaq_rot90_f32(__a, __b, __c) __arm_vcmlaq_rot90_f32(__a, __b, __c)
#define vcvtmq_m_s32_f32(__inactive, __a, __p) __arm_vcvtmq_m_s32_f32(__inactive, __a, __p)
#define vcvtnq_m_s32_f32(__inactive, __a, __p) __arm_vcvtnq_m_s32_f32(__inactive, __a, __p)
#define vcvtpq_m_s32_f32(__inactive, __a, __p) __arm_vcvtpq_m_s32_f32(__inactive, __a, __p)
@@ -344,14 +328,6 @@
#define vmulltq_poly_m_p16(__inactive, __a, __b, __p) __arm_vmulltq_poly_m_p16(__inactive, __a, __b, __p)
#define vbicq_m_f32(__inactive, __a, __b, __p) __arm_vbicq_m_f32(__inactive, __a, __b, __p)
#define vbicq_m_f16(__inactive, __a, __b, __p) __arm_vbicq_m_f16(__inactive, __a, __b, __p)
-#define vcmlaq_m_f32(__a, __b, __c, __p) __arm_vcmlaq_m_f32(__a, __b, __c, __p)
-#define vcmlaq_m_f16(__a, __b, __c, __p) __arm_vcmlaq_m_f16(__a, __b, __c, __p)
-#define vcmlaq_rot180_m_f32(__a, __b, __c, __p) __arm_vcmlaq_rot180_m_f32(__a, __b, __c, __p)
-#define vcmlaq_rot180_m_f16(__a, __b, __c, __p) __arm_vcmlaq_rot180_m_f16(__a, __b, __c, __p)
-#define vcmlaq_rot270_m_f32(__a, __b, __c, __p) __arm_vcmlaq_rot270_m_f32(__a, __b, __c, __p)
-#define vcmlaq_rot270_m_f16(__a, __b, __c, __p) __arm_vcmlaq_rot270_m_f16(__a, __b, __c, __p)
-#define vcmlaq_rot90_m_f32(__a, __b, __c, __p) __arm_vcmlaq_rot90_m_f32(__a, __b, __c, __p)
-#define vcmlaq_rot90_m_f16(__a, __b, __c, __p) __arm_vcmlaq_rot90_m_f16(__a, __b, __c, __p)
#define vcvtq_m_n_s32_f32(__inactive, __a, __imm6, __p) __arm_vcvtq_m_n_s32_f32(__inactive, __a, __imm6, __p)
#define vcvtq_m_n_s16_f16(__inactive, __a, __imm6, __p) __arm_vcvtq_m_n_s16_f16(__inactive, __a, __imm6, __p)
#define vcvtq_m_n_u32_f32(__inactive, __a, __imm6, __p) __arm_vcvtq_m_n_u32_f32(__inactive, __a, __imm6, __p)
@@ -4645,34 +4621,6 @@ __arm_vcvttq_m_f32_f16 (float32x4_t __inactive, float16x8_t __a, mve_pred16_t __
return __builtin_mve_vcvttq_m_f32_f16v4sf (__inactive, __a, __p);
}
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_f16 (float16x8_t __a, float16x8_t __b, float16x8_t __c)
-{
- return __builtin_mve_vcmlaqv8hf (__a, __b, __c);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot180_f16 (float16x8_t __a, float16x8_t __b, float16x8_t __c)
-{
- return __builtin_mve_vcmlaq_rot180v8hf (__a, __b, __c);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot270_f16 (float16x8_t __a, float16x8_t __b, float16x8_t __c)
-{
- return __builtin_mve_vcmlaq_rot270v8hf (__a, __b, __c);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot90_f16 (float16x8_t __a, float16x8_t __b, float16x8_t __c)
-{
- return __builtin_mve_vcmlaq_rot90v8hf (__a, __b, __c);
-}
-
__extension__ extern __inline int16x8_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vcvtmq_m_s16_f16 (int16x8_t __inactive, float16x8_t __a, mve_pred16_t __p)
@@ -4729,34 +4677,6 @@ __arm_vcvtq_m_u16_f16 (uint16x8_t __inactive, float16x8_t __a, mve_pred16_t __p)
return __builtin_mve_vcvtq_m_from_f_uv8hi (__inactive, __a, __p);
}
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
-{
- return __builtin_mve_vcmlaqv4sf (__a, __b, __c);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot180_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
-{
- return __builtin_mve_vcmlaq_rot180v4sf (__a, __b, __c);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot270_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
-{
- return __builtin_mve_vcmlaq_rot270v4sf (__a, __b, __c);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot90_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
-{
- return __builtin_mve_vcmlaq_rot90v4sf (__a, __b, __c);
-}
-
__extension__ extern __inline int32x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vcvtmq_m_s32_f32 (int32x4_t __inactive, float32x4_t __a, mve_pred16_t __p)
@@ -4855,62 +4775,6 @@ __arm_vbicq_m_f16 (float16x8_t __inactive, float16x8_t __a, float16x8_t __b, mve
return __builtin_mve_vbicq_m_fv8hf (__inactive, __a, __b, __p);
}
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_m_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c, mve_pred16_t __p)
-{
- return __builtin_mve_vcmlaq_m_fv4sf (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_m_f16 (float16x8_t __a, float16x8_t __b, float16x8_t __c, mve_pred16_t __p)
-{
- return __builtin_mve_vcmlaq_m_fv8hf (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot180_m_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c, mve_pred16_t __p)
-{
- return __builtin_mve_vcmlaq_rot180_m_fv4sf (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot180_m_f16 (float16x8_t __a, float16x8_t __b, float16x8_t __c, mve_pred16_t __p)
-{
- return __builtin_mve_vcmlaq_rot180_m_fv8hf (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot270_m_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c, mve_pred16_t __p)
-{
- return __builtin_mve_vcmlaq_rot270_m_fv4sf (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot270_m_f16 (float16x8_t __a, float16x8_t __b, float16x8_t __c, mve_pred16_t __p)
-{
- return __builtin_mve_vcmlaq_rot270_m_fv8hf (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot90_m_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c, mve_pred16_t __p)
-{
- return __builtin_mve_vcmlaq_rot90_m_fv4sf (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot90_m_f16 (float16x8_t __a, float16x8_t __b, float16x8_t __c, mve_pred16_t __p)
-{
- return __builtin_mve_vcmlaq_rot90_m_fv8hf (__a, __b, __c, __p);
-}
-
__extension__ extern __inline int32x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vcvtq_m_n_s32_f32 (int32x4_t __inactive, float32x4_t __a, const int __imm6, mve_pred16_t __p)
@@ -8481,34 +8345,6 @@ __arm_vcvttq_m (float32x4_t __inactive, float16x8_t __a, mve_pred16_t __p)
return __arm_vcvttq_m_f32_f16 (__inactive, __a, __p);
}
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq (float16x8_t __a, float16x8_t __b, float16x8_t __c)
-{
- return __arm_vcmlaq_f16 (__a, __b, __c);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot180 (float16x8_t __a, float16x8_t __b, float16x8_t __c)
-{
- return __arm_vcmlaq_rot180_f16 (__a, __b, __c);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot270 (float16x8_t __a, float16x8_t __b, float16x8_t __c)
-{
- return __arm_vcmlaq_rot270_f16 (__a, __b, __c);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot90 (float16x8_t __a, float16x8_t __b, float16x8_t __c)
-{
- return __arm_vcmlaq_rot90_f16 (__a, __b, __c);
-}
-
__extension__ extern __inline int16x8_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vcvtmq_m (int16x8_t __inactive, float16x8_t __a, mve_pred16_t __p)
@@ -8565,34 +8401,6 @@ __arm_vcvtq_m (uint16x8_t __inactive, float16x8_t __a, mve_pred16_t __p)
return __arm_vcvtq_m_u16_f16 (__inactive, __a, __p);
}
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq (float32x4_t __a, float32x4_t __b, float32x4_t __c)
-{
- return __arm_vcmlaq_f32 (__a, __b, __c);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot180 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
-{
- return __arm_vcmlaq_rot180_f32 (__a, __b, __c);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot270 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
-{
- return __arm_vcmlaq_rot270_f32 (__a, __b, __c);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot90 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
-{
- return __arm_vcmlaq_rot90_f32 (__a, __b, __c);
-}
-
__extension__ extern __inline int32x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vcvtmq_m (int32x4_t __inactive, float32x4_t __a, mve_pred16_t __p)
@@ -8691,62 +8499,6 @@ __arm_vbicq_m (float16x8_t __inactive, float16x8_t __a, float16x8_t __b, mve_pre
return __arm_vbicq_m_f16 (__inactive, __a, __b, __p);
}
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_m (float32x4_t __a, float32x4_t __b, float32x4_t __c, mve_pred16_t __p)
-{
- return __arm_vcmlaq_m_f32 (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_m (float16x8_t __a, float16x8_t __b, float16x8_t __c, mve_pred16_t __p)
-{
- return __arm_vcmlaq_m_f16 (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot180_m (float32x4_t __a, float32x4_t __b, float32x4_t __c, mve_pred16_t __p)
-{
- return __arm_vcmlaq_rot180_m_f32 (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot180_m (float16x8_t __a, float16x8_t __b, float16x8_t __c, mve_pred16_t __p)
-{
- return __arm_vcmlaq_rot180_m_f16 (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot270_m (float32x4_t __a, float32x4_t __b, float32x4_t __c, mve_pred16_t __p)
-{
- return __arm_vcmlaq_rot270_m_f32 (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot270_m (float16x8_t __a, float16x8_t __b, float16x8_t __c, mve_pred16_t __p)
-{
- return __arm_vcmlaq_rot270_m_f16 (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot90_m (float32x4_t __a, float32x4_t __b, float32x4_t __c, mve_pred16_t __p)
-{
- return __arm_vcmlaq_rot90_m_f32 (__a, __b, __c, __p);
-}
-
-__extension__ extern __inline float16x8_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vcmlaq_rot90_m (float16x8_t __a, float16x8_t __b, float16x8_t __c, mve_pred16_t __p)
-{
- return __arm_vcmlaq_rot90_m_f16 (__a, __b, __c, __p);
-}
-
__extension__ extern __inline int32x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vcvtq_m_n (int32x4_t __inactive, float32x4_t __a, const int __imm6, mve_pred16_t __p)
@@ -9620,34 +9372,6 @@ extern void *__ARM_undef;
int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vcvtq_m_n_f16_u16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, uint16x8_t), p2, p3), \
int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vcvtq_m_n_f32_u32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, uint32x4_t), p2, p3));})
-#define __arm_vcmlaq(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- __typeof(p2) __p2 = (p2); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
- int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vcmlaq_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t)), \
- int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vcmlaq_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t)));})
-
-#define __arm_vcmlaq_rot180(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- __typeof(p2) __p2 = (p2); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
- int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vcmlaq_rot180_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t)), \
- int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vcmlaq_rot180_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t)));})
-
-#define __arm_vcmlaq_rot270(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- __typeof(p2) __p2 = (p2); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
- int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vcmlaq_rot270_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t)), \
- int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vcmlaq_rot270_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t)));})
-
-#define __arm_vcmlaq_rot90(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- __typeof(p2) __p2 = (p2); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
- int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vcmlaq_rot90_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t)), \
- int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vcmlaq_rot90_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t)));})
-
#define __arm_vcvtbq_m(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
__typeof(p1) __p1 = (p1); \
_Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)])0, \
@@ -9697,34 +9421,6 @@ extern void *__ARM_undef;
int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vbicq_m_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t), p3), \
int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vbicq_m_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t), p3));})
-#define __arm_vcmlaq_m(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- __typeof(p2) __p2 = (p2); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
- int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vcmlaq_m_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t), p3), \
- int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vcmlaq_m_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t), p3));})
-
-#define __arm_vcmlaq_rot180_m(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- __typeof(p2) __p2 = (p2); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
- int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vcmlaq_rot180_m_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t), p3), \
- int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vcmlaq_rot180_m_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t), p3));})
-
-#define __arm_vcmlaq_rot270_m(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- __typeof(p2) __p2 = (p2); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
- int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vcmlaq_rot270_m_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t), p3), \
- int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vcmlaq_rot270_m_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t), p3));})
-
-#define __arm_vcmlaq_rot90_m(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
- __typeof(p1) __p1 = (p1); \
- __typeof(p2) __p2 = (p2); \
- _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
- int (*)[__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t][__ARM_mve_type_float16x8_t]: __arm_vcmlaq_rot90_m_f16 (__ARM_mve_coerce(__p0, float16x8_t), __ARM_mve_coerce(__p1, float16x8_t), __ARM_mve_coerce(__p2, float16x8_t), p3), \
- int (*)[__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t][__ARM_mve_type_float32x4_t]: __arm_vcmlaq_rot90_m_f32 (__ARM_mve_coerce(__p0, float32x4_t), __ARM_mve_coerce(__p1, float32x4_t), __ARM_mve_coerce(__p2, float32x4_t), p3));})
-
#define __arm_vornq_m(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
__typeof(p1) __p1 = (p1); \
__typeof(p2) __p2 = (p2); \
--
2.34.1
next prev parent reply other threads:[~2023-07-13 10:22 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-13 10:22 [PATCH 1/6] arm: [MVE intrinsics] Factorize vcaddq vhcaddq Christophe Lyon
2023-07-13 10:22 ` [PATCH 2/6] arm: [MVE intrinsics] rework " Christophe Lyon
2023-07-13 10:22 ` [PATCH 3/6] arm: [MVE intrinsics factorize vcmulq Christophe Lyon
2023-07-13 10:22 ` [PATCH 4/6] arm: [MVE intrinsics] rework vcmulq Christophe Lyon
2023-07-13 10:22 ` [PATCH 5/6] arm: [MVE intrinsics] factorize vcmlaq Christophe Lyon
2023-07-13 10:22 ` Christophe Lyon [this message]
2023-07-14 16:18 ` [PATCH 1/6] arm: [MVE intrinsics] Factorize vcaddq vhcaddq Kyrylo Tkachov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230713102224.1161596-6-christophe.lyon@linaro.org \
--to=christophe.lyon@linaro.org \
--cc=Kyrylo.Tkachov@arm.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=richard.earnshaw@arm.com \
--cc=richard.sandiford@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).