[clang] 6454ff3 - [Clang Format] emmintrin.h smmintrin.h (NFC)
Xiang1 Zhang via cfe-commits
cfe-commits at lists.llvm.org
Thu Apr 21 18:30:16 PDT 2022
Author: Xiang1 Zhang
Date: 2022-04-22T09:29:52+08:00
New Revision: 6454ff35e0e7b0c0762c640031aa6c2b5d1f16ec
URL: https://github.com/llvm/llvm-project/commit/6454ff35e0e7b0c0762c640031aa6c2b5d1f16ec
DIFF: https://github.com/llvm/llvm-project/commit/6454ff35e0e7b0c0762c640031aa6c2b5d1f16ec.diff
LOG: [Clang Format] emmintrin.h smmintrin.h (NFC)
Added:
Modified:
clang/lib/Headers/emmintrin.h
clang/lib/Headers/smmintrin.h
Removed:
################################################################################
diff --git a/clang/lib/Headers/emmintrin.h b/clang/lib/Headers/emmintrin.h
index 942a0f788a8cd..b177d19f64533 100644
--- a/clang/lib/Headers/emmintrin.h
+++ b/clang/lib/Headers/emmintrin.h
@@ -20,16 +20,17 @@ typedef double __m128d __attribute__((__vector_size__(16), __aligned__(16)));
typedef long long __m128i __attribute__((__vector_size__(16), __aligned__(16)));
typedef double __m128d_u __attribute__((__vector_size__(16), __aligned__(1)));
-typedef long long __m128i_u __attribute__((__vector_size__(16), __aligned__(1)));
+typedef long long __m128i_u
+ __attribute__((__vector_size__(16), __aligned__(1)));
/* Type defines. */
-typedef double __v2df __attribute__ ((__vector_size__ (16)));
-typedef long long __v2di __attribute__ ((__vector_size__ (16)));
+typedef double __v2df __attribute__((__vector_size__(16)));
+typedef long long __v2di __attribute__((__vector_size__(16)));
typedef short __v8hi __attribute__((__vector_size__(16)));
typedef char __v16qi __attribute__((__vector_size__(16)));
/* Unsigned types */
-typedef unsigned long long __v2du __attribute__ ((__vector_size__ (16)));
+typedef unsigned long long __v2du __attribute__((__vector_size__(16)));
typedef unsigned short __v8hu __attribute__((__vector_size__(16)));
typedef unsigned char __v16qu __attribute__((__vector_size__(16)));
@@ -38,8 +39,12 @@ typedef unsigned char __v16qu __attribute__((__vector_size__(16)));
typedef signed char __v16qs __attribute__((__vector_size__(16)));
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse2"), __min_vector_width__(128)))
-#define __DEFAULT_FN_ATTRS_MMX __attribute__((__always_inline__, __nodebug__, __target__("mmx,sse2"), __min_vector_width__(64)))
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("sse2"), \
+ __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS_MMX \
+ __attribute__((__always_inline__, __nodebug__, __target__("mmx,sse2"), \
+ __min_vector_width__(64)))
/// Adds lower double-precision values in both operands and returns the
/// sum in the lower 64 bits of the result. The upper 64 bits of the result
@@ -56,9 +61,8 @@ typedef signed char __v16qs __attribute__((__vector_size__(16)));
/// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
/// sum of the lower 64 bits of both operands. The upper 64 bits are copied
/// from the upper 64 bits of the first source operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_add_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_add_sd(__m128d __a,
+ __m128d __b) {
__a[0] += __b[0];
return __a;
}
@@ -75,9 +79,8 @@ _mm_add_sd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing one of the source operands.
/// \returns A 128-bit vector of [2 x double] containing the sums of both
/// operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_add_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_add_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)((__v2df)__a + (__v2df)__b);
}
@@ -98,9 +101,8 @@ _mm_add_pd(__m128d __a, __m128d __b)
/// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
///
diff erence of the lower 64 bits of both operands. The upper 64 bits are
/// copied from the upper 64 bits of the first source operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_sub_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_sub_sd(__m128d __a,
+ __m128d __b) {
__a[0] -= __b[0];
return __a;
}
@@ -117,9 +119,8 @@ _mm_sub_sd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing the subtrahend.
/// \returns A 128-bit vector of [2 x double] containing the
diff erences between
/// both operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_sub_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_sub_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)((__v2df)__a - (__v2df)__b);
}
@@ -139,9 +140,8 @@ _mm_sub_pd(__m128d __a, __m128d __b)
/// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
/// product of the lower 64 bits of both operands. The upper 64 bits are
/// copied from the upper 64 bits of the first source operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_mul_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_mul_sd(__m128d __a,
+ __m128d __b) {
__a[0] *= __b[0];
return __a;
}
@@ -158,9 +158,8 @@ _mm_mul_sd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing one of the operands.
/// \returns A 128-bit vector of [2 x double] containing the products of both
/// operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_mul_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_mul_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)((__v2df)__a * (__v2df)__b);
}
@@ -181,9 +180,8 @@ _mm_mul_pd(__m128d __a, __m128d __b)
/// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
/// quotient of the lower 64 bits of both operands. The upper 64 bits are
/// copied from the upper 64 bits of the first source operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_div_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_div_sd(__m128d __a,
+ __m128d __b) {
__a[0] /= __b[0];
return __a;
}
@@ -201,9 +199,8 @@ _mm_div_sd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing the divisor.
/// \returns A 128-bit vector of [2 x double] containing the quotients of both
/// operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_div_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_div_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)((__v2df)__a / (__v2df)__b);
}
@@ -226,11 +223,10 @@ _mm_div_pd(__m128d __a, __m128d __b)
/// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
/// square root of the lower 64 bits of operand \a __b, and whose upper 64
/// bits are copied from the upper 64 bits of operand \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_sqrt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_sqrt_sd(__m128d __a,
+ __m128d __b) {
__m128d __c = __builtin_ia32_sqrtsd((__v2df)__b);
- return __extension__ (__m128d) { __c[0], __a[1] };
+ return __extension__(__m128d){__c[0], __a[1]};
}
/// Calculates the square root of the each of two values stored in a
@@ -244,9 +240,7 @@ _mm_sqrt_sd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector of [2 x double] containing the square roots of the
/// values in the operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_sqrt_pd(__m128d __a)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_sqrt_pd(__m128d __a) {
return __builtin_ia32_sqrtpd((__v2df)__a);
}
@@ -268,9 +262,8 @@ _mm_sqrt_pd(__m128d __a)
/// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
/// minimum value between both operands. The upper 64 bits are copied from
/// the upper 64 bits of the first source operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_min_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_min_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_minsd((__v2df)__a, (__v2df)__b);
}
@@ -288,9 +281,8 @@ _mm_min_sd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing one of the operands.
/// \returns A 128-bit vector of [2 x double] containing the minimum values
/// between both operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_min_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_min_pd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_minpd((__v2df)__a, (__v2df)__b);
}
@@ -312,9 +304,8 @@ _mm_min_pd(__m128d __a, __m128d __b)
/// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
/// maximum value between both operands. The upper 64 bits are copied from
/// the upper 64 bits of the first source operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_max_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_max_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_maxsd((__v2df)__a, (__v2df)__b);
}
@@ -332,9 +323,8 @@ _mm_max_sd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing one of the operands.
/// \returns A 128-bit vector of [2 x double] containing the maximum values
/// between both operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_max_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_max_pd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_maxpd((__v2df)__a, (__v2df)__b);
}
@@ -350,9 +340,8 @@ _mm_max_pd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing one of the source operands.
/// \returns A 128-bit vector of [2 x double] containing the bitwise AND of the
/// values between both operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_and_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_and_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)((__v2du)__a & (__v2du)__b);
}
@@ -371,9 +360,8 @@ _mm_and_pd(__m128d __a, __m128d __b)
/// \returns A 128-bit vector of [2 x double] containing the bitwise AND of the
/// values in the second operand and the one's complement of the first
/// operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_andnot_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_andnot_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)(~(__v2du)__a & (__v2du)__b);
}
@@ -389,9 +377,8 @@ _mm_andnot_pd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing one of the source operands.
/// \returns A 128-bit vector of [2 x double] containing the bitwise OR of the
/// values between both operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_or_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_or_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)((__v2du)__a | (__v2du)__b);
}
@@ -407,9 +394,8 @@ _mm_or_pd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double] containing one of the source operands.
/// \returns A 128-bit vector of [2 x double] containing the bitwise XOR of the
/// values between both operands.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_xor_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_xor_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)((__v2du)__a ^ (__v2du)__b);
}
@@ -426,9 +412,8 @@ _mm_xor_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpeq_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpeq_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpeqpd((__v2df)__a, (__v2df)__b);
}
@@ -446,9 +431,8 @@ _mm_cmpeq_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmplt_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmplt_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpltpd((__v2df)__a, (__v2df)__b);
}
@@ -467,9 +451,8 @@ _mm_cmplt_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmple_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmple_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmplepd((__v2df)__a, (__v2df)__b);
}
@@ -488,9 +471,8 @@ _mm_cmple_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpgt_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpgt_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpltpd((__v2df)__b, (__v2df)__a);
}
@@ -509,9 +491,8 @@ _mm_cmpgt_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpge_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpge_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmplepd((__v2df)__b, (__v2df)__a);
}
@@ -532,9 +513,8 @@ _mm_cmpge_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpord_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpord_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpordpd((__v2df)__a, (__v2df)__b);
}
@@ -556,9 +536,8 @@ _mm_cmpord_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpunord_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpunord_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpunordpd((__v2df)__a, (__v2df)__b);
}
@@ -577,9 +556,8 @@ _mm_cmpunord_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpneq_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpneq_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpneqpd((__v2df)__a, (__v2df)__b);
}
@@ -598,9 +576,8 @@ _mm_cmpneq_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpnlt_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpnlt_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpnltpd((__v2df)__a, (__v2df)__b);
}
@@ -619,9 +596,8 @@ _mm_cmpnlt_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpnle_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpnle_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpnlepd((__v2df)__a, (__v2df)__b);
}
@@ -640,9 +616,8 @@ _mm_cmpnle_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpngt_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpngt_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpnltpd((__v2df)__b, (__v2df)__a);
}
@@ -661,9 +636,8 @@ _mm_cmpngt_pd(__m128d __a, __m128d __b)
/// \param __b
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector containing the comparison results.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpnge_pd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpnge_pd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpnlepd((__v2df)__b, (__v2df)__a);
}
@@ -684,9 +658,8 @@ _mm_cmpnge_pd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpeq_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpeq_sd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpeqsd((__v2df)__a, (__v2df)__b);
}
@@ -709,9 +682,8 @@ _mm_cmpeq_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmplt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmplt_sd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpltsd((__v2df)__a, (__v2df)__b);
}
@@ -734,9 +706,8 @@ _mm_cmplt_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmple_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmple_sd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmplesd((__v2df)__a, (__v2df)__b);
}
@@ -759,11 +730,10 @@ _mm_cmple_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpgt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpgt_sd(__m128d __a,
+ __m128d __b) {
__m128d __c = __builtin_ia32_cmpltsd((__v2df)__b, (__v2df)__a);
- return __extension__ (__m128d) { __c[0], __a[1] };
+ return __extension__(__m128d){__c[0], __a[1]};
}
/// Compares the lower double-precision floating-point values in each of
@@ -785,11 +755,10 @@ _mm_cmpgt_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpge_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpge_sd(__m128d __a,
+ __m128d __b) {
__m128d __c = __builtin_ia32_cmplesd((__v2df)__b, (__v2df)__a);
- return __extension__ (__m128d) { __c[0], __a[1] };
+ return __extension__(__m128d){__c[0], __a[1]};
}
/// Compares the lower double-precision floating-point values in each of
@@ -813,9 +782,8 @@ _mm_cmpge_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpord_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpord_sd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpordsd((__v2df)__a, (__v2df)__b);
}
@@ -841,9 +809,8 @@ _mm_cmpord_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpunord_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpunord_sd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpunordsd((__v2df)__a, (__v2df)__b);
}
@@ -866,9 +833,8 @@ _mm_cmpunord_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpneq_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpneq_sd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpneqsd((__v2df)__a, (__v2df)__b);
}
@@ -891,9 +857,8 @@ _mm_cmpneq_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpnlt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpnlt_sd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpnltsd((__v2df)__a, (__v2df)__b);
}
@@ -916,9 +881,8 @@ _mm_cmpnlt_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpnle_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpnle_sd(__m128d __a,
+ __m128d __b) {
return (__m128d)__builtin_ia32_cmpnlesd((__v2df)__a, (__v2df)__b);
}
@@ -941,11 +905,10 @@ _mm_cmpnle_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpngt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpngt_sd(__m128d __a,
+ __m128d __b) {
__m128d __c = __builtin_ia32_cmpnltsd((__v2df)__b, (__v2df)__a);
- return __extension__ (__m128d) { __c[0], __a[1] };
+ return __extension__(__m128d){__c[0], __a[1]};
}
/// Compares the lower double-precision floating-point values in each of
@@ -967,11 +930,10 @@ _mm_cmpngt_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns A 128-bit vector. The lower 64 bits contains the comparison
/// results. The upper 64 bits are copied from the upper 64 bits of \a __a.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cmpnge_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cmpnge_sd(__m128d __a,
+ __m128d __b) {
__m128d __c = __builtin_ia32_cmpnlesd((__v2df)__b, (__v2df)__a);
- return __extension__ (__m128d) { __c[0], __a[1] };
+ return __extension__(__m128d){__c[0], __a[1]};
}
/// Compares the lower double-precision floating-point values in each of
@@ -992,9 +954,8 @@ _mm_cmpnge_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_comieq_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_comieq_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_comisdeq((__v2df)__a, (__v2df)__b);
}
@@ -1018,9 +979,8 @@ _mm_comieq_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_comilt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_comilt_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_comisdlt((__v2df)__a, (__v2df)__b);
}
@@ -1044,9 +1004,8 @@ _mm_comilt_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_comile_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_comile_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_comisdle((__v2df)__a, (__v2df)__b);
}
@@ -1070,9 +1029,8 @@ _mm_comile_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_comigt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_comigt_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_comisdgt((__v2df)__a, (__v2df)__b);
}
@@ -1096,9 +1054,8 @@ _mm_comigt_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_comige_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_comige_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_comisdge((__v2df)__a, (__v2df)__b);
}
@@ -1122,9 +1079,8 @@ _mm_comige_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 1 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_comineq_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_comineq_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_comisdneq((__v2df)__a, (__v2df)__b);
}
@@ -1146,9 +1102,8 @@ _mm_comineq_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_ucomieq_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_ucomieq_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_ucomisdeq((__v2df)__a, (__v2df)__b);
}
@@ -1172,9 +1127,8 @@ _mm_ucomieq_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_ucomilt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_ucomilt_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_ucomisdlt((__v2df)__a, (__v2df)__b);
}
@@ -1198,9 +1152,8 @@ _mm_ucomilt_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_ucomile_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_ucomile_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_ucomisdle((__v2df)__a, (__v2df)__b);
}
@@ -1224,9 +1177,8 @@ _mm_ucomile_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_ucomigt_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_ucomigt_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_ucomisdgt((__v2df)__a, (__v2df)__b);
}
@@ -1250,9 +1202,8 @@ _mm_ucomigt_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison results. If either of the two
/// lower double-precision values is NaN, 0 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_ucomige_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_ucomige_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_ucomisdge((__v2df)__a, (__v2df)__b);
}
@@ -1276,9 +1227,8 @@ _mm_ucomige_sd(__m128d __a, __m128d __b)
/// compared to the lower double-precision value of \a __a.
/// \returns An integer containing the comparison result. If either of the two
/// lower double-precision values is NaN, 1 is returned.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_ucomineq_sd(__m128d __a, __m128d __b)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_ucomineq_sd(__m128d __a,
+ __m128d __b) {
return __builtin_ia32_ucomisdneq((__v2df)__a, (__v2df)__b);
}
@@ -1295,9 +1245,7 @@ _mm_ucomineq_sd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector of [4 x float] whose lower 64 bits contain the
/// converted values. The upper 64 bits are set to zero.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_cvtpd_ps(__m128d __a)
-{
+static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_cvtpd_ps(__m128d __a) {
return __builtin_ia32_cvtpd2ps((__v2df)__a);
}
@@ -1315,9 +1263,7 @@ _mm_cvtpd_ps(__m128d __a)
/// floating-point elements are converted to double-precision values. The
/// upper two elements are unused.
/// \returns A 128-bit vector of [2 x double] containing the converted values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cvtps_pd(__m128 __a)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cvtps_pd(__m128 __a) {
return (__m128d) __builtin_convertvector(
__builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 0, 1), __v2df);
}
@@ -1338,9 +1284,7 @@ _mm_cvtps_pd(__m128 __a)
///
/// The upper two elements are unused.
/// \returns A 128-bit vector of [2 x double] containing the converted values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cvtepi32_pd(__m128i __a)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cvtepi32_pd(__m128i __a) {
return (__m128d) __builtin_convertvector(
__builtin_shufflevector((__v4si)__a, (__v4si)__a, 0, 1), __v2df);
}
@@ -1358,9 +1302,7 @@ _mm_cvtepi32_pd(__m128i __a)
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector of [4 x i32] whose lower 64 bits contain the
/// converted values. The upper 64 bits are set to zero.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtpd_epi32(__m128d __a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtpd_epi32(__m128d __a) {
return __builtin_ia32_cvtpd2dq((__v2df)__a);
}
@@ -1375,9 +1317,7 @@ _mm_cvtpd_epi32(__m128d __a)
/// A 128-bit vector of [2 x double]. The lower 64 bits are used in the
/// conversion.
/// \returns A 32-bit signed integer containing the converted value.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_cvtsd_si32(__m128d __a)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_cvtsd_si32(__m128d __a) {
return __builtin_ia32_cvtsd2si((__v2df)__a);
}
@@ -1400,9 +1340,8 @@ _mm_cvtsd_si32(__m128d __a)
/// \returns A 128-bit vector of [4 x float]. The lower 32 bits contain the
/// converted value from the second parameter. The upper 96 bits are copied
/// from the upper 96 bits of the first parameter.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_cvtsd_ss(__m128 __a, __m128d __b)
-{
+static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_cvtsd_ss(__m128 __a,
+ __m128d __b) {
return (__m128)__builtin_ia32_cvtsd2ss((__v4sf)__a, (__v2df)__b);
}
@@ -1423,9 +1362,8 @@ _mm_cvtsd_ss(__m128 __a, __m128d __b)
/// \returns A 128-bit vector of [2 x double]. The lower 64 bits contain the
/// converted value from the second parameter. The upper 64 bits are copied
/// from the upper 64 bits of the first parameter.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cvtsi32_sd(__m128d __a, int __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cvtsi32_sd(__m128d __a,
+ int __b) {
__a[0] = __b;
return __a;
}
@@ -1449,9 +1387,8 @@ _mm_cvtsi32_sd(__m128d __a, int __b)
/// \returns A 128-bit vector of [2 x double]. The lower 64 bits contain the
/// converted value from the second parameter. The upper 64 bits are copied
/// from the upper 64 bits of the first parameter.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cvtss_sd(__m128d __a, __m128 __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cvtss_sd(__m128d __a,
+ __m128 __b) {
__a[0] = __b[0];
return __a;
}
@@ -1473,9 +1410,7 @@ _mm_cvtss_sd(__m128d __a, __m128 __b)
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector of [4 x i32] whose lower 64 bits contain the
/// converted values. The upper 64 bits are set to zero.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvttpd_epi32(__m128d __a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvttpd_epi32(__m128d __a) {
return (__m128i)__builtin_ia32_cvttpd2dq((__v2df)__a);
}
@@ -1491,9 +1426,7 @@ _mm_cvttpd_epi32(__m128d __a)
/// A 128-bit vector of [2 x double]. The lower 64 bits are used in the
/// conversion.
/// \returns A 32-bit signed integer containing the converted value.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_cvttsd_si32(__m128d __a)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_cvttsd_si32(__m128d __a) {
return __builtin_ia32_cvttsd2si((__v2df)__a);
}
@@ -1508,9 +1441,7 @@ _mm_cvttsd_si32(__m128d __a)
/// \param __a
/// A 128-bit vector of [2 x double].
/// \returns A 64-bit vector of [2 x i32] containing the converted values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
-_mm_cvtpd_pi32(__m128d __a)
-{
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_cvtpd_pi32(__m128d __a) {
return (__m64)__builtin_ia32_cvtpd2pi((__v2df)__a);
}
@@ -1528,9 +1459,7 @@ _mm_cvtpd_pi32(__m128d __a)
/// \param __a
/// A 128-bit vector of [2 x double].
/// \returns A 64-bit vector of [2 x i32] containing the converted values.
-static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
-_mm_cvttpd_pi32(__m128d __a)
-{
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_cvttpd_pi32(__m128d __a) {
return (__m64)__builtin_ia32_cvttpd2pi((__v2df)__a);
}
@@ -1545,9 +1474,7 @@ _mm_cvttpd_pi32(__m128d __a)
/// \param __a
/// A 64-bit vector of [2 x i32].
/// \returns A 128-bit vector of [2 x double] containing the converted values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS_MMX
-_mm_cvtpi32_pd(__m64 __a)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS_MMX _mm_cvtpi32_pd(__m64 __a) {
return __builtin_ia32_cvtpi2pd((__v2si)__a);
}
@@ -1562,9 +1489,7 @@ _mm_cvtpi32_pd(__m64 __a)
/// A 128-bit vector of [2 x double]. The lower 64 bits are returned.
/// \returns A double-precision floating-point value copied from the lower 64
/// bits of \a __a.
-static __inline__ double __DEFAULT_FN_ATTRS
-_mm_cvtsd_f64(__m128d __a)
-{
+static __inline__ double __DEFAULT_FN_ATTRS _mm_cvtsd_f64(__m128d __a) {
return __a[0];
}
@@ -1579,10 +1504,8 @@ _mm_cvtsd_f64(__m128d __a)
/// A pointer to a 128-bit memory location. The address of the memory
/// location has to be 16-byte aligned.
/// \returns A 128-bit vector of [2 x double] containing the loaded values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_load_pd(double const *__dp)
-{
- return *(const __m128d*)__dp;
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_load_pd(double const *__dp) {
+ return *(const __m128d *)__dp;
}
/// Loads a double-precision floating-point value from a specified memory
@@ -1597,17 +1520,15 @@ _mm_load_pd(double const *__dp)
/// A pointer to a memory location containing a double-precision value.
/// \returns A 128-bit vector of [2 x double] containing the loaded and
/// duplicated values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_load1_pd(double const *__dp)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_load1_pd(double const *__dp) {
struct __mm_load1_pd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- double __u = ((const struct __mm_load1_pd_struct*)__dp)->__u;
- return __extension__ (__m128d){ __u, __u };
+ double __u = ((const struct __mm_load1_pd_struct *)__dp)->__u;
+ return __extension__(__m128d){__u, __u};
}
-#define _mm_load_pd1(dp) _mm_load1_pd(dp)
+#define _mm_load_pd1(dp) _mm_load1_pd(dp)
/// Loads two double-precision values, in reverse order, from an aligned
/// memory location into a 128-bit vector of [2 x double].
@@ -1623,10 +1544,8 @@ _mm_load1_pd(double const *__dp)
/// loaded in reverse order.
/// \returns A 128-bit vector of [2 x double] containing the reversed loaded
/// values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_loadr_pd(double const *__dp)
-{
- __m128d __u = *(const __m128d*)__dp;
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_loadr_pd(double const *__dp) {
+ __m128d __u = *(const __m128d *)__dp;
return __builtin_shufflevector((__v2df)__u, (__v2df)__u, 1, 0);
}
@@ -1641,13 +1560,11 @@ _mm_loadr_pd(double const *__dp)
/// A pointer to a 128-bit memory location. The address of the memory
/// location does not have to be aligned.
/// \returns A 128-bit vector of [2 x double] containing the loaded values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_loadu_pd(double const *__dp)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_loadu_pd(double const *__dp) {
struct __loadu_pd {
__m128d_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((const struct __loadu_pd*)__dp)->__v;
+ return ((const struct __loadu_pd *)__dp)->__v;
}
/// Loads a 64-bit integer value to the low element of a 128-bit integer
@@ -1661,14 +1578,12 @@ _mm_loadu_pd(double const *__dp)
/// A pointer to a 64-bit memory location. The address of the memory
/// location does not have to be aligned.
/// \returns A 128-bit vector of [2 x i64] containing the loaded value.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_loadu_si64(void const *__a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_loadu_si64(void const *__a) {
struct __loadu_si64 {
long long __v;
} __attribute__((__packed__, __may_alias__));
- long long __u = ((const struct __loadu_si64*)__a)->__v;
- return __extension__ (__m128i)(__v2di){__u, 0LL};
+ long long __u = ((const struct __loadu_si64 *)__a)->__v;
+ return __extension__(__m128i)(__v2di){__u, 0LL};
}
/// Loads a 32-bit integer value to the low element of a 128-bit integer
@@ -1682,14 +1597,12 @@ _mm_loadu_si64(void const *__a)
/// A pointer to a 32-bit memory location. The address of the memory
/// location does not have to be aligned.
/// \returns A 128-bit vector of [4 x i32] containing the loaded value.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_loadu_si32(void const *__a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_loadu_si32(void const *__a) {
struct __loadu_si32 {
int __v;
} __attribute__((__packed__, __may_alias__));
- int __u = ((const struct __loadu_si32*)__a)->__v;
- return __extension__ (__m128i)(__v4si){__u, 0, 0, 0};
+ int __u = ((const struct __loadu_si32 *)__a)->__v;
+ return __extension__(__m128i)(__v4si){__u, 0, 0, 0};
}
/// Loads a 16-bit integer value to the low element of a 128-bit integer
@@ -1703,14 +1616,12 @@ _mm_loadu_si32(void const *__a)
/// A pointer to a 16-bit memory location. The address of the memory
/// location does not have to be aligned.
/// \returns A 128-bit vector of [8 x i16] containing the loaded value.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_loadu_si16(void const *__a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_loadu_si16(void const *__a) {
struct __loadu_si16 {
short __v;
} __attribute__((__packed__, __may_alias__));
- short __u = ((const struct __loadu_si16*)__a)->__v;
- return __extension__ (__m128i)(__v8hi){__u, 0, 0, 0, 0, 0, 0, 0};
+ short __u = ((const struct __loadu_si16 *)__a)->__v;
+ return __extension__(__m128i)(__v8hi){__u, 0, 0, 0, 0, 0, 0, 0};
}
/// Loads a 64-bit double-precision value to the low element of a
@@ -1724,14 +1635,12 @@ _mm_loadu_si16(void const *__a)
/// A pointer to a memory location containing a double-precision value.
/// The address of the memory location does not have to be aligned.
/// \returns A 128-bit vector of [2 x double] containing the loaded value.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_load_sd(double const *__dp)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_load_sd(double const *__dp) {
struct __mm_load_sd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- double __u = ((const struct __mm_load_sd_struct*)__dp)->__u;
- return __extension__ (__m128d){ __u, 0 };
+ double __u = ((const struct __mm_load_sd_struct *)__dp)->__u;
+ return __extension__(__m128d){__u, 0};
}
/// Loads a double-precision value into the high-order bits of a 128-bit
@@ -1751,14 +1660,13 @@ _mm_load_sd(double const *__dp)
/// [127:64] of the result. The address of the memory location does not have
/// to be aligned.
/// \returns A 128-bit vector of [2 x double] containing the moved values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_loadh_pd(__m128d __a, double const *__dp)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_loadh_pd(__m128d __a,
+ double const *__dp) {
struct __mm_loadh_pd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- double __u = ((const struct __mm_loadh_pd_struct*)__dp)->__u;
- return __extension__ (__m128d){ __a[0], __u };
+ double __u = ((const struct __mm_loadh_pd_struct *)__dp)->__u;
+ return __extension__(__m128d){__a[0], __u};
}
/// Loads a double-precision value into the low-order bits of a 128-bit
@@ -1778,14 +1686,13 @@ _mm_loadh_pd(__m128d __a, double const *__dp)
/// [63:0] of the result. The address of the memory location does not have to
/// be aligned.
/// \returns A 128-bit vector of [2 x double] containing the moved values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_loadl_pd(__m128d __a, double const *__dp)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_loadl_pd(__m128d __a,
+ double const *__dp) {
struct __mm_loadl_pd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- double __u = ((const struct __mm_loadl_pd_struct*)__dp)->__u;
- return __extension__ (__m128d){ __u, __a[1] };
+ double __u = ((const struct __mm_loadl_pd_struct *)__dp)->__u;
+ return __extension__(__m128d){__u, __a[1]};
}
/// Constructs a 128-bit floating-point vector of [2 x double] with
@@ -1799,9 +1706,7 @@ _mm_loadl_pd(__m128d __a, double const *__dp)
///
/// \returns A 128-bit floating-point vector of [2 x double] with unspecified
/// content.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_undefined_pd(void)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_undefined_pd(void) {
return (__m128d)__builtin_ia32_undef128();
}
@@ -1819,10 +1724,8 @@ _mm_undefined_pd(void)
/// \returns An initialized 128-bit floating-point vector of [2 x double]. The
/// lower 64 bits contain the value of the parameter. The upper 64 bits are
/// set to zero.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_set_sd(double __w)
-{
- return __extension__ (__m128d){ __w, 0 };
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_set_sd(double __w) {
+ return __extension__(__m128d){__w, 0};
}
/// Constructs a 128-bit floating-point vector of [2 x double], with each
@@ -1837,10 +1740,8 @@ _mm_set_sd(double __w)
/// A double-precision floating-point value used to initialize each vector
/// element of the result.
/// \returns An initialized 128-bit floating-point vector of [2 x double].
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_set1_pd(double __w)
-{
- return __extension__ (__m128d){ __w, __w };
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_set1_pd(double __w) {
+ return __extension__(__m128d){__w, __w};
}
/// Constructs a 128-bit floating-point vector of [2 x double], with each
@@ -1855,9 +1756,7 @@ _mm_set1_pd(double __w)
/// A double-precision floating-point value used to initialize each vector
/// element of the result.
/// \returns An initialized 128-bit floating-point vector of [2 x double].
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_set_pd1(double __w)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_set_pd1(double __w) {
return _mm_set1_pd(__w);
}
@@ -1875,10 +1774,9 @@ _mm_set_pd1(double __w)
/// A double-precision floating-point value used to initialize the lower 64
/// bits of the result.
/// \returns An initialized 128-bit floating-point vector of [2 x double].
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_set_pd(double __w, double __x)
-{
- return __extension__ (__m128d){ __x, __w };
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_set_pd(double __w,
+ double __x) {
+ return __extension__(__m128d){__x, __w};
}
/// Constructs a 128-bit floating-point vector of [2 x double],
@@ -1896,10 +1794,9 @@ _mm_set_pd(double __w, double __x)
/// A double-precision floating-point value used to initialize the upper 64
/// bits of the result.
/// \returns An initialized 128-bit floating-point vector of [2 x double].
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_setr_pd(double __w, double __x)
-{
- return __extension__ (__m128d){ __w, __x };
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_setr_pd(double __w,
+ double __x) {
+ return __extension__(__m128d){__w, __x};
}
/// Constructs a 128-bit floating-point vector of [2 x double]
@@ -1911,10 +1808,8 @@ _mm_setr_pd(double __w, double __x)
///
/// \returns An initialized 128-bit floating-point vector of [2 x double] with
/// all elements set to zero.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_setzero_pd(void)
-{
- return __extension__ (__m128d){ 0, 0 };
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_setzero_pd(void) {
+ return __extension__(__m128d){0, 0};
}
/// Constructs a 128-bit floating-point vector of [2 x double]. The lower
@@ -1932,9 +1827,8 @@ _mm_setzero_pd(void)
/// A 128-bit vector of [2 x double]. The lower 64 bits are written to the
/// lower 64 bits of the result.
/// \returns A 128-bit vector of [2 x double] containing the moved values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_move_sd(__m128d __a, __m128d __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_move_sd(__m128d __a,
+ __m128d __b) {
__a[0] = __b[0];
return __a;
}
@@ -1950,13 +1844,12 @@ _mm_move_sd(__m128d __a, __m128d __b)
/// A pointer to a 64-bit memory location.
/// \param __a
/// A 128-bit vector of [2 x double] containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_store_sd(double *__dp, __m128d __a)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_store_sd(double *__dp,
+ __m128d __a) {
struct __mm_store_sd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- ((struct __mm_store_sd_struct*)__dp)->__u = __a[0];
+ ((struct __mm_store_sd_struct *)__dp)->__u = __a[0];
}
/// Moves packed double-precision values from a 128-bit vector of
@@ -1972,10 +1865,9 @@ _mm_store_sd(double *__dp, __m128d __a)
/// \param __a
/// A packed 128-bit vector of [2 x double] containing the values to be
/// moved.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_store_pd(double *__dp, __m128d __a)
-{
- *(__m128d*)__dp = __a;
+static __inline__ void __DEFAULT_FN_ATTRS _mm_store_pd(double *__dp,
+ __m128d __a) {
+ *(__m128d *)__dp = __a;
}
/// Moves the lower 64 bits of a 128-bit vector of [2 x double] twice to
@@ -1992,9 +1884,8 @@ _mm_store_pd(double *__dp, __m128d __a)
/// \param __a
/// A 128-bit vector of [2 x double] whose lower 64 bits are copied to each
/// of the values in \a __dp.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_store1_pd(double *__dp, __m128d __a)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_store1_pd(double *__dp,
+ __m128d __a) {
__a = __builtin_shufflevector((__v2df)__a, (__v2df)__a, 0, 0);
_mm_store_pd(__dp, __a);
}
@@ -2013,9 +1904,8 @@ _mm_store1_pd(double *__dp, __m128d __a)
/// \param __a
/// A 128-bit vector of [2 x double] whose lower 64 bits are copied to each
/// of the values in \a __dp.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_store_pd1(double *__dp, __m128d __a)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_store_pd1(double *__dp,
+ __m128d __a) {
_mm_store1_pd(__dp, __a);
}
@@ -2031,13 +1921,12 @@ _mm_store_pd1(double *__dp, __m128d __a)
/// location does not have to be aligned.
/// \param __a
/// A 128-bit vector of [2 x double] containing the values to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storeu_pd(double *__dp, __m128d __a)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storeu_pd(double *__dp,
+ __m128d __a) {
struct __storeu_pd {
__m128d_u __v;
} __attribute__((__packed__, __may_alias__));
- ((struct __storeu_pd*)__dp)->__v = __a;
+ ((struct __storeu_pd *)__dp)->__v = __a;
}
/// Stores two double-precision values, in reverse order, from a 128-bit
@@ -2054,9 +1943,8 @@ _mm_storeu_pd(double *__dp, __m128d __a)
/// \param __a
/// A 128-bit vector of [2 x double] containing the values to be reversed and
/// stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storer_pd(double *__dp, __m128d __a)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storer_pd(double *__dp,
+ __m128d __a) {
__a = __builtin_shufflevector((__v2df)__a, (__v2df)__a, 1, 0);
*(__m128d *)__dp = __a;
}
@@ -2072,13 +1960,12 @@ _mm_storer_pd(double *__dp, __m128d __a)
/// A pointer to a 64-bit memory location.
/// \param __a
/// A 128-bit vector of [2 x double] containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storeh_pd(double *__dp, __m128d __a)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storeh_pd(double *__dp,
+ __m128d __a) {
struct __mm_storeh_pd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- ((struct __mm_storeh_pd_struct*)__dp)->__u = __a[1];
+ ((struct __mm_storeh_pd_struct *)__dp)->__u = __a[1];
}
/// Stores the lower 64 bits of a 128-bit vector of [2 x double] to a
@@ -2092,13 +1979,12 @@ _mm_storeh_pd(double *__dp, __m128d __a)
/// A pointer to a 64-bit memory location.
/// \param __a
/// A 128-bit vector of [2 x double] containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storel_pd(double *__dp, __m128d __a)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storel_pd(double *__dp,
+ __m128d __a) {
struct __mm_storeh_pd_struct {
double __u;
} __attribute__((__packed__, __may_alias__));
- ((struct __mm_storeh_pd_struct*)__dp)->__u = __a[0];
+ ((struct __mm_storeh_pd_struct *)__dp)->__u = __a[0];
}
/// Adds the corresponding elements of two 128-bit vectors of [16 x i8],
@@ -2117,9 +2003,8 @@ _mm_storel_pd(double *__dp, __m128d __a)
/// A 128-bit vector of [16 x i8].
/// \returns A 128-bit vector of [16 x i8] containing the sums of both
/// parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_add_epi8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_add_epi8(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v16qu)__a + (__v16qu)__b);
}
@@ -2139,9 +2024,8 @@ _mm_add_epi8(__m128i __a, __m128i __b)
/// A 128-bit vector of [8 x i16].
/// \returns A 128-bit vector of [8 x i16] containing the sums of both
/// parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_add_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_add_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v8hu)__a + (__v8hu)__b);
}
@@ -2161,9 +2045,8 @@ _mm_add_epi16(__m128i __a, __m128i __b)
/// A 128-bit vector of [4 x i32].
/// \returns A 128-bit vector of [4 x i32] containing the sums of both
/// parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_add_epi32(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_add_epi32(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v4su)__a + (__v4su)__b);
}
@@ -2179,9 +2062,8 @@ _mm_add_epi32(__m128i __a, __m128i __b)
/// \param __b
/// A 64-bit integer.
/// \returns A 64-bit integer containing the sum of both parameters.
-static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
-_mm_add_si64(__m64 __a, __m64 __b)
-{
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_add_si64(__m64 __a,
+ __m64 __b) {
return (__m64)__builtin_ia32_paddq((__v1di)__a, (__v1di)__b);
}
@@ -2201,9 +2083,8 @@ _mm_add_si64(__m64 __a, __m64 __b)
/// A 128-bit vector of [2 x i64].
/// \returns A 128-bit vector of [2 x i64] containing the sums of both
/// parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_add_epi64(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_add_epi64(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v2du)__a + (__v2du)__b);
}
@@ -2222,9 +2103,8 @@ _mm_add_epi64(__m128i __a, __m128i __b)
/// A 128-bit signed [16 x i8] vector.
/// \returns A 128-bit signed [16 x i8] vector containing the saturated sums of
/// both parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_adds_epi8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_adds_epi8(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_add_sat((__v16qs)__a, (__v16qs)__b);
}
@@ -2244,9 +2124,8 @@ _mm_adds_epi8(__m128i __a, __m128i __b)
/// A 128-bit signed [8 x i16] vector.
/// \returns A 128-bit signed [8 x i16] vector containing the saturated sums of
/// both parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_adds_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_adds_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_add_sat((__v8hi)__a, (__v8hi)__b);
}
@@ -2265,9 +2144,8 @@ _mm_adds_epi16(__m128i __a, __m128i __b)
/// A 128-bit unsigned [16 x i8] vector.
/// \returns A 128-bit unsigned [16 x i8] vector containing the saturated sums
/// of both parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_adds_epu8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_adds_epu8(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_add_sat((__v16qu)__a, (__v16qu)__b);
}
@@ -2286,9 +2164,8 @@ _mm_adds_epu8(__m128i __a, __m128i __b)
/// A 128-bit unsigned [8 x i16] vector.
/// \returns A 128-bit unsigned [8 x i16] vector containing the saturated sums
/// of both parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_adds_epu16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_adds_epu16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_add_sat((__v8hu)__a, (__v8hu)__b);
}
@@ -2306,9 +2183,8 @@ _mm_adds_epu16(__m128i __a, __m128i __b)
/// A 128-bit unsigned [16 x i8] vector.
/// \returns A 128-bit unsigned [16 x i8] vector containing the rounded
/// averages of both parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_avg_epu8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_avg_epu8(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_ia32_pavgb128((__v16qi)__a, (__v16qi)__b);
}
@@ -2326,9 +2202,8 @@ _mm_avg_epu8(__m128i __a, __m128i __b)
/// A 128-bit unsigned [8 x i16] vector.
/// \returns A 128-bit unsigned [8 x i16] vector containing the rounded
/// averages of both parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_avg_epu16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_avg_epu16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_ia32_pavgw128((__v8hi)__a, (__v8hi)__b);
}
@@ -2352,9 +2227,8 @@ _mm_avg_epu16(__m128i __a, __m128i __b)
/// A 128-bit signed [8 x i16] vector.
/// \returns A 128-bit signed [4 x i32] vector containing the sums of products
/// of both parameters.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_madd_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_madd_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_ia32_pmaddwd128((__v8hi)__a, (__v8hi)__b);
}
@@ -2372,9 +2246,8 @@ _mm_madd_epi16(__m128i __a, __m128i __b)
/// A 128-bit signed [8 x i16] vector.
/// \returns A 128-bit signed [8 x i16] vector containing the greater value of
/// each comparison.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_max_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_max((__v8hi)__a, (__v8hi)__b);
}
@@ -2392,9 +2265,8 @@ _mm_max_epi16(__m128i __a, __m128i __b)
/// A 128-bit unsigned [16 x i8] vector.
/// \returns A 128-bit unsigned [16 x i8] vector containing the greater value of
/// each comparison.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_max_epu8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu8(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_max((__v16qu)__a, (__v16qu)__b);
}
@@ -2412,9 +2284,8 @@ _mm_max_epu8(__m128i __a, __m128i __b)
/// A 128-bit signed [8 x i16] vector.
/// \returns A 128-bit signed [8 x i16] vector containing the smaller value of
/// each comparison.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_min_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_min((__v8hi)__a, (__v8hi)__b);
}
@@ -2432,9 +2303,8 @@ _mm_min_epi16(__m128i __a, __m128i __b)
/// A 128-bit unsigned [16 x i8] vector.
/// \returns A 128-bit unsigned [16 x i8] vector containing the smaller value of
/// each comparison.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_min_epu8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epu8(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_min((__v16qu)__a, (__v16qu)__b);
}
@@ -2452,9 +2322,8 @@ _mm_min_epu8(__m128i __a, __m128i __b)
/// A 128-bit signed [8 x i16] vector.
/// \returns A 128-bit signed [8 x i16] vector containing the upper 16 bits of
/// each of the eight 32-bit products.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mulhi_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_mulhi_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_ia32_pmulhw128((__v8hi)__a, (__v8hi)__b);
}
@@ -2472,9 +2341,8 @@ _mm_mulhi_epi16(__m128i __a, __m128i __b)
/// A 128-bit unsigned [8 x i16] vector.
/// \returns A 128-bit unsigned [8 x i16] vector containing the upper 16 bits
/// of each of the eight 32-bit products.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mulhi_epu16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_mulhi_epu16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_ia32_pmulhuw128((__v8hi)__a, (__v8hi)__b);
}
@@ -2492,9 +2360,8 @@ _mm_mulhi_epu16(__m128i __a, __m128i __b)
/// A 128-bit signed [8 x i16] vector.
/// \returns A 128-bit signed [8 x i16] vector containing the lower 16 bits of
/// each of the eight 32-bit products.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mullo_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_mullo_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v8hu)__a * (__v8hu)__b);
}
@@ -2511,9 +2378,8 @@ _mm_mullo_epi16(__m128i __a, __m128i __b)
/// \param __b
/// A 64-bit integer containing one of the source operands.
/// \returns A 64-bit integer vector containing the product of both operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
-_mm_mul_su32(__m64 __a, __m64 __b)
-{
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_mul_su32(__m64 __a,
+ __m64 __b) {
return __builtin_ia32_pmuludq((__v2si)__a, (__v2si)__b);
}
@@ -2530,9 +2396,8 @@ _mm_mul_su32(__m64 __a, __m64 __b)
/// \param __b
/// A [2 x i64] vector containing one of the source operands.
/// \returns A [2 x i64] vector containing the product of both operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mul_epu32(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_mul_epu32(__m128i __a,
+ __m128i __b) {
return __builtin_ia32_pmuludq128((__v4si)__a, (__v4si)__b);
}
@@ -2552,9 +2417,8 @@ _mm_mul_epu32(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing one of the source operands.
/// \returns A [2 x i64] vector containing the sums of the sets of absolute
///
diff erences between both operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sad_epu8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sad_epu8(__m128i __a,
+ __m128i __b) {
return __builtin_ia32_psadbw128((__v16qi)__a, (__v16qi)__b);
}
@@ -2570,9 +2434,8 @@ _mm_sad_epu8(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing the subtrahends.
/// \returns A 128-bit integer vector containing the
diff erences of the values
/// in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sub_epi8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sub_epi8(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v16qu)__a - (__v16qu)__b);
}
@@ -2588,9 +2451,8 @@ _mm_sub_epi8(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing the subtrahends.
/// \returns A 128-bit integer vector containing the
diff erences of the values
/// in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sub_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sub_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v8hu)__a - (__v8hu)__b);
}
@@ -2606,9 +2468,8 @@ _mm_sub_epi16(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing the subtrahends.
/// \returns A 128-bit integer vector containing the
diff erences of the values
/// in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sub_epi32(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sub_epi32(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v4su)__a - (__v4su)__b);
}
@@ -2625,9 +2486,8 @@ _mm_sub_epi32(__m128i __a, __m128i __b)
/// A 64-bit integer vector containing the subtrahend.
/// \returns A 64-bit integer vector containing the
diff erence of the values in
/// the operands.
-static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX
-_mm_sub_si64(__m64 __a, __m64 __b)
-{
+static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_sub_si64(__m64 __a,
+ __m64 __b) {
return (__m64)__builtin_ia32_psubq((__v1di)__a, (__v1di)__b);
}
@@ -2643,9 +2503,8 @@ _mm_sub_si64(__m64 __a, __m64 __b)
/// A 128-bit integer vector containing the subtrahends.
/// \returns A 128-bit integer vector containing the
diff erences of the values
/// in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sub_epi64(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sub_epi64(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v2du)__a - (__v2du)__b);
}
@@ -2664,9 +2523,8 @@ _mm_sub_epi64(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing the subtrahends.
/// \returns A 128-bit integer vector containing the
diff erences of the values
/// in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_subs_epi8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_subs_epi8(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_sub_sat((__v16qs)__a, (__v16qs)__b);
}
@@ -2685,9 +2543,8 @@ _mm_subs_epi8(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing the subtrahends.
/// \returns A 128-bit integer vector containing the
diff erences of the values
/// in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_subs_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_subs_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_sub_sat((__v8hi)__a, (__v8hi)__b);
}
@@ -2705,9 +2562,8 @@ _mm_subs_epi16(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing the subtrahends.
/// \returns A 128-bit integer vector containing the unsigned integer
///
diff erences of the values in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_subs_epu8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_subs_epu8(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_sub_sat((__v16qu)__a, (__v16qu)__b);
}
@@ -2725,9 +2581,8 @@ _mm_subs_epu8(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing the subtrahends.
/// \returns A 128-bit integer vector containing the unsigned integer
///
diff erences of the values in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_subs_epu16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_subs_epu16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_elementwise_sub_sat((__v8hu)__a, (__v8hu)__b);
}
@@ -2743,9 +2598,8 @@ _mm_subs_epu16(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing one of the source operands.
/// \returns A 128-bit integer vector containing the bitwise AND of the values
/// in both operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_and_si128(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_and_si128(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v2du)__a & (__v2du)__b);
}
@@ -2763,9 +2617,8 @@ _mm_and_si128(__m128i __a, __m128i __b)
/// A 128-bit vector containing the right source operand.
/// \returns A 128-bit integer vector containing the bitwise AND of the one's
/// complement of the first operand and the values in the second operand.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_andnot_si128(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_andnot_si128(__m128i __a,
+ __m128i __b) {
return (__m128i)(~(__v2du)__a & (__v2du)__b);
}
/// Performs a bitwise OR of two 128-bit integer vectors.
@@ -2780,9 +2633,8 @@ _mm_andnot_si128(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing one of the source operands.
/// \returns A 128-bit integer vector containing the bitwise OR of the values
/// in both operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_or_si128(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_or_si128(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v2du)__a | (__v2du)__b);
}
@@ -2798,9 +2650,8 @@ _mm_or_si128(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing one of the source operands.
/// \returns A 128-bit integer vector containing the bitwise exclusive OR of the
/// values in both operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_xor_si128(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_xor_si128(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v2du)__a ^ (__v2du)__b);
}
@@ -2821,11 +2672,13 @@ _mm_xor_si128(__m128i __a, __m128i __b)
/// An immediate value specifying the number of bytes to left-shift operand
/// \a a.
/// \returns A 128-bit integer vector containing the left-shifted value.
-#define _mm_slli_si128(a, imm) \
- ((__m128i)__builtin_ia32_pslldqi128_byteshift((__v2di)(__m128i)(a), (int)(imm)))
+#define _mm_slli_si128(a, imm) \
+ ((__m128i)__builtin_ia32_pslldqi128_byteshift((__v2di)(__m128i)(a), \
+ (int)(imm)))
-#define _mm_bslli_si128(a, imm) \
- ((__m128i)__builtin_ia32_pslldqi128_byteshift((__v2di)(__m128i)(a), (int)(imm)))
+#define _mm_bslli_si128(a, imm) \
+ ((__m128i)__builtin_ia32_pslldqi128_byteshift((__v2di)(__m128i)(a), \
+ (int)(imm)))
/// Left-shifts each 16-bit value in the 128-bit integer vector operand
/// by the specified number of bits. Low-order bits are cleared.
@@ -2840,9 +2693,8 @@ _mm_xor_si128(__m128i __a, __m128i __b)
/// An integer value specifying the number of bits to left-shift each value
/// in operand \a __a.
/// \returns A 128-bit integer vector containing the left-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_slli_epi16(__m128i __a, int __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_slli_epi16(__m128i __a,
+ int __count) {
return (__m128i)__builtin_ia32_psllwi128((__v8hi)__a, __count);
}
@@ -2859,9 +2711,8 @@ _mm_slli_epi16(__m128i __a, int __count)
/// A 128-bit integer vector in which bits [63:0] specify the number of bits
/// to left-shift each value in operand \a __a.
/// \returns A 128-bit integer vector containing the left-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sll_epi16(__m128i __a, __m128i __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sll_epi16(__m128i __a,
+ __m128i __count) {
return (__m128i)__builtin_ia32_psllw128((__v8hi)__a, (__v8hi)__count);
}
@@ -2878,9 +2729,8 @@ _mm_sll_epi16(__m128i __a, __m128i __count)
/// An integer value specifying the number of bits to left-shift each value
/// in operand \a __a.
/// \returns A 128-bit integer vector containing the left-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_slli_epi32(__m128i __a, int __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_slli_epi32(__m128i __a,
+ int __count) {
return (__m128i)__builtin_ia32_pslldi128((__v4si)__a, __count);
}
@@ -2897,9 +2747,8 @@ _mm_slli_epi32(__m128i __a, int __count)
/// A 128-bit integer vector in which bits [63:0] specify the number of bits
/// to left-shift each value in operand \a __a.
/// \returns A 128-bit integer vector containing the left-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sll_epi32(__m128i __a, __m128i __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sll_epi32(__m128i __a,
+ __m128i __count) {
return (__m128i)__builtin_ia32_pslld128((__v4si)__a, (__v4si)__count);
}
@@ -2916,9 +2765,8 @@ _mm_sll_epi32(__m128i __a, __m128i __count)
/// An integer value specifying the number of bits to left-shift each value
/// in operand \a __a.
/// \returns A 128-bit integer vector containing the left-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_slli_epi64(__m128i __a, int __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_slli_epi64(__m128i __a,
+ int __count) {
return __builtin_ia32_psllqi128((__v2di)__a, __count);
}
@@ -2935,9 +2783,8 @@ _mm_slli_epi64(__m128i __a, int __count)
/// A 128-bit integer vector in which bits [63:0] specify the number of bits
/// to left-shift each value in operand \a __a.
/// \returns A 128-bit integer vector containing the left-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sll_epi64(__m128i __a, __m128i __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sll_epi64(__m128i __a,
+ __m128i __count) {
return __builtin_ia32_psllq128((__v2di)__a, (__v2di)__count);
}
@@ -2955,9 +2802,8 @@ _mm_sll_epi64(__m128i __a, __m128i __count)
/// An integer value specifying the number of bits to right-shift each value
/// in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srai_epi16(__m128i __a, int __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_srai_epi16(__m128i __a,
+ int __count) {
return (__m128i)__builtin_ia32_psrawi128((__v8hi)__a, __count);
}
@@ -2975,9 +2821,8 @@ _mm_srai_epi16(__m128i __a, int __count)
/// A 128-bit integer vector in which bits [63:0] specify the number of bits
/// to right-shift each value in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sra_epi16(__m128i __a, __m128i __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sra_epi16(__m128i __a,
+ __m128i __count) {
return (__m128i)__builtin_ia32_psraw128((__v8hi)__a, (__v8hi)__count);
}
@@ -2995,9 +2840,8 @@ _mm_sra_epi16(__m128i __a, __m128i __count)
/// An integer value specifying the number of bits to right-shift each value
/// in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srai_epi32(__m128i __a, int __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_srai_epi32(__m128i __a,
+ int __count) {
return (__m128i)__builtin_ia32_psradi128((__v4si)__a, __count);
}
@@ -3015,9 +2859,8 @@ _mm_srai_epi32(__m128i __a, int __count)
/// A 128-bit integer vector in which bits [63:0] specify the number of bits
/// to right-shift each value in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sra_epi32(__m128i __a, __m128i __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sra_epi32(__m128i __a,
+ __m128i __count) {
return (__m128i)__builtin_ia32_psrad128((__v4si)__a, (__v4si)__count);
}
@@ -3038,11 +2881,13 @@ _mm_sra_epi32(__m128i __a, __m128i __count)
/// An immediate value specifying the number of bytes to right-shift operand
/// \a a.
/// \returns A 128-bit integer vector containing the right-shifted value.
-#define _mm_srli_si128(a, imm) \
- ((__m128i)__builtin_ia32_psrldqi128_byteshift((__v2di)(__m128i)(a), (int)(imm)))
+#define _mm_srli_si128(a, imm) \
+ ((__m128i)__builtin_ia32_psrldqi128_byteshift((__v2di)(__m128i)(a), \
+ (int)(imm)))
-#define _mm_bsrli_si128(a, imm) \
- ((__m128i)__builtin_ia32_psrldqi128_byteshift((__v2di)(__m128i)(a), (int)(imm)))
+#define _mm_bsrli_si128(a, imm) \
+ ((__m128i)__builtin_ia32_psrldqi128_byteshift((__v2di)(__m128i)(a), \
+ (int)(imm)))
/// Right-shifts each of 16-bit values in the 128-bit integer vector
/// operand by the specified number of bits. High-order bits are cleared.
@@ -3057,9 +2902,8 @@ _mm_sra_epi32(__m128i __a, __m128i __count)
/// An integer value specifying the number of bits to right-shift each value
/// in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srli_epi16(__m128i __a, int __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_srli_epi16(__m128i __a,
+ int __count) {
return (__m128i)__builtin_ia32_psrlwi128((__v8hi)__a, __count);
}
@@ -3076,9 +2920,8 @@ _mm_srli_epi16(__m128i __a, int __count)
/// A 128-bit integer vector in which bits [63:0] specify the number of bits
/// to right-shift each value in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srl_epi16(__m128i __a, __m128i __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_srl_epi16(__m128i __a,
+ __m128i __count) {
return (__m128i)__builtin_ia32_psrlw128((__v8hi)__a, (__v8hi)__count);
}
@@ -3095,9 +2938,8 @@ _mm_srl_epi16(__m128i __a, __m128i __count)
/// An integer value specifying the number of bits to right-shift each value
/// in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srli_epi32(__m128i __a, int __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_srli_epi32(__m128i __a,
+ int __count) {
return (__m128i)__builtin_ia32_psrldi128((__v4si)__a, __count);
}
@@ -3114,9 +2956,8 @@ _mm_srli_epi32(__m128i __a, int __count)
/// A 128-bit integer vector in which bits [63:0] specify the number of bits
/// to right-shift each value in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srl_epi32(__m128i __a, __m128i __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_srl_epi32(__m128i __a,
+ __m128i __count) {
return (__m128i)__builtin_ia32_psrld128((__v4si)__a, (__v4si)__count);
}
@@ -3133,9 +2974,8 @@ _mm_srl_epi32(__m128i __a, __m128i __count)
/// An integer value specifying the number of bits to right-shift each value
/// in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srli_epi64(__m128i __a, int __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_srli_epi64(__m128i __a,
+ int __count) {
return __builtin_ia32_psrlqi128((__v2di)__a, __count);
}
@@ -3152,9 +2992,8 @@ _mm_srli_epi64(__m128i __a, int __count)
/// A 128-bit integer vector in which bits [63:0] specify the number of bits
/// to right-shift each value in operand \a __a.
/// \returns A 128-bit integer vector containing the right-shifted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srl_epi64(__m128i __a, __m128i __count)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_srl_epi64(__m128i __a,
+ __m128i __count) {
return __builtin_ia32_psrlq128((__v2di)__a, (__v2di)__count);
}
@@ -3171,9 +3010,8 @@ _mm_srl_epi64(__m128i __a, __m128i __count)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmpeq_epi8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmpeq_epi8(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v16qi)__a == (__v16qi)__b);
}
@@ -3190,9 +3028,8 @@ _mm_cmpeq_epi8(__m128i __a, __m128i __b)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmpeq_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmpeq_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v8hi)__a == (__v8hi)__b);
}
@@ -3209,9 +3046,8 @@ _mm_cmpeq_epi16(__m128i __a, __m128i __b)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmpeq_epi32(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmpeq_epi32(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v4si)__a == (__v4si)__b);
}
@@ -3229,9 +3065,8 @@ _mm_cmpeq_epi32(__m128i __a, __m128i __b)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmpgt_epi8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmpgt_epi8(__m128i __a,
+ __m128i __b) {
/* This function always performs a signed comparison, but __v16qi is a char
which may be signed or unsigned, so use __v16qs. */
return (__m128i)((__v16qs)__a > (__v16qs)__b);
@@ -3252,9 +3087,8 @@ _mm_cmpgt_epi8(__m128i __a, __m128i __b)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmpgt_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmpgt_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v8hi)__a > (__v8hi)__b);
}
@@ -3273,9 +3107,8 @@ _mm_cmpgt_epi16(__m128i __a, __m128i __b)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmpgt_epi32(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmpgt_epi32(__m128i __a,
+ __m128i __b) {
return (__m128i)((__v4si)__a > (__v4si)__b);
}
@@ -3294,9 +3127,8 @@ _mm_cmpgt_epi32(__m128i __a, __m128i __b)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmplt_epi8(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmplt_epi8(__m128i __a,
+ __m128i __b) {
return _mm_cmpgt_epi8(__b, __a);
}
@@ -3315,9 +3147,8 @@ _mm_cmplt_epi8(__m128i __a, __m128i __b)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmplt_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmplt_epi16(__m128i __a,
+ __m128i __b) {
return _mm_cmpgt_epi16(__b, __a);
}
@@ -3336,9 +3167,8 @@ _mm_cmplt_epi16(__m128i __a, __m128i __b)
/// \param __b
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmplt_epi32(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmplt_epi32(__m128i __a,
+ __m128i __b) {
return _mm_cmpgt_epi32(__b, __a);
}
@@ -3360,9 +3190,8 @@ _mm_cmplt_epi32(__m128i __a, __m128i __b)
/// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
/// converted value of the second operand. The upper 64 bits are copied from
/// the upper 64 bits of the first operand.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_cvtsi64_sd(__m128d __a, long long __b)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cvtsi64_sd(__m128d __a,
+ long long __b) {
__a[0] = __b;
return __a;
}
@@ -3378,9 +3207,7 @@ _mm_cvtsi64_sd(__m128d __a, long long __b)
/// A 128-bit vector of [2 x double]. The lower 64 bits are used in the
/// conversion.
/// \returns A 64-bit signed integer containing the converted value.
-static __inline__ long long __DEFAULT_FN_ATTRS
-_mm_cvtsd_si64(__m128d __a)
-{
+static __inline__ long long __DEFAULT_FN_ATTRS _mm_cvtsd_si64(__m128d __a) {
return __builtin_ia32_cvtsd2si64((__v2df)__a);
}
@@ -3396,9 +3223,7 @@ _mm_cvtsd_si64(__m128d __a)
/// A 128-bit vector of [2 x double]. The lower 64 bits are used in the
/// conversion.
/// \returns A 64-bit signed integer containing the converted value.
-static __inline__ long long __DEFAULT_FN_ATTRS
-_mm_cvttsd_si64(__m128d __a)
-{
+static __inline__ long long __DEFAULT_FN_ATTRS _mm_cvttsd_si64(__m128d __a) {
return __builtin_ia32_cvttsd2si64((__v2df)__a);
}
#endif
@@ -3412,10 +3237,8 @@ _mm_cvttsd_si64(__m128d __a)
/// \param __a
/// A 128-bit integer vector.
/// \returns A 128-bit vector of [4 x float] containing the converted values.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_cvtepi32_ps(__m128i __a)
-{
- return (__m128)__builtin_convertvector((__v4si)__a, __v4sf);
+static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_cvtepi32_ps(__m128i __a) {
+ return (__m128) __builtin_convertvector((__v4si)__a, __v4sf);
}
/// Converts a vector of [4 x float] into a vector of [4 x i32].
@@ -3428,9 +3251,7 @@ _mm_cvtepi32_ps(__m128i __a)
/// A 128-bit vector of [4 x float].
/// \returns A 128-bit integer vector of [4 x i32] containing the converted
/// values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtps_epi32(__m128 __a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtps_epi32(__m128 __a) {
return (__m128i)__builtin_ia32_cvtps2dq((__v4sf)__a);
}
@@ -3445,9 +3266,7 @@ _mm_cvtps_epi32(__m128 __a)
/// \param __a
/// A 128-bit vector of [4 x float].
/// \returns A 128-bit vector of [4 x i32] containing the converted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvttps_epi32(__m128 __a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvttps_epi32(__m128 __a) {
return (__m128i)__builtin_ia32_cvttps2dq((__v4sf)__a);
}
@@ -3461,10 +3280,8 @@ _mm_cvttps_epi32(__m128 __a)
/// \param __a
/// A 32-bit signed integer operand.
/// \returns A 128-bit vector of [4 x i32].
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtsi32_si128(int __a)
-{
- return __extension__ (__m128i)(__v4si){ __a, 0, 0, 0 };
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtsi32_si128(int __a) {
+ return __extension__(__m128i)(__v4si){__a, 0, 0, 0};
}
#ifdef __x86_64__
@@ -3478,10 +3295,8 @@ _mm_cvtsi32_si128(int __a)
/// \param __a
/// A 64-bit signed integer operand containing the value to be converted.
/// \returns A 128-bit vector of [2 x i64] containing the converted value.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtsi64_si128(long long __a)
-{
- return __extension__ (__m128i)(__v2di){ __a, 0 };
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtsi64_si128(long long __a) {
+ return __extension__(__m128i)(__v2di){__a, 0};
}
#endif
@@ -3496,9 +3311,7 @@ _mm_cvtsi64_si128(long long __a)
/// A vector of [4 x i32]. The least significant 32 bits are moved to the
/// destination.
/// \returns A 32-bit signed integer containing the moved value.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_cvtsi128_si32(__m128i __a)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_cvtsi128_si32(__m128i __a) {
__v4si __b = (__v4si)__a;
return __b[0];
}
@@ -3515,9 +3328,7 @@ _mm_cvtsi128_si32(__m128i __a)
/// A vector of [2 x i64]. The least significant 64 bits are moved to the
/// destination.
/// \returns A 64-bit signed integer containing the moved value.
-static __inline__ long long __DEFAULT_FN_ATTRS
-_mm_cvtsi128_si64(__m128i __a)
-{
+static __inline__ long long __DEFAULT_FN_ATTRS _mm_cvtsi128_si64(__m128i __a) {
return __a[0];
}
#endif
@@ -3533,8 +3344,7 @@ _mm_cvtsi128_si64(__m128i __a)
/// An aligned pointer to a memory location containing integer values.
/// \returns A 128-bit integer vector containing the moved values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_load_si128(__m128i const *__p)
-{
+_mm_load_si128(__m128i const *__p) {
return *__p;
}
@@ -3549,12 +3359,11 @@ _mm_load_si128(__m128i const *__p)
/// A pointer to a memory location containing integer values.
/// \returns A 128-bit integer vector containing the moved values.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_loadu_si128(__m128i_u const *__p)
-{
+_mm_loadu_si128(__m128i_u const *__p) {
struct __loadu_si128 {
__m128i_u __v;
} __attribute__((__packed__, __may_alias__));
- return ((const struct __loadu_si128*)__p)->__v;
+ return ((const struct __loadu_si128 *)__p)->__v;
}
/// Returns a vector of [2 x i64] where the lower element is taken from
@@ -3570,12 +3379,12 @@ _mm_loadu_si128(__m128i_u const *__p)
/// \returns A 128-bit vector of [2 x i64]. The lower order bits contain the
/// moved value. The higher order bits are cleared.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_loadl_epi64(__m128i_u const *__p)
-{
+_mm_loadl_epi64(__m128i_u const *__p) {
struct __mm_loadl_epi64_struct {
long long __u;
} __attribute__((__packed__, __may_alias__));
- return __extension__ (__m128i) { ((const struct __mm_loadl_epi64_struct*)__p)->__u, 0};
+ return __extension__(__m128i){
+ ((const struct __mm_loadl_epi64_struct *)__p)->__u, 0};
}
/// Generates a 128-bit vector of [4 x i32] with unspecified content.
@@ -3587,9 +3396,7 @@ _mm_loadl_epi64(__m128i_u const *__p)
/// This intrinsic has no corresponding instruction.
///
/// \returns A 128-bit vector of [4 x i32] with unspecified content.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_undefined_si128(void)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_undefined_si128(void) {
return (__m128i)__builtin_ia32_undef128();
}
@@ -3609,10 +3416,9 @@ _mm_undefined_si128(void)
/// destination vector of [2 x i64].
/// \returns An initialized 128-bit vector of [2 x i64] containing the values
/// provided in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set_epi64x(long long __q1, long long __q0)
-{
- return __extension__ (__m128i)(__v2di){ __q0, __q1 };
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_set_epi64x(long long __q1,
+ long long __q0) {
+ return __extension__(__m128i)(__v2di){__q0, __q1};
}
/// Initializes both 64-bit values in a 128-bit vector of [2 x i64] with
@@ -3631,9 +3437,8 @@ _mm_set_epi64x(long long __q1, long long __q0)
/// destination vector of [2 x i64].
/// \returns An initialized 128-bit vector of [2 x i64] containing the values
/// provided in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set_epi64(__m64 __q1, __m64 __q0)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_set_epi64(__m64 __q1,
+ __m64 __q0) {
return _mm_set_epi64x((long long)__q1, (long long)__q0);
}
@@ -3659,10 +3464,9 @@ _mm_set_epi64(__m64 __q1, __m64 __q0)
/// vector.
/// \returns An initialized 128-bit vector of [4 x i32] containing the values
/// provided in the operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set_epi32(int __i3, int __i2, int __i1, int __i0)
-{
- return __extension__ (__m128i)(__v4si){ __i0, __i1, __i2, __i3};
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_set_epi32(int __i3, int __i2,
+ int __i1, int __i0) {
+ return __extension__(__m128i)(__v4si){__i0, __i1, __i2, __i3};
}
/// Initializes the 16-bit values in a 128-bit vector of [8 x i16] with
@@ -3700,9 +3504,10 @@ _mm_set_epi32(int __i3, int __i2, int __i1, int __i0)
/// \returns An initialized 128-bit vector of [8 x i16] containing the values
/// provided in the operands.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set_epi16(short __w7, short __w6, short __w5, short __w4, short __w3, short __w2, short __w1, short __w0)
-{
- return __extension__ (__m128i)(__v8hi){ __w0, __w1, __w2, __w3, __w4, __w5, __w6, __w7 };
+_mm_set_epi16(short __w7, short __w6, short __w5, short __w4, short __w3,
+ short __w2, short __w1, short __w0) {
+ return __extension__(__m128i)(__v8hi){__w0, __w1, __w2, __w3,
+ __w4, __w5, __w6, __w7};
}
/// Initializes the 8-bit values in a 128-bit vector of [16 x i8] with
@@ -3748,9 +3553,12 @@ _mm_set_epi16(short __w7, short __w6, short __w5, short __w4, short __w3, short
/// \returns An initialized 128-bit vector of [16 x i8] containing the values
/// provided in the operands.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set_epi8(char __b15, char __b14, char __b13, char __b12, char __b11, char __b10, char __b9, char __b8, char __b7, char __b6, char __b5, char __b4, char __b3, char __b2, char __b1, char __b0)
-{
- return __extension__ (__m128i)(__v16qi){ __b0, __b1, __b2, __b3, __b4, __b5, __b6, __b7, __b8, __b9, __b10, __b11, __b12, __b13, __b14, __b15 };
+_mm_set_epi8(char __b15, char __b14, char __b13, char __b12, char __b11,
+ char __b10, char __b9, char __b8, char __b7, char __b6, char __b5,
+ char __b4, char __b3, char __b2, char __b1, char __b0) {
+ return __extension__(__m128i)(__v16qi){
+ __b0, __b1, __b2, __b3, __b4, __b5, __b6, __b7,
+ __b8, __b9, __b10, __b11, __b12, __b13, __b14, __b15};
}
/// Initializes both values in a 128-bit integer vector with the
@@ -3766,9 +3574,7 @@ _mm_set_epi8(char __b15, char __b14, char __b13, char __b12, char __b11, char __
/// vector.
/// \returns An initialized 128-bit integer vector of [2 x i64] with both
/// elements containing the value provided in the operand.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set1_epi64x(long long __q)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_set1_epi64x(long long __q) {
return _mm_set_epi64x(__q, __q);
}
@@ -3785,9 +3591,7 @@ _mm_set1_epi64x(long long __q)
/// vector.
/// \returns An initialized 128-bit vector of [2 x i64] with all elements
/// containing the value provided in the operand.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set1_epi64(__m64 __q)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_set1_epi64(__m64 __q) {
return _mm_set_epi64(__q, __q);
}
@@ -3804,9 +3608,7 @@ _mm_set1_epi64(__m64 __q)
/// vector.
/// \returns An initialized 128-bit vector of [4 x i32] with all elements
/// containing the value provided in the operand.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set1_epi32(int __i)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_set1_epi32(int __i) {
return _mm_set_epi32(__i, __i, __i, __i);
}
@@ -3823,9 +3625,7 @@ _mm_set1_epi32(int __i)
/// vector.
/// \returns An initialized 128-bit vector of [8 x i16] with all elements
/// containing the value provided in the operand.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set1_epi16(short __w)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_set1_epi16(short __w) {
return _mm_set_epi16(__w, __w, __w, __w, __w, __w, __w, __w);
}
@@ -3842,10 +3642,9 @@ _mm_set1_epi16(short __w)
/// vector.
/// \returns An initialized 128-bit vector of [16 x i8] with all elements
/// containing the value provided in the operand.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_set1_epi8(char __b)
-{
- return _mm_set_epi8(__b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_set1_epi8(char __b) {
+ return _mm_set_epi8(__b, __b, __b, __b, __b, __b, __b, __b, __b, __b, __b,
+ __b, __b, __b, __b, __b);
}
/// Constructs a 128-bit integer vector, initialized in reverse order
@@ -3862,9 +3661,8 @@ _mm_set1_epi8(char __b)
/// A 64-bit integral value used to initialize the upper 64 bits of the
/// result.
/// \returns An initialized 128-bit integer vector.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_setr_epi64(__m64 __q0, __m64 __q1)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_setr_epi64(__m64 __q0,
+ __m64 __q1) {
return _mm_set_epi64(__q1, __q0);
}
@@ -3885,9 +3683,9 @@ _mm_setr_epi64(__m64 __q0, __m64 __q1)
/// \param __i3
/// A 32-bit integral value used to initialize bits [127:96] of the result.
/// \returns An initialized 128-bit integer vector.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_setr_epi32(int __i0, int __i1, int __i2, int __i3)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_setr_epi32(int __i0, int __i1,
+ int __i2,
+ int __i3) {
return _mm_set_epi32(__i3, __i2, __i1, __i0);
}
@@ -3917,8 +3715,8 @@ _mm_setr_epi32(int __i0, int __i1, int __i2, int __i3)
/// A 16-bit integral value used to initialize bits [127:112] of the result.
/// \returns An initialized 128-bit integer vector.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_setr_epi16(short __w0, short __w1, short __w2, short __w3, short __w4, short __w5, short __w6, short __w7)
-{
+_mm_setr_epi16(short __w0, short __w1, short __w2, short __w3, short __w4,
+ short __w5, short __w6, short __w7) {
return _mm_set_epi16(__w7, __w6, __w5, __w4, __w3, __w2, __w1, __w0);
}
@@ -3964,9 +3762,11 @@ _mm_setr_epi16(short __w0, short __w1, short __w2, short __w3, short __w4, short
/// An 8-bit integral value used to initialize bits [127:120] of the result.
/// \returns An initialized 128-bit integer vector.
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_setr_epi8(char __b0, char __b1, char __b2, char __b3, char __b4, char __b5, char __b6, char __b7, char __b8, char __b9, char __b10, char __b11, char __b12, char __b13, char __b14, char __b15)
-{
- return _mm_set_epi8(__b15, __b14, __b13, __b12, __b11, __b10, __b9, __b8, __b7, __b6, __b5, __b4, __b3, __b2, __b1, __b0);
+_mm_setr_epi8(char __b0, char __b1, char __b2, char __b3, char __b4, char __b5,
+ char __b6, char __b7, char __b8, char __b9, char __b10,
+ char __b11, char __b12, char __b13, char __b14, char __b15) {
+ return _mm_set_epi8(__b15, __b14, __b13, __b12, __b11, __b10, __b9, __b8,
+ __b7, __b6, __b5, __b4, __b3, __b2, __b1, __b0);
}
/// Creates a 128-bit integer vector initialized to zero.
@@ -3977,10 +3777,8 @@ _mm_setr_epi8(char __b0, char __b1, char __b2, char __b3, char __b4, char __b5,
///
/// \returns An initialized 128-bit integer vector with all elements set to
/// zero.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_setzero_si128(void)
-{
- return __extension__ (__m128i)(__v2di){ 0LL, 0LL };
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_setzero_si128(void) {
+ return __extension__(__m128i)(__v2di){0LL, 0LL};
}
/// Stores a 128-bit integer vector to a memory location aligned on a
@@ -3995,9 +3793,8 @@ _mm_setzero_si128(void)
/// values.
/// \param __b
/// A 128-bit integer vector containing the values to be moved.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_store_si128(__m128i *__p, __m128i __b)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_store_si128(__m128i *__p,
+ __m128i __b) {
*__p = __b;
}
@@ -4011,13 +3808,12 @@ _mm_store_si128(__m128i *__p, __m128i __b)
/// A pointer to a memory location that will receive the integer values.
/// \param __b
/// A 128-bit integer vector containing the values to be moved.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storeu_si128(__m128i_u *__p, __m128i __b)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storeu_si128(__m128i_u *__p,
+ __m128i __b) {
struct __storeu_si128 {
__m128i_u __v;
} __attribute__((__packed__, __may_alias__));
- ((struct __storeu_si128*)__p)->__v = __b;
+ ((struct __storeu_si128 *)__p)->__v = __b;
}
/// Stores a 64-bit integer value from the low element of a 128-bit integer
@@ -4032,13 +3828,12 @@ _mm_storeu_si128(__m128i_u *__p, __m128i __b)
/// location does not have to be aligned.
/// \param __b
/// A 128-bit integer vector containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storeu_si64(void *__p, __m128i __b)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storeu_si64(void *__p,
+ __m128i __b) {
struct __storeu_si64 {
long long __v;
} __attribute__((__packed__, __may_alias__));
- ((struct __storeu_si64*)__p)->__v = ((__v2di)__b)[0];
+ ((struct __storeu_si64 *)__p)->__v = ((__v2di)__b)[0];
}
/// Stores a 32-bit integer value from the low element of a 128-bit integer
@@ -4053,13 +3848,12 @@ _mm_storeu_si64(void *__p, __m128i __b)
/// location does not have to be aligned.
/// \param __b
/// A 128-bit integer vector containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storeu_si32(void *__p, __m128i __b)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storeu_si32(void *__p,
+ __m128i __b) {
struct __storeu_si32 {
int __v;
} __attribute__((__packed__, __may_alias__));
- ((struct __storeu_si32*)__p)->__v = ((__v4si)__b)[0];
+ ((struct __storeu_si32 *)__p)->__v = ((__v4si)__b)[0];
}
/// Stores a 16-bit integer value from the low element of a 128-bit integer
@@ -4074,13 +3868,12 @@ _mm_storeu_si32(void *__p, __m128i __b)
/// location does not have to be aligned.
/// \param __b
/// A 128-bit integer vector containing the value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storeu_si16(void *__p, __m128i __b)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storeu_si16(void *__p,
+ __m128i __b) {
struct __storeu_si16 {
short __v;
} __attribute__((__packed__, __may_alias__));
- ((struct __storeu_si16*)__p)->__v = ((__v8hi)__b)[0];
+ ((struct __storeu_si16 *)__p)->__v = ((__v8hi)__b)[0];
}
/// Moves bytes selected by the mask from the first operand to the
@@ -4104,9 +3897,9 @@ _mm_storeu_si16(void *__p, __m128i __b)
/// \param __p
/// A pointer to an unaligned 128-bit memory location where the specified
/// values are moved.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_maskmoveu_si128(__m128i __d, __m128i __n, char *__p)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_maskmoveu_si128(__m128i __d,
+ __m128i __n,
+ char *__p) {
__builtin_ia32_maskmovdqu((__v16qi)__d, (__v16qi)__n, __p);
}
@@ -4123,13 +3916,12 @@ _mm_maskmoveu_si128(__m128i __d, __m128i __n, char *__p)
/// \param __a
/// A 128-bit integer vector of [2 x i64]. The lower 64 bits contain the
/// value to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_storel_epi64(__m128i_u *__p, __m128i __a)
-{
+static __inline__ void __DEFAULT_FN_ATTRS _mm_storel_epi64(__m128i_u *__p,
+ __m128i __a) {
struct __mm_storel_epi64_struct {
long long __u;
} __attribute__((__packed__, __may_alias__));
- ((struct __mm_storel_epi64_struct*)__p)->__u = __a[0];
+ ((struct __mm_storel_epi64_struct *)__p)->__u = __a[0];
}
/// Stores a 128-bit floating point vector of [2 x double] to a 128-bit
@@ -4146,10 +3938,9 @@ _mm_storel_epi64(__m128i_u *__p, __m128i __a)
/// A pointer to the 128-bit aligned memory location used to store the value.
/// \param __a
/// A vector of [2 x double] containing the 64-bit values to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_stream_pd(double *__p, __m128d __a)
-{
- __builtin_nontemporal_store((__v2df)__a, (__v2df*)__p);
+static __inline__ void __DEFAULT_FN_ATTRS _mm_stream_pd(double *__p,
+ __m128d __a) {
+ __builtin_nontemporal_store((__v2df)__a, (__v2df *)__p);
}
/// Stores a 128-bit integer vector to a 128-bit aligned memory location.
@@ -4165,10 +3956,9 @@ _mm_stream_pd(double *__p, __m128d __a)
/// A pointer to the 128-bit aligned memory location used to store the value.
/// \param __a
/// A 128-bit integer vector containing the values to be stored.
-static __inline__ void __DEFAULT_FN_ATTRS
-_mm_stream_si128(__m128i *__p, __m128i __a)
-{
- __builtin_nontemporal_store((__v2di)__a, (__v2di*)__p);
+static __inline__ void __DEFAULT_FN_ATTRS _mm_stream_si128(__m128i *__p,
+ __m128i __a) {
+ __builtin_nontemporal_store((__v2di)__a, (__v2di *)__p);
}
/// Stores a 32-bit integer value in the specified memory location.
@@ -4184,9 +3974,9 @@ _mm_stream_si128(__m128i *__p, __m128i __a)
/// A pointer to the 32-bit memory location used to store the value.
/// \param __a
/// A 32-bit integer containing the value to be stored.
-static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
-_mm_stream_si32(int *__p, int __a)
-{
+static __inline__ void
+ __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
+ _mm_stream_si32(int *__p, int __a) {
__builtin_ia32_movnti(__p, __a);
}
@@ -4204,9 +3994,9 @@ _mm_stream_si32(int *__p, int __a)
/// A pointer to the 64-bit memory location used to store the value.
/// \param __a
/// A 64-bit integer containing the value to be stored.
-static __inline__ void __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
-_mm_stream_si64(long long *__p, long long __a)
-{
+static __inline__ void
+ __attribute__((__always_inline__, __nodebug__, __target__("sse2")))
+ _mm_stream_si64(long long *__p, long long __a) {
__builtin_ia32_movnti64(__p, __a);
}
#endif
@@ -4225,7 +4015,7 @@ extern "C" {
/// \param __p
/// A pointer to the memory location used to identify the cache line to be
/// flushed.
-void _mm_clflush(void const * __p);
+void _mm_clflush(void const *__p);
/// Forces strong memory ordering (serialization) between load
/// instructions preceding this instruction and load instructions following
@@ -4275,9 +4065,8 @@ void _mm_mfence(void);
/// than 0x80 are saturated to 0x80. The converted [8 x i8] values are
/// written to the higher 64 bits of the result.
/// \returns A 128-bit vector of [16 x i8] containing the converted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_packs_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_packs_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_ia32_packsswb128((__v8hi)__a, (__v8hi)__b);
}
@@ -4303,9 +4092,8 @@ _mm_packs_epi16(__m128i __a, __m128i __b)
/// less than 0x8000 are saturated to 0x8000. The converted [4 x i16] values
/// are written to the higher 64 bits of the result.
/// \returns A 128-bit vector of [8 x i16] containing the converted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_packs_epi32(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_packs_epi32(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_ia32_packssdw128((__v4si)__a, (__v4si)__b);
}
@@ -4331,9 +4119,8 @@ _mm_packs_epi32(__m128i __a, __m128i __b)
/// than 0x00 are saturated to 0x00. The converted [8 x i8] values are
/// written to the higher 64 bits of the result.
/// \returns A 128-bit vector of [16 x i8] containing the converted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_packus_epi16(__m128i __a, __m128i __b)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_packus_epi16(__m128i __a,
+ __m128i __b) {
return (__m128i)__builtin_ia32_packuswb128((__v8hi)__a, (__v8hi)__b);
}
@@ -4359,8 +4146,8 @@ _mm_packus_epi16(__m128i __a, __m128i __b)
/// 111: assign values from bits [127:112] of \a __a.
/// \returns An integer, whose lower 16 bits are selected from the 128-bit
/// integer vector parameter and the remaining bits are assigned zeros.
-#define _mm_extract_epi16(a, imm) \
- ((int)(unsigned short)__builtin_ia32_vec_ext_v8hi((__v8hi)(__m128i)(a), \
+#define _mm_extract_epi16(a, imm) \
+ ((int)(unsigned short)__builtin_ia32_vec_ext_v8hi((__v8hi)(__m128i)(a), \
(int)(imm)))
/// Constructs a 128-bit integer vector by first making a copy of the
@@ -4383,8 +4170,8 @@ _mm_packus_epi16(__m128i __a, __m128i __b)
/// An immediate value specifying the bit offset in the result at which the
/// lower 16 bits of \a __b are written.
/// \returns A 128-bit integer vector containing the constructed values.
-#define _mm_insert_epi16(a, b, imm) \
- ((__m128i)__builtin_ia32_vec_set_v8hi((__v8hi)(__m128i)(a), (int)(b), \
+#define _mm_insert_epi16(a, b, imm) \
+ ((__m128i)__builtin_ia32_vec_set_v8hi((__v8hi)(__m128i)(a), (int)(b), \
(int)(imm)))
/// Copies the values of the most significant bits from each 8-bit
@@ -4399,9 +4186,7 @@ _mm_packus_epi16(__m128i __a, __m128i __b)
/// A 128-bit integer vector containing the values with bits to be extracted.
/// \returns The most significant bits from each 8-bit element in \a __a,
/// written to bits [15:0]. The other bits are assigned zeros.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_movemask_epi8(__m128i __a)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_movemask_epi8(__m128i __a) {
return __builtin_ia32_pmovmskb128((__v16qi)__a);
}
@@ -4433,7 +4218,7 @@ _mm_movemask_epi8(__m128i __a)
/// 10: assign values from bits [95:64] of \a a. \n
/// 11: assign values from bits [127:96] of \a a.
/// \returns A 128-bit integer vector containing the shuffled values.
-#define _mm_shuffle_epi32(a, imm) \
+#define _mm_shuffle_epi32(a, imm) \
((__m128i)__builtin_ia32_pshufd((__v4si)(__m128i)(a), (int)(imm)))
/// Constructs a 128-bit integer vector by shuffling four lower 16-bit
@@ -4463,7 +4248,7 @@ _mm_movemask_epi8(__m128i __a)
/// 10: assign values from bits [47:32] of \a a. \n
/// 11: assign values from bits [63:48] of \a a. \n
/// \returns A 128-bit integer vector containing the shuffled values.
-#define _mm_shufflelo_epi16(a, imm) \
+#define _mm_shufflelo_epi16(a, imm) \
((__m128i)__builtin_ia32_pshuflw((__v8hi)(__m128i)(a), (int)(imm)))
/// Constructs a 128-bit integer vector by shuffling four upper 16-bit
@@ -4493,7 +4278,7 @@ _mm_movemask_epi8(__m128i __a)
/// 10: assign values from bits [111:96] of \a a. \n
/// 11: assign values from bits [127:112] of \a a. \n
/// \returns A 128-bit integer vector containing the shuffled values.
-#define _mm_shufflehi_epi16(a, imm) \
+#define _mm_shufflehi_epi16(a, imm) \
((__m128i)__builtin_ia32_pshufhw((__v8hi)(__m128i)(a), (int)(imm)))
/// Unpacks the high-order (index 8-15) values from two 128-bit vectors
@@ -4525,10 +4310,11 @@ _mm_movemask_epi8(__m128i __a)
/// Bits [119:112] are written to bits [111:104] of the result. \n
/// Bits [127:120] are written to bits [127:120] of the result.
/// \returns A 128-bit vector of [16 x i8] containing the interleaved values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_unpackhi_epi8(__m128i __a, __m128i __b)
-{
- return (__m128i)__builtin_shufflevector((__v16qi)__a, (__v16qi)__b, 8, 16+8, 9, 16+9, 10, 16+10, 11, 16+11, 12, 16+12, 13, 16+13, 14, 16+14, 15, 16+15);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi8(__m128i __a,
+ __m128i __b) {
+ return (__m128i)__builtin_shufflevector(
+ (__v16qi)__a, (__v16qi)__b, 8, 16 + 8, 9, 16 + 9, 10, 16 + 10, 11,
+ 16 + 11, 12, 16 + 12, 13, 16 + 13, 14, 16 + 14, 15, 16 + 15);
}
/// Unpacks the high-order (index 4-7) values from two 128-bit vectors of
@@ -4552,10 +4338,10 @@ _mm_unpackhi_epi8(__m128i __a, __m128i __b)
/// Bits [111:96] are written to bits [95:80] of the result. \n
/// Bits [127:112] are written to bits [127:112] of the result.
/// \returns A 128-bit vector of [8 x i16] containing the interleaved values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_unpackhi_epi16(__m128i __a, __m128i __b)
-{
- return (__m128i)__builtin_shufflevector((__v8hi)__a, (__v8hi)__b, 4, 8+4, 5, 8+5, 6, 8+6, 7, 8+7);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi16(__m128i __a,
+ __m128i __b) {
+ return (__m128i)__builtin_shufflevector((__v8hi)__a, (__v8hi)__b, 4, 8 + 4, 5,
+ 8 + 5, 6, 8 + 6, 7, 8 + 7);
}
/// Unpacks the high-order (index 2,3) values from two 128-bit vectors of
@@ -4575,10 +4361,10 @@ _mm_unpackhi_epi16(__m128i __a, __m128i __b)
/// Bits [95:64] are written to bits [64:32] of the destination. \n
/// Bits [127:96] are written to bits [127:96] of the destination.
/// \returns A 128-bit vector of [4 x i32] containing the interleaved values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_unpackhi_epi32(__m128i __a, __m128i __b)
-{
- return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 2, 4+2, 3, 4+3);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi32(__m128i __a,
+ __m128i __b) {
+ return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 2, 4 + 2, 3,
+ 4 + 3);
}
/// Unpacks the high-order 64-bit elements from two 128-bit vectors of
@@ -4596,10 +4382,9 @@ _mm_unpackhi_epi32(__m128i __a, __m128i __b)
/// A 128-bit vector of [2 x i64]. \n
/// Bits [127:64] are written to bits [127:64] of the destination.
/// \returns A 128-bit vector of [2 x i64] containing the interleaved values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_unpackhi_epi64(__m128i __a, __m128i __b)
-{
- return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 1, 2+1);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi64(__m128i __a,
+ __m128i __b) {
+ return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 1, 2 + 1);
}
/// Unpacks the low-order (index 0-7) values from two 128-bit vectors of
@@ -4631,10 +4416,11 @@ _mm_unpackhi_epi64(__m128i __a, __m128i __b)
/// Bits [55:48] are written to bits [111:104] of the result. \n
/// Bits [63:56] are written to bits [127:120] of the result.
/// \returns A 128-bit vector of [16 x i8] containing the interleaved values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_unpacklo_epi8(__m128i __a, __m128i __b)
-{
- return (__m128i)__builtin_shufflevector((__v16qi)__a, (__v16qi)__b, 0, 16+0, 1, 16+1, 2, 16+2, 3, 16+3, 4, 16+4, 5, 16+5, 6, 16+6, 7, 16+7);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi8(__m128i __a,
+ __m128i __b) {
+ return (__m128i)__builtin_shufflevector(
+ (__v16qi)__a, (__v16qi)__b, 0, 16 + 0, 1, 16 + 1, 2, 16 + 2, 3, 16 + 3, 4,
+ 16 + 4, 5, 16 + 5, 6, 16 + 6, 7, 16 + 7);
}
/// Unpacks the low-order (index 0-3) values from each of the two 128-bit
@@ -4659,10 +4445,10 @@ _mm_unpacklo_epi8(__m128i __a, __m128i __b)
/// Bits [47:32] are written to bits [95:80] of the result. \n
/// Bits [63:48] are written to bits [127:112] of the result.
/// \returns A 128-bit vector of [8 x i16] containing the interleaved values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_unpacklo_epi16(__m128i __a, __m128i __b)
-{
- return (__m128i)__builtin_shufflevector((__v8hi)__a, (__v8hi)__b, 0, 8+0, 1, 8+1, 2, 8+2, 3, 8+3);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi16(__m128i __a,
+ __m128i __b) {
+ return (__m128i)__builtin_shufflevector((__v8hi)__a, (__v8hi)__b, 0, 8 + 0, 1,
+ 8 + 1, 2, 8 + 2, 3, 8 + 3);
}
/// Unpacks the low-order (index 0,1) values from two 128-bit vectors of
@@ -4682,10 +4468,10 @@ _mm_unpacklo_epi16(__m128i __a, __m128i __b)
/// Bits [31:0] are written to bits [64:32] of the destination. \n
/// Bits [63:32] are written to bits [127:96] of the destination.
/// \returns A 128-bit vector of [4 x i32] containing the interleaved values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_unpacklo_epi32(__m128i __a, __m128i __b)
-{
- return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 0, 4+0, 1, 4+1);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi32(__m128i __a,
+ __m128i __b) {
+ return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 0, 4 + 0, 1,
+ 4 + 1);
}
/// Unpacks the low-order 64-bit elements from two 128-bit vectors of
@@ -4703,10 +4489,9 @@ _mm_unpacklo_epi32(__m128i __a, __m128i __b)
/// A 128-bit vector of [2 x i64]. \n
/// Bits [63:0] are written to bits [127:64] of the destination. \n
/// \returns A 128-bit vector of [2 x i64] containing the interleaved values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_unpacklo_epi64(__m128i __a, __m128i __b)
-{
- return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 0, 2+0);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi64(__m128i __a,
+ __m128i __b) {
+ return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 0, 2 + 0);
}
/// Returns the lower 64 bits of a 128-bit integer vector as a 64-bit
@@ -4720,9 +4505,7 @@ _mm_unpacklo_epi64(__m128i __a, __m128i __b)
/// A 128-bit integer vector operand. The lower 64 bits are moved to the
/// destination.
/// \returns A 64-bit integer containing the lower 64 bits of the parameter.
-static __inline__ __m64 __DEFAULT_FN_ATTRS
-_mm_movepi64_pi64(__m128i __a)
-{
+static __inline__ __m64 __DEFAULT_FN_ATTRS _mm_movepi64_pi64(__m128i __a) {
return (__m64)__a[0];
}
@@ -4737,10 +4520,8 @@ _mm_movepi64_pi64(__m128i __a)
/// A 64-bit value.
/// \returns A 128-bit integer vector. The lower 64 bits contain the value from
/// the operand. The upper 64 bits are assigned zeros.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_movpi64_epi64(__m64 __a)
-{
- return __extension__ (__m128i)(__v2di){ (long long)__a, 0 };
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_movpi64_epi64(__m64 __a) {
+ return __extension__(__m128i)(__v2di){(long long)__a, 0};
}
/// Moves the lower 64 bits of a 128-bit integer vector to a 128-bit
@@ -4755,9 +4536,7 @@ _mm_movpi64_epi64(__m64 __a)
/// destination.
/// \returns A 128-bit integer vector. The lower 64 bits contain the value from
/// the operand. The upper 64 bits are assigned zeros.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_move_epi64(__m128i __a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_move_epi64(__m128i __a) {
return __builtin_shufflevector((__v2di)__a, _mm_setzero_si128(), 0, 2);
}
@@ -4776,10 +4555,9 @@ _mm_move_epi64(__m128i __a)
/// A 128-bit vector of [2 x double]. \n
/// Bits [127:64] are written to bits [127:64] of the destination.
/// \returns A 128-bit vector of [2 x double] containing the interleaved values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_unpackhi_pd(__m128d __a, __m128d __b)
-{
- return __builtin_shufflevector((__v2df)__a, (__v2df)__b, 1, 2+1);
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_unpackhi_pd(__m128d __a,
+ __m128d __b) {
+ return __builtin_shufflevector((__v2df)__a, (__v2df)__b, 1, 2 + 1);
}
/// Unpacks the low-order 64-bit elements from two 128-bit vectors
@@ -4797,10 +4575,9 @@ _mm_unpackhi_pd(__m128d __a, __m128d __b)
/// A 128-bit vector of [2 x double]. \n
/// Bits [63:0] are written to bits [127:64] of the destination.
/// \returns A 128-bit vector of [2 x double] containing the interleaved values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_unpacklo_pd(__m128d __a, __m128d __b)
-{
- return __builtin_shufflevector((__v2df)__a, (__v2df)__b, 0, 2+0);
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_unpacklo_pd(__m128d __a,
+ __m128d __b) {
+ return __builtin_shufflevector((__v2df)__a, (__v2df)__b, 0, 2 + 0);
}
/// Extracts the sign bits of the double-precision values in the 128-bit
@@ -4816,13 +4593,10 @@ _mm_unpacklo_pd(__m128d __a, __m128d __b)
/// be extracted.
/// \returns The sign bits from each of the double-precision elements in \a __a,
/// written to bits [1:0]. The remaining bits are assigned values of zero.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_movemask_pd(__m128d __a)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_movemask_pd(__m128d __a) {
return __builtin_ia32_movmskpd((__v2df)__a);
}
-
/// Constructs a 128-bit floating-point vector of [2 x double] from two
/// 128-bit vector parameters of [2 x double], using the immediate-value
/// parameter as a specifier.
@@ -4847,8 +4621,8 @@ _mm_movemask_pd(__m128d __a)
/// Bit[1] = 0: lower element of \a b copied to upper element of result. \n
/// Bit[1] = 1: upper element of \a b copied to upper element of result. \n
/// \returns A 128-bit vector of [2 x double] containing the shuffled values.
-#define _mm_shuffle_pd(a, b, i) \
- ((__m128d)__builtin_ia32_shufpd((__v2df)(__m128d)(a), (__v2df)(__m128d)(b), \
+#define _mm_shuffle_pd(a, b, i) \
+ ((__m128d)__builtin_ia32_shufpd((__v2df)(__m128d)(a), (__v2df)(__m128d)(b), \
(int)(i)))
/// Casts a 128-bit floating-point vector of [2 x double] into a 128-bit
@@ -4862,9 +4636,7 @@ _mm_movemask_pd(__m128d __a)
/// A 128-bit floating-point vector of [2 x double].
/// \returns A 128-bit floating-point vector of [4 x float] containing the same
/// bitwise pattern as the parameter.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_castpd_ps(__m128d __a)
-{
+static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_castpd_ps(__m128d __a) {
return (__m128)__a;
}
@@ -4879,9 +4651,7 @@ _mm_castpd_ps(__m128d __a)
/// A 128-bit floating-point vector of [2 x double].
/// \returns A 128-bit integer vector containing the same bitwise pattern as the
/// parameter.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_castpd_si128(__m128d __a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_castpd_si128(__m128d __a) {
return (__m128i)__a;
}
@@ -4896,9 +4666,7 @@ _mm_castpd_si128(__m128d __a)
/// A 128-bit floating-point vector of [4 x float].
/// \returns A 128-bit floating-point vector of [2 x double] containing the same
/// bitwise pattern as the parameter.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_castps_pd(__m128 __a)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_castps_pd(__m128 __a) {
return (__m128d)__a;
}
@@ -4913,9 +4681,7 @@ _mm_castps_pd(__m128 __a)
/// A 128-bit floating-point vector of [4 x float].
/// \returns A 128-bit integer vector containing the same bitwise pattern as the
/// parameter.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_castps_si128(__m128 __a)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_castps_si128(__m128 __a) {
return (__m128i)__a;
}
@@ -4930,9 +4696,7 @@ _mm_castps_si128(__m128 __a)
/// A 128-bit integer vector.
/// \returns A 128-bit floating-point vector of [4 x float] containing the same
/// bitwise pattern as the parameter.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_castsi128_ps(__m128i __a)
-{
+static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_castsi128_ps(__m128i __a) {
return (__m128)__a;
}
@@ -4947,9 +4711,7 @@ _mm_castsi128_ps(__m128i __a)
/// A 128-bit integer vector.
/// \returns A 128-bit floating-point vector of [2 x double] containing the same
/// bitwise pattern as the parameter.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_castsi128_pd(__m128i __a)
-{
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_castsi128_pd(__m128i __a) {
return (__m128d)__a;
}
@@ -4974,12 +4736,13 @@ void _mm_pause(void);
#define _MM_SHUFFLE2(x, y) (((x) << 1) | (y))
-#define _MM_DENORMALS_ZERO_ON (0x0040U)
-#define _MM_DENORMALS_ZERO_OFF (0x0000U)
+#define _MM_DENORMALS_ZERO_ON (0x0040U)
+#define _MM_DENORMALS_ZERO_OFF (0x0000U)
#define _MM_DENORMALS_ZERO_MASK (0x0040U)
#define _MM_GET_DENORMALS_ZERO_MODE() (_mm_getcsr() & _MM_DENORMALS_ZERO_MASK)
-#define _MM_SET_DENORMALS_ZERO_MODE(x) (_mm_setcsr((_mm_getcsr() & ~_MM_DENORMALS_ZERO_MASK) | (x)))
+#define _MM_SET_DENORMALS_ZERO_MODE(x) \
+ (_mm_setcsr((_mm_getcsr() & ~_MM_DENORMALS_ZERO_MASK) | (x)))
#endif /* __EMMINTRIN_H */
diff --git a/clang/lib/Headers/smmintrin.h b/clang/lib/Headers/smmintrin.h
index 0df59c5fcc592..4f0b04fcb3831 100644
--- a/clang/lib/Headers/smmintrin.h
+++ b/clang/lib/Headers/smmintrin.h
@@ -17,23 +17,25 @@
#include <tmmintrin.h>
/* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse4.1"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("sse4.1"), \
+ __min_vector_width__(128)))
/* SSE4 Rounding macros. */
-#define _MM_FROUND_TO_NEAREST_INT 0x00
-#define _MM_FROUND_TO_NEG_INF 0x01
-#define _MM_FROUND_TO_POS_INF 0x02
-#define _MM_FROUND_TO_ZERO 0x03
-#define _MM_FROUND_CUR_DIRECTION 0x04
-
-#define _MM_FROUND_RAISE_EXC 0x00
-#define _MM_FROUND_NO_EXC 0x08
-
-#define _MM_FROUND_NINT (_MM_FROUND_RAISE_EXC | _MM_FROUND_TO_NEAREST_INT)
-#define _MM_FROUND_FLOOR (_MM_FROUND_RAISE_EXC | _MM_FROUND_TO_NEG_INF)
-#define _MM_FROUND_CEIL (_MM_FROUND_RAISE_EXC | _MM_FROUND_TO_POS_INF)
-#define _MM_FROUND_TRUNC (_MM_FROUND_RAISE_EXC | _MM_FROUND_TO_ZERO)
-#define _MM_FROUND_RINT (_MM_FROUND_RAISE_EXC | _MM_FROUND_CUR_DIRECTION)
+#define _MM_FROUND_TO_NEAREST_INT 0x00
+#define _MM_FROUND_TO_NEG_INF 0x01
+#define _MM_FROUND_TO_POS_INF 0x02
+#define _MM_FROUND_TO_ZERO 0x03
+#define _MM_FROUND_CUR_DIRECTION 0x04
+
+#define _MM_FROUND_RAISE_EXC 0x00
+#define _MM_FROUND_NO_EXC 0x08
+
+#define _MM_FROUND_NINT (_MM_FROUND_RAISE_EXC | _MM_FROUND_TO_NEAREST_INT)
+#define _MM_FROUND_FLOOR (_MM_FROUND_RAISE_EXC | _MM_FROUND_TO_NEG_INF)
+#define _MM_FROUND_CEIL (_MM_FROUND_RAISE_EXC | _MM_FROUND_TO_POS_INF)
+#define _MM_FROUND_TRUNC (_MM_FROUND_RAISE_EXC | _MM_FROUND_TO_ZERO)
+#define _MM_FROUND_RINT (_MM_FROUND_RAISE_EXC | _MM_FROUND_CUR_DIRECTION)
#define _MM_FROUND_NEARBYINT (_MM_FROUND_NO_EXC | _MM_FROUND_CUR_DIRECTION)
/// Rounds up each element of the 128-bit vector of [4 x float] to an
@@ -51,7 +53,7 @@
/// \param X
/// A 128-bit vector of [4 x float] values to be rounded up.
/// \returns A 128-bit vector of [4 x float] containing the rounded values.
-#define _mm_ceil_ps(X) _mm_round_ps((X), _MM_FROUND_CEIL)
+#define _mm_ceil_ps(X) _mm_round_ps((X), _MM_FROUND_CEIL)
/// Rounds up each element of the 128-bit vector of [2 x double] to an
/// integer and returns the rounded values in a 128-bit vector of
@@ -68,7 +70,7 @@
/// \param X
/// A 128-bit vector of [2 x double] values to be rounded up.
/// \returns A 128-bit vector of [2 x double] containing the rounded values.
-#define _mm_ceil_pd(X) _mm_round_pd((X), _MM_FROUND_CEIL)
+#define _mm_ceil_pd(X) _mm_round_pd((X), _MM_FROUND_CEIL)
/// Copies three upper elements of the first 128-bit vector operand to
/// the corresponding three upper elements of the 128-bit result vector of
@@ -93,7 +95,7 @@
/// of the result.
/// \returns A 128-bit vector of [4 x float] containing the copied and rounded
/// values.
-#define _mm_ceil_ss(X, Y) _mm_round_ss((X), (Y), _MM_FROUND_CEIL)
+#define _mm_ceil_ss(X, Y) _mm_round_ss((X), (Y), _MM_FROUND_CEIL)
/// Copies the upper element of the first 128-bit vector operand to the
/// corresponding upper element of the 128-bit result vector of [2 x double].
@@ -118,7 +120,7 @@
/// of the result.
/// \returns A 128-bit vector of [2 x double] containing the copied and rounded
/// values.
-#define _mm_ceil_sd(X, Y) _mm_round_sd((X), (Y), _MM_FROUND_CEIL)
+#define _mm_ceil_sd(X, Y) _mm_round_sd((X), (Y), _MM_FROUND_CEIL)
/// Rounds down each element of the 128-bit vector of [4 x float] to an
/// an integer and returns the rounded values in a 128-bit vector of
@@ -135,7 +137,7 @@
/// \param X
/// A 128-bit vector of [4 x float] values to be rounded down.
/// \returns A 128-bit vector of [4 x float] containing the rounded values.
-#define _mm_floor_ps(X) _mm_round_ps((X), _MM_FROUND_FLOOR)
+#define _mm_floor_ps(X) _mm_round_ps((X), _MM_FROUND_FLOOR)
/// Rounds down each element of the 128-bit vector of [2 x double] to an
/// integer and returns the rounded values in a 128-bit vector of
@@ -152,7 +154,7 @@
/// \param X
/// A 128-bit vector of [2 x double].
/// \returns A 128-bit vector of [2 x double] containing the rounded values.
-#define _mm_floor_pd(X) _mm_round_pd((X), _MM_FROUND_FLOOR)
+#define _mm_floor_pd(X) _mm_round_pd((X), _MM_FROUND_FLOOR)
/// Copies three upper elements of the first 128-bit vector operand to
/// the corresponding three upper elements of the 128-bit result vector of
@@ -177,7 +179,7 @@
/// of the result.
/// \returns A 128-bit vector of [4 x float] containing the copied and rounded
/// values.
-#define _mm_floor_ss(X, Y) _mm_round_ss((X), (Y), _MM_FROUND_FLOOR)
+#define _mm_floor_ss(X, Y) _mm_round_ss((X), (Y), _MM_FROUND_FLOOR)
/// Copies the upper element of the first 128-bit vector operand to the
/// corresponding upper element of the 128-bit result vector of [2 x double].
@@ -202,7 +204,7 @@
/// of the result.
/// \returns A 128-bit vector of [2 x double] containing the copied and rounded
/// values.
-#define _mm_floor_sd(X, Y) _mm_round_sd((X), (Y), _MM_FROUND_FLOOR)
+#define _mm_floor_sd(X, Y) _mm_round_sd((X), (Y), _MM_FROUND_FLOOR)
/// Rounds each element of the 128-bit vector of [4 x float] to an
/// integer value according to the rounding control specified by the second
@@ -234,7 +236,7 @@
/// 10: Upward (toward positive infinity) \n
/// 11: Truncated
/// \returns A 128-bit vector of [4 x float] containing the rounded values.
-#define _mm_round_ps(X, M) \
+#define _mm_round_ps(X, M) \
((__m128)__builtin_ia32_roundps((__v4sf)(__m128)(X), (M)))
/// Copies three upper elements of the first 128-bit vector operand to
@@ -275,9 +277,9 @@
/// 11: Truncated
/// \returns A 128-bit vector of [4 x float] containing the copied and rounded
/// values.
-#define _mm_round_ss(X, Y, M) \
- ((__m128)__builtin_ia32_roundss((__v4sf)(__m128)(X), \
- (__v4sf)(__m128)(Y), (M)))
+#define _mm_round_ss(X, Y, M) \
+ ((__m128)__builtin_ia32_roundss((__v4sf)(__m128)(X), (__v4sf)(__m128)(Y), \
+ (M)))
/// Rounds each element of the 128-bit vector of [2 x double] to an
/// integer value according to the rounding control specified by the second
@@ -309,7 +311,7 @@
/// 10: Upward (toward positive infinity) \n
/// 11: Truncated
/// \returns A 128-bit vector of [2 x double] containing the rounded values.
-#define _mm_round_pd(X, M) \
+#define _mm_round_pd(X, M) \
((__m128d)__builtin_ia32_roundpd((__v2df)(__m128d)(X), (M)))
/// Copies the upper element of the first 128-bit vector operand to the
@@ -350,9 +352,9 @@
/// 11: Truncated
/// \returns A 128-bit vector of [2 x double] containing the copied and rounded
/// values.
-#define _mm_round_sd(X, Y, M) \
- ((__m128d)__builtin_ia32_roundsd((__v2df)(__m128d)(X), \
- (__v2df)(__m128d)(Y), (M)))
+#define _mm_round_sd(X, Y, M) \
+ ((__m128d)__builtin_ia32_roundsd((__v2df)(__m128d)(X), (__v2df)(__m128d)(Y), \
+ (M)))
/* SSE4 Packed Blending Intrinsics. */
/// Returns a 128-bit vector of [2 x double] where the values are
@@ -379,9 +381,9 @@
/// When a mask bit is 1, the corresponding 64-bit element in operand \a V2
/// is copied to the same position in the result.
/// \returns A 128-bit vector of [2 x double] containing the copied values.
-#define _mm_blend_pd(V1, V2, M) \
- ((__m128d) __builtin_ia32_blendpd ((__v2df)(__m128d)(V1), \
- (__v2df)(__m128d)(V2), (int)(M)))
+#define _mm_blend_pd(V1, V2, M) \
+ ((__m128d)__builtin_ia32_blendpd((__v2df)(__m128d)(V1), \
+ (__v2df)(__m128d)(V2), (int)(M)))
/// Returns a 128-bit vector of [4 x float] where the values are selected
/// from either the first or second operand as specified by the third
@@ -407,9 +409,9 @@
/// When a mask bit is 1, the corresponding 32-bit element in operand \a V2
/// is copied to the same position in the result.
/// \returns A 128-bit vector of [4 x float] containing the copied values.
-#define _mm_blend_ps(V1, V2, M) \
- ((__m128) __builtin_ia32_blendps ((__v4sf)(__m128)(V1), \
- (__v4sf)(__m128)(V2), (int)(M)))
+#define _mm_blend_ps(V1, V2, M) \
+ ((__m128)__builtin_ia32_blendps((__v4sf)(__m128)(V1), (__v4sf)(__m128)(V2), \
+ (int)(M)))
/// Returns a 128-bit vector of [2 x double] where the values are
/// selected from either the first or second operand as specified by the
@@ -431,11 +433,11 @@
/// position in the result. When a mask bit is 1, the corresponding 64-bit
/// element in operand \a __V2 is copied to the same position in the result.
/// \returns A 128-bit vector of [2 x double] containing the copied values.
-static __inline__ __m128d __DEFAULT_FN_ATTRS
-_mm_blendv_pd (__m128d __V1, __m128d __V2, __m128d __M)
-{
- return (__m128d) __builtin_ia32_blendvpd ((__v2df)__V1, (__v2df)__V2,
- (__v2df)__M);
+static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_blendv_pd(__m128d __V1,
+ __m128d __V2,
+ __m128d __M) {
+ return (__m128d)__builtin_ia32_blendvpd((__v2df)__V1, (__v2df)__V2,
+ (__v2df)__M);
}
/// Returns a 128-bit vector of [4 x float] where the values are
@@ -458,11 +460,11 @@ _mm_blendv_pd (__m128d __V1, __m128d __V2, __m128d __M)
/// position in the result. When a mask bit is 1, the corresponding 32-bit
/// element in operand \a __V2 is copied to the same position in the result.
/// \returns A 128-bit vector of [4 x float] containing the copied values.
-static __inline__ __m128 __DEFAULT_FN_ATTRS
-_mm_blendv_ps (__m128 __V1, __m128 __V2, __m128 __M)
-{
- return (__m128) __builtin_ia32_blendvps ((__v4sf)__V1, (__v4sf)__V2,
- (__v4sf)__M);
+static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_blendv_ps(__m128 __V1,
+ __m128 __V2,
+ __m128 __M) {
+ return (__m128)__builtin_ia32_blendvps((__v4sf)__V1, (__v4sf)__V2,
+ (__v4sf)__M);
}
/// Returns a 128-bit vector of [16 x i8] where the values are selected
@@ -485,11 +487,11 @@ _mm_blendv_ps (__m128 __V1, __m128 __V2, __m128 __M)
/// position in the result. When a mask bit is 1, the corresponding 8-bit
/// element in operand \a __V2 is copied to the same position in the result.
/// \returns A 128-bit vector of [16 x i8] containing the copied values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_blendv_epi8 (__m128i __V1, __m128i __V2, __m128i __M)
-{
- return (__m128i) __builtin_ia32_pblendvb128 ((__v16qi)__V1, (__v16qi)__V2,
- (__v16qi)__M);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_blendv_epi8(__m128i __V1,
+ __m128i __V2,
+ __m128i __M) {
+ return (__m128i)__builtin_ia32_pblendvb128((__v16qi)__V1, (__v16qi)__V2,
+ (__v16qi)__M);
}
/// Returns a 128-bit vector of [8 x i16] where the values are selected
@@ -516,9 +518,9 @@ _mm_blendv_epi8 (__m128i __V1, __m128i __V2, __m128i __M)
/// When a mask bit is 1, the corresponding 16-bit element in operand \a V2
/// is copied to the same position in the result.
/// \returns A 128-bit vector of [8 x i16] containing the copied values.
-#define _mm_blend_epi16(V1, V2, M) \
- ((__m128i) __builtin_ia32_pblendw128 ((__v8hi)(__m128i)(V1), \
- (__v8hi)(__m128i)(V2), (int)(M)))
+#define _mm_blend_epi16(V1, V2, M) \
+ ((__m128i)__builtin_ia32_pblendw128((__v8hi)(__m128i)(V1), \
+ (__v8hi)(__m128i)(V2), (int)(M)))
/* SSE4 Dword Multiply Instructions. */
/// Multiples corresponding elements of two 128-bit vectors of [4 x i32]
@@ -534,10 +536,9 @@ _mm_blendv_epi8 (__m128i __V1, __m128i __V2, __m128i __M)
/// \param __V2
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the products of both operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mullo_epi32 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) ((__v4su)__V1 * (__v4su)__V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_mullo_epi32(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)((__v4su)__V1 * (__v4su)__V2);
}
/// Multiplies corresponding even-indexed elements of two 128-bit
@@ -554,10 +555,9 @@ _mm_mullo_epi32 (__m128i __V1, __m128i __V2)
/// A 128-bit vector of [4 x i32].
/// \returns A 128-bit vector of [2 x i64] containing the products of both
/// operands.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mul_epi32 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_ia32_pmuldq128 ((__v4si)__V1, (__v4si)__V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_mul_epi32(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_ia32_pmuldq128((__v4si)__V1, (__v4si)__V2);
}
/* SSE4 Floating Point Dot Product Instructions. */
@@ -593,9 +593,8 @@ _mm_mul_epi32 (__m128i __V1, __m128i __V2)
/// each [4 x float] subvector. If a bit is set, the dot product is returned
/// in the corresponding element; otherwise that element is set to zero.
/// \returns A 128-bit vector of [4 x float] containing the dot product.
-#define _mm_dp_ps(X, Y, M) \
- ((__m128) __builtin_ia32_dpps((__v4sf)(__m128)(X), \
- (__v4sf)(__m128)(Y), (M)))
+#define _mm_dp_ps(X, Y, M) \
+ ((__m128)__builtin_ia32_dpps((__v4sf)(__m128)(X), (__v4sf)(__m128)(Y), (M)))
/// Computes the dot product of the two 128-bit vectors of [2 x double]
/// and returns it in the elements of the 128-bit result vector of
@@ -628,9 +627,9 @@ _mm_mul_epi32 (__m128i __V1, __m128i __V2)
/// to the lowest element and bit [1] corresponding to the highest element of
/// each [2 x double] vector. If a bit is set, the dot product is returned in
/// the corresponding element; otherwise that element is set to zero.
-#define _mm_dp_pd(X, Y, M) \
- ((__m128d) __builtin_ia32_dppd((__v2df)(__m128d)(X), \
- (__v2df)(__m128d)(Y), (M)))
+#define _mm_dp_pd(X, Y, M) \
+ ((__m128d)__builtin_ia32_dppd((__v2df)(__m128d)(X), (__v2df)(__m128d)(Y), \
+ (M)))
/* SSE4 Streaming Load Hint Instruction. */
/// Loads integer values from a 128-bit aligned memory location to a
@@ -645,10 +644,9 @@ _mm_mul_epi32 (__m128i __V1, __m128i __V2)
/// values.
/// \returns A 128-bit integer vector containing the data stored at the
/// specified memory location.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_stream_load_si128 (__m128i const *__V)
-{
- return (__m128i) __builtin_nontemporal_load ((const __v2di *) __V);
+static __inline__ __m128i __DEFAULT_FN_ATTRS
+_mm_stream_load_si128(__m128i const *__V) {
+ return (__m128i)__builtin_nontemporal_load((const __v2di *)__V);
}
/* SSE4 Packed Integer Min/Max Instructions. */
@@ -665,10 +663,9 @@ _mm_stream_load_si128 (__m128i const *__V)
/// \param __V2
/// A 128-bit vector of [16 x i8]
/// \returns A 128-bit vector of [16 x i8] containing the lesser values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_min_epi8 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_elementwise_min((__v16qs) __V1, (__v16qs) __V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi8(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_elementwise_min((__v16qs)__V1, (__v16qs)__V2);
}
/// Compares the corresponding elements of two 128-bit vectors of
@@ -684,10 +681,9 @@ _mm_min_epi8 (__m128i __V1, __m128i __V2)
/// \param __V2
/// A 128-bit vector of [16 x i8].
/// \returns A 128-bit vector of [16 x i8] containing the greater values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_max_epi8 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_elementwise_max((__v16qs) __V1, (__v16qs) __V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi8(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_elementwise_max((__v16qs)__V1, (__v16qs)__V2);
}
/// Compares the corresponding elements of two 128-bit vectors of
@@ -703,10 +699,9 @@ _mm_max_epi8 (__m128i __V1, __m128i __V2)
/// \param __V2
/// A 128-bit vector of [8 x u16].
/// \returns A 128-bit vector of [8 x u16] containing the lesser values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_min_epu16 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_elementwise_min((__v8hu) __V1, (__v8hu) __V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epu16(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_elementwise_min((__v8hu)__V1, (__v8hu)__V2);
}
/// Compares the corresponding elements of two 128-bit vectors of
@@ -722,10 +717,9 @@ _mm_min_epu16 (__m128i __V1, __m128i __V2)
/// \param __V2
/// A 128-bit vector of [8 x u16].
/// \returns A 128-bit vector of [8 x u16] containing the greater values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_max_epu16 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_elementwise_max((__v8hu) __V1, (__v8hu) __V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu16(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_elementwise_max((__v8hu)__V1, (__v8hu)__V2);
}
/// Compares the corresponding elements of two 128-bit vectors of
@@ -741,10 +735,9 @@ _mm_max_epu16 (__m128i __V1, __m128i __V2)
/// \param __V2
/// A 128-bit vector of [4 x i32].
/// \returns A 128-bit vector of [4 x i32] containing the lesser values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_min_epi32 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_elementwise_min((__v4si) __V1, (__v4si) __V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epi32(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_elementwise_min((__v4si)__V1, (__v4si)__V2);
}
/// Compares the corresponding elements of two 128-bit vectors of
@@ -760,10 +753,9 @@ _mm_min_epi32 (__m128i __V1, __m128i __V2)
/// \param __V2
/// A 128-bit vector of [4 x i32].
/// \returns A 128-bit vector of [4 x i32] containing the greater values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_max_epi32 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_elementwise_max((__v4si) __V1, (__v4si) __V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epi32(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_elementwise_max((__v4si)__V1, (__v4si)__V2);
}
/// Compares the corresponding elements of two 128-bit vectors of
@@ -779,10 +771,9 @@ _mm_max_epi32 (__m128i __V1, __m128i __V2)
/// \param __V2
/// A 128-bit vector of [4 x u32].
/// \returns A 128-bit vector of [4 x u32] containing the lesser values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_min_epu32 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_elementwise_min((__v4su) __V1, (__v4su) __V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_min_epu32(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_elementwise_min((__v4su)__V1, (__v4su)__V2);
}
/// Compares the corresponding elements of two 128-bit vectors of
@@ -798,10 +789,9 @@ _mm_min_epu32 (__m128i __V1, __m128i __V2)
/// \param __V2
/// A 128-bit vector of [4 x u32].
/// \returns A 128-bit vector of [4 x u32] containing the greater values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_max_epu32 (__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_elementwise_max((__v4su) __V1, (__v4su) __V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu32(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_elementwise_max((__v4su)__V1, (__v4su)__V2);
}
/* SSE4 Insertion and Extraction from XMM Register Instructions. */
@@ -869,21 +859,24 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2)
/// 10: Bits [95:64] of parameter \a X are returned. \n
/// 11: Bits [127:96] of parameter \a X are returned.
/// \returns A 32-bit integer containing the extracted 32 bits of float data.
-#define _mm_extract_ps(X, N) \
- __builtin_bit_cast(int, __builtin_ia32_vec_ext_v4sf((__v4sf)(__m128)(X), (int)(N)))
+#define _mm_extract_ps(X, N) \
+ __builtin_bit_cast( \
+ int, __builtin_ia32_vec_ext_v4sf((__v4sf)(__m128)(X), (int)(N)))
/* Miscellaneous insert and extract macros. */
/* Extract a single-precision float from X at index N into D. */
-#define _MM_EXTRACT_FLOAT(D, X, N) \
- do { (D) = __builtin_ia32_vec_ext_v4sf((__v4sf)(__m128)(X), (int)(N)); } while (0)
+#define _MM_EXTRACT_FLOAT(D, X, N) \
+ do { \
+ (D) = __builtin_ia32_vec_ext_v4sf((__v4sf)(__m128)(X), (int)(N)); \
+ } while (0)
/* Or together 2 sets of indexes (X and Y) with the zeroing bits (Z) to create
an index suitable for _mm_insert_ps. */
#define _MM_MK_INSERTPS_NDX(X, Y, Z) (((X) << 6) | ((Y) << 4) | (Z))
/* Extract a float from X at index N into the first index of the return. */
-#define _MM_PICK_OUT_PS(X, N) _mm_insert_ps (_mm_setzero_ps(), (X), \
- _MM_MK_INSERTPS_NDX((N), 0, 0x0e))
+#define _MM_PICK_OUT_PS(X, N) \
+ _mm_insert_ps(_mm_setzero_ps(), (X), _MM_MK_INSERTPS_NDX((N), 0, 0x0e))
/* Insert int into packed integer array at index. */
/// Constructs a 128-bit vector of [16 x i8] by first making a copy of
@@ -926,9 +919,9 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2)
/// 1110: Bits [119:112] of the result are used for insertion. \n
/// 1111: Bits [127:120] of the result are used for insertion.
/// \returns A 128-bit integer vector containing the constructed values.
-#define _mm_insert_epi8(X, I, N) \
- ((__m128i)__builtin_ia32_vec_set_v16qi((__v16qi)(__m128i)(X), \
- (int)(I), (int)(N)))
+#define _mm_insert_epi8(X, I, N) \
+ ((__m128i)__builtin_ia32_vec_set_v16qi((__v16qi)(__m128i)(X), (int)(I), \
+ (int)(N)))
/// Constructs a 128-bit vector of [4 x i32] by first making a copy of
/// the 128-bit integer vector parameter, and then inserting the 32-bit
@@ -958,9 +951,9 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2)
/// 10: Bits [95:64] of the result are used for insertion. \n
/// 11: Bits [127:96] of the result are used for insertion.
/// \returns A 128-bit integer vector containing the constructed values.
-#define _mm_insert_epi32(X, I, N) \
- ((__m128i)__builtin_ia32_vec_set_v4si((__v4si)(__m128i)(X), \
- (int)(I), (int)(N)))
+#define _mm_insert_epi32(X, I, N) \
+ ((__m128i)__builtin_ia32_vec_set_v4si((__v4si)(__m128i)(X), (int)(I), \
+ (int)(N)))
#ifdef __x86_64__
/// Constructs a 128-bit vector of [2 x i64] by first making a copy of
@@ -989,9 +982,9 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2)
/// 0: Bits [63:0] of the result are used for insertion. \n
/// 1: Bits [127:64] of the result are used for insertion. \n
/// \returns A 128-bit integer vector containing the constructed values.
-#define _mm_insert_epi64(X, I, N) \
- ((__m128i)__builtin_ia32_vec_set_v2di((__v2di)(__m128i)(X), \
- (long long)(I), (int)(N)))
+#define _mm_insert_epi64(X, I, N) \
+ ((__m128i)__builtin_ia32_vec_set_v2di((__v2di)(__m128i)(X), (long long)(I), \
+ (int)(N)))
#endif /* __x86_64__ */
/* Extract int from packed integer array at index. This returns the element
@@ -1032,8 +1025,8 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2)
/// \returns An unsigned integer, whose lower 8 bits are selected from the
/// 128-bit integer vector parameter and the remaining bits are assigned
/// zeros.
-#define _mm_extract_epi8(X, N) \
- ((int)(unsigned char)__builtin_ia32_vec_ext_v16qi((__v16qi)(__m128i)(X), \
+#define _mm_extract_epi8(X, N) \
+ ((int)(unsigned char)__builtin_ia32_vec_ext_v16qi((__v16qi)(__m128i)(X), \
(int)(N)))
/// Extracts a 32-bit element from the 128-bit integer vector of
@@ -1058,7 +1051,7 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2)
/// 11: Bits [127:96] of the parameter \a X are exracted.
/// \returns An integer, whose lower 32 bits are selected from the 128-bit
/// integer vector parameter and the remaining bits are assigned zeros.
-#define _mm_extract_epi32(X, N) \
+#define _mm_extract_epi32(X, N) \
((int)__builtin_ia32_vec_ext_v4si((__v4si)(__m128i)(X), (int)(N)))
#ifdef __x86_64__
@@ -1081,7 +1074,7 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2)
/// 0: Bits [63:0] are returned. \n
/// 1: Bits [127:64] are returned. \n
/// \returns A 64-bit integer.
-#define _mm_extract_epi64(X, N) \
+#define _mm_extract_epi64(X, N) \
((long long)__builtin_ia32_vec_ext_v2di((__v2di)(__m128i)(X), (int)(N)))
#endif /* __x86_64 */
@@ -1098,9 +1091,8 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2)
/// \param __V
/// A 128-bit integer vector selecting which bits to test in operand \a __M.
/// \returns TRUE if the specified bits are all zeros; FALSE otherwise.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_testz_si128(__m128i __M, __m128i __V)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_testz_si128(__m128i __M,
+ __m128i __V) {
return __builtin_ia32_ptestz128((__v2di)__M, (__v2di)__V);
}
@@ -1116,9 +1108,8 @@ _mm_testz_si128(__m128i __M, __m128i __V)
/// \param __V
/// A 128-bit integer vector selecting which bits to test in operand \a __M.
/// \returns TRUE if the specified bits are all ones; FALSE otherwise.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_testc_si128(__m128i __M, __m128i __V)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_testc_si128(__m128i __M,
+ __m128i __V) {
return __builtin_ia32_ptestc128((__v2di)__M, (__v2di)__V);
}
@@ -1135,9 +1126,8 @@ _mm_testc_si128(__m128i __M, __m128i __V)
/// A 128-bit integer vector selecting which bits to test in operand \a __M.
/// \returns TRUE if the specified bits are neither all zeros nor all ones;
/// FALSE otherwise.
-static __inline__ int __DEFAULT_FN_ATTRS
-_mm_testnzc_si128(__m128i __M, __m128i __V)
-{
+static __inline__ int __DEFAULT_FN_ATTRS _mm_testnzc_si128(__m128i __M,
+ __m128i __V) {
return __builtin_ia32_ptestnzc128((__v2di)__M, (__v2di)__V);
}
@@ -1193,7 +1183,7 @@ _mm_testnzc_si128(__m128i __M, __m128i __V)
/// \param V
/// A 128-bit integer vector selecting which bits to test in operand \a M.
/// \returns TRUE if the specified bits are all zeros; FALSE otherwise.
-#define _mm_test_all_zeros(M, V) _mm_testz_si128 ((M), (V))
+#define _mm_test_all_zeros(M, V) _mm_testz_si128((M), (V))
/* SSE4 64-bit Packed Integer Comparisons. */
/// Compares each of the corresponding 64-bit values of the 128-bit
@@ -1208,9 +1198,8 @@ _mm_testnzc_si128(__m128i __M, __m128i __V)
/// \param __V2
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmpeq_epi64(__m128i __V1, __m128i __V2)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmpeq_epi64(__m128i __V1,
+ __m128i __V2) {
return (__m128i)((__v2di)__V1 == (__v2di)__V2);
}
@@ -1228,12 +1217,13 @@ _mm_cmpeq_epi64(__m128i __V1, __m128i __V2)
/// A 128-bit vector of [16 x i8]. The lower eight 8-bit elements are sign-
/// extended to 16-bit values.
/// \returns A 128-bit vector of [8 x i16] containing the sign-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepi8_epi16(__m128i __V)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepi8_epi16(__m128i __V) {
/* This function always performs a signed extension, but __v16qi is a char
which may be signed or unsigned, so use __v16qs. */
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1, 2, 3, 4, 5, 6, 7), __v8hi);
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1, 2, 3, 4, 5, 6,
+ 7),
+ __v8hi);
}
/// Sign-extends each of the lower four 8-bit integer elements of a
@@ -1249,12 +1239,11 @@ _mm_cvtepi8_epi16(__m128i __V)
/// A 128-bit vector of [16 x i8]. The lower four 8-bit elements are
/// sign-extended to 32-bit values.
/// \returns A 128-bit vector of [4 x i32] containing the sign-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepi8_epi32(__m128i __V)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepi8_epi32(__m128i __V) {
/* This function always performs a signed extension, but __v16qi is a char
which may be signed or unsigned, so use __v16qs. */
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1, 2, 3), __v4si);
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1, 2, 3), __v4si);
}
/// Sign-extends each of the lower two 8-bit integer elements of a
@@ -1270,12 +1259,11 @@ _mm_cvtepi8_epi32(__m128i __V)
/// A 128-bit vector of [16 x i8]. The lower two 8-bit elements are
/// sign-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the sign-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepi8_epi64(__m128i __V)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepi8_epi64(__m128i __V) {
/* This function always performs a signed extension, but __v16qi is a char
which may be signed or unsigned, so use __v16qs. */
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1), __v2di);
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v16qs)__V, (__v16qs)__V, 0, 1), __v2di);
}
/// Sign-extends each of the lower four 16-bit integer elements of a
@@ -1291,10 +1279,9 @@ _mm_cvtepi8_epi64(__m128i __V)
/// A 128-bit vector of [8 x i16]. The lower four 16-bit elements are
/// sign-extended to 32-bit values.
/// \returns A 128-bit vector of [4 x i32] containing the sign-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepi16_epi32(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v8hi)__V, (__v8hi)__V, 0, 1, 2, 3), __v4si);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepi16_epi32(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v8hi)__V, (__v8hi)__V, 0, 1, 2, 3), __v4si);
}
/// Sign-extends each of the lower two 16-bit integer elements of a
@@ -1310,10 +1297,9 @@ _mm_cvtepi16_epi32(__m128i __V)
/// A 128-bit vector of [8 x i16]. The lower two 16-bit elements are
/// sign-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the sign-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepi16_epi64(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v8hi)__V, (__v8hi)__V, 0, 1), __v2di);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepi16_epi64(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v8hi)__V, (__v8hi)__V, 0, 1), __v2di);
}
/// Sign-extends each of the lower two 32-bit integer elements of a
@@ -1329,10 +1315,9 @@ _mm_cvtepi16_epi64(__m128i __V)
/// A 128-bit vector of [4 x i32]. The lower two 32-bit elements are
/// sign-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the sign-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepi32_epi64(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v4si)__V, (__v4si)__V, 0, 1), __v2di);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepi32_epi64(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v4si)__V, (__v4si)__V, 0, 1), __v2di);
}
/* SSE4 Packed Integer Zero-Extension. */
@@ -1349,10 +1334,11 @@ _mm_cvtepi32_epi64(__m128i __V)
/// A 128-bit vector of [16 x i8]. The lower eight 8-bit elements are
/// zero-extended to 16-bit values.
/// \returns A 128-bit vector of [8 x i16] containing the zero-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepu8_epi16(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1, 2, 3, 4, 5, 6, 7), __v8hi);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepu8_epi16(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1, 2, 3, 4, 5, 6,
+ 7),
+ __v8hi);
}
/// Zero-extends each of the lower four 8-bit integer elements of a
@@ -1368,10 +1354,9 @@ _mm_cvtepu8_epi16(__m128i __V)
/// A 128-bit vector of [16 x i8]. The lower four 8-bit elements are
/// zero-extended to 32-bit values.
/// \returns A 128-bit vector of [4 x i32] containing the zero-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepu8_epi32(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1, 2, 3), __v4si);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepu8_epi32(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1, 2, 3), __v4si);
}
/// Zero-extends each of the lower two 8-bit integer elements of a
@@ -1387,10 +1372,9 @@ _mm_cvtepu8_epi32(__m128i __V)
/// A 128-bit vector of [16 x i8]. The lower two 8-bit elements are
/// zero-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the zero-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepu8_epi64(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1), __v2di);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepu8_epi64(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v16qu)__V, (__v16qu)__V, 0, 1), __v2di);
}
/// Zero-extends each of the lower four 16-bit integer elements of a
@@ -1406,10 +1390,9 @@ _mm_cvtepu8_epi64(__m128i __V)
/// A 128-bit vector of [8 x i16]. The lower four 16-bit elements are
/// zero-extended to 32-bit values.
/// \returns A 128-bit vector of [4 x i32] containing the zero-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepu16_epi32(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v8hu)__V, (__v8hu)__V, 0, 1, 2, 3), __v4si);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepu16_epi32(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v8hu)__V, (__v8hu)__V, 0, 1, 2, 3), __v4si);
}
/// Zero-extends each of the lower two 16-bit integer elements of a
@@ -1425,10 +1408,9 @@ _mm_cvtepu16_epi32(__m128i __V)
/// A 128-bit vector of [8 x i16]. The lower two 16-bit elements are
/// zero-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the zero-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepu16_epi64(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v8hu)__V, (__v8hu)__V, 0, 1), __v2di);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepu16_epi64(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v8hu)__V, (__v8hu)__V, 0, 1), __v2di);
}
/// Zero-extends each of the lower two 32-bit integer elements of a
@@ -1444,10 +1426,9 @@ _mm_cvtepu16_epi64(__m128i __V)
/// A 128-bit vector of [4 x i32]. The lower two 32-bit elements are
/// zero-extended to 64-bit values.
/// \returns A 128-bit vector of [2 x i64] containing the zero-extended values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cvtepu32_epi64(__m128i __V)
-{
- return (__m128i)__builtin_convertvector(__builtin_shufflevector((__v4su)__V, (__v4su)__V, 0, 1), __v2di);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cvtepu32_epi64(__m128i __V) {
+ return (__m128i) __builtin_convertvector(
+ __builtin_shufflevector((__v4su)__V, (__v4su)__V, 0, 1), __v2di);
}
/* SSE4 Pack with Unsigned Saturation. */
@@ -1473,10 +1454,9 @@ _mm_cvtepu32_epi64(__m128i __V)
/// less than 0x0000 are saturated to 0x0000. The converted [4 x i16] values
/// are written to the higher 64 bits of the result.
/// \returns A 128-bit vector of [8 x i16] containing the converted values.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_packus_epi32(__m128i __V1, __m128i __V2)
-{
- return (__m128i) __builtin_ia32_packusdw128((__v4si)__V1, (__v4si)__V2);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_packus_epi32(__m128i __V1,
+ __m128i __V2) {
+ return (__m128i)__builtin_ia32_packusdw128((__v4si)__V1, (__v4si)__V2);
}
/* SSE4 Multiple Packed Sums of Absolute Difference. */
@@ -1515,9 +1495,9 @@ _mm_packus_epi32(__m128i __V1, __m128i __V2)
/// \endcode
/// \returns A 128-bit integer vector containing the sums of the sets of
/// absolute
diff erences between both operands.
-#define _mm_mpsadbw_epu8(X, Y, M) \
- ((__m128i) __builtin_ia32_mpsadbw128((__v16qi)(__m128i)(X), \
- (__v16qi)(__m128i)(Y), (M)))
+#define _mm_mpsadbw_epu8(X, Y, M) \
+ ((__m128i)__builtin_ia32_mpsadbw128((__v16qi)(__m128i)(X), \
+ (__v16qi)(__m128i)(Y), (M)))
/// Finds the minimum unsigned 16-bit element in the input 128-bit
/// vector of [8 x u16] and returns it and along with its index.
@@ -1532,10 +1512,8 @@ _mm_packus_epi32(__m128i __V1, __m128i __V2)
/// \returns A 128-bit value where bits [15:0] contain the minimum value found
/// in parameter \a __V, bits [18:16] contain the index of the minimum value
/// and the remaining bits are set to 0.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_minpos_epu16(__m128i __V)
-{
- return (__m128i) __builtin_ia32_phminposuw128((__v8hi)__V);
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_minpos_epu16(__m128i __V) {
+ return (__m128i)__builtin_ia32_phminposuw128((__v8hi)__V);
}
/* Handle the sse4.2 definitions here. */
@@ -1544,33 +1522,34 @@ _mm_minpos_epu16(__m128i __V)
so we'll do the same. */
#undef __DEFAULT_FN_ATTRS
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("sse4.2")))
+#define __DEFAULT_FN_ATTRS \
+ __attribute__((__always_inline__, __nodebug__, __target__("sse4.2")))
/* These specify the type of data that we're comparing. */
-#define _SIDD_UBYTE_OPS 0x00
-#define _SIDD_UWORD_OPS 0x01
-#define _SIDD_SBYTE_OPS 0x02
-#define _SIDD_SWORD_OPS 0x03
+#define _SIDD_UBYTE_OPS 0x00
+#define _SIDD_UWORD_OPS 0x01
+#define _SIDD_SBYTE_OPS 0x02
+#define _SIDD_SWORD_OPS 0x03
/* These specify the type of comparison operation. */
-#define _SIDD_CMP_EQUAL_ANY 0x00
-#define _SIDD_CMP_RANGES 0x04
-#define _SIDD_CMP_EQUAL_EACH 0x08
-#define _SIDD_CMP_EQUAL_ORDERED 0x0c
+#define _SIDD_CMP_EQUAL_ANY 0x00
+#define _SIDD_CMP_RANGES 0x04
+#define _SIDD_CMP_EQUAL_EACH 0x08
+#define _SIDD_CMP_EQUAL_ORDERED 0x0c
/* These macros specify the polarity of the operation. */
-#define _SIDD_POSITIVE_POLARITY 0x00
-#define _SIDD_NEGATIVE_POLARITY 0x10
-#define _SIDD_MASKED_POSITIVE_POLARITY 0x20
-#define _SIDD_MASKED_NEGATIVE_POLARITY 0x30
+#define _SIDD_POSITIVE_POLARITY 0x00
+#define _SIDD_NEGATIVE_POLARITY 0x10
+#define _SIDD_MASKED_POSITIVE_POLARITY 0x20
+#define _SIDD_MASKED_NEGATIVE_POLARITY 0x30
/* These macros are used in _mm_cmpXstri() to specify the return. */
-#define _SIDD_LEAST_SIGNIFICANT 0x00
-#define _SIDD_MOST_SIGNIFICANT 0x40
+#define _SIDD_LEAST_SIGNIFICANT 0x00
+#define _SIDD_MOST_SIGNIFICANT 0x40
/* These macros are used in _mm_cmpXstri() to specify the return. */
-#define _SIDD_BIT_MASK 0x00
-#define _SIDD_UNIT_MASK 0x40
+#define _SIDD_BIT_MASK 0x00
+#define _SIDD_UNIT_MASK 0x40
/* SSE4.2 Packed Comparison Intrinsics. */
/// Uses the immediate operand \a M to perform a comparison of string
@@ -1625,8 +1604,8 @@ _mm_minpos_epu16(__m128i __V)
/// repeating each bit 8 or 16 times).
/// \returns Returns a 128-bit integer vector representing the result mask of
/// the comparison.
-#define _mm_cmpistrm(A, B, M) \
- ((__m128i)__builtin_ia32_pcmpistrm128((__v16qi)(__m128i)(A), \
+#define _mm_cmpistrm(A, B, M) \
+ ((__m128i)__builtin_ia32_pcmpistrm128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -1679,9 +1658,9 @@ _mm_minpos_epu16(__m128i __V)
/// 0: The index of the least significant set bit. \n
/// 1: The index of the most significant set bit. \n
/// \returns Returns an integer representing the result index of the comparison.
-#define _mm_cmpistri(A, B, M) \
- ((int)__builtin_ia32_pcmpistri128((__v16qi)(__m128i)(A), \
- (__v16qi)(__m128i)(B), (int)(M)))
+#define _mm_cmpistri(A, B, M) \
+ ((int)__builtin_ia32_pcmpistri128((__v16qi)(__m128i)(A), \
+ (__v16qi)(__m128i)(B), (int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
/// data with explicitly defined lengths that is contained in source operands
@@ -1739,9 +1718,9 @@ _mm_minpos_epu16(__m128i __V)
/// repeating each bit 8 or 16 times). \n
/// \returns Returns a 128-bit integer vector representing the result mask of
/// the comparison.
-#define _mm_cmpestrm(A, LA, B, LB, M) \
- ((__m128i)__builtin_ia32_pcmpestrm128((__v16qi)(__m128i)(A), (int)(LA), \
- (__v16qi)(__m128i)(B), (int)(LB), \
+#define _mm_cmpestrm(A, LA, B, LB, M) \
+ ((__m128i)__builtin_ia32_pcmpestrm128((__v16qi)(__m128i)(A), (int)(LA), \
+ (__v16qi)(__m128i)(B), (int)(LB), \
(int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -1798,9 +1777,9 @@ _mm_minpos_epu16(__m128i __V)
/// 0: The index of the least significant set bit. \n
/// 1: The index of the most significant set bit. \n
/// \returns Returns an integer representing the result index of the comparison.
-#define _mm_cmpestri(A, LA, B, LB, M) \
- ((int)__builtin_ia32_pcmpestri128((__v16qi)(__m128i)(A), (int)(LA), \
- (__v16qi)(__m128i)(B), (int)(LB), \
+#define _mm_cmpestri(A, LA, B, LB, M) \
+ ((int)__builtin_ia32_pcmpestri128((__v16qi)(__m128i)(A), (int)(LA), \
+ (__v16qi)(__m128i)(B), (int)(LB), \
(int)(M)))
/* SSE4.2 Packed Comparison Intrinsics and EFlag Reading. */
@@ -1850,8 +1829,8 @@ _mm_minpos_epu16(__m128i __V)
/// to the size of \a A or \a B. \n
/// \returns Returns 1 if the bit mask is zero and the length of the string in
/// \a B is the maximum; otherwise, returns 0.
-#define _mm_cmpistra(A, B, M) \
- ((int)__builtin_ia32_pcmpistria128((__v16qi)(__m128i)(A), \
+#define _mm_cmpistra(A, B, M) \
+ ((int)__builtin_ia32_pcmpistria128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -1899,8 +1878,8 @@ _mm_minpos_epu16(__m128i __V)
/// 11: Negate the bit mask only for bits with an index less than or equal
/// to the size of \a A or \a B.
/// \returns Returns 1 if the bit mask is non-zero, otherwise, returns 0.
-#define _mm_cmpistrc(A, B, M) \
- ((int)__builtin_ia32_pcmpistric128((__v16qi)(__m128i)(A), \
+#define _mm_cmpistrc(A, B, M) \
+ ((int)__builtin_ia32_pcmpistric128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -1947,8 +1926,8 @@ _mm_minpos_epu16(__m128i __V)
/// 11: Negate the bit mask only for bits with an index less than or equal
/// to the size of \a A or \a B. \n
/// \returns Returns bit 0 of the resulting bit mask.
-#define _mm_cmpistro(A, B, M) \
- ((int)__builtin_ia32_pcmpistrio128((__v16qi)(__m128i)(A), \
+#define _mm_cmpistro(A, B, M) \
+ ((int)__builtin_ia32_pcmpistrio128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -1997,8 +1976,8 @@ _mm_minpos_epu16(__m128i __V)
/// to the size of \a A or \a B. \n
/// \returns Returns 1 if the length of the string in \a A is less than the
/// maximum, otherwise, returns 0.
-#define _mm_cmpistrs(A, B, M) \
- ((int)__builtin_ia32_pcmpistris128((__v16qi)(__m128i)(A), \
+#define _mm_cmpistrs(A, B, M) \
+ ((int)__builtin_ia32_pcmpistris128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -2047,8 +2026,8 @@ _mm_minpos_epu16(__m128i __V)
/// to the size of \a A or \a B.
/// \returns Returns 1 if the length of the string in \a B is less than the
/// maximum, otherwise, returns 0.
-#define _mm_cmpistrz(A, B, M) \
- ((int)__builtin_ia32_pcmpistriz128((__v16qi)(__m128i)(A), \
+#define _mm_cmpistrz(A, B, M) \
+ ((int)__builtin_ia32_pcmpistriz128((__v16qi)(__m128i)(A), \
(__v16qi)(__m128i)(B), (int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -2101,9 +2080,9 @@ _mm_minpos_epu16(__m128i __V)
/// to the size of \a A or \a B.
/// \returns Returns 1 if the bit mask is zero and the length of the string in
/// \a B is the maximum, otherwise, returns 0.
-#define _mm_cmpestra(A, LA, B, LB, M) \
- ((int)__builtin_ia32_pcmpestria128((__v16qi)(__m128i)(A), (int)(LA), \
- (__v16qi)(__m128i)(B), (int)(LB), \
+#define _mm_cmpestra(A, LA, B, LB, M) \
+ ((int)__builtin_ia32_pcmpestria128((__v16qi)(__m128i)(A), (int)(LA), \
+ (__v16qi)(__m128i)(B), (int)(LB), \
(int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -2155,9 +2134,9 @@ _mm_minpos_epu16(__m128i __V)
/// 11: Negate the bit mask only for bits with an index less than or equal
/// to the size of \a A or \a B. \n
/// \returns Returns 1 if the resulting mask is non-zero, otherwise, returns 0.
-#define _mm_cmpestrc(A, LA, B, LB, M) \
- ((int)__builtin_ia32_pcmpestric128((__v16qi)(__m128i)(A), (int)(LA), \
- (__v16qi)(__m128i)(B), (int)(LB), \
+#define _mm_cmpestrc(A, LA, B, LB, M) \
+ ((int)__builtin_ia32_pcmpestric128((__v16qi)(__m128i)(A), (int)(LA), \
+ (__v16qi)(__m128i)(B), (int)(LB), \
(int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -2208,9 +2187,9 @@ _mm_minpos_epu16(__m128i __V)
/// 11: Negate the bit mask only for bits with an index less than or equal
/// to the size of \a A or \a B.
/// \returns Returns bit 0 of the resulting bit mask.
-#define _mm_cmpestro(A, LA, B, LB, M) \
- ((int)__builtin_ia32_pcmpestrio128((__v16qi)(__m128i)(A), (int)(LA), \
- (__v16qi)(__m128i)(B), (int)(LB), \
+#define _mm_cmpestro(A, LA, B, LB, M) \
+ ((int)__builtin_ia32_pcmpestrio128((__v16qi)(__m128i)(A), (int)(LA), \
+ (__v16qi)(__m128i)(B), (int)(LB), \
(int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -2263,9 +2242,9 @@ _mm_minpos_epu16(__m128i __V)
/// to the size of \a A or \a B. \n
/// \returns Returns 1 if the length of the string in \a A is less than the
/// maximum, otherwise, returns 0.
-#define _mm_cmpestrs(A, LA, B, LB, M) \
- ((int)__builtin_ia32_pcmpestris128((__v16qi)(__m128i)(A), (int)(LA), \
- (__v16qi)(__m128i)(B), (int)(LB), \
+#define _mm_cmpestrs(A, LA, B, LB, M) \
+ ((int)__builtin_ia32_pcmpestris128((__v16qi)(__m128i)(A), (int)(LA), \
+ (__v16qi)(__m128i)(B), (int)(LB), \
(int)(M)))
/// Uses the immediate operand \a M to perform a comparison of string
@@ -2317,9 +2296,9 @@ _mm_minpos_epu16(__m128i __V)
/// to the size of \a A or \a B.
/// \returns Returns 1 if the length of the string in \a B is less than the
/// maximum, otherwise, returns 0.
-#define _mm_cmpestrz(A, LA, B, LB, M) \
- ((int)__builtin_ia32_pcmpestriz128((__v16qi)(__m128i)(A), (int)(LA), \
- (__v16qi)(__m128i)(B), (int)(LB), \
+#define _mm_cmpestrz(A, LA, B, LB, M) \
+ ((int)__builtin_ia32_pcmpestriz128((__v16qi)(__m128i)(A), (int)(LA), \
+ (__v16qi)(__m128i)(B), (int)(LB), \
(int)(M)))
/* SSE4.2 Compare Packed Data -- Greater Than. */
@@ -2336,9 +2315,8 @@ _mm_minpos_epu16(__m128i __V)
/// \param __V2
/// A 128-bit integer vector.
/// \returns A 128-bit integer vector containing the comparison results.
-static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_cmpgt_epi64(__m128i __V1, __m128i __V2)
-{
+static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_cmpgt_epi64(__m128i __V1,
+ __m128i __V2) {
return (__m128i)((__v2di)__V1 > (__v2di)__V2);
}
More information about the cfe-commits
mailing list