@@ -28,258 +28,257 @@
/* Offsets for data table __svml_dacos_data_internal
*/
-#define SgnBit 0
-#define OneHalf 32
-#define SmallNorm 64
-#define MOne 96
-#define Two 128
-#define sqrt_coeff 160
-#define poly_coeff 288
-#define PiH 672
-#define Pi2H 704
+#define SgnBit 0
+#define OneHalf 32
+#define SmallNorm 64
+#define MOne 96
+#define Two 128
+#define sqrt_coeff 160
+#define poly_coeff 288
+#define PiH 672
+#define Pi2H 704
#include <sysdep.h>
- .text
- .section .text.avx2,"ax",@progbits
+ .section .text.avx2, "ax", @progbits
ENTRY(_ZGVdN4v_acos_avx2)
- pushq %rbp
- cfi_def_cfa_offset(16)
- movq %rsp, %rbp
- cfi_def_cfa(6, 16)
- cfi_offset(6, -16)
- andq $-32, %rsp
- subq $96, %rsp
- vmovupd __svml_dacos_data_internal(%rip), %ymm6
- vmovupd OneHalf+__svml_dacos_data_internal(%rip), %ymm7
- vmovapd %ymm0, %ymm5
-
-/* x = -|arg| */
- vorpd %ymm5, %ymm6, %ymm4
-
-/* Y = 0.5 + 0.5*(-x) */
- vfmadd231pd %ymm4, %ymm7, %ymm7
-
-/* x^2 */
- vmulpd %ymm4, %ymm4, %ymm8
-
-/* S ~ 2*sqrt(Y) */
- vmovupd sqrt_coeff+__svml_dacos_data_internal(%rip), %ymm0
- vcmplt_oqpd SmallNorm+__svml_dacos_data_internal(%rip), %ymm7, %ymm12
- vminpd %ymm7, %ymm8, %ymm2
-
-/* NaN processed in special branch (so wind test passed) */
- vcmpnge_uqpd MOne+__svml_dacos_data_internal(%rip), %ymm4, %ymm9
- vcvtpd2ps %ymm7, %xmm10
- vmovupd poly_coeff+64+__svml_dacos_data_internal(%rip), %ymm8
- vcmpnlt_uqpd %ymm7, %ymm2, %ymm1
- vrsqrtps %xmm10, %xmm11
- vfmadd213pd poly_coeff+96+__svml_dacos_data_internal(%rip), %ymm2, %ymm8
- vcvtps2pd %xmm11, %ymm13
- vmovupd poly_coeff+128+__svml_dacos_data_internal(%rip), %ymm11
- vandnpd %ymm13, %ymm12, %ymm14
- vmulpd %ymm14, %ymm14, %ymm15
- vfmadd213pd poly_coeff+160+__svml_dacos_data_internal(%rip), %ymm2, %ymm11
- vmulpd %ymm2, %ymm2, %ymm13
- vmovupd poly_coeff+256+__svml_dacos_data_internal(%rip), %ymm12
- vmulpd %ymm13, %ymm13, %ymm10
- vfmadd213pd poly_coeff+288+__svml_dacos_data_internal(%rip), %ymm2, %ymm12
- vandpd %ymm5, %ymm6, %ymm3
- vaddpd %ymm7, %ymm7, %ymm6
- vmulpd %ymm6, %ymm14, %ymm7
- vfmsub213pd Two+__svml_dacos_data_internal(%rip), %ymm15, %ymm6
- vmovupd poly_coeff+320+__svml_dacos_data_internal(%rip), %ymm14
- vfmadd213pd sqrt_coeff+32+__svml_dacos_data_internal(%rip), %ymm6, %ymm0
- vmulpd %ymm6, %ymm7, %ymm15
- vfmadd213pd poly_coeff+352+__svml_dacos_data_internal(%rip), %ymm2, %ymm14
- vfmadd213pd sqrt_coeff+64+__svml_dacos_data_internal(%rip), %ymm6, %ymm0
- vfmadd213pd sqrt_coeff+96+__svml_dacos_data_internal(%rip), %ymm6, %ymm0
-
-/* polynomial */
- vmovupd poly_coeff+__svml_dacos_data_internal(%rip), %ymm6
- vfnmadd213pd %ymm7, %ymm15, %ymm0
- vfmadd213pd poly_coeff+32+__svml_dacos_data_internal(%rip), %ymm2, %ymm6
- vblendvpd %ymm1, %ymm0, %ymm4, %ymm0
- vfmadd213pd %ymm8, %ymm13, %ymm6
- vmovmskpd %ymm9, %edx
- vmovupd poly_coeff+192+__svml_dacos_data_internal(%rip), %ymm9
- vfmadd213pd poly_coeff+224+__svml_dacos_data_internal(%rip), %ymm2, %ymm9
- vfmadd213pd %ymm9, %ymm13, %ymm11
- vfmadd213pd %ymm11, %ymm10, %ymm6
- vfmadd213pd %ymm12, %ymm13, %ymm6
- vfmadd213pd %ymm14, %ymm13, %ymm6
- vmulpd %ymm6, %ymm2, %ymm9
-
-/* X<X^2 iff X<0 */
- vcmplt_oqpd %ymm2, %ymm5, %ymm6
- vandpd PiH+__svml_dacos_data_internal(%rip), %ymm1, %ymm2
- vandnpd Pi2H+__svml_dacos_data_internal(%rip), %ymm1, %ymm7
- vxorpd %ymm3, %ymm0, %ymm1
- vfmadd213pd %ymm1, %ymm1, %ymm9
- vandpd %ymm6, %ymm2, %ymm2
- vaddpd %ymm7, %ymm2, %ymm8
- vaddpd %ymm9, %ymm8, %ymm0
- testl %edx, %edx
-
-/* Go to special inputs processing branch */
- jne L(SPECIAL_VALUES_BRANCH)
- # LOE rbx r12 r13 r14 r15 edx ymm0 ymm5
-
-/* Restore registers
- * and exit the function
- */
+ pushq %rbp
+ cfi_def_cfa_offset(16)
+ movq %rsp, %rbp
+ cfi_def_cfa(6, 16)
+ cfi_offset(6, -16)
+ andq $-32, %rsp
+ subq $96, %rsp
+ vmovupd __svml_dacos_data_internal(%rip), %ymm6
+ vmovupd OneHalf+__svml_dacos_data_internal(%rip), %ymm7
+ vmovapd %ymm0, %ymm5
+
+ /* x = -|arg| */
+ vorpd %ymm5, %ymm6, %ymm4
+
+ /* Y = 0.5 + 0.5*(-x) */
+ vfmadd231pd %ymm4, %ymm7, %ymm7
+
+ /* x^2 */
+ vmulpd %ymm4, %ymm4, %ymm8
+
+ /* S ~ 2*sqrt(Y) */
+ vmovupd sqrt_coeff+__svml_dacos_data_internal(%rip), %ymm0
+ vcmplt_oqpd SmallNorm+__svml_dacos_data_internal(%rip), %ymm7, %ymm12
+ vminpd %ymm7, %ymm8, %ymm2
+
+ /* NaN processed in special branch (so wind test passed) */
+ vcmpnge_uqpd MOne+__svml_dacos_data_internal(%rip), %ymm4, %ymm9
+ vcvtpd2ps %ymm7, %xmm10
+ vmovupd poly_coeff+64+__svml_dacos_data_internal(%rip), %ymm8
+ vcmpnlt_uqpd %ymm7, %ymm2, %ymm1
+ vrsqrtps %xmm10, %xmm11
+ vfmadd213pd poly_coeff+96+__svml_dacos_data_internal(%rip), %ymm2, %ymm8
+ vcvtps2pd %xmm11, %ymm13
+ vmovupd poly_coeff+128+__svml_dacos_data_internal(%rip), %ymm11
+ vandnpd %ymm13, %ymm12, %ymm14
+ vmulpd %ymm14, %ymm14, %ymm15
+ vfmadd213pd poly_coeff+160+__svml_dacos_data_internal(%rip), %ymm2, %ymm11
+ vmulpd %ymm2, %ymm2, %ymm13
+ vmovupd poly_coeff+256+__svml_dacos_data_internal(%rip), %ymm12
+ vmulpd %ymm13, %ymm13, %ymm10
+ vfmadd213pd poly_coeff+288+__svml_dacos_data_internal(%rip), %ymm2, %ymm12
+ vandpd %ymm5, %ymm6, %ymm3
+ vaddpd %ymm7, %ymm7, %ymm6
+ vmulpd %ymm6, %ymm14, %ymm7
+ vfmsub213pd Two+__svml_dacos_data_internal(%rip), %ymm15, %ymm6
+ vmovupd poly_coeff+320+__svml_dacos_data_internal(%rip), %ymm14
+ vfmadd213pd sqrt_coeff+32+__svml_dacos_data_internal(%rip), %ymm6, %ymm0
+ vmulpd %ymm6, %ymm7, %ymm15
+ vfmadd213pd poly_coeff+352+__svml_dacos_data_internal(%rip), %ymm2, %ymm14
+ vfmadd213pd sqrt_coeff+64+__svml_dacos_data_internal(%rip), %ymm6, %ymm0
+ vfmadd213pd sqrt_coeff+96+__svml_dacos_data_internal(%rip), %ymm6, %ymm0
+
+ /* polynomial */
+ vmovupd poly_coeff+__svml_dacos_data_internal(%rip), %ymm6
+ vfnmadd213pd %ymm7, %ymm15, %ymm0
+ vfmadd213pd poly_coeff+32+__svml_dacos_data_internal(%rip), %ymm2, %ymm6
+ vblendvpd %ymm1, %ymm0, %ymm4, %ymm0
+ vfmadd213pd %ymm8, %ymm13, %ymm6
+ vmovmskpd %ymm9, %edx
+ vmovupd poly_coeff+192+__svml_dacos_data_internal(%rip), %ymm9
+ vfmadd213pd poly_coeff+224+__svml_dacos_data_internal(%rip), %ymm2, %ymm9
+ vfmadd213pd %ymm9, %ymm13, %ymm11
+ vfmadd213pd %ymm11, %ymm10, %ymm6
+ vfmadd213pd %ymm12, %ymm13, %ymm6
+ vfmadd213pd %ymm14, %ymm13, %ymm6
+ vmulpd %ymm6, %ymm2, %ymm9
+
+ /* X<X^2 iff X<0 */
+ vcmplt_oqpd %ymm2, %ymm5, %ymm6
+ vandpd PiH+__svml_dacos_data_internal(%rip), %ymm1, %ymm2
+ vandnpd Pi2H+__svml_dacos_data_internal(%rip), %ymm1, %ymm7
+ vxorpd %ymm3, %ymm0, %ymm1
+ vfmadd213pd %ymm1, %ymm1, %ymm9
+ vandpd %ymm6, %ymm2, %ymm2
+ vaddpd %ymm7, %ymm2, %ymm8
+ vaddpd %ymm9, %ymm8, %ymm0
+ testl %edx, %edx
+
+ /* Go to special inputs processing branch */
+ jne L(SPECIAL_VALUES_BRANCH)
+ # LOE rbx r12 r13 r14 r15 edx ymm0 ymm5
+
+ /* Restore registers
+ * and exit the function
+ */
L(EXIT):
- movq %rbp, %rsp
- popq %rbp
- cfi_def_cfa(7, 8)
- cfi_restore(6)
- ret
- cfi_def_cfa(6, 16)
- cfi_offset(6, -16)
-
-/* Branch to process
- * special inputs
- */
+ movq %rbp, %rsp
+ popq %rbp
+ cfi_def_cfa(7, 8)
+ cfi_restore(6)
+ ret
+ cfi_def_cfa(6, 16)
+ cfi_offset(6, -16)
+
+ /* Branch to process
+ * special inputs
+ */
L(SPECIAL_VALUES_BRANCH):
- vmovupd %ymm5, 32(%rsp)
- vmovupd %ymm0, 64(%rsp)
- # LOE rbx r12 r13 r14 r15 edx ymm0
-
- xorl %eax, %eax
- # LOE rbx r12 r13 r14 r15 eax edx
-
- vzeroupper
- movq %r12, 16(%rsp)
- /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */
- .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22
- movl %eax, %r12d
- movq %r13, 8(%rsp)
- /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */
- .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22
- movl %edx, %r13d
- movq %r14, (%rsp)
- /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */
- .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22
- # LOE rbx r15 r12d r13d
-
-/* Range mask
- * bits check
- */
+ vmovupd %ymm5, 32(%rsp)
+ vmovupd %ymm0, 64(%rsp)
+ # LOE rbx r12 r13 r14 r15 edx ymm0
+
+ xorl %eax, %eax
+ # LOE rbx r12 r13 r14 r15 eax edx
+
+ vzeroupper
+ movq %r12, 16(%rsp)
+ /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22
+ movl %eax, %r12d
+ movq %r13, 8(%rsp)
+ /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22
+ movl %edx, %r13d
+ movq %r14, (%rsp)
+ /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22
+ # LOE rbx r15 r12d r13d
+
+ /* Range mask
+ * bits check
+ */
L(RANGEMASK_CHECK):
- btl %r12d, %r13d
+ btl %r12d, %r13d
-/* Call scalar math function */
- jc L(SCALAR_MATH_CALL)
- # LOE rbx r15 r12d r13d
+ /* Call scalar math function */
+ jc L(SCALAR_MATH_CALL)
+ # LOE rbx r15 r12d r13d
-/* Special inputs
- * processing loop
- */
+ /* Special inputs
+ * processing loop
+ */
L(SPECIAL_VALUES_LOOP):
- incl %r12d
- cmpl $4, %r12d
-
-/* Check bits in range mask */
- jl L(RANGEMASK_CHECK)
- # LOE rbx r15 r12d r13d
-
- movq 16(%rsp), %r12
- cfi_restore(12)
- movq 8(%rsp), %r13
- cfi_restore(13)
- movq (%rsp), %r14
- cfi_restore(14)
- vmovupd 64(%rsp), %ymm0
-
-/* Go to exit */
- jmp L(EXIT)
- /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */
- .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22
- /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */
- .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22
- /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */
- .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22
- # LOE rbx r12 r13 r14 r15 ymm0
-
-/* Scalar math fucntion call
- * to process special input
- */
+ incl %r12d
+ cmpl $4, %r12d
+
+ /* Check bits in range mask */
+ jl L(RANGEMASK_CHECK)
+ # LOE rbx r15 r12d r13d
+
+ movq 16(%rsp), %r12
+ cfi_restore(12)
+ movq 8(%rsp), %r13
+ cfi_restore(13)
+ movq (%rsp), %r14
+ cfi_restore(14)
+ vmovupd 64(%rsp), %ymm0
+
+ /* Go to exit */
+ jmp L(EXIT)
+ /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22
+ /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22
+ /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22
+ # LOE rbx r12 r13 r14 r15 ymm0
+
+ /* Scalar math fucntion call
+ * to process special input
+ */
L(SCALAR_MATH_CALL):
- movl %r12d, %r14d
- movsd 32(%rsp,%r14,8), %xmm0
- call acos@PLT
- # LOE rbx r14 r15 r12d r13d xmm0
+ movl %r12d, %r14d
+ movsd 32(%rsp, %r14, 8), %xmm0
+ call acos@PLT
+ # LOE rbx r14 r15 r12d r13d xmm0
- movsd %xmm0, 64(%rsp,%r14,8)
+ movsd %xmm0, 64(%rsp, %r14, 8)
-/* Process special inputs in loop */
- jmp L(SPECIAL_VALUES_LOOP)
- # LOE rbx r15 r12d r13d
+ /* Process special inputs in loop */
+ jmp L(SPECIAL_VALUES_LOOP)
+ # LOE rbx r15 r12d r13d
END(_ZGVdN4v_acos_avx2)
- .section .rodata, "a"
- .align 32
+ .section .rodata, "a"
+ .align 32
#ifdef __svml_dacos_data_internal_typedef
typedef unsigned int VUINT32;
typedef struct {
- __declspec(align(32)) VUINT32 SgnBit[4][2];
- __declspec(align(32)) VUINT32 OneHalf[4][2];
- __declspec(align(32)) VUINT32 SmallNorm[4][2];
- __declspec(align(32)) VUINT32 MOne[4][2];
- __declspec(align(32)) VUINT32 Two[4][2];
- __declspec(align(32)) VUINT32 sqrt_coeff[4][4][2];
- __declspec(align(32)) VUINT32 poly_coeff[12][4][2];
- __declspec(align(32)) VUINT32 PiH[4][2];
- __declspec(align(32)) VUINT32 Pi2H[4][2];
+ __declspec(align(32)) VUINT32 SgnBit[4][2];
+ __declspec(align(32)) VUINT32 OneHalf[4][2];
+ __declspec(align(32)) VUINT32 SmallNorm[4][2];
+ __declspec(align(32)) VUINT32 MOne[4][2];
+ __declspec(align(32)) VUINT32 Two[4][2];
+ __declspec(align(32)) VUINT32 sqrt_coeff[4][4][2];
+ __declspec(align(32)) VUINT32 poly_coeff[12][4][2];
+ __declspec(align(32)) VUINT32 PiH[4][2];
+ __declspec(align(32)) VUINT32 Pi2H[4][2];
} __svml_dacos_data_internal;
#endif
__svml_dacos_data_internal:
- /*== SgnBit ==*/
- .quad 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000
- /*== OneHalf ==*/
- .align 32
- .quad 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000
- /*== SmallNorm ==*/
- .align 32
- .quad 0x3000000000000000, 0x3000000000000000, 0x3000000000000000, 0x3000000000000000
- /*== MOne ==*/
- .align 32
- .quad 0xbff0000000000000, 0xbff0000000000000, 0xbff0000000000000, 0xbff0000000000000
- /*== Two ==*/
- .align 32
- .quad 0x4000000000000000, 0x4000000000000000, 0x4000000000000000, 0x4000000000000000
- /*== sqrt_coeff[4] ==*/
- .align 32
- .quad 0xbf918000993B24C3, 0xbf918000993B24C3, 0xbf918000993B24C3, 0xbf918000993B24C3 /* sqrt_coeff4 */
- .quad 0x3fa400006F70D42D, 0x3fa400006F70D42D, 0x3fa400006F70D42D, 0x3fa400006F70D42D /* sqrt_coeff3 */
- .quad 0xbfb7FFFFFFFFFE97, 0xbfb7FFFFFFFFFE97, 0xbfb7FFFFFFFFFE97, 0xbfb7FFFFFFFFFE97 /* sqrt_coeff2 */
- .quad 0x3fcFFFFFFFFFFF9D, 0x3fcFFFFFFFFFFF9D, 0x3fcFFFFFFFFFFF9D, 0x3fcFFFFFFFFFFF9D /* sqrt_coeff1 */
- /*== poly_coeff[12] ==*/
- .align 32
- .quad 0x3fa07520C70EB909, 0x3fa07520C70EB909, 0x3fa07520C70EB909, 0x3fa07520C70EB909 /* poly_coeff12 */
- .quad 0xbf90FB17F7DBB0ED, 0xbf90FB17F7DBB0ED, 0xbf90FB17F7DBB0ED, 0xbf90FB17F7DBB0ED /* poly_coeff11 */
- .quad 0x3f943F44BFBC3BAE, 0x3f943F44BFBC3BAE, 0x3f943F44BFBC3BAE, 0x3f943F44BFBC3BAE /* poly_coeff10 */
- .quad 0x3f7A583395D45ED5, 0x3f7A583395D45ED5, 0x3f7A583395D45ED5, 0x3f7A583395D45ED5 /* poly_coeff9 */
- .quad 0x3f88F8DC2AFCCAD6, 0x3f88F8DC2AFCCAD6, 0x3f88F8DC2AFCCAD6, 0x3f88F8DC2AFCCAD6 /* poly_coeff8 */
- .quad 0x3f8C6DBBCB88BD57, 0x3f8C6DBBCB88BD57, 0x3f8C6DBBCB88BD57, 0x3f8C6DBBCB88BD57 /* poly_coeff7 */
- .quad 0x3f91C6DCF538AD2E, 0x3f91C6DCF538AD2E, 0x3f91C6DCF538AD2E, 0x3f91C6DCF538AD2E /* poly_coeff6 */
- .quad 0x3f96E89CEBDEFadd, 0x3f96E89CEBDEFadd, 0x3f96E89CEBDEFadd, 0x3f96E89CEBDEFadd /* poly_coeff5 */
- .quad 0x3f9F1C72E13AD8BE, 0x3f9F1C72E13AD8BE, 0x3f9F1C72E13AD8BE, 0x3f9F1C72E13AD8BE /* poly_coeff4 */
- .quad 0x3fa6DB6DB3B445F8, 0x3fa6DB6DB3B445F8, 0x3fa6DB6DB3B445F8, 0x3fa6DB6DB3B445F8 /* poly_coeff3 */
- .quad 0x3fb333333337E0DE, 0x3fb333333337E0DE, 0x3fb333333337E0DE, 0x3fb333333337E0DE /* poly_coeff2 */
- .quad 0x3fc555555555529C, 0x3fc555555555529C, 0x3fc555555555529C, 0x3fc555555555529C /* poly_coeff1 */
- /*== PiH ==*/
- .align 32
- .quad 0x400921fb54442d18, 0x400921fb54442d18, 0x400921fb54442d18, 0x400921fb54442d18
- /*== Pi2H ==*/
- .align 32
- .quad 0x3ff921fb54442d18, 0x3ff921fb54442d18, 0x3ff921fb54442d18, 0x3ff921fb54442d18
- .align 32
- .type __svml_dacos_data_internal,@object
- .size __svml_dacos_data_internal,.-__svml_dacos_data_internal
+ /* SgnBit */
+ .quad 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000
+ /* OneHalf */
+ .align 32
+ .quad 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000
+ /* SmallNorm */
+ .align 32
+ .quad 0x3000000000000000, 0x3000000000000000, 0x3000000000000000, 0x3000000000000000
+ /* MOne */
+ .align 32
+ .quad 0xbff0000000000000, 0xbff0000000000000, 0xbff0000000000000, 0xbff0000000000000
+ /* Two */
+ .align 32
+ .quad 0x4000000000000000, 0x4000000000000000, 0x4000000000000000, 0x4000000000000000
+ /* sqrt_coeff[4] */
+ .align 32
+ .quad 0xbf918000993B24C3, 0xbf918000993B24C3, 0xbf918000993B24C3, 0xbf918000993B24C3 /* sqrt_coeff4 */
+ .quad 0x3fa400006F70D42D, 0x3fa400006F70D42D, 0x3fa400006F70D42D, 0x3fa400006F70D42D /* sqrt_coeff3 */
+ .quad 0xbfb7FFFFFFFFFE97, 0xbfb7FFFFFFFFFE97, 0xbfb7FFFFFFFFFE97, 0xbfb7FFFFFFFFFE97 /* sqrt_coeff2 */
+ .quad 0x3fcFFFFFFFFFFF9D, 0x3fcFFFFFFFFFFF9D, 0x3fcFFFFFFFFFFF9D, 0x3fcFFFFFFFFFFF9D /* sqrt_coeff1 */
+ /* poly_coeff[12] */
+ .align 32
+ .quad 0x3fa07520C70EB909, 0x3fa07520C70EB909, 0x3fa07520C70EB909, 0x3fa07520C70EB909 /* poly_coeff12 */
+ .quad 0xbf90FB17F7DBB0ED, 0xbf90FB17F7DBB0ED, 0xbf90FB17F7DBB0ED, 0xbf90FB17F7DBB0ED /* poly_coeff11 */
+ .quad 0x3f943F44BFBC3BAE, 0x3f943F44BFBC3BAE, 0x3f943F44BFBC3BAE, 0x3f943F44BFBC3BAE /* poly_coeff10 */
+ .quad 0x3f7A583395D45ED5, 0x3f7A583395D45ED5, 0x3f7A583395D45ED5, 0x3f7A583395D45ED5 /* poly_coeff9 */
+ .quad 0x3f88F8DC2AFCCAD6, 0x3f88F8DC2AFCCAD6, 0x3f88F8DC2AFCCAD6, 0x3f88F8DC2AFCCAD6 /* poly_coeff8 */
+ .quad 0x3f8C6DBBCB88BD57, 0x3f8C6DBBCB88BD57, 0x3f8C6DBBCB88BD57, 0x3f8C6DBBCB88BD57 /* poly_coeff7 */
+ .quad 0x3f91C6DCF538AD2E, 0x3f91C6DCF538AD2E, 0x3f91C6DCF538AD2E, 0x3f91C6DCF538AD2E /* poly_coeff6 */
+ .quad 0x3f96E89CEBDEFadd, 0x3f96E89CEBDEFadd, 0x3f96E89CEBDEFadd, 0x3f96E89CEBDEFadd /* poly_coeff5 */
+ .quad 0x3f9F1C72E13AD8BE, 0x3f9F1C72E13AD8BE, 0x3f9F1C72E13AD8BE, 0x3f9F1C72E13AD8BE /* poly_coeff4 */
+ .quad 0x3fa6DB6DB3B445F8, 0x3fa6DB6DB3B445F8, 0x3fa6DB6DB3B445F8, 0x3fa6DB6DB3B445F8 /* poly_coeff3 */
+ .quad 0x3fb333333337E0DE, 0x3fb333333337E0DE, 0x3fb333333337E0DE, 0x3fb333333337E0DE /* poly_coeff2 */
+ .quad 0x3fc555555555529C, 0x3fc555555555529C, 0x3fc555555555529C, 0x3fc555555555529C /* poly_coeff1 */
+ /* PiH */
+ .align 32
+ .quad 0x400921fb54442d18, 0x400921fb54442d18, 0x400921fb54442d18, 0x400921fb54442d18
+ /* Pi2H */
+ .align 32
+ .quad 0x3ff921fb54442d18, 0x3ff921fb54442d18, 0x3ff921fb54442d18, 0x3ff921fb54442d18
+ .align 32
+ .type __svml_dacos_data_internal, @object
+ .size __svml_dacos_data_internal, .-__svml_dacos_data_internal