aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S549
1 files changed, 274 insertions, 275 deletions
diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S
index 4878603..efae1f8 100644
--- a/sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S
+++ b/sysdeps/x86_64/fpu/multiarch/svml_d_expm18_core_avx512.S
@@ -31,304 +31,303 @@
/* Offsets for data table __svml_dexpm1_data_internal_avx512
*/
-#define Exp_tbl_H 0
-#define Exp_tbl_L 128
-#define L2E 256
-#define Shifter 320
-#define Threshold 384
-#define SgnMask 448
-#define L2H 512
-#define L2L 576
-#define ZThres 640
-#define EMask 704
-#define poly_coeff7 768
-#define poly_coeff6 832
-#define poly_coeff5 896
-#define poly_coeff4 960
-#define poly_coeff3 1024
-#define poly_coeff2 1088
-#define One 1152
+#define Exp_tbl_H 0
+#define Exp_tbl_L 128
+#define L2E 256
+#define Shifter 320
+#define Threshold 384
+#define SgnMask 448
+#define L2H 512
+#define L2L 576
+#define ZThres 640
+#define EMask 704
+#define poly_coeff7 768
+#define poly_coeff6 832
+#define poly_coeff5 896
+#define poly_coeff4 960
+#define poly_coeff3 1024
+#define poly_coeff2 1088
+#define One 1152
#include <sysdep.h>
- .text
- .section .text.evex512,"ax",@progbits
+ .section .text.evex512, "ax", @progbits
ENTRY(_ZGVeN8v_expm1_skx)
- pushq %rbp
- cfi_def_cfa_offset(16)
- movq %rsp, %rbp
- cfi_def_cfa(6, 16)
- cfi_offset(6, -16)
- andq $-64, %rsp
- subq $192, %rsp
- vmovups L2E+__svml_dexpm1_data_internal_avx512(%rip), %zmm6
- vmovups Shifter+__svml_dexpm1_data_internal_avx512(%rip), %zmm4
- vmovups L2H+__svml_dexpm1_data_internal_avx512(%rip), %zmm11
- vmovups L2L+__svml_dexpm1_data_internal_avx512(%rip), %zmm5
- vmovups Threshold+__svml_dexpm1_data_internal_avx512(%rip), %zmm3
- vmovups poly_coeff5+__svml_dexpm1_data_internal_avx512(%rip), %zmm13
- vmovups poly_coeff4+__svml_dexpm1_data_internal_avx512(%rip), %zmm15
-
-/* polynomial */
- vmovups poly_coeff7+__svml_dexpm1_data_internal_avx512(%rip), %zmm12
-
-/* set Z0=max(Z0, -128.0) */
- vmovups ZThres+__svml_dexpm1_data_internal_avx512(%rip), %zmm8
- vmovups poly_coeff3+__svml_dexpm1_data_internal_avx512(%rip), %zmm14
- vmovups __svml_dexpm1_data_internal_avx512(%rip), %zmm9
- vmovaps %zmm0, %zmm2
-
-/* 2^(52-4)*1.5 + x * log2(e) */
- vfmadd213pd {rn-sae}, %zmm4, %zmm2, %zmm6
- vmovups Exp_tbl_L+__svml_dexpm1_data_internal_avx512(%rip), %zmm0
- vcmppd $21, {sae}, %zmm3, %zmm2, %k0
-
-/* Z0 ~ x*log2(e), rounded to 4 fractional bits */
- vsubpd {rn-sae}, %zmm4, %zmm6, %zmm7
- vpermt2pd Exp_tbl_H+64+__svml_dexpm1_data_internal_avx512(%rip), %zmm6, %zmm9
- vpermt2pd Exp_tbl_L+64+__svml_dexpm1_data_internal_avx512(%rip), %zmm6, %zmm0
- vandpd SgnMask+__svml_dexpm1_data_internal_avx512(%rip), %zmm2, %zmm1
-
-/* R = x - Z0*log(2) */
- vfnmadd213pd {rn-sae}, %zmm2, %zmm7, %zmm11
- vmaxpd {sae}, %zmm8, %zmm7, %zmm10
- vfnmadd231pd {rn-sae}, %zmm7, %zmm5, %zmm11
- kmovw %k0, %edx
-
-/* ensure |R|<2 even for special cases */
- vandpd EMask+__svml_dexpm1_data_internal_avx512(%rip), %zmm11, %zmm3
- vmovups poly_coeff6+__svml_dexpm1_data_internal_avx512(%rip), %zmm11
-
-/* scale Th */
- vscalefpd {rn-sae}, %zmm10, %zmm9, %zmm4
- vfmadd231pd {rn-sae}, %zmm3, %zmm13, %zmm15
- vfmadd231pd {rn-sae}, %zmm3, %zmm12, %zmm11
- vmovups poly_coeff2+__svml_dexpm1_data_internal_avx512(%rip), %zmm12
- vmulpd {rn-sae}, %zmm3, %zmm3, %zmm13
- vfmadd231pd {rn-sae}, %zmm3, %zmm14, %zmm12
- vfmadd213pd {rn-sae}, %zmm15, %zmm13, %zmm11
- vfmadd213pd {rn-sae}, %zmm12, %zmm13, %zmm11
-
-/* Tlr + R+ R*Poly */
- vfmadd213pd {rn-sae}, %zmm0, %zmm13, %zmm11
-
-/* Th - 1 */
- vmovups One+__svml_dexpm1_data_internal_avx512(%rip), %zmm0
- vaddpd {rn-sae}, %zmm3, %zmm11, %zmm14
- vsubpd {rn-sae}, %zmm0, %zmm4, %zmm15
-
-/* (Th-1)+Th*(Tlr + R+ R*Poly) */
- vfmadd213pd {rn-sae}, %zmm15, %zmm14, %zmm4
- vorpd %zmm1, %zmm4, %zmm0
- testl %edx, %edx
-
-/* Go to special inputs processing branch */
- jne L(SPECIAL_VALUES_BRANCH)
- # LOE rbx r12 r13 r14 r15 edx zmm0 zmm2
-
-/* 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 $-64, %rsp
+ subq $192, %rsp
+ vmovups L2E+__svml_dexpm1_data_internal_avx512(%rip), %zmm6
+ vmovups Shifter+__svml_dexpm1_data_internal_avx512(%rip), %zmm4
+ vmovups L2H+__svml_dexpm1_data_internal_avx512(%rip), %zmm11
+ vmovups L2L+__svml_dexpm1_data_internal_avx512(%rip), %zmm5
+ vmovups Threshold+__svml_dexpm1_data_internal_avx512(%rip), %zmm3
+ vmovups poly_coeff5+__svml_dexpm1_data_internal_avx512(%rip), %zmm13
+ vmovups poly_coeff4+__svml_dexpm1_data_internal_avx512(%rip), %zmm15
+
+ /* polynomial */
+ vmovups poly_coeff7+__svml_dexpm1_data_internal_avx512(%rip), %zmm12
+
+ /* set Z0=max(Z0, -128.0) */
+ vmovups ZThres+__svml_dexpm1_data_internal_avx512(%rip), %zmm8
+ vmovups poly_coeff3+__svml_dexpm1_data_internal_avx512(%rip), %zmm14
+ vmovups __svml_dexpm1_data_internal_avx512(%rip), %zmm9
+ vmovaps %zmm0, %zmm2
+
+ /* 2^(52-4)*1.5 + x * log2(e) */
+ vfmadd213pd {rn-sae}, %zmm4, %zmm2, %zmm6
+ vmovups Exp_tbl_L+__svml_dexpm1_data_internal_avx512(%rip), %zmm0
+ vcmppd $21, {sae}, %zmm3, %zmm2, %k0
+
+ /* Z0 ~ x*log2(e), rounded to 4 fractional bits */
+ vsubpd {rn-sae}, %zmm4, %zmm6, %zmm7
+ vpermt2pd Exp_tbl_H+64+__svml_dexpm1_data_internal_avx512(%rip), %zmm6, %zmm9
+ vpermt2pd Exp_tbl_L+64+__svml_dexpm1_data_internal_avx512(%rip), %zmm6, %zmm0
+ vandpd SgnMask+__svml_dexpm1_data_internal_avx512(%rip), %zmm2, %zmm1
+
+ /* R = x - Z0*log(2) */
+ vfnmadd213pd {rn-sae}, %zmm2, %zmm7, %zmm11
+ vmaxpd {sae}, %zmm8, %zmm7, %zmm10
+ vfnmadd231pd {rn-sae}, %zmm7, %zmm5, %zmm11
+ kmovw %k0, %edx
+
+ /* ensure |R|<2 even for special cases */
+ vandpd EMask+__svml_dexpm1_data_internal_avx512(%rip), %zmm11, %zmm3
+ vmovups poly_coeff6+__svml_dexpm1_data_internal_avx512(%rip), %zmm11
+
+ /* scale Th */
+ vscalefpd {rn-sae}, %zmm10, %zmm9, %zmm4
+ vfmadd231pd {rn-sae}, %zmm3, %zmm13, %zmm15
+ vfmadd231pd {rn-sae}, %zmm3, %zmm12, %zmm11
+ vmovups poly_coeff2+__svml_dexpm1_data_internal_avx512(%rip), %zmm12
+ vmulpd {rn-sae}, %zmm3, %zmm3, %zmm13
+ vfmadd231pd {rn-sae}, %zmm3, %zmm14, %zmm12
+ vfmadd213pd {rn-sae}, %zmm15, %zmm13, %zmm11
+ vfmadd213pd {rn-sae}, %zmm12, %zmm13, %zmm11
+
+ /* Tlr + R+ R*Poly */
+ vfmadd213pd {rn-sae}, %zmm0, %zmm13, %zmm11
+
+ /* Th - 1 */
+ vmovups One+__svml_dexpm1_data_internal_avx512(%rip), %zmm0
+ vaddpd {rn-sae}, %zmm3, %zmm11, %zmm14
+ vsubpd {rn-sae}, %zmm0, %zmm4, %zmm15
+
+ /* (Th-1)+Th*(Tlr + R+ R*Poly) */
+ vfmadd213pd {rn-sae}, %zmm15, %zmm14, %zmm4
+ vorpd %zmm1, %zmm4, %zmm0
+ testl %edx, %edx
+
+ /* Go to special inputs processing branch */
+ jne L(SPECIAL_VALUES_BRANCH)
+ # LOE rbx r12 r13 r14 r15 edx zmm0 zmm2
+
+ /* 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):
- vmovups %zmm2, 64(%rsp)
- vmovups %zmm0, 128(%rsp)
- # LOE rbx r12 r13 r14 r15 edx zmm0
-
- 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: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */
- .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 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: -64; DW_OP_and; DW_OP_const4s: -184; DW_OP_plus) */
- .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x48, 0xff, 0xff, 0xff, 0x22
- movl %edx, %r13d
- movq %r14, (%rsp)
- /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -192; DW_OP_plus) */
- .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22
- # LOE rbx r15 r12d r13d
-
-/* Range mask
- * bits check
- */
+ vmovups %zmm2, 64(%rsp)
+ vmovups %zmm0, 128(%rsp)
+ # LOE rbx r12 r13 r14 r15 edx zmm0
+
+ 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: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 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: -64; DW_OP_and; DW_OP_const4s: -184; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x48, 0xff, 0xff, 0xff, 0x22
+ movl %edx, %r13d
+ movq %r14, (%rsp)
+ /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -192; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 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 $8, %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)
- vmovups 128(%rsp), %zmm0
-
-/* Go to exit */
- jmp L(EXIT)
- /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */
- .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 0xff, 0xff, 0xff, 0x22
- /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -184; DW_OP_plus) */
- .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x48, 0xff, 0xff, 0xff, 0x22
- /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -192; DW_OP_plus) */
- .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22
- # LOE rbx r12 r13 r14 r15 zmm0
-
-/* Scalar math fucntion call
- * to process special input
- */
+ incl %r12d
+ cmpl $8, %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)
+ vmovups 128(%rsp), %zmm0
+
+ /* Go to exit */
+ jmp L(EXIT)
+ /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 0xff, 0xff, 0xff, 0x22
+ /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -184; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x48, 0xff, 0xff, 0xff, 0x22
+ /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -192; DW_OP_plus) */
+ .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22
+ # LOE rbx r12 r13 r14 r15 zmm0
+
+ /* Scalar math fucntion call
+ * to process special input
+ */
L(SCALAR_MATH_CALL):
- movl %r12d, %r14d
- movsd 64(%rsp,%r14,8), %xmm0
- call expm1@PLT
- # LOE rbx r14 r15 r12d r13d xmm0
+ movl %r12d, %r14d
+ movsd 64(%rsp, %r14, 8), %xmm0
+ call expm1@PLT
+ # LOE rbx r14 r15 r12d r13d xmm0
- movsd %xmm0, 128(%rsp,%r14,8)
+ movsd %xmm0, 128(%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(_ZGVeN8v_expm1_skx)
- .section .rodata, "a"
- .align 64
+ .section .rodata, "a"
+ .align 64
#ifdef __svml_dexpm1_data_internal_avx512_typedef
typedef unsigned int VUINT32;
typedef struct {
- __declspec(align(64)) VUINT32 Exp_tbl_H[16][2];
- __declspec(align(64)) VUINT32 Exp_tbl_L[16][2];
- __declspec(align(64)) VUINT32 L2E[8][2];
- __declspec(align(64)) VUINT32 Shifter[8][2];
- __declspec(align(64)) VUINT32 Threshold[8][2];
- __declspec(align(64)) VUINT32 SgnMask[8][2];
- __declspec(align(64)) VUINT32 L2H[8][2];
- __declspec(align(64)) VUINT32 L2L[8][2];
- __declspec(align(64)) VUINT32 ZThres[8][2];
- __declspec(align(64)) VUINT32 EMask[8][2];
- __declspec(align(64)) VUINT32 poly_coeff7[8][2];
- __declspec(align(64)) VUINT32 poly_coeff6[8][2];
- __declspec(align(64)) VUINT32 poly_coeff5[8][2];
- __declspec(align(64)) VUINT32 poly_coeff4[8][2];
- __declspec(align(64)) VUINT32 poly_coeff3[8][2];
- __declspec(align(64)) VUINT32 poly_coeff2[8][2];
- __declspec(align(64)) VUINT32 One[8][2];
- } __svml_dexpm1_data_internal_avx512;
+ __declspec(align(64)) VUINT32 Exp_tbl_H[16][2];
+ __declspec(align(64)) VUINT32 Exp_tbl_L[16][2];
+ __declspec(align(64)) VUINT32 L2E[8][2];
+ __declspec(align(64)) VUINT32 Shifter[8][2];
+ __declspec(align(64)) VUINT32 Threshold[8][2];
+ __declspec(align(64)) VUINT32 SgnMask[8][2];
+ __declspec(align(64)) VUINT32 L2H[8][2];
+ __declspec(align(64)) VUINT32 L2L[8][2];
+ __declspec(align(64)) VUINT32 ZThres[8][2];
+ __declspec(align(64)) VUINT32 EMask[8][2];
+ __declspec(align(64)) VUINT32 poly_coeff7[8][2];
+ __declspec(align(64)) VUINT32 poly_coeff6[8][2];
+ __declspec(align(64)) VUINT32 poly_coeff5[8][2];
+ __declspec(align(64)) VUINT32 poly_coeff4[8][2];
+ __declspec(align(64)) VUINT32 poly_coeff3[8][2];
+ __declspec(align(64)) VUINT32 poly_coeff2[8][2];
+ __declspec(align(64)) VUINT32 One[8][2];
+} __svml_dexpm1_data_internal_avx512;
#endif
__svml_dexpm1_data_internal_avx512:
- /*== Exp_tbl_H ==*/
- .quad 0x3ff0000000000000
- .quad 0x3ff0b5586cf9890f
- .quad 0x3ff172b83c7d517b
- .quad 0x3ff2387a6e756238
- .quad 0x3ff306fe0a31b715
- .quad 0x3ff3dea64c123422
- .quad 0x3ff4bfdad5362a27
- .quad 0x3ff5ab07dd485429
- .quad 0x3ff6a09e667f3bcd
- .quad 0x3ff7a11473eb0187
- .quad 0x3ff8ace5422aa0db
- .quad 0x3ff9c49182a3f090
- .quad 0x3ffae89f995ad3ad
- .quad 0x3ffc199bdd85529c
- .quad 0x3ffd5818dcfba487
- .quad 0x3ffea4afa2a490da
- /*== Exp_tbl_L ==*/
- .align 64
- .quad 0x0000000000000000
- .quad 0x3c979aa65d837b6d
- .quad 0xbc801b15eaa59348
- .quad 0x3c968efde3a8a894
- .quad 0x3c834d754db0abb6
- .quad 0x3c859f48a72a4c6d
- .quad 0x3c7690cebb7aafb0
- .quad 0x3c9063e1e21c5409
- .quad 0xbc93b3efbf5e2228
- .quad 0xbc7b32dcb94da51d
- .quad 0x3c8db72fc1f0eab4
- .quad 0x3c71affc2b91ce27
- .quad 0x3c8c1a7792cb3387
- .quad 0x3c736eae30af0cb3
- .quad 0x3c74a385a63d07a7
- .quad 0xbc8ff7128fd391f0
- /*== log2(e) ==*/
- .align 64
- .quad 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE
- /*== Shifter=2^(52-4)*1.5 ==*/
- .align 64
- .quad 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0
- /*== Threshold ==*/
- .align 64
- .quad 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44
- /*== Sgn ==*/
- .align 64
- .quad 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000
- /*== L2H = log(2)_high ==*/
- .align 64
- .quad 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef
- /*== L2L = log(2)_low ==*/
- .align 64
- .quad 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f
- /*== ZThres ==*/
- .align 64
- .quad 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000
- /*== EMask ==*/
- .align 64
- .quad 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff
- /*== poly_coeff7 ==*/
- .align 64
- .quad 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a
- /*== poly_coeff6 ==*/
- .align 64
- .quad 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f
- /*== poly_coeff5 ==*/
- .align 64
- .quad 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214
- /*== poly_coeff4 ==*/
- .align 64
- .quad 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06
- /*== poly_coeff3 ==*/
- .align 64
- .quad 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656
- /*== poly_coeff2 ==*/
- .align 64
- .quad 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2
- /*== One ==*/
- .align 64
- .quad 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000
- .align 64
- .type __svml_dexpm1_data_internal_avx512,@object
- .size __svml_dexpm1_data_internal_avx512,.-__svml_dexpm1_data_internal_avx512
+ /* Exp_tbl_H */
+ .quad 0x3ff0000000000000
+ .quad 0x3ff0b5586cf9890f
+ .quad 0x3ff172b83c7d517b
+ .quad 0x3ff2387a6e756238
+ .quad 0x3ff306fe0a31b715
+ .quad 0x3ff3dea64c123422
+ .quad 0x3ff4bfdad5362a27
+ .quad 0x3ff5ab07dd485429
+ .quad 0x3ff6a09e667f3bcd
+ .quad 0x3ff7a11473eb0187
+ .quad 0x3ff8ace5422aa0db
+ .quad 0x3ff9c49182a3f090
+ .quad 0x3ffae89f995ad3ad
+ .quad 0x3ffc199bdd85529c
+ .quad 0x3ffd5818dcfba487
+ .quad 0x3ffea4afa2a490da
+ /* Exp_tbl_L */
+ .align 64
+ .quad 0x0000000000000000
+ .quad 0x3c979aa65d837b6d
+ .quad 0xbc801b15eaa59348
+ .quad 0x3c968efde3a8a894
+ .quad 0x3c834d754db0abb6
+ .quad 0x3c859f48a72a4c6d
+ .quad 0x3c7690cebb7aafb0
+ .quad 0x3c9063e1e21c5409
+ .quad 0xbc93b3efbf5e2228
+ .quad 0xbc7b32dcb94da51d
+ .quad 0x3c8db72fc1f0eab4
+ .quad 0x3c71affc2b91ce27
+ .quad 0x3c8c1a7792cb3387
+ .quad 0x3c736eae30af0cb3
+ .quad 0x3c74a385a63d07a7
+ .quad 0xbc8ff7128fd391f0
+ /* log2(e) */
+ .align 64
+ .quad 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE, 0x3ff71547652B82FE
+ /* Shifter=2^(52-4)*1.5 */
+ .align 64
+ .quad 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0, 0x42f8000000003ff0
+ /* Threshold */
+ .align 64
+ .quad 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44, 0x40861DA04CBAFE44
+ /* Sgn */
+ .align 64
+ .quad 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000
+ /* L2H = log(2)_high */
+ .align 64
+ .quad 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef, 0x3fe62e42fefa39ef
+ /* L2L = log(2)_low */
+ .align 64
+ .quad 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f, 0x3c7abc9e3b39803f
+ /* ZThres */
+ .align 64
+ .quad 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000, 0xc060000000000000
+ /* EMask */
+ .align 64
+ .quad 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff, 0xbfffffffffffffff
+ /* poly_coeff7 */
+ .align 64
+ .quad 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a, 0x3f2a020410303d8a
+ /* poly_coeff6 */
+ .align 64
+ .quad 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f, 0x3f56c1c38e164a2f
+ /* poly_coeff5 */
+ .align 64
+ .quad 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214, 0x3f81111110865214
+ /* poly_coeff4 */
+ .align 64
+ .quad 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06, 0x3fa5555554ad3d06
+ /* poly_coeff3 */
+ .align 64
+ .quad 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656, 0x3fc5555555555656
+ /* poly_coeff2 */
+ .align 64
+ .quad 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2, 0x3fe00000000000a2
+ /* One */
+ .align 64
+ .quad 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000
+ .align 64
+ .type __svml_dexpm1_data_internal_avx512, @object
+ .size __svml_dexpm1_data_internal_avx512, .-__svml_dexpm1_data_internal_avx512