From 994266f5019560f26e8d07be7fdf8621903339a1 Mon Sep 17 00:00:00 2001 From: Sunil K Pandey Date: Mon, 7 Mar 2022 10:47:10 -0800 Subject: [PATCH] x86_64: Fix svml_d_atanh8_core_avx512.S code formatting This commit contains following formatting changes 1. Instructions proceeded by a tab. 2. Instruction less than 8 characters in length have a tab between it and the first operand. 3. Instruction greater than 7 characters in length have a space between it and the first operand. 4. Tabs after `#define`d names and their value. 5. 8 space at the beginning of line replaced by tab. 6. Indent comments with code. 7. Remove redundent .text section. 8. 1 space between line content and line comment. 9. Space after all commas. Reviewed-by: Noah Goldstein --- .../fpu/multiarch/svml_d_atanh8_core_avx512.S | 679 ++++++++++----------- 1 file changed, 339 insertions(+), 340 deletions(-) diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_atanh8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_atanh8_core_avx512.S index 797d9d4..3193c02 100644 --- a/sysdeps/x86_64/fpu/multiarch/svml_d_atanh8_core_avx512.S +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_atanh8_core_avx512.S @@ -33,369 +33,368 @@ /* Offsets for data table __svml_datanh_data_internal_avx512 */ -#define Log_tbl_H 0 -#define Log_tbl_L 128 -#define One 256 -#define AbsMask 320 -#define AddB5 384 -#define RcpBitMask 448 -#define poly_coeff8 512 -#define poly_coeff7 576 -#define poly_coeff6 640 -#define poly_coeff5 704 -#define poly_coeff4 768 -#define poly_coeff3 832 -#define poly_coeff2 896 -#define poly_coeff1 960 -#define poly_coeff0 1024 -#define Half 1088 -#define L2H 1152 -#define L2L 1216 +#define Log_tbl_H 0 +#define Log_tbl_L 128 +#define One 256 +#define AbsMask 320 +#define AddB5 384 +#define RcpBitMask 448 +#define poly_coeff8 512 +#define poly_coeff7 576 +#define poly_coeff6 640 +#define poly_coeff5 704 +#define poly_coeff4 768 +#define poly_coeff3 832 +#define poly_coeff2 896 +#define poly_coeff1 960 +#define poly_coeff0 1024 +#define Half 1088 +#define L2H 1152 +#define L2L 1216 #include - .text - .section .text.evex512,"ax",@progbits + .section .text.evex512, "ax", @progbits ENTRY(_ZGVeN8v_atanh_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 One+__svml_datanh_data_internal_avx512(%rip), %zmm15 - -/* round reciprocals to 1+4b mantissas */ - vmovups AddB5+__svml_datanh_data_internal_avx512(%rip), %zmm6 - vmovups RcpBitMask+__svml_datanh_data_internal_avx512(%rip), %zmm9 - vmovaps %zmm0, %zmm2 - vandpd AbsMask+__svml_datanh_data_internal_avx512(%rip), %zmm2, %zmm13 - -/* 1+y */ - vaddpd {rn-sae}, %zmm15, %zmm13, %zmm0 - -/* 1-y */ - vsubpd {rn-sae}, %zmm13, %zmm15, %zmm4 - vxorpd %zmm13, %zmm2, %zmm1 - -/* Yp_high */ - vsubpd {rn-sae}, %zmm15, %zmm0, %zmm7 - -/* -Ym_high */ - vsubpd {rn-sae}, %zmm15, %zmm4, %zmm12 - -/* RcpP ~ 1/Yp */ - vrcp14pd %zmm0, %zmm3 - -/* RcpM ~ 1/Ym */ - vrcp14pd %zmm4, %zmm5 - -/* input outside (-1, 1) ? */ - vcmppd $21, {sae}, %zmm15, %zmm13, %k0 - vpaddq %zmm6, %zmm3, %zmm11 - vpaddq %zmm6, %zmm5, %zmm10 - -/* Yp_low */ - vsubpd {rn-sae}, %zmm7, %zmm13, %zmm8 - vandpd %zmm9, %zmm11, %zmm14 - vandpd %zmm9, %zmm10, %zmm3 - -/* Ym_low */ - vaddpd {rn-sae}, %zmm12, %zmm13, %zmm12 - -/* Reduced argument: Rp = (RcpP*Yp - 1)+RcpP*Yp_low */ - vfmsub213pd {rn-sae}, %zmm15, %zmm14, %zmm0 - -/* Reduced argument: Rm = (RcpM*Ym - 1)+RcpM*Ym_low */ - vfmsub231pd {rn-sae}, %zmm3, %zmm4, %zmm15 - -/* exponents */ - vgetexppd {sae}, %zmm14, %zmm5 - vgetexppd {sae}, %zmm3, %zmm4 - -/* Table lookups */ - vmovups __svml_datanh_data_internal_avx512(%rip), %zmm9 - vmovups Log_tbl_H+64+__svml_datanh_data_internal_avx512(%rip), %zmm13 - vmovups Log_tbl_L+__svml_datanh_data_internal_avx512(%rip), %zmm7 - vfmadd231pd {rn-sae}, %zmm14, %zmm8, %zmm0 - vfnmadd231pd {rn-sae}, %zmm3, %zmm12, %zmm15 - -/* Prepare table index */ - vpsrlq $48, %zmm14, %zmm11 - vpsrlq $48, %zmm3, %zmm8 - vmovups Log_tbl_L+64+__svml_datanh_data_internal_avx512(%rip), %zmm14 - -/* polynomials */ - vmovups poly_coeff8+__svml_datanh_data_internal_avx512(%rip), %zmm3 - -/* Km-Kp */ - vsubpd {rn-sae}, %zmm5, %zmm4, %zmm5 - vmovups poly_coeff7+__svml_datanh_data_internal_avx512(%rip), %zmm4 - kmovw %k0, %edx - vmovaps %zmm11, %zmm10 - vmovaps %zmm4, %zmm6 - vpermi2pd %zmm13, %zmm9, %zmm10 - vpermi2pd %zmm14, %zmm7, %zmm11 - vpermt2pd %zmm13, %zmm8, %zmm9 - vpermt2pd %zmm14, %zmm8, %zmm7 - vmovups poly_coeff6+__svml_datanh_data_internal_avx512(%rip), %zmm8 - vfmadd231pd {rn-sae}, %zmm0, %zmm3, %zmm6 - vfmadd231pd {rn-sae}, %zmm15, %zmm3, %zmm4 - vmovups poly_coeff3+__svml_datanh_data_internal_avx512(%rip), %zmm13 - vmovups poly_coeff2+__svml_datanh_data_internal_avx512(%rip), %zmm14 - vfmadd213pd {rn-sae}, %zmm8, %zmm0, %zmm6 - vfmadd213pd {rn-sae}, %zmm8, %zmm15, %zmm4 - vmovups poly_coeff0+__svml_datanh_data_internal_avx512(%rip), %zmm8 - vsubpd {rn-sae}, %zmm11, %zmm7, %zmm12 - -/* table values */ - vsubpd {rn-sae}, %zmm10, %zmm9, %zmm3 - vmovups poly_coeff5+__svml_datanh_data_internal_avx512(%rip), %zmm7 - vmovups poly_coeff4+__svml_datanh_data_internal_avx512(%rip), %zmm9 - -/* K*L2H + Th */ - vmovups L2H+__svml_datanh_data_internal_avx512(%rip), %zmm10 - -/* K*L2L + Tl */ - vmovups L2L+__svml_datanh_data_internal_avx512(%rip), %zmm11 - vfmadd213pd {rn-sae}, %zmm7, %zmm0, %zmm6 - vfmadd213pd {rn-sae}, %zmm7, %zmm15, %zmm4 - vmovups poly_coeff1+__svml_datanh_data_internal_avx512(%rip), %zmm7 - vfmadd231pd {rn-sae}, %zmm5, %zmm10, %zmm3 - vfmadd213pd {rn-sae}, %zmm12, %zmm11, %zmm5 - vfmadd213pd {rn-sae}, %zmm9, %zmm0, %zmm6 - vfmadd213pd {rn-sae}, %zmm9, %zmm15, %zmm4 - vfmadd213pd {rn-sae}, %zmm13, %zmm0, %zmm6 - vfmadd213pd {rn-sae}, %zmm13, %zmm15, %zmm4 - vfmadd213pd {rn-sae}, %zmm14, %zmm0, %zmm6 - vfmadd213pd {rn-sae}, %zmm14, %zmm15, %zmm4 - vfmadd213pd {rn-sae}, %zmm7, %zmm0, %zmm6 - vfmadd213pd {rn-sae}, %zmm7, %zmm15, %zmm4 - vfmadd213pd {rn-sae}, %zmm8, %zmm0, %zmm6 - vfmadd213pd {rn-sae}, %zmm8, %zmm15, %zmm4 - -/* (K*L2L + Tl) + Rp*PolyP */ - vfmadd213pd {rn-sae}, %zmm5, %zmm0, %zmm6 - vorpd Half+__svml_datanh_data_internal_avx512(%rip), %zmm1, %zmm0 - -/* (K*L2L + Tl) + Rp*PolyP -Rm*PolyM */ - vfnmadd213pd {rn-sae}, %zmm6, %zmm15, %zmm4 - vaddpd {rn-sae}, %zmm4, %zmm3, %zmm1 - vmulpd {rn-sae}, %zmm0, %zmm1, %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 One+__svml_datanh_data_internal_avx512(%rip), %zmm15 + + /* round reciprocals to 1+4b mantissas */ + vmovups AddB5+__svml_datanh_data_internal_avx512(%rip), %zmm6 + vmovups RcpBitMask+__svml_datanh_data_internal_avx512(%rip), %zmm9 + vmovaps %zmm0, %zmm2 + vandpd AbsMask+__svml_datanh_data_internal_avx512(%rip), %zmm2, %zmm13 + + /* 1+y */ + vaddpd {rn-sae}, %zmm15, %zmm13, %zmm0 + + /* 1-y */ + vsubpd {rn-sae}, %zmm13, %zmm15, %zmm4 + vxorpd %zmm13, %zmm2, %zmm1 + + /* Yp_high */ + vsubpd {rn-sae}, %zmm15, %zmm0, %zmm7 + + /* -Ym_high */ + vsubpd {rn-sae}, %zmm15, %zmm4, %zmm12 + + /* RcpP ~ 1/Yp */ + vrcp14pd %zmm0, %zmm3 + + /* RcpM ~ 1/Ym */ + vrcp14pd %zmm4, %zmm5 + + /* input outside (-1, 1) ? */ + vcmppd $21, {sae}, %zmm15, %zmm13, %k0 + vpaddq %zmm6, %zmm3, %zmm11 + vpaddq %zmm6, %zmm5, %zmm10 + + /* Yp_low */ + vsubpd {rn-sae}, %zmm7, %zmm13, %zmm8 + vandpd %zmm9, %zmm11, %zmm14 + vandpd %zmm9, %zmm10, %zmm3 + + /* Ym_low */ + vaddpd {rn-sae}, %zmm12, %zmm13, %zmm12 + + /* Reduced argument: Rp = (RcpP*Yp - 1)+RcpP*Yp_low */ + vfmsub213pd {rn-sae}, %zmm15, %zmm14, %zmm0 + + /* Reduced argument: Rm = (RcpM*Ym - 1)+RcpM*Ym_low */ + vfmsub231pd {rn-sae}, %zmm3, %zmm4, %zmm15 + + /* exponents */ + vgetexppd {sae}, %zmm14, %zmm5 + vgetexppd {sae}, %zmm3, %zmm4 + + /* Table lookups */ + vmovups __svml_datanh_data_internal_avx512(%rip), %zmm9 + vmovups Log_tbl_H+64+__svml_datanh_data_internal_avx512(%rip), %zmm13 + vmovups Log_tbl_L+__svml_datanh_data_internal_avx512(%rip), %zmm7 + vfmadd231pd {rn-sae}, %zmm14, %zmm8, %zmm0 + vfnmadd231pd {rn-sae}, %zmm3, %zmm12, %zmm15 + + /* Prepare table index */ + vpsrlq $48, %zmm14, %zmm11 + vpsrlq $48, %zmm3, %zmm8 + vmovups Log_tbl_L+64+__svml_datanh_data_internal_avx512(%rip), %zmm14 + + /* polynomials */ + vmovups poly_coeff8+__svml_datanh_data_internal_avx512(%rip), %zmm3 + + /* Km-Kp */ + vsubpd {rn-sae}, %zmm5, %zmm4, %zmm5 + vmovups poly_coeff7+__svml_datanh_data_internal_avx512(%rip), %zmm4 + kmovw %k0, %edx + vmovaps %zmm11, %zmm10 + vmovaps %zmm4, %zmm6 + vpermi2pd %zmm13, %zmm9, %zmm10 + vpermi2pd %zmm14, %zmm7, %zmm11 + vpermt2pd %zmm13, %zmm8, %zmm9 + vpermt2pd %zmm14, %zmm8, %zmm7 + vmovups poly_coeff6+__svml_datanh_data_internal_avx512(%rip), %zmm8 + vfmadd231pd {rn-sae}, %zmm0, %zmm3, %zmm6 + vfmadd231pd {rn-sae}, %zmm15, %zmm3, %zmm4 + vmovups poly_coeff3+__svml_datanh_data_internal_avx512(%rip), %zmm13 + vmovups poly_coeff2+__svml_datanh_data_internal_avx512(%rip), %zmm14 + vfmadd213pd {rn-sae}, %zmm8, %zmm0, %zmm6 + vfmadd213pd {rn-sae}, %zmm8, %zmm15, %zmm4 + vmovups poly_coeff0+__svml_datanh_data_internal_avx512(%rip), %zmm8 + vsubpd {rn-sae}, %zmm11, %zmm7, %zmm12 + + /* table values */ + vsubpd {rn-sae}, %zmm10, %zmm9, %zmm3 + vmovups poly_coeff5+__svml_datanh_data_internal_avx512(%rip), %zmm7 + vmovups poly_coeff4+__svml_datanh_data_internal_avx512(%rip), %zmm9 + + /* K*L2H + Th */ + vmovups L2H+__svml_datanh_data_internal_avx512(%rip), %zmm10 + + /* K*L2L + Tl */ + vmovups L2L+__svml_datanh_data_internal_avx512(%rip), %zmm11 + vfmadd213pd {rn-sae}, %zmm7, %zmm0, %zmm6 + vfmadd213pd {rn-sae}, %zmm7, %zmm15, %zmm4 + vmovups poly_coeff1+__svml_datanh_data_internal_avx512(%rip), %zmm7 + vfmadd231pd {rn-sae}, %zmm5, %zmm10, %zmm3 + vfmadd213pd {rn-sae}, %zmm12, %zmm11, %zmm5 + vfmadd213pd {rn-sae}, %zmm9, %zmm0, %zmm6 + vfmadd213pd {rn-sae}, %zmm9, %zmm15, %zmm4 + vfmadd213pd {rn-sae}, %zmm13, %zmm0, %zmm6 + vfmadd213pd {rn-sae}, %zmm13, %zmm15, %zmm4 + vfmadd213pd {rn-sae}, %zmm14, %zmm0, %zmm6 + vfmadd213pd {rn-sae}, %zmm14, %zmm15, %zmm4 + vfmadd213pd {rn-sae}, %zmm7, %zmm0, %zmm6 + vfmadd213pd {rn-sae}, %zmm7, %zmm15, %zmm4 + vfmadd213pd {rn-sae}, %zmm8, %zmm0, %zmm6 + vfmadd213pd {rn-sae}, %zmm8, %zmm15, %zmm4 + + /* (K*L2L + Tl) + Rp*PolyP */ + vfmadd213pd {rn-sae}, %zmm5, %zmm0, %zmm6 + vorpd Half+__svml_datanh_data_internal_avx512(%rip), %zmm1, %zmm0 + + /* (K*L2L + Tl) + Rp*PolyP -Rm*PolyM */ + vfnmadd213pd {rn-sae}, %zmm6, %zmm15, %zmm4 + vaddpd {rn-sae}, %zmm4, %zmm3, %zmm1 + vmulpd {rn-sae}, %zmm0, %zmm1, %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 atanh@PLT - # LOE rbx r14 r15 r12d r13d xmm0 + movl %r12d, %r14d + movsd 64(%rsp, %r14, 8), %xmm0 + call atanh@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_atanh_skx) - .section .rodata, "a" - .align 64 + .section .rodata, "a" + .align 64 #ifdef __svml_datanh_data_internal_avx512_typedef typedef unsigned int VUINT32; typedef struct { - __declspec(align(64)) VUINT32 Log_tbl_H[16][2]; - __declspec(align(64)) VUINT32 Log_tbl_L[16][2]; - __declspec(align(64)) VUINT32 One[8][2]; - __declspec(align(64)) VUINT32 AbsMask[8][2]; - __declspec(align(64)) VUINT32 AddB5[8][2]; - __declspec(align(64)) VUINT32 RcpBitMask[8][2]; - __declspec(align(64)) VUINT32 poly_coeff8[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 poly_coeff1[8][2]; - __declspec(align(64)) VUINT32 poly_coeff0[8][2]; - __declspec(align(64)) VUINT32 Half[8][2]; - __declspec(align(64)) VUINT32 L2H[8][2]; - __declspec(align(64)) VUINT32 L2L[8][2]; - } __svml_datanh_data_internal_avx512; + __declspec(align(64)) VUINT32 Log_tbl_H[16][2]; + __declspec(align(64)) VUINT32 Log_tbl_L[16][2]; + __declspec(align(64)) VUINT32 One[8][2]; + __declspec(align(64)) VUINT32 AbsMask[8][2]; + __declspec(align(64)) VUINT32 AddB5[8][2]; + __declspec(align(64)) VUINT32 RcpBitMask[8][2]; + __declspec(align(64)) VUINT32 poly_coeff8[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 poly_coeff1[8][2]; + __declspec(align(64)) VUINT32 poly_coeff0[8][2]; + __declspec(align(64)) VUINT32 Half[8][2]; + __declspec(align(64)) VUINT32 L2H[8][2]; + __declspec(align(64)) VUINT32 L2L[8][2]; +} __svml_datanh_data_internal_avx512; #endif __svml_datanh_data_internal_avx512: - /*== Log_tbl_H ==*/ - .quad 0x0000000000000000 - .quad 0x3faf0a30c0100000 - .quad 0x3fbe27076e2a0000 - .quad 0x3fc5ff3070a80000 - .quad 0x3fcc8ff7c79b0000 - .quad 0x3fd1675cabab8000 - .quad 0x3fd4618bc21c8000 - .quad 0x3fd739d7f6bc0000 - .quad 0x3fd9f323ecbf8000 - .quad 0x3fdc8ff7c79a8000 - .quad 0x3fdf128f5faf0000 - .quad 0x3fe0be72e4254000 - .quad 0x3fe1e85f5e704000 - .quad 0x3fe307d7334f0000 - .quad 0x3fe41d8fe8468000 - .quad 0x3fe52a2d265bc000 - /*== Log_tbl_L ==*/ - .align 64 - .quad 0x0000000000000000 - .quad 0x3d662a6617cc9717 - .quad 0x3d6e5cbd3d50fffc - .quad 0xbd6b0b0de3077d7e - .quad 0xbd697794f689f843 - .quad 0x3d630701ce63eab9 - .quad 0xbd609ec17a426426 - .quad 0xbd67fcb18ed9d603 - .quad 0x3d584bf2b68d766f - .quad 0x3d5a21ac25d81ef3 - .quad 0x3d3bb2cd720ec44c - .quad 0xbd657d49676844cc - .quad 0x3d1a07bd8b34be7c - .quad 0x3d60be1fb590a1f5 - .quad 0xbd5aa33736867a17 - .quad 0x3d46abb9df22bc57 - /*== One ==*/ - .align 64 - .quad 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000 - /*== AbsMask ==*/ - .align 64 - .quad 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff - /*== AddB5 ==*/ - .align 64 - .quad 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000 - /*== RcpBitMask ==*/ - .align 64 - .quad 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000 - /*== poly_coeff8 ==*/ - .align 64 - .quad 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142 - /*== poly_coeff7 ==*/ - .align 64 - .quad 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70 - /*== poly_coeff6 ==*/ - .align 64 - .quad 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8 - /*== poly_coeff5 ==*/ - .align 64 - .quad 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5 - /*== poly_coeff4 ==*/ - .align 64 - .quad 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a - /*== poly_coeff3 ==*/ - .align 64 - .quad 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01 - /*== poly_coeff2 ==*/ - .align 64 - .quad 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462 - /*== poly_coeff1 ==*/ - .align 64 - .quad 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5 - /*== poly_coeff0 ==*/ - .align 64 - .quad 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000 - /*== Half ==*/ - .align 64 - .quad 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000 - /*== L2H = log(2)_high ==*/ - .align 64 - .quad 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000 - /*== L2L = log(2)_low ==*/ - .align 64 - .quad 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000 - .align 64 - .type __svml_datanh_data_internal_avx512,@object - .size __svml_datanh_data_internal_avx512,.-__svml_datanh_data_internal_avx512 + /* Log_tbl_H */ + .quad 0x0000000000000000 + .quad 0x3faf0a30c0100000 + .quad 0x3fbe27076e2a0000 + .quad 0x3fc5ff3070a80000 + .quad 0x3fcc8ff7c79b0000 + .quad 0x3fd1675cabab8000 + .quad 0x3fd4618bc21c8000 + .quad 0x3fd739d7f6bc0000 + .quad 0x3fd9f323ecbf8000 + .quad 0x3fdc8ff7c79a8000 + .quad 0x3fdf128f5faf0000 + .quad 0x3fe0be72e4254000 + .quad 0x3fe1e85f5e704000 + .quad 0x3fe307d7334f0000 + .quad 0x3fe41d8fe8468000 + .quad 0x3fe52a2d265bc000 + /* Log_tbl_L */ + .align 64 + .quad 0x0000000000000000 + .quad 0x3d662a6617cc9717 + .quad 0x3d6e5cbd3d50fffc + .quad 0xbd6b0b0de3077d7e + .quad 0xbd697794f689f843 + .quad 0x3d630701ce63eab9 + .quad 0xbd609ec17a426426 + .quad 0xbd67fcb18ed9d603 + .quad 0x3d584bf2b68d766f + .quad 0x3d5a21ac25d81ef3 + .quad 0x3d3bb2cd720ec44c + .quad 0xbd657d49676844cc + .quad 0x3d1a07bd8b34be7c + .quad 0x3d60be1fb590a1f5 + .quad 0xbd5aa33736867a17 + .quad 0x3d46abb9df22bc57 + /* One */ + .align 64 + .quad 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000 + /* AbsMask */ + .align 64 + .quad 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff, 0x7fffffffffffffff + /* AddB5 */ + .align 64 + .quad 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000, 0x0000800000000000 + /* RcpBitMask */ + .align 64 + .quad 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000, 0xffff000000000000 + /* poly_coeff8 */ + .align 64 + .quad 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142, 0x3fbc81dd40d38142 + /* poly_coeff7 */ + .align 64 + .quad 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70, 0xbfc0073cb82e8b70 + /* poly_coeff6 */ + .align 64 + .quad 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8, 0x3fc2492298ffdae8 + /* poly_coeff5 */ + .align 64 + .quad 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5, 0xbfc55553f871e5c5 + /* poly_coeff4 */ + .align 64 + .quad 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a, 0x3fc9999999cd394a + /* poly_coeff3 */ + .align 64 + .quad 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01, 0xbfd00000000c2a01 + /* poly_coeff2 */ + .align 64 + .quad 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462, 0x3fd5555555555462 + /* poly_coeff1 */ + .align 64 + .quad 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5, 0xbfdfffffffffffc5 + /* poly_coeff0 */ + .align 64 + .quad 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000, 0x3ff0000000000000 + /* Half */ + .align 64 + .quad 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000, 0x3fe0000000000000 + /* L2H = log(2)_high */ + .align 64 + .quad 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000, 0x3fe62E42FEFA0000 + /* L2L = log(2)_low */ + .align 64 + .quad 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000, 0x3d7cf79abc9e0000 + .align 64 + .type __svml_datanh_data_internal_avx512, @object + .size __svml_datanh_data_internal_avx512, .-__svml_datanh_data_internal_avx512 -- 2.7.4