Improve the OSACA macros

[eggrobin]

• Don’t inline the functions.
• fix the garbled argument reduction in UNDER_OSACA_HYPOTHESES;
• carry the loop through registers by default now that ordering the graph is dealt with by Graph improvements OSACA#21;
• use an outer loop to pin the markers where the loop-carried dependencies are still valid;
• make the OSACA_loop_terminator volatile. Reading it early did not seem to allow any unwanted optimizations, but it made the code less readable, with a load outside the loop and a mysterious test at the end. The extra movzx is irrelevant to latency analysis.

Now works as expected for other quadrants, e.g., θ = 3 (CSX; Mind the missing roundsd latency; something else seems to be missing, compare other *CAs on the same code):

Open Source Architecture Code Analyzer (OSACA) - 0.6.1
Analyzed file:      ..\Principia\numerics\Release\x64\sin_cos.asm
Architecture:       CSX
Timestamp:          2025-01-01 04:28:52


 P - Throughput of LOAD operation can be hidden behind a past or future STORE instruction
 * - Instruction micro-ops not bound to a port
 X - No throughput/latency information for this instruction in data file


Combined Analysis Report
------------------------
                                      Port pressure in cycles

     |  0   - 0DV  |  1   |  2   -  2D  |  3   -  3D  |  4   |  5   |  6   |  7   ||  CP  | LCD  |
--------------------------------------------------------------------------------------------------
2718 |             |      |             |             |      |      |      |      ||      |      | X npad 7
2719 |             |      |             |             |      |      |      |      ||      |      |   $OSACA_loop$154:
2720 |             |      |             |             |      |      |      |      ||      |      |   ; Line 333
2721 | 1.00        |      |             |             |      |      |      |      ||      |      |   comisd xmm15, xmm4
2722 |             |      |             |             |      |      |      |      ||      |      | * jbe SHORT $LN16@Cos
2723 | 1.00        |      |             |             |      |      |      |      ||      |      |   comisd xmm4, xmm13
2724 |             |      |             |             |      |      |      |      ||      |      | * jbe SHORT $LN16@Cos
2725 | 0.50        | 0.00 |             |             |      | 0.00 | 0.50 |      ||      |      |   mov al, 1
2726 | 0.50        |      | 0.50   0.50 | 0.50   0.50 |      |      | 0.50 |      ||      |      |   jmp SHORT $LN17@Cos
2727 |             |      |             |             |      |      |      |      ||      |      |   $LN16@Cos:
2728 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   xor eax, eax
2729 |             |      |             |             |      |      |      |      ||      |      |   $LN17@Cos:
2730 |             |      |             |             |      |      |      |      ||      |      |   ; Line 338
2731 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movsd xmm0, QWORD PTR __real@411921fb54442d18
2732 | 1.00        |      |             |             |      |      |      |      ||      |      |   comisd xmm0, xmm4
2733 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$13[rsp], al
2734 |             |      |             |             |      |      |      |      ||      |      | * jb SHORT $LN18@Cos
2735 | 1.00        |      |             |             |      |      |      |      ||      |      |   comisd xmm4, xmm14
2736 |             |      |             |             |      |      |      |      ||      |      | * jb SHORT $LN18@Cos
2737 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   mov al, 1
2738 | 0.00        |      | 0.50   0.50 | 0.50   0.50 |      |      | 1.00 |      ||      |      |   jmp SHORT $LN19@Cos
2739 |             |      |             |             |      |      |      |      ||      |      |   $LN18@Cos:
2740 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   xor eax, eax
2741 |             |      |             |             |      |      |      |      ||      |      |   $LN19@Cos:
2742 |             |      |             |             |      |      |      |      ||      |      |   ; Line 343
2743 |             |      |             |             |      |      |      |      ||  0.0 |  0.0 | * movaps xmm0, xmm4
2744 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$12[rsp], al
2745 | 0.33        | 0.33 | 0.50   0.50 | 0.50   0.50 |      | 0.33 |      |      ||  4.0 |  4.0 |   mulsd xmm0, QWORD PTR __real@3fe45f306dc9c883
2746 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||      |      |   xorps xmm1, xmm1
2747 |             |      |             |             |      | 1.00 |      |      ||  1.0 |  1.0 |   movsd xmm1, xmm0
2748 |             |      |             |             |      |      |      |      ||  0.0 |  0.0 | X roundsd xmm2, xmm1, 4
2749 |             |      |             |             |      |      |      |      ||      |      |   ; Line 347
2750 |             |      |             |             |      |      |      |      ||  0.0 |  0.0 | * movaps xmm0, xmm2
2751 | 0.33        | 0.33 | 0.50   0.50 | 0.50   0.50 |      | 0.33 |      |      ||  4.0 |  4.0 |   mulsd xmm0, QWORD PTR __real@3ff921fb54440000
2752 | 1.33        | 0.33 |             |             |      | 0.33 |      |      ||      |      |   cvtsd2si rdx, xmm2
2753 |             |      |             |             |      |      |      |      ||      |      |   ; Line 348
2754 | 0.33        | 0.33 | 0.50   0.50 | 0.50   0.50 |      | 0.33 |      |      ||      |      |   mulsd xmm2, QWORD PTR __real@3d868c234c4c6629
2755 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||  4.0 |  4.0 |   subsd xmm4, xmm0
2756 |             |      |             |             |      |      |      |      ||      |      |   ; Line 351
2757 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movsd xmm0, QWORD PTR __real@be10000000000000
2758 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\double_precision_body.hpp
2759 |             |      |             |             |      |      |      |      ||      |      |   ; Line 370
2760 |             |      |             |             |      |      |      |      ||  0.0 |  0.0 | * movaps xmm11, xmm4
2761 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||  4.0 |  4.0 |   subsd xmm11, xmm2
2762 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2763 |             |      |             |             |      |      |      |      ||      |      |   ; Line 351
2764 | 1.00        |      |             |             |      |      |      |      ||      |      |   comisd xmm0, xmm11
2765 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\double_precision_body.hpp
2766 |             |      |             |             |      |      |      |      ||      |      |   ; Line 371
2767 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||      |      |   subsd xmm4, xmm11
2768 | 0.67        | 0.16 |             |             |      | 0.16 |      |      ||      |      |   subsd xmm4, xmm2
2769 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2770 |             |      |             |             |      |      |      |      ||      |      |   ; Line 351
2771 |             |      |             |             |      |      |      |      ||      |      | * ja SHORT $LN20@Cos
2772 | 1.00        |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   comisd xmm11, QWORD PTR __real@3e10000000000000
2773 |             |      |             |             |      |      |      |      ||      |      | * ja SHORT $LN20@Cos
2774 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   xor eax, eax
2775 | 0.00        |      | 0.50   0.50 | 0.50   0.50 |      |      | 1.00 |      ||      |      |   jmp SHORT $LN21@Cos
2776 |             |      |             |             |      |      |      |      ||      |      |   $LN20@Cos:
2777 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   mov al, 1
2778 |             |      |             |             |      |      |      |      ||      |      |   $LN21@Cos:
2779 |             |      |             |             |      |      |      |      ||      |      |   ; Line 431
2780 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movaps xmm12, XMMWORD PTR ?sign_bit@masks@internal@_sin_cos@numerics@principia@@3U__m128d@@B
2781 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||      |      |   xorps xmm3, xmm3
2782 |             |      |             |             |      |      |      |      ||      |      |   ; Line 351
2783 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$11[rsp], al
2784 |             |      |             |             |      |      |      |      ||      |      |   ; Line 431
2785 |             |      |             |             |      | 1.00 |      |      ||      |      |   movsd xmm3, xmm11
2786 |             |      |             |             |      |      |      |      ||      |      |   ; Line 499
2787 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movzx eax, BYTE PTR ?UseHardwareFMA@internal@_fma@numerics@principia@@3_NB ; principia::numerics::_fma::internal::UseHardwareFMA
2788 |             |      |             |             |      |      |      |      ||      |      |   ; Line 431
2789 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||      |      |   andps xmm3, xmm12
2790 |             |      |             |             |      |      |      |      ||      |      |   ; Line 499
2791 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$18[rsp], al
2792 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||      |      |   xorps xmm10, xmm10
2793 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.38.33130\include\cstdlib
2794 |             |      |             |             |      |      |      |      ||      |      |   ; Line 23
2795 |             |      |             |             |      |      |      |      ||  0.0 |  0.0 | * movaps xmm0, xmm11
2796 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2797 |             |      |             |             |      |      |      |      ||      |      |   ; Line 352
2798 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   mov rcx, rdx
2799 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.38.33130\include\cstdlib
2800 |             |      |             |             |      |      |      |      ||      |      |   ; Line 23
2801 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||  1.0 |  1.0 |   andps xmm0, xmm6
2802 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2803 |             |      |             |             |      |      |      |      ||      |      |   ; Line 500
2804 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   and dl, 1
2805 |             |      |             |             |      |      |      |      ||      |      |   ; Line 275
2806 | 0.33        | 0.33 | 0.50   0.50 | 0.50   0.50 |      | 0.33 |      |      ||  4.0 |  4.0 |   addsd xmm0, QWORD PTR __real@42a0000000000000
2807 |             |      |             |             |      |      |      |      ||      |      |   ; Line 448
2808 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm6, xmm4
2809 |             |      |             |             |      |      |      |      ||      |      |   ; Line 500
2810 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$17[rsp], dl
2811 |             |      |             |             |      |      |      |      ||      |      |   ; Line 448
2812 | 0.50        | 0.50 |             |             |      |      |      |      ||      |      |   addsd xmm6, xmm4
2813 |             |      |             |             |      |      |      |      ||      |      |   ; Line 352
2814 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   and ecx, 3
2815 |             |      |             |             |      |      |      |      ||      |      |   ; Line 275
2816 | 0.67        | 0.16 | 0.50   0.50 | 0.50   0.50 |      | 0.16 |      |      ||  1.0 |  1.0 |   andps xmm0, XMMWORD PTR ?mantissa_index_bits@masks@internal@_sin_cos@numerics@principia@@3U__m128d@@B
2817 | 1.00        |      |             |             |      |      |      |      ||  1.0 |  1.0 |   movq rax, xmm0
2818 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.38.33130\include\array
2819 |             |      |             |             |      |      |      |      ||      |      |   ; Line 545
2820 | 0.00        |      |             |             |      |      | 1.00 |      ||  1.0 |  1.0 |   shl rax, 5
2821 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\fma_body.hpp
2822 |             |      |             |             |      |      |      |      ||      |      |   ; Line 35
2823 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movsd xmm8, QWORD PTR [rax+r8+16]
2824 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2825 |             |      |             |             |      |      |      |      ||      |      |   ; Line 436
2826 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movsd xmm0, QWORD PTR [rax+r8+8]
2827 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||  4.0 |  4.0 |   movsd xmm1, QWORD PTR [rax+r8]
2828 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm2, xmm0
2829 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||  1.0 |  1.0 |   xorps xmm1, xmm3
2830 |             |      |             |             |      |      |      |      ||      |      |   ; Line 283
2831 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$10[rsp], 1
2832 |             |      |             |             |      |      |      |      ||      |      |   ; Line 436
2833 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||      |      |   xorps xmm2, xmm3
2834 |             |      |             |             |      |      |      |      ||      |      |   ; Line 294
2835 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$9[rsp], 1
2836 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\fma_body.hpp
2837 |             |      |             |             |      |      |      |      ||      |      |   ; Line 35
2838 |             |      |             |             |      | 1.00 |      |      ||      |      |   movsd xmm10, xmm2
2839 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2840 |             |      |             |             |      |      |      |      ||      |      |   ; Line 444
2841 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||  4.0 |  4.0 |   subsd xmm11, xmm1
2842 | 0.33        | 0.33 |             |             |      | 0.33 |      |      ||      |      |   xorps xmm1, xmm1
2843 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\fma_body.hpp
2844 |             |      |             |             |      |      |      |      ||      |      |   ; Line 35
2845 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm0, xmm8
2846 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm9, xmm10
2847 |             |      |             |             |      |      |      |      ||      |      |   ; Line 46
2848 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm7, xmm10
2849 | 0.11        | 0.13 |             |             |      | 0.75 |      |      ||      |      |   xorps xmm3, xmm3
2850 |             |      |             |             |      |      |      |      ||      |      |   ; Line 35
2851 |             |      |             |             |      | 1.00 |      |      ||      |      |   movsd xmm3, xmm11
2852 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2853 |             |      |             |             |      |      |      |      ||      |      |   ; Line 448
2854 | 0.50        | 0.50 |             |             |      |      |      |      ||  4.0 |  4.0 |   addsd xmm6, xmm11
2855 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\fma_body.hpp
2856 |             |      |             |             |      |      |      |      ||      |      |   ; Line 35
2857 | 0.50        | 0.50 |             |             |      |      |      |      ||      |      |   vfnmadd213sd xmm9, xmm3, xmm0
2858 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\double_precision_body.hpp
2859 |             |      |             |             |      |      |      |      ||      |      |   ; Line 310
2860 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm0, xmm9
2861 | 0.00        | 0.00 |             |             |      | 1.00 |      |      ||      |      |   subsd xmm0, xmm8
2862 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2863 |             |      |             |             |      |      |      |      ||      |      |   ; Line 448
2864 | 0.50        | 0.50 |             |             |      |      |      |      ||  4.0 |  4.0 |   mulsd xmm6, xmm11
2865 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\fma_body.hpp
2866 |             |      |             |             |      |      |      |      ||      |      |   ; Line 46
2867 | 0.50        | 0.50 |             |             |      |      |      |      ||      |      |   vfnmsub213sd xmm7, xmm3, xmm0
2868 |             |      |             |             |      |      |      |      ||      |      |   ; Line 14
2869 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movaps xmm0, XMMWORD PTR tv946[rsp]
2870 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm2, xmm0
2871 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2872 |             |      |             |             |      |      |      |      ||      |      |   ; Line 450
2873 | 0.38        | 0.62 |             |             |      |      |      |      ||      |      |   mulsd xmm8, xmm6
2874 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\fma_body.hpp
2875 |             |      |             |             |      |      |      |      ||      |      |   ; Line 14
2876 |             |      |             |             |      | 1.00 |      |      ||  1.0 |  1.0 |   movsd xmm1, xmm6
2877 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm0, xmm4
2878 | 0.00        | 1.00 | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||  4.0 |  4.0 |   vfmadd213sd xmm2, xmm1, XMMWORD PTR tv947[rsp]
2879 | 0.00        | 1.00 | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   vfmadd213sd xmm5, xmm1, XMMWORD PTR tv949[rsp]
2880 |             |      |             |             |      |      |      |      ||  0.0 |  0.0 | * movaps xmm1, xmm2
2881 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2882 |             |      |             |             |      |      |      |      ||      |      |   ; Line 450
2883 | 0.00        | 1.00 |             |             |      |      |      |      ||      |      |   mulsd xmm8, xmm5
2884 | 0.00        | 0.25 |             |             |      | 0.75 |      |      ||      |      |   xorps xmm2, xmm2
2885 | 0.00        | 0.00 |             |             |      | 1.00 |      |      ||      |      |   xorps xmm4, xmm4
2886 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\fma_body.hpp
2887 |             |      |             |             |      |      |      |      ||      |      |   ; Line 14
2888 |             |      |             |             |      | 1.00 |      |      ||      |      |   movsd xmm4, xmm0
2889 |             |      |             |             |      | 1.00 |      |      ||  1.0 |  1.0 |   movsd xmm2, xmm1
2890 | 0.00        | 0.00 |             |             |      | 1.00 |      |      ||      |      |   xorps xmm1, xmm1
2891 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2892 |             |      |             |             |      |      |      |      ||      |      |   ; Line 449
2893 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm3, xmm6
2894 | 0.00        | 1.00 |             |             |      |      |      |      ||      |      |   mulsd xmm3, xmm11
2895 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\fma_body.hpp
2896 |             |      |             |             |      |      |      |      ||      |      |   ; Line 14
2897 |             |      |             |             |      | 1.00 |      |      ||      |      |   movsd xmm1, xmm3
2898 | 0.00        | 1.00 |             |             |      |      |      |      ||  4.0 |  4.0 |   vfmadd213sd xmm1, xmm2, xmm4
2899 |             |      |             |             |      |      |      |      ||      |      |   ; Line 35
2900 |             |      |             |             |      |      |      |      ||  0.0 |  0.0 | * movaps xmm0, xmm1
2901 | 0.00        | 1.00 |             |             |      |      |      |      ||  4.0 |  4.0 |   vfnmadd213sd xmm10, xmm0, xmm8
2902 | 0.00        | 1.00 |             |             |      |      |      |      ||  4.0 |  4.0 |   addsd xmm10, xmm7
2903 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\double_precision_body.hpp
2904 |             |      |             |             |      |      |      |      ||      |      |   ; Line 352
2905 |             |      |             |             |      |      |      |      ||  0.0 |  0.0 | * movaps xmm4, xmm10
2906 | 0.00        | 1.00 |             |             |      |      |      |      ||  4.0 |  4.0 |   addsd xmm4, xmm9
2907 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2908 |             |      |             |             |      |      |      |      ||      |      |   ; Line 310
2909 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm0, xmm4
2910 |             |      |             |             |      |      |      |      ||      |      | * movaps xmm2, xmm0
2911 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\double_precision_body.hpp
2912 |             |      |             |             |      |      |      |      ||      |      |   ; Line 353
2913 |             |      |             |             |      |      |      |      ||  0.0 |      | * movaps xmm0, xmm4
2914 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2915 |             |      |             |             |      |      |      |      ||      |      |   ; Line 310
2916 | 0.00        | 0.50 | 0.50   0.50 | 0.50   0.50 |      | 0.50 |      |      ||      |      |   andps xmm2, XMMWORD PTR ?exponent_bits@masks@internal@_sin_cos@numerics@principia@@3U__m128d@@B
2917 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\double_precision_body.hpp
2918 |             |      |             |             |      |      |      |      ||      |      |   ; Line 353
2919 | 0.00        | 0.50 |             |             |      | 0.50 |      |      ||  4.0 |      |   subsd xmm0, xmm9
2920 | 0.00        | 0.50 |             |             |      | 0.50 |      |      ||  4.0 |      |   subsd xmm10, xmm0
2921 |             |      |             |             |      |      |      |      ||      |      |   ; File C:\Users\robin\Projects\mockingbirdnest\Principia\numerics\sin_cos.cpp
2922 |             |      |             |             |      |      |      |      ||      |      |   ; Line 318
2923 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movsd xmm0, QWORD PTR __real@fcaffff000000000
2924 | 0.00        | 0.49 |             |             |      | 0.51 |      |      ||  1.0 |      |   andnps xmm12, xmm10
2925 | 0.00        | 0.51 |             |             |      | 0.49 |      |      ||  1.0 |      |   psubq xmm12, xmm2
2926 | 1.00        |      |             |             |      |      |      |      ||  3.0 |      |   comisd xmm0, xmm12
2927 |             |      |             |             |      |      |      |      ||      |      | * jbe SHORT $LN128@Cos
2928 | 1.00        |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   comisd xmm12, QWORD PTR __real@fcb0000800000000
2929 |             |      |             |             |      |      |      |      ||      |      | * jbe SHORT $LN128@Cos
2930 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   mov al, 1
2931 | 0.00        |      | 0.50   0.50 | 0.50   0.50 |      |      | 1.00 |      ||      |      |   jmp SHORT $LN129@Cos
2932 |             |      |             |             |      |      |      |      ||      |      |   $LN128@Cos:
2933 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   xor eax, eax
2934 |             |      |             |             |      |      |      |      ||      |      |   $LN129@Cos:
2935 |             |      |             |             |      |      |      |      ||      |      |   ; Line 512
2936 | 1.00        |      |             |             |      |      |      |      ||      |      |   ucomisd xmm4, xmm4
2937 |             |      |             |             |      |      |      |      ||      |      |   ; Line 318
2938 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$8[rsp], al
2939 |             |      |             |             |      |      |      |      ||      |      |   ; Line 512
2940 |             |      |             |             |      |      |      |      ||      |      | * jnp SHORT $LN8@Cos
2941 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   mov al, 1
2942 | 0.00        |      | 0.50   0.50 | 0.50   0.50 |      |      | 1.00 |      ||      |      |   jmp SHORT $LN9@Cos
2943 |             |      |             |             |      |      |      |      ||      |      |   $LN8@Cos:
2944 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   xor eax, eax
2945 |             |      |             |             |      |      |      |      ||      |      |   $LN9@Cos:
2946 |             |      |             |             |      |      |      |      ||      |      |   ; Line 515
2947 | 0.00        | 0.50 | 0.50   0.50 | 0.50   0.50 |      | 0.50 |      |      ||      |  1.0 |   xorps xmm4, QWORD PTR __xmm@80000000000000008000000000000000
2948 | 0.00        | 0.50 |             |             |      | 0.50 |      |      ||      |      |   xorps xmm2, xmm2
2949 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movaps xmm5, XMMWORD PTR tv948[rsp]
2950 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movsd xmm6, QWORD PTR __xmm@7fffffffffffffff7fffffffffffffff
2951 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   mov BYTE PTR OSACA_computed_condition$16[rsp], al
2952 |             | 1.00 |             |             |      | 0.00 |      |      ||      |      |   lea rax, QWORD PTR [rcx-1]
2953 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   cmp rax, 1
2954 | 0.00        |      | 0.00        | 0.00        | 1.00 |      | 2.00 | 1.00 ||      |      |   setbe BYTE PTR OSACA_computed_condition$15[rsp]
2955 |             |      | 0.50   0.50 | 0.50   0.50 |      |      |      |      ||      |      |   movzx eax, BYTE PTR ?OSACA_loop_terminator@@3_NC
2956 | 0.00        | 0.00 |             |             |      | 0.00 | 1.00 |      ||      |      |   test al, al
2957 |             |      |             |             |      |      |      |      ||      |      | * je $OSACA_loop$154
2958 |             |      | 0.00        | 0.00        | 1.00 |      |      | 1.00 ||      |      |   movsd QWORD PTR OSACA_result$14[rsp], xmm4

       22.0          22.0   14.0   14.0   14.0   14.0   11.0   22.0   22.0   11.0    77.0   65.0

[pleroy]

The references to this symbol in sin_cos.cpp must be removed too.