Provide consistent alignment to swap_ranges benchmark

[AlexGuteniev]

Towards #5035

C:\Project\STL>start /B /WAIT /HIGH /AFFINITY 2 out\bench\benchmark-swap_ranges.exe --benchmark_min_time=2s
2024-10-26T19:35:11+03:00
Running out\bench\benchmark-swap_ranges.exe
Run on (12 X 2496 MHz CPU s)
CPU Caches:
  L1 Data 48 KiB (x6)
  L1 Instruction 32 KiB (x6)
  L2 Unified 1280 KiB (x6)
  L3 Unified 12288 KiB (x1)
---------------------------------------------------------------------------------------------
Benchmark                                                   Time             CPU   Iterations
---------------------------------------------------------------------------------------------
std_swap<1, uint8_t, aligner>                            1.64 ns         1.63 ns   1739805825
std_swap<5, uint8_t, aligner>                            1.64 ns         1.61 ns   1739805825
std_swap<15, uint8_t, aligner>                           2.11 ns         2.10 ns   1378461538
std_swap<26, uint8_t, aligner>                           2.12 ns         2.11 ns   1194666667
std_swap<38, uint8_t, aligner>                           2.12 ns         2.12 ns   1194666667
std_swap<60, uint8_t, aligner>                           2.59 ns         2.56 ns    995555556
std_swap<125, uint8_t, aligner>                          3.53 ns         3.51 ns    716800000
std_swap<800, uint8_t, aligner>                          13.0 ns         12.9 ns    215903614
std_swap<3000, uint8_t, aligner>                         47.8 ns         46.8 ns     57806452
std_swap<9000, uint8_t, aligner>                          141 ns          140 ns     20134831
std_swap<1, uint8_t, unaligner>                          1.65 ns         1.64 ns   1723076923
std_swap<5, uint8_t, unaligner>                          1.67 ns         1.67 ns   1644036697
std_swap<15, uint8_t, unaligner>                         2.12 ns         2.11 ns   1194666667
std_swap<26, uint8_t, unaligner>                         2.12 ns         2.09 ns   1194666667
std_swap<38, uint8_t, unaligner>                         2.12 ns         2.12 ns   1194666667
std_swap<60, uint8_t, unaligner>                         2.59 ns         2.58 ns   1054117647
std_swap<125, uint8_t, unaligner>                        4.47 ns         4.50 ns    663703704
std_swap<800, uint8_t, unaligner>                        24.2 ns         23.7 ns    112000000
std_swap<3000, uint8_t, unaligner>                       88.8 ns         88.2 ns     32581818
std_swap<9000, uint8_t, unaligner>                        287 ns          285 ns      9431579
std_swap_ranges<uint8_t, aligned_allocator>/1            1.88 ns         1.87 ns   1280000000
std_swap_ranges<uint8_t, aligned_allocator>/5            3.53 ns         3.54 ns    746666667
std_swap_ranges<uint8_t, aligned_allocator>/15           4.94 ns         4.88 ns    560000000
std_swap_ranges<uint8_t, aligned_allocator>/26           3.06 ns         3.03 ns    814545455
std_swap_ranges<uint8_t, aligned_allocator>/38           4.47 ns         4.43 ns    663703704
std_swap_ranges<uint8_t, aligned_allocator>/60           4.47 ns         4.47 ns    640000000
std_swap_ranges<uint8_t, aligned_allocator>/125          5.88 ns         5.84 ns    484324324
std_swap_ranges<uint8_t, aligned_allocator>/800          12.9 ns         13.0 ns    215903614
std_swap_ranges<uint8_t, aligned_allocator>/3000         45.9 ns         45.8 ns     61793103
std_swap_ranges<uint8_t, aligned_allocator>/9000          134 ns          131 ns     20837209
std_swap_ranges<uint8_t, unaligned_allocator>/1          1.88 ns         1.87 ns   1493333333
std_swap_ranges<uint8_t, unaligned_allocator>/5          3.53 ns         3.54 ns    746666667
std_swap_ranges<uint8_t, unaligned_allocator>/15         4.94 ns         4.91 ns    560000000
std_swap_ranges<uint8_t, unaligned_allocator>/26         3.06 ns         3.06 ns    853333333
std_swap_ranges<uint8_t, unaligned_allocator>/38         4.47 ns         4.44 ns    640000000
std_swap_ranges<uint8_t, unaligned_allocator>/60         4.47 ns         4.44 ns    640000000
std_swap_ranges<uint8_t, unaligned_allocator>/125        6.37 ns         6.39 ns    459487179
std_swap_ranges<uint8_t, unaligned_allocator>/800        18.6 ns         18.5 ns    137846154
std_swap_ranges<uint8_t, unaligned_allocator>/3000       67.7 ns         67.5 ns     40727273
std_swap_ranges<uint8_t, unaligned_allocator>/9000        221 ns          217 ns     11946667

[StephanTLavavej]

Thanks! 😻 I pushed several commits, including a couple of bugfixes. FYI @CaseyCarter.

[StephanTLavavej]

I'm mirroring this to the MSVC-internal repo - please notify me if any further changes are pushed.

[StephanTLavavej]

Thanks for realigning the benchmark! ⏱️ ↔️ ⏳