Skip to content

Instantly share code, notes, and snippets.

@zanieb
Last active January 11, 2025 00:02

Revisions

  1. zanieb revised this gist Jan 11, 2025. 1 changed file with 943 additions and 0 deletions.
    943 changes: 943 additions & 0 deletions bolt-benches.md
    Original file line number Diff line number Diff line change
    @@ -847,4 +847,947 @@ Turbo Boost (MSR): Failed to read MSR 0x1a0 from /dev/cpu/0/msr: [Errno 13] Perm
    ERROR: At least one operation failed with permission error, retry as root
    OK! System ready for benchmarking
    ```





    ```
    ❯ uvx pyperformance compare bolt-baseline.json bolt-hugify.json
    bolt-baseline.json
    ==================
    Performance version: 1.11.0
    Report on Linux-6.1.0-26-amd64-x86_64-with-glibc2.36
    Number of logical CPUs: 12
    Start date: 2025-01-10 21:55:03.035465
    End date: 2025-01-10 22:32:07.098269
    bolt-hugify.json
    ================
    Performance version: 1.11.0
    Report on Linux-6.1.0-26-amd64-x86_64-with-glibc2.36
    Number of logical CPUs: 12
    Start date: 2025-01-10 23:10:53.371551
    End date: 2025-01-10 23:46:30.321131
    ### 2to3 ###
    Mean +- std dev: 266 ms +- 1 ms -> 258 ms +- 1 ms: 1.03x faster
    Significant (t=47.13)
    ### async_generators ###
    Mean +- std dev: 437 ms +- 4 ms -> 408 ms +- 4 ms: 1.07x faster
    Significant (t=39.38)
    ### asyncio_tcp ###
    Mean +- std dev: 352 ms +- 4 ms -> 352 ms +- 4 ms: 1.00x slower
    Not significant
    ### asyncio_tcp_ssl ###
    Mean +- std dev: 1.38 sec +- 0.01 sec -> 1.37 sec +- 0.01 sec: 1.01x faster
    Not significant
    ### asyncio_websockets ###
    Mean +- std dev: 502 ms +- 15 ms -> 501 ms +- 9 ms: 1.00x faster
    Not significant
    ### bench_mp_pool ###
    Mean +- std dev: 919 ms +- 613 ms -> 116 ms +- 75 ms: 7.95x faster
    Significant (t=10.08)
    ### bench_thread_pool ###
    Mean +- std dev: 974 us +- 39 us -> 947 us +- 38 us: 1.03x faster
    Significant (t=3.90)
    ### chaos ###
    Mean +- std dev: 72.2 ms +- 0.6 ms -> 69.6 ms +- 0.7 ms: 1.04x faster
    Significant (t=21.66)
    ### comprehensions ###
    Mean +- std dev: 19.6 us +- 0.2 us -> 18.1 us +- 0.1 us: 1.08x faster
    Significant (t=51.87)
    ### coroutines ###
    Mean +- std dev: 26.9 ms +- 0.2 ms -> 23.0 ms +- 0.2 ms: 1.17x faster
    Significant (t=105.43)
    ### create_gc_cycles ###
    Mean +- std dev: 969 us +- 6 us -> 971 us +- 6 us: 1.00x slower
    Not significant
    ### crypto_pyaes ###
    Mean +- std dev: 78.5 ms +- 0.6 ms -> 76.2 ms +- 0.6 ms: 1.03x faster
    Significant (t=22.28)
    ### deepcopy ###
    Mean +- std dev: 301 us +- 2 us -> 277 us +- 3 us: 1.09x faster
    Significant (t=55.07)
    ### deepcopy_memo ###
    Mean +- std dev: 35.0 us +- 0.5 us -> 32.8 us +- 0.8 us: 1.07x faster
    Significant (t=19.23)
    ### deepcopy_reduce ###
    Mean +- std dev: 3.09 us +- 0.04 us -> 2.88 us +- 0.02 us: 1.07x faster
    Significant (t=34.61)
    ### deltablue ###
    Mean +- std dev: 4.03 ms +- 0.03 ms -> 3.86 ms +- 0.04 ms: 1.04x faster
    Significant (t=25.72)
    ### docutils ###
    Mean +- std dev: 2.38 sec +- 0.01 sec -> 2.30 sec +- 0.02 sec: 1.03x faster
    Significant (t=29.37)
    ### dulwich_log ###
    Mean +- std dev: 66.5 ms +- 0.3 ms -> 64.4 ms +- 0.3 ms: 1.03x faster
    Significant (t=41.08)
    ### fannkuch ###
    Mean +- std dev: 485 ms +- 6 ms -> 427 ms +- 5 ms: 1.13x faster
    Significant (t=61.03)
    ### float ###
    Mean +- std dev: 77.6 ms +- 1.1 ms -> 75.4 ms +- 1.2 ms: 1.03x faster
    Significant (t=10.28)
    ### gc_traversal ###
    Mean +- std dev: 3.32 ms +- 0.05 ms -> 3.79 ms +- 0.12 ms: 1.14x slower
    Significant (t=-27.78)
    ### generators ###
    Mean +- std dev: 32.9 ms +- 0.4 ms -> 31.2 ms +- 0.3 ms: 1.06x faster
    Significant (t=28.13)
    ### genshi_text ###
    Mean +- std dev: 25.6 ms +- 0.3 ms -> 24.4 ms +- 0.2 ms: 1.05x faster
    Significant (t=29.76)
    ### genshi_xml ###
    Mean +- std dev: 61.7 ms +- 0.4 ms -> 57.0 ms +- 0.4 ms: 1.08x faster
    Significant (t=64.38)
    ### go ###
    Mean +- std dev: 146 ms +- 1 ms -> 136 ms +- 1 ms: 1.07x faster
    Significant (t=55.24)
    ### hexiom ###
    Mean +- std dev: 7.65 ms +- 0.12 ms -> 7.06 ms +- 0.10 ms: 1.08x faster
    Significant (t=28.43)
    ### html5lib ###
    Mean +- std dev: 65.3 ms +- 0.6 ms -> 63.4 ms +- 0.7 ms: 1.03x faster
    Significant (t=16.36)
    ### json_dumps ###
    Mean +- std dev: 11.7 ms +- 0.1 ms -> 11.1 ms +- 0.2 ms: 1.05x faster
    Significant (t=20.75)
    ### json_loads ###
    Mean +- std dev: 22.2 us +- 0.2 us -> 21.6 us +- 0.3 us: 1.03x faster
    Significant (t=13.36)
    ### logging_format ###
    Mean +- std dev: 8.38 us +- 0.12 us -> 8.21 us +- 0.08 us: 1.02x faster
    Significant (t=9.57)
    ### logging_silent ###
    Mean +- std dev: 120 ns +- 2 ns -> 120 ns +- 3 ns: 1.00x faster
    Not significant
    ### logging_simple ###
    Mean +- std dev: 7.56 us +- 0.12 us -> 7.50 us +- 0.11 us: 1.01x faster
    Not significant
    ### mako ###
    Mean +- std dev: 12.9 ms +- 0.2 ms -> 12.4 ms +- 0.3 ms: 1.04x faster
    Significant (t=12.79)
    ### mdp ###
    Mean +- std dev: 2.48 sec +- 0.07 sec -> 2.48 sec +- 0.07 sec: 1.00x slower
    Not significant
    ### meteor_contest ###
    Mean +- std dev: 105 ms +- 1 ms -> 100 ms +- 1 ms: 1.04x faster
    Significant (t=39.95)
    ### nbody ###
    Mean +- std dev: 114 ms +- 1 ms -> 109 ms +- 1 ms: 1.05x faster
    Significant (t=34.87)
    ### nqueens ###
    Mean +- std dev: 92.6 ms +- 1.1 ms -> 86.1 ms +- 0.9 ms: 1.08x faster
    Significant (t=36.46)
    ### pathlib ###
    Mean +- std dev: 19.9 ms +- 0.1 ms -> 19.3 ms +- 0.1 ms: 1.03x faster
    Significant (t=35.42)
    ### pickle ###
    Mean +- std dev: 11.0 us +- 0.1 us -> 10.7 us +- 0.2 us: 1.03x faster
    Significant (t=10.03)
    ### pickle_dict ###
    Mean +- std dev: 24.1 us +- 0.5 us -> 24.0 us +- 0.2 us: 1.00x faster
    Not significant
    ### pickle_list ###
    Mean +- std dev: 3.88 us +- 0.11 us -> 3.87 us +- 0.09 us: 1.00x faster
    Not significant
    ### pickle_pure_python ###
    Mean +- std dev: 366 us +- 2 us -> 348 us +- 4 us: 1.05x faster
    Significant (t=32.25)
    ### pidigits ###
    Mean +- std dev: 182 ms +- 1 ms -> 182 ms +- 1 ms: 1.00x faster
    Not significant
    ### pprint_pformat ###
    Mean +- std dev: 1.80 sec +- 0.02 sec -> 1.70 sec +- 0.02 sec: 1.06x faster
    Significant (t=29.06)
    ### pprint_safe_repr ###
    Mean +- std dev: 860 ms +- 9 ms -> 808 ms +- 13 ms: 1.06x faster
    Significant (t=25.72)
    ### pyflate ###
    Mean +- std dev: 488 ms +- 4 ms -> 447 ms +- 3 ms: 1.09x faster
    Significant (t=63.96)
    ### python_startup ###
    Mean +- std dev: 9.40 ms +- 0.03 ms -> 11.42 ms +- 0.04 ms: 1.22x slower
    Significant (t=-530.58)
    ### python_startup_no_site ###
    Mean +- std dev: 6.63 ms +- 0.02 ms -> 8.66 ms +- 0.04 ms: 1.31x slower
    Significant (t=-620.18)
    ### raytrace ###
    Mean +- std dev: 304 ms +- 3 ms -> 299 ms +- 4 ms: 1.01x faster
    Not significant
    ### regex_compile ###
    Mean +- std dev: 149 ms +- 1 ms -> 141 ms +- 1 ms: 1.06x faster
    Significant (t=59.68)
    ### regex_dna ###
    Mean +- std dev: 158 ms +- 2 ms -> 154 ms +- 2 ms: 1.03x faster
    Significant (t=13.64)
    ### regex_effbot ###
    Mean +- std dev: 2.65 ms +- 0.06 ms -> 2.65 ms +- 0.06 ms: 1.00x faster
    Not significant
    ### regex_v8 ###
    Mean +- std dev: 22.9 ms +- 0.3 ms -> 22.1 ms +- 0.2 ms: 1.04x faster
    Significant (t=17.14)
    ### richards ###
    Mean +- std dev: 58.1 ms +- 0.7 ms -> 55.1 ms +- 0.5 ms: 1.05x faster
    Significant (t=26.12)
    ### richards_super ###
    Mean +- std dev: 66.1 ms +- 0.8 ms -> 62.0 ms +- 0.5 ms: 1.07x faster
    Significant (t=32.58)
    ### scimark_fft ###
    Mean +- std dev: 404 ms +- 5 ms -> 373 ms +- 4 ms: 1.08x faster
    Significant (t=38.54)
    ### scimark_lu ###
    Mean +- std dev: 127 ms +- 3 ms -> 119 ms +- 2 ms: 1.06x faster
    Significant (t=17.57)
    ### scimark_monte_carlo ###
    Mean +- std dev: 74.6 ms +- 0.8 ms -> 69.8 ms +- 0.9 ms: 1.07x faster
    Significant (t=30.76)
    ### scimark_sor ###
    Mean +- std dev: 157 ms +- 1 ms -> 147 ms +- 1 ms: 1.07x faster
    Significant (t=41.84)
    ### scimark_sparse_mat_mult ###
    Mean +- std dev: 5.63 ms +- 0.07 ms -> 5.21 ms +- 0.19 ms: 1.08x faster
    Significant (t=15.73)
    ### spectral_norm ###
    Mean +- std dev: 120 ms +- 2 ms -> 105 ms +- 1 ms: 1.14x faster
    Significant (t=56.11)
    ### sqlglot_normalize ###
    Mean +- std dev: 119 ms +- 1 ms -> 113 ms +- 1 ms: 1.05x faster
    Significant (t=27.70)
    ### telco ###
    Mean +- std dev: 7.65 ms +- 0.13 ms -> 7.13 ms +- 0.08 ms: 1.07x faster
    Significant (t=25.45)
    ### tomli_loads ###
    Mean +- std dev: 2.81 sec +- 0.03 sec -> 2.58 sec +- 0.02 sec: 1.09x faster
    Significant (t=48.19)
    ### typing_runtime_protocols ###
    Mean +- std dev: 177 us +- 3 us -> 164 us +- 3 us: 1.08x faster
    Significant (t=21.62)
    ### unpack_sequence ###
    Mean +- std dev: 44.5 ns +- 1.0 ns -> 40.8 ns +- 0.4 ns: 1.09x faster
    Significant (t=27.08)
    ### unpickle ###
    Mean +- std dev: 12.7 us +- 0.1 us -> 12.8 us +- 0.1 us: 1.00x slower
    Not significant
    ### unpickle_list ###
    Mean +- std dev: 4.09 us +- 0.11 us -> 4.01 us +- 0.07 us: 1.02x faster
    Not significant
    ### unpickle_pure_python ###
    Mean +- std dev: 251 us +- 2 us -> 240 us +- 3 us: 1.05x faster
    Significant (t=25.08)
    ### xml_etree_generate ###
    Mean +- std dev: 88.7 ms +- 0.5 ms -> 85.2 ms +- 0.7 ms: 1.04x faster
    Significant (t=31.43)
    ### xml_etree_iterparse ###
    Mean +- std dev: 97.2 ms +- 1.1 ms -> 95.7 ms +- 1.2 ms: 1.02x faster
    Not significant
    ### xml_etree_parse ###
    Mean +- std dev: 150 ms +- 3 ms -> 148 ms +- 2 ms: 1.01x faster
    Not significant
    ### xml_etree_process ###
    Mean +- std dev: 63.3 ms +- 0.5 ms -> 60.0 ms +- 0.6 ms: 1.05x faster
    Significant (t=33.38)
    ```

    ```
    ❯ uvx pyperformance compare bolt-baseline.json bolt-cdsplit.json
    bolt-baseline.json
    ==================
    Performance version: 1.11.0
    Report on Linux-6.1.0-26-amd64-x86_64-with-glibc2.36
    Number of logical CPUs: 12
    Start date: 2025-01-10 21:55:03.035465
    End date: 2025-01-10 22:32:07.098269
    bolt-cdsplit.json
    =================
    Performance version: 1.11.0
    Report on Linux-6.1.0-26-amd64-x86_64-with-glibc2.36
    Number of logical CPUs: 12
    Start date: 2025-01-10 22:33:11.759573
    End date: 2025-01-10 23:09:49.109957
    ### 2to3 ###
    Mean +- std dev: 266 ms +- 1 ms -> 263 ms +- 1 ms: 1.01x faster
    Not significant
    ### async_generators ###
    Mean +- std dev: 437 ms +- 4 ms -> 430 ms +- 5 ms: 1.02x faster
    Not significant
    ### asyncio_tcp ###
    Mean +- std dev: 352 ms +- 4 ms -> 351 ms +- 3 ms: 1.00x faster
    Not significant
    ### asyncio_tcp_ssl ###
    Mean +- std dev: 1.38 sec +- 0.01 sec -> 1.39 sec +- 0.01 sec: 1.01x slower
    Not significant
    ### asyncio_websockets ###
    Mean +- std dev: 502 ms +- 15 ms -> 504 ms +- 23 ms: 1.00x slower
    Not significant
    ### bench_mp_pool ###
    Mean +- std dev: 919 ms +- 613 ms -> 616 ms +- 413 ms: 1.49x faster
    Significant (t=3.18)
    ### bench_thread_pool ###
    Mean +- std dev: 974 us +- 39 us -> 980 us +- 36 us: 1.01x slower
    Not significant
    ### chaos ###
    Mean +- std dev: 72.2 ms +- 0.6 ms -> 73.3 ms +- 0.5 ms: 1.01x slower
    Not significant
    ### comprehensions ###
    Mean +- std dev: 19.6 us +- 0.2 us -> 19.3 us +- 0.2 us: 1.02x faster
    Not significant
    ### coroutines ###
    Mean +- std dev: 26.9 ms +- 0.2 ms -> 24.0 ms +- 0.2 ms: 1.12x faster
    Significant (t=74.54)
    ### create_gc_cycles ###
    Mean +- std dev: 969 us +- 6 us -> 974 us +- 10 us: 1.00x slower
    Not significant
    ### crypto_pyaes ###
    Mean +- std dev: 78.5 ms +- 0.6 ms -> 78.3 ms +- 0.7 ms: 1.00x faster
    Not significant
    ### deepcopy ###
    Mean +- std dev: 301 us +- 2 us -> 295 us +- 3 us: 1.02x faster
    Significant (t=12.92)
    ### deepcopy_memo ###
    Mean +- std dev: 35.0 us +- 0.5 us -> 34.2 us +- 0.4 us: 1.02x faster
    Significant (t=10.33)
    ### deepcopy_reduce ###
    Mean +- std dev: 3.09 us +- 0.04 us -> 3.08 us +- 0.07 us: 1.01x faster
    Not significant
    ### deltablue ###
    Mean +- std dev: 4.03 ms +- 0.03 ms -> 3.96 ms +- 0.03 ms: 1.02x faster
    Not significant
    ### docutils ###
    Mean +- std dev: 2.38 sec +- 0.01 sec -> 2.34 sec +- 0.02 sec: 1.02x faster
    Not significant
    ### dulwich_log ###
    Mean +- std dev: 66.5 ms +- 0.3 ms -> 66.6 ms +- 0.3 ms: 1.00x slower
    Not significant
    ### fannkuch ###
    Mean +- std dev: 485 ms +- 6 ms -> 444 ms +- 3 ms: 1.09x faster
    Significant (t=49.21)
    ### float ###
    Mean +- std dev: 77.6 ms +- 1.1 ms -> 77.9 ms +- 1.0 ms: 1.00x slower
    Not significant
    ### gc_traversal ###
    Mean +- std dev: 3.32 ms +- 0.05 ms -> 3.78 ms +- 0.17 ms: 1.14x slower
    Significant (t=-19.76)
    ### generators ###
    Mean +- std dev: 32.9 ms +- 0.4 ms -> 32.0 ms +- 0.3 ms: 1.03x faster
    Significant (t=15.63)
    ### genshi_text ###
    Mean +- std dev: 25.6 ms +- 0.3 ms -> 24.9 ms +- 0.2 ms: 1.03x faster
    Significant (t=15.22)
    ### genshi_xml ###
    Mean +- std dev: 61.7 ms +- 0.4 ms -> 62.5 ms +- 0.5 ms: 1.01x slower
    Not significant
    ### go ###
    Mean +- std dev: 146 ms +- 1 ms -> 141 ms +- 1 ms: 1.03x faster
    Significant (t=26.05)
    ### hexiom ###
    Mean +- std dev: 7.65 ms +- 0.12 ms -> 7.31 ms +- 0.03 ms: 1.05x faster
    Significant (t=20.51)
    ### html5lib ###
    Mean +- std dev: 65.3 ms +- 0.6 ms -> 64.6 ms +- 0.5 ms: 1.01x faster
    Not significant
    ### json_dumps ###
    Mean +- std dev: 11.7 ms +- 0.1 ms -> 11.7 ms +- 0.2 ms: 1.00x slower
    Not significant
    ### json_loads ###
    Mean +- std dev: 22.2 us +- 0.2 us -> 22.9 us +- 0.2 us: 1.03x slower
    Significant (t=-15.05)
    ### logging_format ###
    Mean +- std dev: 8.38 us +- 0.12 us -> 8.47 us +- 0.17 us: 1.01x slower
    Not significant
    ### logging_silent ###
    Mean +- std dev: 120 ns +- 2 ns -> 111 ns +- 2 ns: 1.09x faster
    Significant (t=25.41)
    ### logging_simple ###
    Mean +- std dev: 7.56 us +- 0.12 us -> 7.56 us +- 0.11 us: 1.00x faster
    Not significant
    ### mako ###
    Mean +- std dev: 12.9 ms +- 0.2 ms -> 12.8 ms +- 0.1 ms: 1.01x faster
    Not significant
    ### mdp ###
    Mean +- std dev: 2.48 sec +- 0.07 sec -> 2.47 sec +- 0.04 sec: 1.00x faster
    Not significant
    ### meteor_contest ###
    Mean +- std dev: 105 ms +- 1 ms -> 105 ms +- 1 ms: 1.00x slower
    Not significant
    ### nbody ###
    Mean +- std dev: 114 ms +- 1 ms -> 109 ms +- 1 ms: 1.05x faster
    Significant (t=41.07)
    ### nqueens ###
    Mean +- std dev: 92.6 ms +- 1.1 ms -> 91.7 ms +- 0.6 ms: 1.01x faster
    Not significant
    ### pathlib ###
    Mean +- std dev: 19.9 ms +- 0.1 ms -> 19.9 ms +- 0.1 ms: 1.00x faster
    Not significant
    ### pickle ###
    Mean +- std dev: 11.0 us +- 0.1 us -> 11.0 us +- 0.1 us: 1.00x slower
    Not significant
    ### pickle_dict ###
    Mean +- std dev: 24.1 us +- 0.5 us -> 23.5 us +- 1.2 us: 1.03x faster
    Significant (t=3.74)
    ### pickle_list ###
    Mean +- std dev: 3.88 us +- 0.11 us -> 3.84 us +- 0.05 us: 1.01x faster
    Not significant
    ### pickle_pure_python ###
    Mean +- std dev: 366 us +- 2 us -> 356 us +- 3 us: 1.03x faster
    Significant (t=21.08)
    ### pidigits ###
    Mean +- std dev: 182 ms +- 1 ms -> 180 ms +- 1 ms: 1.01x faster
    Not significant
    ### pprint_pformat ###
    Mean +- std dev: 1.80 sec +- 0.02 sec -> 1.80 sec +- 0.02 sec: 1.00x faster
    Not significant
    ### pprint_safe_repr ###
    Mean +- std dev: 860 ms +- 9 ms -> 861 ms +- 8 ms: 1.00x slower
    Not significant
    ### pyflate ###
    Mean +- std dev: 488 ms +- 4 ms -> 465 ms +- 3 ms: 1.05x faster
    Significant (t=37.83)
    ### python_startup ###
    Mean +- std dev: 9.40 ms +- 0.03 ms -> 9.42 ms +- 0.02 ms: 1.00x slower
    Not significant
    ### python_startup_no_site ###
    Mean +- std dev: 6.63 ms +- 0.02 ms -> 6.64 ms +- 0.03 ms: 1.00x slower
    Not significant
    ### raytrace ###
    Mean +- std dev: 304 ms +- 3 ms -> 306 ms +- 4 ms: 1.01x slower
    Not significant
    ### regex_compile ###
    Mean +- std dev: 149 ms +- 1 ms -> 147 ms +- 1 ms: 1.02x faster
    Not significant
    ### regex_dna ###
    Mean +- std dev: 158 ms +- 2 ms -> 164 ms +- 2 ms: 1.03x slower
    Significant (t=-16.73)
    ### regex_effbot ###
    Mean +- std dev: 2.65 ms +- 0.06 ms -> 2.76 ms +- 0.06 ms: 1.04x slower
    Significant (t=-9.73)
    ### regex_v8 ###
    Mean +- std dev: 22.9 ms +- 0.3 ms -> 22.9 ms +- 0.4 ms: 1.00x faster
    Not significant
    ### richards ###
    Mean +- std dev: 58.1 ms +- 0.7 ms -> 55.3 ms +- 0.7 ms: 1.05x faster
    Significant (t=21.71)
    ### richards_super ###
    Mean +- std dev: 66.1 ms +- 0.8 ms -> 63.0 ms +- 0.6 ms: 1.05x faster
    Significant (t=23.11)
    ### scimark_fft ###
    Mean +- std dev: 404 ms +- 5 ms -> 396 ms +- 7 ms: 1.02x faster
    Significant (t=7.51)
    ### scimark_lu ###
    Mean +- std dev: 127 ms +- 3 ms -> 124 ms +- 1 ms: 1.02x faster
    Not significant
    ### scimark_monte_carlo ###
    Mean +- std dev: 74.6 ms +- 0.8 ms -> 71.7 ms +- 1.0 ms: 1.04x faster
    Significant (t=18.09)
    ### scimark_sor ###
    Mean +- std dev: 157 ms +- 1 ms -> 155 ms +- 2 ms: 1.01x faster
    Not significant
    ### scimark_sparse_mat_mult ###
    Mean +- std dev: 5.63 ms +- 0.07 ms -> 5.50 ms +- 0.07 ms: 1.02x faster
    Significant (t=9.74)
    ### spectral_norm ###
    Mean +- std dev: 120 ms +- 2 ms -> 115 ms +- 1 ms: 1.04x faster
    Significant (t=20.82)
    ### sqlglot_normalize ###
    Mean +- std dev: 119 ms +- 1 ms -> 118 ms +- 1 ms: 1.00x faster
    Not significant
    ### telco ###
    Mean +- std dev: 7.65 ms +- 0.13 ms -> 7.64 ms +- 0.18 ms: 1.00x faster
    Not significant
    ### tomli_loads ###
    Mean +- std dev: 2.81 sec +- 0.03 sec -> 2.69 sec +- 0.02 sec: 1.04x faster
    Significant (t=26.73)
    ### typing_runtime_protocols ###
    Mean +- std dev: 177 us +- 3 us -> 179 us +- 4 us: 1.01x slower
    Not significant
    ### unpack_sequence ###
    Mean +- std dev: 44.5 ns +- 1.0 ns -> 42.2 ns +- 0.3 ns: 1.05x faster
    Significant (t=16.90)
    ### unpickle ###
    Mean +- std dev: 12.7 us +- 0.1 us -> 13.2 us +- 0.1 us: 1.04x slower
    Significant (t=-21.58)
    ### unpickle_list ###
    Mean +- std dev: 4.09 us +- 0.11 us -> 4.00 us +- 0.04 us: 1.02x faster
    Significant (t=6.09)
    ### unpickle_pure_python ###
    Mean +- std dev: 251 us +- 2 us -> 244 us +- 2 us: 1.03x faster
    Significant (t=18.64)
    ### xml_etree_generate ###
    Mean +- std dev: 88.7 ms +- 0.5 ms -> 89.9 ms +- 0.8 ms: 1.01x slower
    Not significant
    ### xml_etree_iterparse ###
    Mean +- std dev: 97.2 ms +- 1.1 ms -> 96.8 ms +- 0.9 ms: 1.00x faster
    Not significant
    ### xml_etree_parse ###
    Mean +- std dev: 150 ms +- 3 ms -> 142 ms +- 2 ms: 1.06x faster
    Significant (t=16.24)
    ### xml_etree_process ###
    Mean +- std dev: 63.3 ms +- 0.5 ms -> 63.7 ms +- 0.4 ms: 1.01x slower
    Not significant
    ```

    ```
    ❯ uvx pyperformance compare bolt-baseline.json bolt-split-all-cold.json
    bolt-baseline.json
    ==================
    Performance version: 1.11.0
    Report on Linux-6.1.0-26-amd64-x86_64-with-glibc2.36
    Number of logical CPUs: 12
    Start date: 2025-01-10 21:55:03.035465
    End date: 2025-01-10 22:32:07.098269
    bolt-split-all-cold.json
    ========================
    Performance version: 1.11.0
    Report on Linux-6.1.0-26-amd64-x86_64-with-glibc2.36
    Number of logical CPUs: 12
    Start date: 2025-01-10 23:47:34.385198
    End date: 2025-01-11 00:24:28.147446
    ### 2to3 ###
    Mean +- std dev: 266 ms +- 1 ms -> 260 ms +- 1 ms: 1.02x faster
    Significant (t=38.19)
    ### async_generators ###
    Mean +- std dev: 437 ms +- 4 ms -> 434 ms +- 5 ms: 1.01x faster
    Not significant
    ### asyncio_tcp ###
    Mean +- std dev: 352 ms +- 4 ms -> 371 ms +- 3 ms: 1.06x slower
    Significant (t=-34.70)
    ### asyncio_tcp_ssl ###
    Mean +- std dev: 1.38 sec +- 0.01 sec -> 1.38 sec +- 0.01 sec: 1.00x slower
    Not significant
    ### asyncio_websockets ###
    Mean +- std dev: 502 ms +- 15 ms -> 505 ms +- 7 ms: 1.01x slower
    Not significant
    ### bench_mp_pool ###
    Mean +- std dev: 919 ms +- 613 ms -> 339 ms +- 221 ms: 2.71x faster
    Significant (t=6.89)
    ### bench_thread_pool ###
    Mean +- std dev: 974 us +- 39 us -> 974 us +- 40 us: 1.00x slower
    Not significant
    ### chaos ###
    Mean +- std dev: 72.2 ms +- 0.6 ms -> 71.2 ms +- 0.5 ms: 1.02x faster
    Not significant
    ### comprehensions ###
    Mean +- std dev: 19.6 us +- 0.2 us -> 18.8 us +- 0.2 us: 1.04x faster
    Significant (t=24.98)
    ### coroutines ###
    Mean +- std dev: 26.9 ms +- 0.2 ms -> 23.7 ms +- 0.2 ms: 1.13x faster
    Significant (t=88.04)
    ### create_gc_cycles ###
    Mean +- std dev: 969 us +- 6 us -> 1030 us +- 7 us: 1.06x slower
    Significant (t=-51.00)
    ### crypto_pyaes ###
    Mean +- std dev: 78.5 ms +- 0.6 ms -> 76.4 ms +- 0.6 ms: 1.03x faster
    Significant (t=18.52)
    ### deepcopy ###
    Mean +- std dev: 301 us +- 2 us -> 290 us +- 3 us: 1.04x faster
    Significant (t=21.89)
    ### deepcopy_memo ###
    Mean +- std dev: 35.0 us +- 0.5 us -> 34.9 us +- 0.4 us: 1.00x faster
    Not significant
    ### deepcopy_reduce ###
    Mean +- std dev: 3.09 us +- 0.04 us -> 3.01 us +- 0.05 us: 1.03x faster
    Significant (t=8.88)
    ### deltablue ###
    Mean +- std dev: 4.03 ms +- 0.03 ms -> 3.85 ms +- 0.04 ms: 1.05x faster
    Significant (t=28.18)
    ### docutils ###
    Mean +- std dev: 2.38 sec +- 0.01 sec -> 2.31 sec +- 0.02 sec: 1.03x faster
    Significant (t=22.87)
    ### dulwich_log ###
    Mean +- std dev: 66.5 ms +- 0.3 ms -> 65.5 ms +- 0.2 ms: 1.02x faster
    Not significant
    ### fannkuch ###
    Mean +- std dev: 485 ms +- 6 ms -> 430 ms +- 3 ms: 1.13x faster
    Significant (t=65.12)
    ### float ###
    Mean +- std dev: 77.6 ms +- 1.1 ms -> 77.5 ms +- 0.8 ms: 1.00x faster
    Not significant
    ### gc_traversal ###
    Mean +- std dev: 3.32 ms +- 0.05 ms -> 3.99 ms +- 0.17 ms: 1.20x slower
    Significant (t=-29.94)
    ### generators ###
    Mean +- std dev: 32.9 ms +- 0.4 ms -> 32.1 ms +- 0.4 ms: 1.03x faster
    Significant (t=12.05)
    ### genshi_text ###
    Mean +- std dev: 25.6 ms +- 0.3 ms -> 25.1 ms +- 0.3 ms: 1.02x faster
    Significant (t=11.72)
    ### genshi_xml ###
    Mean +- std dev: 61.7 ms +- 0.4 ms -> 60.8 ms +- 0.4 ms: 1.01x faster
    Not significant
    ### go ###
    Mean +- std dev: 146 ms +- 1 ms -> 141 ms +- 1 ms: 1.04x faster
    Significant (t=32.95)
    ### hexiom ###
    Mean +- std dev: 7.65 ms +- 0.12 ms -> 7.27 ms +- 0.10 ms: 1.05x faster
    Significant (t=19.01)
    ### html5lib ###
    Mean +- std dev: 65.3 ms +- 0.6 ms -> 64.0 ms +- 0.5 ms: 1.02x faster
    Significant (t=13.88)
    ### json_dumps ###
    Mean +- std dev: 11.7 ms +- 0.1 ms -> 11.4 ms +- 0.1 ms: 1.03x faster
    Significant (t=15.45)
    ### json_loads ###
    Mean +- std dev: 22.2 us +- 0.2 us -> 23.2 us +- 0.3 us: 1.05x slower
    Significant (t=-22.31)
    ### logging_format ###
    Mean +- std dev: 8.38 us +- 0.12 us -> 8.30 us +- 0.10 us: 1.01x faster
    Not significant
    ### logging_silent ###
    Mean +- std dev: 120 ns +- 2 ns -> 114 ns +- 2 ns: 1.06x faster
    Significant (t=16.89)
    ### logging_simple ###
    Mean +- std dev: 7.56 us +- 0.12 us -> 7.67 us +- 0.14 us: 1.01x slower
    Not significant
    ### mako ###
    Mean +- std dev: 12.9 ms +- 0.2 ms -> 12.4 ms +- 0.2 ms: 1.04x faster
    Significant (t=14.24)
    ### mdp ###
    Mean +- std dev: 2.48 sec +- 0.07 sec -> 2.49 sec +- 0.03 sec: 1.00x slower
    Not significant
    ### meteor_contest ###
    Mean +- std dev: 105 ms +- 1 ms -> 102 ms +- 0 ms: 1.02x faster
    Significant (t=23.79)
    ### nbody ###
    Mean +- std dev: 114 ms +- 1 ms -> 117 ms +- 1 ms: 1.02x slower
    Not significant
    ### nqueens ###
    Mean +- std dev: 92.6 ms +- 1.1 ms -> 88.9 ms +- 0.6 ms: 1.04x faster
    Significant (t=23.06)
    ### pathlib ###
    Mean +- std dev: 19.9 ms +- 0.1 ms -> 19.8 ms +- 0.1 ms: 1.01x faster
    Not significant
    ### pickle ###
    Mean +- std dev: 11.0 us +- 0.1 us -> 10.7 us +- 0.1 us: 1.03x faster
    Significant (t=11.29)
    ### pickle_dict ###
    Mean +- std dev: 24.1 us +- 0.5 us -> 22.9 us +- 0.4 us: 1.05x faster
    Significant (t=14.70)
    ### pickle_list ###
    Mean +- std dev: 3.88 us +- 0.11 us -> 3.84 us +- 0.05 us: 1.01x faster
    Not significant
    ### pickle_pure_python ###
    Mean +- std dev: 366 us +- 2 us -> 361 us +- 2 us: 1.01x faster
    Not significant
    ### pidigits ###
    Mean +- std dev: 182 ms +- 1 ms -> 182 ms +- 1 ms: 1.00x faster
    Not significant
    ### pprint_pformat ###
    Mean +- std dev: 1.80 sec +- 0.02 sec -> 1.78 sec +- 0.01 sec: 1.01x faster
    Not significant
    ### pprint_safe_repr ###
    Mean +- std dev: 860 ms +- 9 ms -> 853 ms +- 5 ms: 1.01x faster
    Not significant
    ### pyflate ###
    Mean +- std dev: 488 ms +- 4 ms -> 467 ms +- 2 ms: 1.04x faster
    Significant (t=37.37)
    ### python_startup ###
    Mean +- std dev: 9.40 ms +- 0.03 ms -> 9.41 ms +- 0.02 ms: 1.00x slower
    Not significant
    ### python_startup_no_site ###
    Mean +- std dev: 6.63 ms +- 0.02 ms -> 6.63 ms +- 0.02 ms: 1.00x slower
    Not significant
    ### raytrace ###
    Mean +- std dev: 304 ms +- 3 ms -> 304 ms +- 2 ms: 1.00x slower
    Not significant
    ### regex_compile ###
    Mean +- std dev: 149 ms +- 1 ms -> 144 ms +- 1 ms: 1.03x faster
    Significant (t=33.15)
    ### regex_dna ###
    Mean +- std dev: 158 ms +- 2 ms -> 154 ms +- 2 ms: 1.03x faster
    Significant (t=10.70)
    ### regex_effbot ###
    Mean +- std dev: 2.65 ms +- 0.06 ms -> 2.70 ms +- 0.05 ms: 1.02x slower
    Not significant
    ### regex_v8 ###
    Mean +- std dev: 22.9 ms +- 0.3 ms -> 22.9 ms +- 0.3 ms: 1.00x faster
    Not significant
    ### richards ###
    Mean +- std dev: 58.1 ms +- 0.7 ms -> 55.3 ms +- 0.6 ms: 1.05x faster
    Significant (t=22.69)
    ### richards_super ###
    Mean +- std dev: 66.1 ms +- 0.8 ms -> 61.5 ms +- 0.6 ms: 1.07x faster
    Significant (t=33.47)
    ### scimark_fft ###
    Mean +- std dev: 404 ms +- 5 ms -> 400 ms +- 3 ms: 1.01x faster
    Not significant
    ### scimark_lu ###
    Mean +- std dev: 127 ms +- 3 ms -> 123 ms +- 1 ms: 1.03x faster
    Significant (t=10.41)
    ### scimark_monte_carlo ###
    Mean +- std dev: 74.6 ms +- 0.8 ms -> 71.6 ms +- 0.7 ms: 1.04x faster
    Significant (t=21.67)
    ### scimark_sor ###
    Mean +- std dev: 157 ms +- 1 ms -> 153 ms +- 2 ms: 1.02x faster
    Significant (t=12.70)
    ### scimark_sparse_mat_mult ###
    Mean +- std dev: 5.63 ms +- 0.07 ms -> 5.37 ms +- 0.08 ms: 1.05x faster
    Significant (t=18.34)
    ### spectral_norm ###
    Mean +- std dev: 120 ms +- 2 ms -> 116 ms +- 1 ms: 1.04x faster
    Significant (t=15.84)
    ### sqlglot_normalize ###
    Mean +- std dev: 119 ms +- 1 ms -> 116 ms +- 1 ms: 1.02x faster
    Significant (t=14.01)
    ### telco ###
    Mean +- std dev: 7.65 ms +- 0.13 ms -> 7.38 ms +- 0.06 ms: 1.04x faster
    Significant (t=14.17)
    ### tomli_loads ###
    Mean +- std dev: 2.81 sec +- 0.03 sec -> 2.70 sec +- 0.02 sec: 1.04x faster
    Significant (t=23.08)
    ### typing_runtime_protocols ###
    Mean +- std dev: 177 us +- 3 us -> 176 us +- 3 us: 1.00x faster
    Not significant
    ### unpack_sequence ###
    Mean +- std dev: 44.5 ns +- 1.0 ns -> 48.0 ns +- 1.6 ns: 1.08x slower
    Significant (t=-14.83)
    ### unpickle ###
    Mean +- std dev: 12.7 us +- 0.1 us -> 13.0 us +- 0.1 us: 1.02x slower
    Significant (t=-12.79)
    ### unpickle_list ###
    Mean +- std dev: 4.09 us +- 0.11 us -> 4.54 us +- 0.04 us: 1.11x slower
    Significant (t=-30.29)
    ### unpickle_pure_python ###
    Mean +- std dev: 251 us +- 2 us -> 248 us +- 2 us: 1.01x faster
    Not significant
    ### xml_etree_generate ###
    Mean +- std dev: 88.7 ms +- 0.5 ms -> 89.2 ms +- 0.6 ms: 1.01x slower
    Not significant
    ### xml_etree_iterparse ###
    Mean +- std dev: 97.2 ms +- 1.1 ms -> 94.1 ms +- 0.5 ms: 1.03x faster
    Significant (t=20.96)
    ### xml_etree_parse ###
    Mean +- std dev: 150 ms +- 3 ms -> 139 ms +- 2 ms: 1.08x faster
    Significant (t=22.02)
    ### xml_etree_process ###
    Mean +- std dev: 63.3 ms +- 0.5 ms -> 63.1 ms +- 0.5 ms: 1.00x faster
    Not significant
    ```
  2. zanieb created this gist Jan 11, 2025.
    850 changes: 850 additions & 0 deletions bolt-benches.md
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,850 @@
    On a Hetzner bare metal machine with AMD Ryzen 5 3600, 64 GB DDR4 RAM, 2 x 512 GB NVMe SSD

    ```
    ❯ uvx pyperf compare_to bolt-baseline.json bolt-hugify.json bolt-cdsplit.json bolt-split-all-cold.json
    Benchmarks with tag 'apps':
    ===========================
    2to3
    ====
    Mean +- std dev: [bolt-baseline] 266 ms +- 1 ms -> [bolt-hugify] 258 ms +- 1 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 266 ms +- 1 ms -> [bolt-cdsplit] 263 ms +- 1 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 266 ms +- 1 ms -> [bolt-split-all-cold] 260 ms +- 1 ms: 1.02x faster
    docutils
    ========
    Mean +- std dev: [bolt-baseline] 2.38 sec +- 0.01 sec -> [bolt-hugify] 2.30 sec +- 0.02 sec: 1.03x faster
    Mean +- std dev: [bolt-baseline] 2.38 sec +- 0.01 sec -> [bolt-cdsplit] 2.34 sec +- 0.02 sec: 1.02x faster
    Mean +- std dev: [bolt-baseline] 2.38 sec +- 0.01 sec -> [bolt-split-all-cold] 2.31 sec +- 0.02 sec: 1.03x faster
    html5lib
    ========
    Mean +- std dev: [bolt-baseline] 65.3 ms +- 0.6 ms -> [bolt-hugify] 63.4 ms +- 0.7 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 65.3 ms +- 0.6 ms -> [bolt-cdsplit] 64.6 ms +- 0.5 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 65.3 ms +- 0.6 ms -> [bolt-split-all-cold] 64.0 ms +- 0.5 ms: 1.02x faster
    Geometric mean
    ==============
    bolt-hugify: 1.03x faster
    bolt-cdsplit: 1.01x faster
    bolt-split-all-cold: 1.02x faster
    Benchmarks with tag 'math':
    ===========================
    float
    =====
    Mean +- std dev: [bolt-baseline] 77.6 ms +- 1.1 ms -> [bolt-hugify] 75.4 ms +- 1.2 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 77.6 ms +- 1.1 ms -> [bolt-cdsplit] 77.9 ms +- 1.0 ms: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 77.6 ms +- 1.1 ms -> [bolt-split-all-cold] 77.5 ms +- 0.8 ms: 1.00x faster
    Not significant!
    nbody
    =====
    Mean +- std dev: [bolt-baseline] 114 ms +- 1 ms -> [bolt-hugify] 109 ms +- 1 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 114 ms +- 1 ms -> [bolt-cdsplit] 109 ms +- 1 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 114 ms +- 1 ms -> [bolt-split-all-cold] 117 ms +- 1 ms: 1.02x slower
    pidigits
    ========
    Mean +- std dev: [bolt-baseline] 182 ms +- 1 ms -> [bolt-hugify] 182 ms +- 1 ms: 1.00x faster
    Mean +- std dev: [bolt-baseline] 182 ms +- 1 ms -> [bolt-cdsplit] 180 ms +- 1 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 182 ms +- 1 ms -> [bolt-split-all-cold] 182 ms +- 1 ms: 1.00x faster
    Geometric mean
    ==============
    bolt-hugify: 1.03x faster
    bolt-cdsplit: 1.02x faster
    bolt-split-all-cold: 1.00x slower
    Benchmarks with tag 'regex':
    ============================
    regex_compile
    =============
    Mean +- std dev: [bolt-baseline] 149 ms +- 1 ms -> [bolt-hugify] 141 ms +- 1 ms: 1.06x faster
    Mean +- std dev: [bolt-baseline] 149 ms +- 1 ms -> [bolt-cdsplit] 147 ms +- 1 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 149 ms +- 1 ms -> [bolt-split-all-cold] 144 ms +- 1 ms: 1.03x faster
    regex_dna
    =========
    Mean +- std dev: [bolt-baseline] 158 ms +- 2 ms -> [bolt-hugify] 154 ms +- 2 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 158 ms +- 2 ms -> [bolt-cdsplit] 164 ms +- 2 ms: 1.03x slower
    Mean +- std dev: [bolt-baseline] 158 ms +- 2 ms -> [bolt-split-all-cold] 154 ms +- 2 ms: 1.03x faster
    regex_effbot
    ============
    Mean +- std dev: [bolt-baseline] 2.65 ms +- 0.06 ms -> [bolt-hugify] 2.65 ms +- 0.06 ms: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 2.65 ms +- 0.06 ms -> [bolt-cdsplit] 2.76 ms +- 0.06 ms: 1.04x slower
    Mean +- std dev: [bolt-baseline] 2.65 ms +- 0.06 ms -> [bolt-split-all-cold] 2.70 ms +- 0.05 ms: 1.02x slower
    regex_v8
    ========
    Mean +- std dev: [bolt-baseline] 22.9 ms +- 0.3 ms -> [bolt-hugify] 22.1 ms +- 0.2 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 22.9 ms +- 0.3 ms -> [bolt-cdsplit] 22.9 ms +- 0.4 ms: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 22.9 ms +- 0.3 ms -> [bolt-split-all-cold] 22.9 ms +- 0.3 ms: 1.00x faster
    Not significant!
    Geometric mean
    ==============
    bolt-hugify: 1.03x faster
    bolt-cdsplit: 1.01x slower
    bolt-split-all-cold: 1.01x faster
    Benchmarks with tag 'serialize':
    ================================
    json_dumps
    ==========
    Mean +- std dev: [bolt-baseline] 11.7 ms +- 0.1 ms -> [bolt-hugify] 11.1 ms +- 0.2 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 11.7 ms +- 0.1 ms -> [bolt-cdsplit] 11.7 ms +- 0.2 ms: 1.00x slower
    Mean +- std dev: [bolt-baseline] 11.7 ms +- 0.1 ms -> [bolt-split-all-cold] 11.4 ms +- 0.1 ms: 1.03x faster
    json_loads
    ==========
    Mean +- std dev: [bolt-baseline] 22.2 us +- 0.2 us -> [bolt-hugify] 21.6 us +- 0.3 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 22.2 us +- 0.2 us -> [bolt-cdsplit] 22.9 us +- 0.2 us: 1.03x slower
    Mean +- std dev: [bolt-baseline] 22.2 us +- 0.2 us -> [bolt-split-all-cold] 23.2 us +- 0.3 us: 1.05x slower
    pickle
    ======
    Mean +- std dev: [bolt-baseline] 11.0 us +- 0.1 us -> [bolt-hugify] 10.7 us +- 0.2 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 11.0 us +- 0.1 us -> [bolt-cdsplit] 11.0 us +- 0.1 us: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 11.0 us +- 0.1 us -> [bolt-split-all-cold] 10.7 us +- 0.1 us: 1.03x faster
    pickle_dict
    ===========
    Mean +- std dev: [bolt-baseline] 24.1 us +- 0.5 us -> [bolt-hugify] 24.0 us +- 0.2 us: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 24.1 us +- 0.5 us -> [bolt-cdsplit] 23.5 us +- 1.2 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 24.1 us +- 0.5 us -> [bolt-split-all-cold] 22.9 us +- 0.4 us: 1.05x faster
    pickle_list
    ===========
    Mean +- std dev: [bolt-baseline] 3.88 us +- 0.11 us -> [bolt-hugify] 3.87 us +- 0.09 us: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 3.88 us +- 0.11 us -> [bolt-cdsplit] 3.84 us +- 0.05 us: 1.01x faster
    Mean +- std dev: [bolt-baseline] 3.88 us +- 0.11 us -> [bolt-split-all-cold] 3.84 us +- 0.05 us: 1.01x faster
    pickle_pure_python
    ==================
    Mean +- std dev: [bolt-baseline] 366 us +- 2 us -> [bolt-hugify] 348 us +- 4 us: 1.05x faster
    Mean +- std dev: [bolt-baseline] 366 us +- 2 us -> [bolt-cdsplit] 356 us +- 3 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 366 us +- 2 us -> [bolt-split-all-cold] 361 us +- 2 us: 1.01x faster
    tomli_loads
    ===========
    Mean +- std dev: [bolt-baseline] 2.81 sec +- 0.03 sec -> [bolt-hugify] 2.58 sec +- 0.02 sec: 1.09x faster
    Mean +- std dev: [bolt-baseline] 2.81 sec +- 0.03 sec -> [bolt-cdsplit] 2.69 sec +- 0.02 sec: 1.04x faster
    Mean +- std dev: [bolt-baseline] 2.81 sec +- 0.03 sec -> [bolt-split-all-cold] 2.70 sec +- 0.02 sec: 1.04x faster
    unpickle
    ========
    Mean +- std dev: [bolt-baseline] 12.7 us +- 0.1 us -> [bolt-hugify] 12.8 us +- 0.1 us: 1.00x slower
    Mean +- std dev: [bolt-baseline] 12.7 us +- 0.1 us -> [bolt-cdsplit] 13.2 us +- 0.1 us: 1.04x slower
    Mean +- std dev: [bolt-baseline] 12.7 us +- 0.1 us -> [bolt-split-all-cold] 13.0 us +- 0.1 us: 1.02x slower
    unpickle_list
    =============
    Mean +- std dev: [bolt-baseline] 4.09 us +- 0.11 us -> [bolt-hugify] 4.01 us +- 0.07 us: 1.02x faster
    Mean +- std dev: [bolt-baseline] 4.09 us +- 0.11 us -> [bolt-cdsplit] 4.00 us +- 0.04 us: 1.02x faster
    Mean +- std dev: [bolt-baseline] 4.09 us +- 0.11 us -> [bolt-split-all-cold] 4.54 us +- 0.04 us: 1.11x slower
    unpickle_pure_python
    ====================
    Mean +- std dev: [bolt-baseline] 251 us +- 2 us -> [bolt-hugify] 240 us +- 3 us: 1.05x faster
    Mean +- std dev: [bolt-baseline] 251 us +- 2 us -> [bolt-cdsplit] 244 us +- 2 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 251 us +- 2 us -> [bolt-split-all-cold] 248 us +- 2 us: 1.01x faster
    xml_etree_parse
    ===============
    Mean +- std dev: [bolt-baseline] 150 ms +- 3 ms -> [bolt-hugify] 148 ms +- 2 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 150 ms +- 3 ms -> [bolt-cdsplit] 142 ms +- 2 ms: 1.06x faster
    Mean +- std dev: [bolt-baseline] 150 ms +- 3 ms -> [bolt-split-all-cold] 139 ms +- 2 ms: 1.08x faster
    xml_etree_iterparse
    ===================
    Mean +- std dev: [bolt-baseline] 97.2 ms +- 1.1 ms -> [bolt-hugify] 95.7 ms +- 1.2 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 97.2 ms +- 1.1 ms -> [bolt-cdsplit] 96.8 ms +- 0.9 ms: 1.00x faster
    Mean +- std dev: [bolt-baseline] 97.2 ms +- 1.1 ms -> [bolt-split-all-cold] 94.1 ms +- 0.5 ms: 1.03x faster
    xml_etree_generate
    ==================
    Mean +- std dev: [bolt-baseline] 88.7 ms +- 0.5 ms -> [bolt-hugify] 85.2 ms +- 0.7 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 88.7 ms +- 0.5 ms -> [bolt-cdsplit] 89.9 ms +- 0.8 ms: 1.01x slower
    Mean +- std dev: [bolt-baseline] 88.7 ms +- 0.5 ms -> [bolt-split-all-cold] 89.2 ms +- 0.6 ms: 1.01x slower
    xml_etree_process
    =================
    Mean +- std dev: [bolt-baseline] 63.3 ms +- 0.5 ms -> [bolt-hugify] 60.0 ms +- 0.6 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 63.3 ms +- 0.5 ms -> [bolt-cdsplit] 63.7 ms +- 0.4 ms: 1.01x slower
    Mean +- std dev: [bolt-baseline] 63.3 ms +- 0.5 ms -> [bolt-split-all-cold] 63.1 ms +- 0.5 ms: 1.00x faster
    Geometric mean
    ==============
    bolt-hugify: 1.03x faster
    bolt-cdsplit: 1.01x faster
    bolt-split-all-cold: 1.01x faster
    Benchmarks with tag 'startup':
    ==============================
    python_startup
    ==============
    Mean +- std dev: [bolt-baseline] 9.40 ms +- 0.03 ms -> [bolt-hugify] 11.4 ms +- 0.0 ms: 1.22x slower
    Mean +- std dev: [bolt-baseline] 9.40 ms +- 0.03 ms -> [bolt-cdsplit] 9.42 ms +- 0.02 ms: 1.00x slower
    Mean +- std dev: [bolt-baseline] 9.40 ms +- 0.03 ms -> [bolt-split-all-cold] 9.41 ms +- 0.02 ms: 1.00x slower
    python_startup_no_site
    ======================
    Mean +- std dev: [bolt-baseline] 6.63 ms +- 0.02 ms -> [bolt-hugify] 8.66 ms +- 0.04 ms: 1.31x slower
    Mean +- std dev: [bolt-baseline] 6.63 ms +- 0.02 ms -> [bolt-cdsplit] 6.64 ms +- 0.03 ms: 1.00x slower
    Mean +- std dev: [bolt-baseline] 6.63 ms +- 0.02 ms -> [bolt-split-all-cold] 6.63 ms +- 0.02 ms: 1.00x slower
    Not significant!
    Geometric mean
    ==============
    bolt-hugify: 1.26x slower
    bolt-cdsplit: 1.00x slower
    bolt-split-all-cold: 1.00x slower
    Benchmarks with tag 'template':
    ===============================
    genshi_text
    ===========
    Mean +- std dev: [bolt-baseline] 25.6 ms +- 0.3 ms -> [bolt-hugify] 24.4 ms +- 0.2 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 25.6 ms +- 0.3 ms -> [bolt-cdsplit] 24.9 ms +- 0.2 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 25.6 ms +- 0.3 ms -> [bolt-split-all-cold] 25.1 ms +- 0.3 ms: 1.02x faster
    genshi_xml
    ==========
    Mean +- std dev: [bolt-baseline] 61.7 ms +- 0.4 ms -> [bolt-hugify] 57.0 ms +- 0.4 ms: 1.08x faster
    Mean +- std dev: [bolt-baseline] 61.7 ms +- 0.4 ms -> [bolt-cdsplit] 62.5 ms +- 0.5 ms: 1.01x slower
    Mean +- std dev: [bolt-baseline] 61.7 ms +- 0.4 ms -> [bolt-split-all-cold] 60.8 ms +- 0.4 ms: 1.01x faster
    mako
    ====
    Mean +- std dev: [bolt-baseline] 12.9 ms +- 0.2 ms -> [bolt-hugify] 12.4 ms +- 0.3 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 12.9 ms +- 0.2 ms -> [bolt-cdsplit] 12.8 ms +- 0.1 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 12.9 ms +- 0.2 ms -> [bolt-split-all-cold] 12.4 ms +- 0.2 ms: 1.04x faster
    Geometric mean
    ==============
    bolt-hugify: 1.06x faster
    bolt-cdsplit: 1.01x faster
    bolt-split-all-cold: 1.03x faster
    All benchmarks:
    ===============
    2to3
    ====
    Mean +- std dev: [bolt-baseline] 266 ms +- 1 ms -> [bolt-hugify] 258 ms +- 1 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 266 ms +- 1 ms -> [bolt-cdsplit] 263 ms +- 1 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 266 ms +- 1 ms -> [bolt-split-all-cold] 260 ms +- 1 ms: 1.02x faster
    async_generators
    ================
    Mean +- std dev: [bolt-baseline] 437 ms +- 4 ms -> [bolt-hugify] 408 ms +- 4 ms: 1.07x faster
    Mean +- std dev: [bolt-baseline] 437 ms +- 4 ms -> [bolt-cdsplit] 430 ms +- 5 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 437 ms +- 4 ms -> [bolt-split-all-cold] 434 ms +- 5 ms: 1.01x faster
    asyncio_tcp
    ===========
    Mean +- std dev: [bolt-baseline] 352 ms +- 4 ms -> [bolt-hugify] 352 ms +- 4 ms: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 352 ms +- 4 ms -> [bolt-cdsplit] 351 ms +- 3 ms: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 352 ms +- 4 ms -> [bolt-split-all-cold] 371 ms +- 3 ms: 1.06x slower
    asyncio_tcp_ssl
    ===============
    Mean +- std dev: [bolt-baseline] 1.38 sec +- 0.01 sec -> [bolt-hugify] 1.37 sec +- 0.01 sec: 1.01x faster
    Mean +- std dev: [bolt-baseline] 1.38 sec +- 0.01 sec -> [bolt-cdsplit] 1.39 sec +- 0.01 sec: 1.01x slower
    Mean +- std dev: [bolt-baseline] 1.38 sec +- 0.01 sec -> [bolt-split-all-cold] 1.38 sec +- 0.01 sec: 1.00x slower
    Not significant!
    chaos
    =====
    Mean +- std dev: [bolt-baseline] 72.2 ms +- 0.6 ms -> [bolt-hugify] 69.6 ms +- 0.7 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 72.2 ms +- 0.6 ms -> [bolt-cdsplit] 73.3 ms +- 0.5 ms: 1.01x slower
    Mean +- std dev: [bolt-baseline] 72.2 ms +- 0.6 ms -> [bolt-split-all-cold] 71.2 ms +- 0.5 ms: 1.02x faster
    comprehensions
    ==============
    Mean +- std dev: [bolt-baseline] 19.6 us +- 0.2 us -> [bolt-hugify] 18.1 us +- 0.1 us: 1.08x faster
    Mean +- std dev: [bolt-baseline] 19.6 us +- 0.2 us -> [bolt-cdsplit] 19.3 us +- 0.2 us: 1.02x faster
    Mean +- std dev: [bolt-baseline] 19.6 us +- 0.2 us -> [bolt-split-all-cold] 18.8 us +- 0.2 us: 1.04x faster
    bench_mp_pool
    =============
    Mean +- std dev: [bolt-baseline] 919 ms +- 613 ms -> [bolt-hugify] 116 ms +- 75 ms: 7.95x faster
    Mean +- std dev: [bolt-baseline] 919 ms +- 613 ms -> [bolt-cdsplit] 616 ms +- 413 ms: 1.49x faster
    Mean +- std dev: [bolt-baseline] 919 ms +- 613 ms -> [bolt-split-all-cold] 339 ms +- 221 ms: 2.71x faster
    bench_thread_pool
    =================
    Mean +- std dev: [bolt-baseline] 974 us +- 39 us -> [bolt-hugify] 947 us +- 38 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 974 us +- 39 us -> [bolt-cdsplit] 980 us +- 36 us: 1.01x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 974 us +- 39 us -> [bolt-split-all-cold] 974 us +- 40 us: 1.00x slower
    Not significant!
    coroutines
    ==========
    Mean +- std dev: [bolt-baseline] 26.9 ms +- 0.2 ms -> [bolt-hugify] 23.0 ms +- 0.2 ms: 1.17x faster
    Mean +- std dev: [bolt-baseline] 26.9 ms +- 0.2 ms -> [bolt-cdsplit] 24.0 ms +- 0.2 ms: 1.12x faster
    Mean +- std dev: [bolt-baseline] 26.9 ms +- 0.2 ms -> [bolt-split-all-cold] 23.7 ms +- 0.2 ms: 1.13x faster
    crypto_pyaes
    ============
    Mean +- std dev: [bolt-baseline] 78.5 ms +- 0.6 ms -> [bolt-hugify] 76.2 ms +- 0.6 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 78.5 ms +- 0.6 ms -> [bolt-cdsplit] 78.3 ms +- 0.7 ms: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 78.5 ms +- 0.6 ms -> [bolt-split-all-cold] 76.4 ms +- 0.6 ms: 1.03x faster
    deepcopy
    ========
    Mean +- std dev: [bolt-baseline] 301 us +- 2 us -> [bolt-hugify] 277 us +- 3 us: 1.09x faster
    Mean +- std dev: [bolt-baseline] 301 us +- 2 us -> [bolt-cdsplit] 295 us +- 3 us: 1.02x faster
    Mean +- std dev: [bolt-baseline] 301 us +- 2 us -> [bolt-split-all-cold] 290 us +- 3 us: 1.04x faster
    deepcopy_reduce
    ===============
    Mean +- std dev: [bolt-baseline] 3.09 us +- 0.04 us -> [bolt-hugify] 2.88 us +- 0.02 us: 1.07x faster
    Mean +- std dev: [bolt-baseline] 3.09 us +- 0.04 us -> [bolt-cdsplit] 3.08 us +- 0.07 us: 1.01x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 3.09 us +- 0.04 us -> [bolt-split-all-cold] 3.01 us +- 0.05 us: 1.03x faster
    deepcopy_memo
    =============
    Mean +- std dev: [bolt-baseline] 35.0 us +- 0.5 us -> [bolt-hugify] 32.8 us +- 0.8 us: 1.07x faster
    Mean +- std dev: [bolt-baseline] 35.0 us +- 0.5 us -> [bolt-cdsplit] 34.2 us +- 0.4 us: 1.02x faster
    Mean +- std dev: [bolt-baseline] 35.0 us +- 0.5 us -> [bolt-split-all-cold] 34.9 us +- 0.4 us: 1.00x faster
    Not significant!
    deltablue
    =========
    Mean +- std dev: [bolt-baseline] 4.03 ms +- 0.03 ms -> [bolt-hugify] 3.86 ms +- 0.04 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 4.03 ms +- 0.03 ms -> [bolt-cdsplit] 3.96 ms +- 0.03 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 4.03 ms +- 0.03 ms -> [bolt-split-all-cold] 3.85 ms +- 0.04 ms: 1.05x faster
    docutils
    ========
    Mean +- std dev: [bolt-baseline] 2.38 sec +- 0.01 sec -> [bolt-hugify] 2.30 sec +- 0.02 sec: 1.03x faster
    Mean +- std dev: [bolt-baseline] 2.38 sec +- 0.01 sec -> [bolt-cdsplit] 2.34 sec +- 0.02 sec: 1.02x faster
    Mean +- std dev: [bolt-baseline] 2.38 sec +- 0.01 sec -> [bolt-split-all-cold] 2.31 sec +- 0.02 sec: 1.03x faster
    dulwich_log
    ===========
    Mean +- std dev: [bolt-baseline] 66.5 ms +- 0.3 ms -> [bolt-hugify] 64.4 ms +- 0.3 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 66.5 ms +- 0.3 ms -> [bolt-cdsplit] 66.6 ms +- 0.3 ms: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 66.5 ms +- 0.3 ms -> [bolt-split-all-cold] 65.5 ms +- 0.2 ms: 1.02x faster
    fannkuch
    ========
    Mean +- std dev: [bolt-baseline] 485 ms +- 6 ms -> [bolt-hugify] 427 ms +- 5 ms: 1.13x faster
    Mean +- std dev: [bolt-baseline] 485 ms +- 6 ms -> [bolt-cdsplit] 444 ms +- 3 ms: 1.09x faster
    Mean +- std dev: [bolt-baseline] 485 ms +- 6 ms -> [bolt-split-all-cold] 430 ms +- 3 ms: 1.13x faster
    float
    =====
    Mean +- std dev: [bolt-baseline] 77.6 ms +- 1.1 ms -> [bolt-hugify] 75.4 ms +- 1.2 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 77.6 ms +- 1.1 ms -> [bolt-cdsplit] 77.9 ms +- 1.0 ms: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 77.6 ms +- 1.1 ms -> [bolt-split-all-cold] 77.5 ms +- 0.8 ms: 1.00x faster
    Not significant!
    create_gc_cycles
    ================
    Mean +- std dev: [bolt-baseline] 969 us +- 6 us -> [bolt-hugify] 971 us +- 6 us: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 969 us +- 6 us -> [bolt-cdsplit] 974 us +- 10 us: 1.00x slower
    Mean +- std dev: [bolt-baseline] 969 us +- 6 us -> [bolt-split-all-cold] 1.03 ms +- 0.01 ms: 1.06x slower
    gc_traversal
    ============
    Mean +- std dev: [bolt-baseline] 3.32 ms +- 0.05 ms -> [bolt-hugify] 3.79 ms +- 0.12 ms: 1.14x slower
    Mean +- std dev: [bolt-baseline] 3.32 ms +- 0.05 ms -> [bolt-cdsplit] 3.78 ms +- 0.17 ms: 1.14x slower
    Mean +- std dev: [bolt-baseline] 3.32 ms +- 0.05 ms -> [bolt-split-all-cold] 3.99 ms +- 0.17 ms: 1.20x slower
    generators
    ==========
    Mean +- std dev: [bolt-baseline] 32.9 ms +- 0.4 ms -> [bolt-hugify] 31.2 ms +- 0.3 ms: 1.06x faster
    Mean +- std dev: [bolt-baseline] 32.9 ms +- 0.4 ms -> [bolt-cdsplit] 32.0 ms +- 0.3 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 32.9 ms +- 0.4 ms -> [bolt-split-all-cold] 32.1 ms +- 0.4 ms: 1.03x faster
    genshi_text
    ===========
    Mean +- std dev: [bolt-baseline] 25.6 ms +- 0.3 ms -> [bolt-hugify] 24.4 ms +- 0.2 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 25.6 ms +- 0.3 ms -> [bolt-cdsplit] 24.9 ms +- 0.2 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 25.6 ms +- 0.3 ms -> [bolt-split-all-cold] 25.1 ms +- 0.3 ms: 1.02x faster
    genshi_xml
    ==========
    Mean +- std dev: [bolt-baseline] 61.7 ms +- 0.4 ms -> [bolt-hugify] 57.0 ms +- 0.4 ms: 1.08x faster
    Mean +- std dev: [bolt-baseline] 61.7 ms +- 0.4 ms -> [bolt-cdsplit] 62.5 ms +- 0.5 ms: 1.01x slower
    Mean +- std dev: [bolt-baseline] 61.7 ms +- 0.4 ms -> [bolt-split-all-cold] 60.8 ms +- 0.4 ms: 1.01x faster
    go
    ==
    Mean +- std dev: [bolt-baseline] 146 ms +- 1 ms -> [bolt-hugify] 136 ms +- 1 ms: 1.07x faster
    Mean +- std dev: [bolt-baseline] 146 ms +- 1 ms -> [bolt-cdsplit] 141 ms +- 1 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 146 ms +- 1 ms -> [bolt-split-all-cold] 141 ms +- 1 ms: 1.04x faster
    hexiom
    ======
    Mean +- std dev: [bolt-baseline] 7.65 ms +- 0.12 ms -> [bolt-hugify] 7.06 ms +- 0.10 ms: 1.08x faster
    Mean +- std dev: [bolt-baseline] 7.65 ms +- 0.12 ms -> [bolt-cdsplit] 7.31 ms +- 0.03 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 7.65 ms +- 0.12 ms -> [bolt-split-all-cold] 7.27 ms +- 0.10 ms: 1.05x faster
    html5lib
    ========
    Mean +- std dev: [bolt-baseline] 65.3 ms +- 0.6 ms -> [bolt-hugify] 63.4 ms +- 0.7 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 65.3 ms +- 0.6 ms -> [bolt-cdsplit] 64.6 ms +- 0.5 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 65.3 ms +- 0.6 ms -> [bolt-split-all-cold] 64.0 ms +- 0.5 ms: 1.02x faster
    json_dumps
    ==========
    Mean +- std dev: [bolt-baseline] 11.7 ms +- 0.1 ms -> [bolt-hugify] 11.1 ms +- 0.2 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 11.7 ms +- 0.1 ms -> [bolt-cdsplit] 11.7 ms +- 0.2 ms: 1.00x slower
    Mean +- std dev: [bolt-baseline] 11.7 ms +- 0.1 ms -> [bolt-split-all-cold] 11.4 ms +- 0.1 ms: 1.03x faster
    json_loads
    ==========
    Mean +- std dev: [bolt-baseline] 22.2 us +- 0.2 us -> [bolt-hugify] 21.6 us +- 0.3 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 22.2 us +- 0.2 us -> [bolt-cdsplit] 22.9 us +- 0.2 us: 1.03x slower
    Mean +- std dev: [bolt-baseline] 22.2 us +- 0.2 us -> [bolt-split-all-cold] 23.2 us +- 0.3 us: 1.05x slower
    logging_format
    ==============
    Mean +- std dev: [bolt-baseline] 8.38 us +- 0.12 us -> [bolt-hugify] 8.21 us +- 0.08 us: 1.02x faster
    Mean +- std dev: [bolt-baseline] 8.38 us +- 0.12 us -> [bolt-cdsplit] 8.47 us +- 0.17 us: 1.01x slower
    Mean +- std dev: [bolt-baseline] 8.38 us +- 0.12 us -> [bolt-split-all-cold] 8.30 us +- 0.10 us: 1.01x faster
    logging_silent
    ==============
    Mean +- std dev: [bolt-baseline] 120 ns +- 2 ns -> [bolt-hugify] 120 ns +- 3 ns: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 120 ns +- 2 ns -> [bolt-cdsplit] 111 ns +- 2 ns: 1.09x faster
    Mean +- std dev: [bolt-baseline] 120 ns +- 2 ns -> [bolt-split-all-cold] 114 ns +- 2 ns: 1.06x faster
    logging_simple
    ==============
    Mean +- std dev: [bolt-baseline] 7.56 us +- 0.12 us -> [bolt-hugify] 7.50 us +- 0.11 us: 1.01x faster
    Mean +- std dev: [bolt-baseline] 7.56 us +- 0.12 us -> [bolt-cdsplit] 7.56 us +- 0.11 us: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 7.56 us +- 0.12 us -> [bolt-split-all-cold] 7.67 us +- 0.14 us: 1.01x slower
    mako
    ====
    Mean +- std dev: [bolt-baseline] 12.9 ms +- 0.2 ms -> [bolt-hugify] 12.4 ms +- 0.3 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 12.9 ms +- 0.2 ms -> [bolt-cdsplit] 12.8 ms +- 0.1 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 12.9 ms +- 0.2 ms -> [bolt-split-all-cold] 12.4 ms +- 0.2 ms: 1.04x faster
    meteor_contest
    ==============
    Mean +- std dev: [bolt-baseline] 105 ms +- 1 ms -> [bolt-hugify] 100 ms +- 1 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 105 ms +- 1 ms -> [bolt-cdsplit] 105 ms +- 1 ms: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 105 ms +- 1 ms -> [bolt-split-all-cold] 102 ms +- 0 ms: 1.02x faster
    nbody
    =====
    Mean +- std dev: [bolt-baseline] 114 ms +- 1 ms -> [bolt-hugify] 109 ms +- 1 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 114 ms +- 1 ms -> [bolt-cdsplit] 109 ms +- 1 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 114 ms +- 1 ms -> [bolt-split-all-cold] 117 ms +- 1 ms: 1.02x slower
    nqueens
    =======
    Mean +- std dev: [bolt-baseline] 92.6 ms +- 1.1 ms -> [bolt-hugify] 86.1 ms +- 0.9 ms: 1.08x faster
    Mean +- std dev: [bolt-baseline] 92.6 ms +- 1.1 ms -> [bolt-cdsplit] 91.7 ms +- 0.6 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 92.6 ms +- 1.1 ms -> [bolt-split-all-cold] 88.9 ms +- 0.6 ms: 1.04x faster
    pathlib
    =======
    Mean +- std dev: [bolt-baseline] 19.9 ms +- 0.1 ms -> [bolt-hugify] 19.3 ms +- 0.1 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 19.9 ms +- 0.1 ms -> [bolt-cdsplit] 19.9 ms +- 0.1 ms: 1.00x faster
    Mean +- std dev: [bolt-baseline] 19.9 ms +- 0.1 ms -> [bolt-split-all-cold] 19.8 ms +- 0.1 ms: 1.01x faster
    pickle
    ======
    Mean +- std dev: [bolt-baseline] 11.0 us +- 0.1 us -> [bolt-hugify] 10.7 us +- 0.2 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 11.0 us +- 0.1 us -> [bolt-cdsplit] 11.0 us +- 0.1 us: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 11.0 us +- 0.1 us -> [bolt-split-all-cold] 10.7 us +- 0.1 us: 1.03x faster
    pickle_dict
    ===========
    Mean +- std dev: [bolt-baseline] 24.1 us +- 0.5 us -> [bolt-hugify] 24.0 us +- 0.2 us: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 24.1 us +- 0.5 us -> [bolt-cdsplit] 23.5 us +- 1.2 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 24.1 us +- 0.5 us -> [bolt-split-all-cold] 22.9 us +- 0.4 us: 1.05x faster
    pickle_list
    ===========
    Mean +- std dev: [bolt-baseline] 3.88 us +- 0.11 us -> [bolt-hugify] 3.87 us +- 0.09 us: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 3.88 us +- 0.11 us -> [bolt-cdsplit] 3.84 us +- 0.05 us: 1.01x faster
    Mean +- std dev: [bolt-baseline] 3.88 us +- 0.11 us -> [bolt-split-all-cold] 3.84 us +- 0.05 us: 1.01x faster
    pickle_pure_python
    ==================
    Mean +- std dev: [bolt-baseline] 366 us +- 2 us -> [bolt-hugify] 348 us +- 4 us: 1.05x faster
    Mean +- std dev: [bolt-baseline] 366 us +- 2 us -> [bolt-cdsplit] 356 us +- 3 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 366 us +- 2 us -> [bolt-split-all-cold] 361 us +- 2 us: 1.01x faster
    pidigits
    ========
    Mean +- std dev: [bolt-baseline] 182 ms +- 1 ms -> [bolt-hugify] 182 ms +- 1 ms: 1.00x faster
    Mean +- std dev: [bolt-baseline] 182 ms +- 1 ms -> [bolt-cdsplit] 180 ms +- 1 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 182 ms +- 1 ms -> [bolt-split-all-cold] 182 ms +- 1 ms: 1.00x faster
    pprint_safe_repr
    ================
    Mean +- std dev: [bolt-baseline] 860 ms +- 9 ms -> [bolt-hugify] 808 ms +- 13 ms: 1.06x faster
    Mean +- std dev: [bolt-baseline] 860 ms +- 9 ms -> [bolt-cdsplit] 861 ms +- 8 ms: 1.00x slower
    Not significant!
    Mean +- std dev: [bolt-baseline] 860 ms +- 9 ms -> [bolt-split-all-cold] 853 ms +- 5 ms: 1.01x faster
    pprint_pformat
    ==============
    Mean +- std dev: [bolt-baseline] 1.80 sec +- 0.02 sec -> [bolt-hugify] 1.70 sec +- 0.02 sec: 1.06x faster
    Mean +- std dev: [bolt-baseline] 1.80 sec +- 0.02 sec -> [bolt-cdsplit] 1.80 sec +- 0.02 sec: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 1.80 sec +- 0.02 sec -> [bolt-split-all-cold] 1.78 sec +- 0.01 sec: 1.01x faster
    pyflate
    =======
    Mean +- std dev: [bolt-baseline] 488 ms +- 4 ms -> [bolt-hugify] 447 ms +- 3 ms: 1.09x faster
    Mean +- std dev: [bolt-baseline] 488 ms +- 4 ms -> [bolt-cdsplit] 465 ms +- 3 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 488 ms +- 4 ms -> [bolt-split-all-cold] 467 ms +- 2 ms: 1.04x faster
    python_startup
    ==============
    Mean +- std dev: [bolt-baseline] 9.40 ms +- 0.03 ms -> [bolt-hugify] 11.4 ms +- 0.0 ms: 1.22x slower
    Mean +- std dev: [bolt-baseline] 9.40 ms +- 0.03 ms -> [bolt-cdsplit] 9.42 ms +- 0.02 ms: 1.00x slower
    Mean +- std dev: [bolt-baseline] 9.40 ms +- 0.03 ms -> [bolt-split-all-cold] 9.41 ms +- 0.02 ms: 1.00x slower
    python_startup_no_site
    ======================
    Mean +- std dev: [bolt-baseline] 6.63 ms +- 0.02 ms -> [bolt-hugify] 8.66 ms +- 0.04 ms: 1.31x slower
    Mean +- std dev: [bolt-baseline] 6.63 ms +- 0.02 ms -> [bolt-cdsplit] 6.64 ms +- 0.03 ms: 1.00x slower
    Mean +- std dev: [bolt-baseline] 6.63 ms +- 0.02 ms -> [bolt-split-all-cold] 6.63 ms +- 0.02 ms: 1.00x slower
    Not significant!
    raytrace
    ========
    Mean +- std dev: [bolt-baseline] 304 ms +- 3 ms -> [bolt-hugify] 299 ms +- 4 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 304 ms +- 3 ms -> [bolt-cdsplit] 306 ms +- 4 ms: 1.01x slower
    Mean +- std dev: [bolt-baseline] 304 ms +- 3 ms -> [bolt-split-all-cold] 304 ms +- 2 ms: 1.00x slower
    Not significant!
    regex_compile
    =============
    Mean +- std dev: [bolt-baseline] 149 ms +- 1 ms -> [bolt-hugify] 141 ms +- 1 ms: 1.06x faster
    Mean +- std dev: [bolt-baseline] 149 ms +- 1 ms -> [bolt-cdsplit] 147 ms +- 1 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 149 ms +- 1 ms -> [bolt-split-all-cold] 144 ms +- 1 ms: 1.03x faster
    regex_dna
    =========
    Mean +- std dev: [bolt-baseline] 158 ms +- 2 ms -> [bolt-hugify] 154 ms +- 2 ms: 1.03x faster
    Mean +- std dev: [bolt-baseline] 158 ms +- 2 ms -> [bolt-cdsplit] 164 ms +- 2 ms: 1.03x slower
    Mean +- std dev: [bolt-baseline] 158 ms +- 2 ms -> [bolt-split-all-cold] 154 ms +- 2 ms: 1.03x faster
    regex_effbot
    ============
    Mean +- std dev: [bolt-baseline] 2.65 ms +- 0.06 ms -> [bolt-hugify] 2.65 ms +- 0.06 ms: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 2.65 ms +- 0.06 ms -> [bolt-cdsplit] 2.76 ms +- 0.06 ms: 1.04x slower
    Mean +- std dev: [bolt-baseline] 2.65 ms +- 0.06 ms -> [bolt-split-all-cold] 2.70 ms +- 0.05 ms: 1.02x slower
    regex_v8
    ========
    Mean +- std dev: [bolt-baseline] 22.9 ms +- 0.3 ms -> [bolt-hugify] 22.1 ms +- 0.2 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 22.9 ms +- 0.3 ms -> [bolt-cdsplit] 22.9 ms +- 0.4 ms: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 22.9 ms +- 0.3 ms -> [bolt-split-all-cold] 22.9 ms +- 0.3 ms: 1.00x faster
    Not significant!
    richards
    ========
    Mean +- std dev: [bolt-baseline] 58.1 ms +- 0.7 ms -> [bolt-hugify] 55.1 ms +- 0.5 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 58.1 ms +- 0.7 ms -> [bolt-cdsplit] 55.3 ms +- 0.7 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 58.1 ms +- 0.7 ms -> [bolt-split-all-cold] 55.3 ms +- 0.6 ms: 1.05x faster
    richards_super
    ==============
    Mean +- std dev: [bolt-baseline] 66.1 ms +- 0.8 ms -> [bolt-hugify] 62.0 ms +- 0.5 ms: 1.07x faster
    Mean +- std dev: [bolt-baseline] 66.1 ms +- 0.8 ms -> [bolt-cdsplit] 63.0 ms +- 0.6 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 66.1 ms +- 0.8 ms -> [bolt-split-all-cold] 61.5 ms +- 0.6 ms: 1.07x faster
    scimark_fft
    ===========
    Mean +- std dev: [bolt-baseline] 404 ms +- 5 ms -> [bolt-hugify] 373 ms +- 4 ms: 1.08x faster
    Mean +- std dev: [bolt-baseline] 404 ms +- 5 ms -> [bolt-cdsplit] 396 ms +- 7 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 404 ms +- 5 ms -> [bolt-split-all-cold] 400 ms +- 3 ms: 1.01x faster
    scimark_lu
    ==========
    Mean +- std dev: [bolt-baseline] 127 ms +- 3 ms -> [bolt-hugify] 119 ms +- 2 ms: 1.06x faster
    Mean +- std dev: [bolt-baseline] 127 ms +- 3 ms -> [bolt-cdsplit] 124 ms +- 1 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 127 ms +- 3 ms -> [bolt-split-all-cold] 123 ms +- 1 ms: 1.03x faster
    scimark_monte_carlo
    ===================
    Mean +- std dev: [bolt-baseline] 74.6 ms +- 0.8 ms -> [bolt-hugify] 69.8 ms +- 0.9 ms: 1.07x faster
    Mean +- std dev: [bolt-baseline] 74.6 ms +- 0.8 ms -> [bolt-cdsplit] 71.7 ms +- 1.0 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 74.6 ms +- 0.8 ms -> [bolt-split-all-cold] 71.6 ms +- 0.7 ms: 1.04x faster
    scimark_sor
    ===========
    Mean +- std dev: [bolt-baseline] 157 ms +- 1 ms -> [bolt-hugify] 147 ms +- 1 ms: 1.07x faster
    Mean +- std dev: [bolt-baseline] 157 ms +- 1 ms -> [bolt-cdsplit] 155 ms +- 2 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 157 ms +- 1 ms -> [bolt-split-all-cold] 153 ms +- 2 ms: 1.02x faster
    scimark_sparse_mat_mult
    =======================
    Mean +- std dev: [bolt-baseline] 5.63 ms +- 0.07 ms -> [bolt-hugify] 5.21 ms +- 0.19 ms: 1.08x faster
    Mean +- std dev: [bolt-baseline] 5.63 ms +- 0.07 ms -> [bolt-cdsplit] 5.50 ms +- 0.07 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 5.63 ms +- 0.07 ms -> [bolt-split-all-cold] 5.37 ms +- 0.08 ms: 1.05x faster
    spectral_norm
    =============
    Mean +- std dev: [bolt-baseline] 120 ms +- 2 ms -> [bolt-hugify] 105 ms +- 1 ms: 1.14x faster
    Mean +- std dev: [bolt-baseline] 120 ms +- 2 ms -> [bolt-cdsplit] 115 ms +- 1 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 120 ms +- 2 ms -> [bolt-split-all-cold] 116 ms +- 1 ms: 1.04x faster
    sqlglot_normalize
    =================
    Mean +- std dev: [bolt-baseline] 119 ms +- 1 ms -> [bolt-hugify] 113 ms +- 1 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 119 ms +- 1 ms -> [bolt-cdsplit] 118 ms +- 1 ms: 1.00x faster
    Mean +- std dev: [bolt-baseline] 119 ms +- 1 ms -> [bolt-split-all-cold] 116 ms +- 1 ms: 1.02x faster
    telco
    =====
    Mean +- std dev: [bolt-baseline] 7.65 ms +- 0.13 ms -> [bolt-hugify] 7.13 ms +- 0.08 ms: 1.07x faster
    Mean +- std dev: [bolt-baseline] 7.65 ms +- 0.13 ms -> [bolt-cdsplit] 7.64 ms +- 0.18 ms: 1.00x faster
    Not significant!
    Mean +- std dev: [bolt-baseline] 7.65 ms +- 0.13 ms -> [bolt-split-all-cold] 7.38 ms +- 0.06 ms: 1.04x faster
    tomli_loads
    ===========
    Mean +- std dev: [bolt-baseline] 2.81 sec +- 0.03 sec -> [bolt-hugify] 2.58 sec +- 0.02 sec: 1.09x faster
    Mean +- std dev: [bolt-baseline] 2.81 sec +- 0.03 sec -> [bolt-cdsplit] 2.69 sec +- 0.02 sec: 1.04x faster
    Mean +- std dev: [bolt-baseline] 2.81 sec +- 0.03 sec -> [bolt-split-all-cold] 2.70 sec +- 0.02 sec: 1.04x faster
    typing_runtime_protocols
    ========================
    Mean +- std dev: [bolt-baseline] 177 us +- 3 us -> [bolt-hugify] 164 us +- 3 us: 1.08x faster
    Mean +- std dev: [bolt-baseline] 177 us +- 3 us -> [bolt-cdsplit] 179 us +- 4 us: 1.01x slower
    Mean +- std dev: [bolt-baseline] 177 us +- 3 us -> [bolt-split-all-cold] 176 us +- 3 us: 1.00x faster
    Not significant!
    unpack_sequence
    ===============
    Mean +- std dev: [bolt-baseline] 44.5 ns +- 1.0 ns -> [bolt-hugify] 40.8 ns +- 0.4 ns: 1.09x faster
    Mean +- std dev: [bolt-baseline] 44.5 ns +- 1.0 ns -> [bolt-cdsplit] 42.2 ns +- 0.3 ns: 1.05x faster
    Mean +- std dev: [bolt-baseline] 44.5 ns +- 1.0 ns -> [bolt-split-all-cold] 48.0 ns +- 1.6 ns: 1.08x slower
    unpickle
    ========
    Mean +- std dev: [bolt-baseline] 12.7 us +- 0.1 us -> [bolt-hugify] 12.8 us +- 0.1 us: 1.00x slower
    Mean +- std dev: [bolt-baseline] 12.7 us +- 0.1 us -> [bolt-cdsplit] 13.2 us +- 0.1 us: 1.04x slower
    Mean +- std dev: [bolt-baseline] 12.7 us +- 0.1 us -> [bolt-split-all-cold] 13.0 us +- 0.1 us: 1.02x slower
    unpickle_list
    =============
    Mean +- std dev: [bolt-baseline] 4.09 us +- 0.11 us -> [bolt-hugify] 4.01 us +- 0.07 us: 1.02x faster
    Mean +- std dev: [bolt-baseline] 4.09 us +- 0.11 us -> [bolt-cdsplit] 4.00 us +- 0.04 us: 1.02x faster
    Mean +- std dev: [bolt-baseline] 4.09 us +- 0.11 us -> [bolt-split-all-cold] 4.54 us +- 0.04 us: 1.11x slower
    unpickle_pure_python
    ====================
    Mean +- std dev: [bolt-baseline] 251 us +- 2 us -> [bolt-hugify] 240 us +- 3 us: 1.05x faster
    Mean +- std dev: [bolt-baseline] 251 us +- 2 us -> [bolt-cdsplit] 244 us +- 2 us: 1.03x faster
    Mean +- std dev: [bolt-baseline] 251 us +- 2 us -> [bolt-split-all-cold] 248 us +- 2 us: 1.01x faster
    xml_etree_parse
    ===============
    Mean +- std dev: [bolt-baseline] 150 ms +- 3 ms -> [bolt-hugify] 148 ms +- 2 ms: 1.01x faster
    Mean +- std dev: [bolt-baseline] 150 ms +- 3 ms -> [bolt-cdsplit] 142 ms +- 2 ms: 1.06x faster
    Mean +- std dev: [bolt-baseline] 150 ms +- 3 ms -> [bolt-split-all-cold] 139 ms +- 2 ms: 1.08x faster
    xml_etree_iterparse
    ===================
    Mean +- std dev: [bolt-baseline] 97.2 ms +- 1.1 ms -> [bolt-hugify] 95.7 ms +- 1.2 ms: 1.02x faster
    Mean +- std dev: [bolt-baseline] 97.2 ms +- 1.1 ms -> [bolt-cdsplit] 96.8 ms +- 0.9 ms: 1.00x faster
    Mean +- std dev: [bolt-baseline] 97.2 ms +- 1.1 ms -> [bolt-split-all-cold] 94.1 ms +- 0.5 ms: 1.03x faster
    xml_etree_generate
    ==================
    Mean +- std dev: [bolt-baseline] 88.7 ms +- 0.5 ms -> [bolt-hugify] 85.2 ms +- 0.7 ms: 1.04x faster
    Mean +- std dev: [bolt-baseline] 88.7 ms +- 0.5 ms -> [bolt-cdsplit] 89.9 ms +- 0.8 ms: 1.01x slower
    Mean +- std dev: [bolt-baseline] 88.7 ms +- 0.5 ms -> [bolt-split-all-cold] 89.2 ms +- 0.6 ms: 1.01x slower
    xml_etree_process
    =================
    Mean +- std dev: [bolt-baseline] 63.3 ms +- 0.5 ms -> [bolt-hugify] 60.0 ms +- 0.6 ms: 1.05x faster
    Mean +- std dev: [bolt-baseline] 63.3 ms +- 0.5 ms -> [bolt-cdsplit] 63.7 ms +- 0.4 ms: 1.01x slower
    Mean +- std dev: [bolt-baseline] 63.3 ms +- 0.5 ms -> [bolt-split-all-cold] 63.1 ms +- 0.5 ms: 1.00x faster
    Benchmark hidden because not significant (2): asyncio_websockets, mdp
    Geometric mean
    ==============
    bolt-hugify: 1.07x faster
    bolt-cdsplit: 1.02x faster
    bolt-split-all-cold: 1.03x faster
    ```

    ```
    ❯ pyperf system show
    Show the system configuration
    System state
    ============
    CPU: use 12 logical CPUs: 0-11
    Perf event: Maximum sample rate: 1 per second
    ASLR: Full randomization
    Linux scheduler: No CPU is isolated
    CPU Frequency: 0-11=min=max=3600 MHz
    IRQ affinity: irqbalance service: inactive
    IRQ affinity: Default IRQ affinity: CPU 0-11
    IRQ affinity: IRQ affinity: IRQ 0-15,25-29,31-32,43-44,46-48,113,115-122,124,127-128=CPU 0-11; IRQ 49,81=CPU 0; IRQ 50,82=CPU 1; IRQ 51,83=CPU 2; IRQ 52,84=CPU 3; IRQ 53,85=CPU 4; IRQ 54,86=CPU 5; IRQ 55,87=CPU 6; IRQ 56,88=CPU 7; IRQ 57,89=CPU 8; IRQ 58,90=CPU 9; IRQ 59,91=CPU 10; IRQ 60,92=CPU 11; IRQ 61-80,93-112=CPU None
    CPU scaling governor: 0-11=performance
    Advices
    =======
    Linux scheduler: Use isolcpus=<cpu list> kernel parameter to isolate CPUs
    Linux scheduler: Use rcu_nocbs=<cpu list> kernel parameter (with isolcpus) to not schedule RCU on isolated CPUs
    Warnings
    ========
    Turbo Boost (MSR): Failed to read MSR 0x1a0 from /dev/cpu/0/msr: [Errno 13] Permission denied: '/dev/cpu/0/msr'
    ERROR: At least one operation failed with permission error, retry as root
    OK! System ready for benchmarking
    ```