Exploring benchmarks for CPython's tail-call JIT

gistfile1.txt

❯ PYTHONHASHSEED=11 taskset -c 0-11 perf stat $tailcall -u venv/lib/python3.12/site-packages/pyperformance/data-files/benchmarks/bm_unpack_sequence/run_benchmark.py tuple -p1 -n1 -l 5000
.
unpack_sequence_tuple: 22.0 ns

 Performance counter stats for 'venv/cpython3.14-fbced5d02230-compat-af96b9431081/bin/python -u venv/lib/python3.12/site-packages/pyperformance/data-files/benchmarks/bm_unpack_sequence/run_benchmark.py tuple -p1 -n1 -l 5000':

            244.38 msec task-clock:u                     #    0.998 CPUs utilized
                 0      context-switches:u               #    0.000 /sec
                 0      cpu-migrations:u                 #    0.000 /sec
             6,340      page-faults:u                    #   25.943 K/sec
     <not counted>      cpu_atom/instructions/u                                                 (0.00%)
     1,860,506,925      cpu_core/instructions/u          #    3.43  insn per cycle
     <not counted>      cpu_atom/cycles/u                                                       (0.00%)
       541,688,315      cpu_core/cycles/u                #    2.217 GHz
     <not counted>      cpu_atom/branches/u                                                     (0.00%)
       316,346,365      cpu_core/branches/u              #    1.294 G/sec
     <not counted>      cpu_atom/branch-misses/u                                                (0.00%)
         3,523,902      cpu_core/branch-misses/u         #    1.11% of all branches
             TopdownL1 (cpu_core)                 #      6.5 %  tma_backend_bound
                                                  #     13.3 %  tma_bad_speculation
                                                  #     27.7 %  tma_frontend_bound
                                                  #     52.5 %  tma_retiring

       0.244815704 seconds time elapsed

       0.219950000 seconds user
       0.023853000 seconds sys


❯ PYTHONHASHSEED=11 taskset -c 0-11 perf stat $goto -u venv/lib/python3.12/site-packages/pyperformance/data-files/benchmarks/bm_unpack_sequence/run_benchmark.py tuple -p1 -n1 -l 5000
.
unpack_sequence_tuple: 27.5 ns

 Performance counter stats for 'venv/cpython3.14-0386bad6388d-compat-af96b9431081/bin/python -u venv/lib/python3.12/site-packages/pyperformance/data-files/benchmarks/bm_unpack_sequence/run_benchmark.py tuple -p1 -n1 -l 5000':

            267.09 msec task-clock:u                     #    0.998 CPUs utilized
                 0      context-switches:u               #    0.000 /sec
                 0      cpu-migrations:u                 #    0.000 /sec
             6,329      page-faults:u                    #   23.697 K/sec
     <not counted>      cpu_atom/instructions/u                                                 (0.00%)
     1,824,738,642      cpu_core/instructions/u          #    3.05  insn per cycle
     <not counted>      cpu_atom/cycles/u                                                       (0.00%)
       597,365,501      cpu_core/cycles/u                #    2.237 GHz
     <not counted>      cpu_atom/branches/u                                                     (0.00%)
       391,773,160      cpu_core/branches/u              #    1.467 G/sec
     <not counted>      cpu_atom/branch-misses/u                                                (0.00%)
         3,528,953      cpu_core/branch-misses/u         #    0.90% of all branches
             TopdownL1 (cpu_core)                 #      5.9 %  tma_backend_bound
                                                  #     11.8 %  tma_bad_speculation
                                                  #     35.5 %  tma_frontend_bound
                                                  #     46.7 %  tma_retiring

       0.267620493 seconds time elapsed

       0.235486000 seconds user
       0.031021000 seconds sys