Branchless Binary Search
Perform binary search with conditional moves or arithmetic updates instead of unpredictable branches.
Tag
Performance
17 notes
95. JIT Work in CPython
CPython 3.13 copy-and-patch JIT: template JIT design, trace selection, and the roadmap toward full JIT compilation.
81. Benchmarking CPython
pyperformance benchmark suite, microbenchmark pitfalls, timer resolution, and interpreter warm-up effects.
80. Profiling CPython
Using perf, Instruments, and py-spy to profile CPython; reading perf maps generated by the JIT.
79. Memory Locality
Object allocation locality, arena page placement, freelists for common types, and cache-line–aware design.
78. Attribute Access Performance
Type version tags, LOAD_ATTR inline cache hits, and the attribute specialization guards for slots and descriptors.
77. Dictionary Performance
Compact dict memory layout, hash collision probing strategy, split-table sharing, and lookup specialization.
76. Vectorcall
PEP 590 vectorcall protocol, _Py_TPFLAGS_HAVE_VECTORCALL, and stack-based argument passing for zero-overhead calls.
75. Function Call Fast Paths
CALL_PY_EXACT_ARGS, CALL_BUILTIN_FAST, and the fast-path conditions that bypass the generic call machinery.
74. Specializing Adaptive Interpreter
Adaptive counter logic, specialization guards, and how CPython 3.11+ rewrites opcodes to LOAD_ATTR_SLOT and friends.
73. Inline Caches
Inline cache entries appended to CACHE instructions in the bytecode array and their layout per opcode family.
72. Interpreter Dispatch
Computed goto dispatch table, the switch fallback, and how opcode prediction reduces branch mispredictions.
5.24 Real-World Hash Table Design
Choose and implement hash tables that perform reliably under mixed key types, uneven access patterns, and adversarial input.
5.23 Cache-Aware Hashing
Design hash table layouts that minimize cache misses and align memory access patterns with hardware behavior.
5.21 Consistent Performance Guarantees
Achieve predictable worst-case bounds for hash-based structures rather than relying solely on average-case expectations.
5.17 Cache Behavior
Understand how memory hierarchy effects cause hash table performance to deviate from asymptotic expectations.
5.4 Load Factor and Resizing
Control hash table performance by monitoring the load factor and growing the bucket array before collisions accumulate.