OS: Darwin 24.6.0 Darwin Kernel Version 24.6.0: Wed Nov 5 21:28:03 PST 2025; root:xnu-11417.140.69.705.2~1/RELEASE_ARM64_T8122 arm64
CPU: arm
Timing: Python perf_counter
Levels: 0-9
Runs: 70 Trim worst: 40
Nathan Moinvaziri nmoinvaz
nmoinvaz
/ quick-bench-count-matches.cc
Last active
January 26, 2026 04:06
Benchmark count matching bytes
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #include <benchmark/benchmark.h> | |
| #include <cstdint> | |
| static inline uint32_t count_matching_bytes_ctzll(uint64_t mask) { | |
| return __builtin_ctzll(mask); | |
| } | |
| static inline uint32_t count_matching_bytes_ctz32(uint64_t mask) { | |
| uint32_t lo = (uint32_t)mask; | |
| if (lo) |
nmoinvaz
/ benchmark_crc32_tail_copy.cc
Last active
January 26, 2026 03:26
Benchmark zlib crc32 tail copy
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| /* benchmark_crc32_tail_copy.cc -- benchmark different copy strategies for CRC32 tail handling | |
| * Copyright (C) 2022 Nathan Moinvaziri | |
| * For conditions of distribution and use, see copyright notice in zlib.h | |
| */ | |
| #include <benchmark/benchmark.h> | |
| #include <cstring> | |
| #include <cstdint> | |
| extern "C" { |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| /* benchmark_tailcopy.cc -- benchmark different copy strategies for tail handling | |
| * Copyright (C) 2022 Nathan Moinvaziri | |
| * For conditions of distribution and use, see copyright notice in zlib.h | |
| */ | |
| #include <benchmark/benchmark.h> | |
| #include <cstring> | |
| #include <cstdint> | |
| extern "C" { |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| { | |
| // Use IntelliSense to learn about possible attributes. | |
| // Hover to view descriptions of existing attributes. | |
| // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387 | |
| "version": "0.2.0", | |
| "configurations": [ | |
| { | |
| "name": "fuzzer_example_small", | |
| "type": "lldb", | |
| "request": "launch", |
nmoinvaz
/ crc32_armv8_pmull_pclmulqdq.c
Created
December 29, 2025 00:13
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| // Polynomial for CRC32 (IEEE 802.3): 0x1EDC6F41 | |
| // Intel whitepaper uses reflected polynomial: 0x82F63B78 | |
| // We'll use 0x1EDC6F41 for compatibility with zlib-ng | |
| Z_INTERNAL Z_TARGET_CRC uint32_t crc32_armv8_pmull(uint32_t crc, const uint8_t *buf, size_t len) { | |
| uint32_t c = ~crc; | |
| // Constants for PMULL folding (from Intel whitepaper) | |
| const uint64x2_t k1 = {0x0154442bd4ULL, 0x00000001ULL}; |
nmoinvaz
/ crc32_armv8_pmull_single_lane.c
Created
December 9, 2025 20:34
crc32_armv8_pmull_single_lane
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Z_INTERNAL Z_TARGET_PMULL uint32_t crc32_armv8_pmull_single_lane(uint32_t crc, const uint8_t *buf, size_t len) { | |
| uint32_t crc0 = ~crc; | |
| /* 1. Alignment (Scalar) */ | |
| for (; len && ((uintptr_t)buf & 7); --len) { | |
| crc0 = __crc32b(crc0, *buf++); | |
| } | |
| /* 2. Alignment to 16-byte boundary (8-byte scalar CRC) */ | |
| if (((uintptr_t)buf & 8) && len >= 8) { |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| /* arm_cpu_id.c -- ARM CPU identification for microarchitecture detection | |
| * Copyright (C) 2025 Nathan Moinvaziri | |
| * For conditions of distribution and use, see copyright notice in zlib.h | |
| */ | |
| #include "zbuild.h" | |
| #include "arm_cpu_id.h" | |
| #if defined(__linux__) | |
| # include <stdio.h> |
nmoinvaz
/ dougallj-benchmarks.md
Last active
December 1, 2025 22:20
Benchmarks for zlib-ng issue #1998
https://gist.github.com/dougallj/1fec1fb1efd03cf559b4626ff71cb35a
zlib-ng/zlib-ng#1998 (comment)
OS: Darwin 24.6.0 Darwin Kernel Version 24.6.0: Mon Jul 14 11:29:54 PDT 2025; root:xnu-11417.140.69~1/RELEASE_ARM64_T8122 arm64
CPU: arm
Tool: /Users/nathan/Source/zlib-ng/build-develop/minigzip Size: 159,688 B
Timing: GNU time (gtime)
nmoinvaz
/ insert_string_distributions.md
Last active
August 23, 2025 00:11
insert_string distributions
NewerOlder