Hello 算法 C++ 语言版 靳宇栋（Krahets） Release 1.0.0b4 2023‑07‑26 序 两年前，我在力扣上分享了《剑指 Offer》系列题解，受到了许多朋友的喜爱与支持。在此期间，我回答了众 多读者的评论问题，其中最常见的一个问题是“如何入门学习算法”。我逐渐也对这个问题产生了浓厚的兴 趣。 两眼一抹黑地刷题似乎是最受欢迎的方法，简单直接且有效。然而，刷题就如同玩“扫雷”游戏，自学能力 wangwang105, ZongYangL, shanghai‑Jerry, curly210102, Suremotoo, linzeyan, yi427, boloboloda, huawuque404, 4yDX3906, ZJKung, xb534, siqyka, ZnYang2018, beintentional, luluxia, GaochaoZhu, weibk, dshlstarr, ShiMaRing 45 3.5. 小结 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 4. 数组与链表 50 4.1. 数组 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

build from source by the user, with that configuration 4. Pass information to the build system so it can locate and use it successfully 4#include int main(void){ … deflateInit(&defstream metadata: { compiled: 2023-10-01 checksum: … } 91 … … Creates CPS Read CPSZLib Proposal: Lean MVP build: { "includedirs": ["include"], "libdirs": ["lib"], "libs": ["zlib"] }, abi: { dynamic_runtime:

research and functional programming Love performance, software abstractions, and clean APIs Why me 4What is reflection Reflection in other languages (Go, Python, Java) Reflection in C++ as per P2996 join(attrs)) result = orig_copy(obj) return result cls.copy = _wrapped_copy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 class MyClass: def __init__(self, x): self.x = x def copy(self): >>> MyClass(2).copy() Calling wrapped copy Attributes: x <__main__.MyClass object at 0x7f1a9 1 2 3 4 5 6 7 8 9 10 11 13GoLang 14Golang is a compiled but duck-typed language Well, structurally typed

C++: A Practical Guide CppCon 2024 September 18, 2024 Pete Muldoon Senior Engineering LeadBloomberg 4 Questions #include HereWho Am I? Bloomberg 5 • Started using C++ professionally independent of its underlying dependencies 3. Improves the reusability – and so testability - of code. 4. Better long term maintainability of code. Encyclopedia Muldoonica Basic Definitions11 Bloomberg testing ❑ Lack of specificity ❑ More difficult setup / error investigation ❑ Longer feedback loop ❑ Can lean entirely on regression A/B testing Testing without Dependency Injection12 Bloomberg What is the

achieved goal doesn’t justify the design compromise. - There may be a solution that meets both needs. 4. Because there are tradeoffs 17 Hiding your Implementation Details, Amir Kirsh @ CppCon, 2024Investigate std::cout << squarepair << std::endl; (doable but not straightforward: http://coliru.stacked-crooked.com/a/4c31320c394bcbb5) 22 Hiding your Implementation Details, Amir Kirsh @ CppCon, 2024 std::pair23 Hiding overloads), a wrong option would be selected at some point. Less options means also less required tests.Be Lean and Mean 41 Hiding your Implementation Details, Amir Kirsh @ CppCon, 2024 1. Keep your API short

X X X XMicrocontroller Constraints: Limited Storage & Memory 7 < 1 MB = 33,725 devices >= 1 MB = 4,964 devicesMicrocontroller Constraints: Limited Storage & Memory 8 < 100kB = 31,098 devices >= 100kB arm-none-eabi-g++ -o except.elf except.cpp -std=c++20 -Os -g -fexceptions -frtti -mcpu=cortex-m4 -mfloat-abi=hard -mthumb --specs=nosys.specs --specs=nano.specs --Wl,-T linker.ld Let's start simple 0000800c : 800c: b508 push {r3, lr} 800e: 2004 movs r0, #4 8010: f000 f93c bl 828c <__cxa_allocate_exception> 8014: 2305 movs r3, #5

std::endl; } ~Foo() { std::cout << "~Foo()" << std::endl; } }; int main() { Foo a; { Foo b; } } 1 2 3 4 5 6 7 8 9 10 11 21 Constructors & Destructorshttps://godbolt.org/z/ffzG8hPT8 struct Foo { Foo() std::endl; } ~Foo() { std::cout << "~Foo()" << std::endl; } }; int main() { Foo a; { Foo b; } } 1 2 3 4 5 6 7 8 9 10 11 22 Constructors & Destructorshttps://godbolt.org/z/ffzG8hPT8 struct Foo { Foo() ~Foo() { std::cout << "~Foo()" << std::endl; } }; int main() { Foo a; // Foo() { Foo b; } } 1 2 3 4 5 6 7 8 9 10 11 23 Constructors & Destructorshttps://godbolt.org/z/ffzG8hPT8 struct Foo { Foo()

Data is all you need for fusion N 1int x = 4; callee(x); // do work } #include #include #include "benchmark.h" #include "matrix_lib.h" int main(..){ std::vector a; a.reserve(100); cblas_sgemm(....); return 0; } High Performance code is about Hardware Matrix Multiply 2int x = 4; callee(x); // do work } #include #include #include "benchmark.h" #include "matrix_lib k++) { result[i][j] += A[i][k] * B[k][j]; } } } caller: push r4 ; do work mov r0, #4 bl callee pop r4 ; do work callee: 3int x = 4; callee(x); // do work } #include #include

__m512i; using msk_512 = uint32_t; ... }Copyright © 2020 Bob Steagall K E W B C O M P U T I N G 4 CppCon 2020 - Adventures in SIMD Thinking Function load_value() KEWB_FORCE_INLINE rf_512 load_value(float load_from(float const* psrc) { return _mm512_loadu_ps(psrc); } m0 m1 m2 m3 m4 m5 m6 m7 m8 m9 m10 m11 m12 m13 m14 m15 m0 m1 m2 m3 m4 m5 m6 m7 m8 m9 m10 m11 m12 m13 m14 m15 psrcCopyright © 2020 Bob Steagall mask) { return _mm512_mask_loadu_ps(_mm512_set1_ps(fill), (__mmask16) mask, psrc); } m0 m1 m2 m3 m4 m5 m6 m7 m8 m9 m10 m11 m12 m13 m14 m15 f f m2 m3 f f m6 m7 f f f f m12 m13 m14 m15 f 0 0 1 1 0 0

Image Processing Programming Models Outlook Summary std::simd is for you! Matthias Kretz CppCon ’23 4 GSI Helmholtz Center for Heavy Ion ResearchMotivation std::simd Overview Example: Image Processing Programming … unless you don’t care for speed-up measured in factors not percentages Matthias Kretz CppCon ’23 4 GSI Helmholtz Center for Heavy Ion ResearchMotivation std::simd Overview Example: Image Processing Programming the same work in less time x0 + a x1 + a x2 + a x3 + a x4 + a x5 + a x6 + a x7 + a x8 + a x9 + a · · · Time x0 + a x1 + a x2 + a x3 + a x4 + a x5 + a x6 + a x7 + a x8 + a x9 + a · · · · ·