Constant-time arithmetic operations

Name: CppCon 2015: Chandler Carruth "Tuning C++: Benchmarks, and CPUs, and Compilers! Oh My!"
Uploaded: 2020-04-19T05:48:26Z
Description: IIUC the compiler may optimize the inline assembly if it can. GCC and Clang has asm volatile for the programmer to explicitly tell them to not touch it. At least that’s what learnt from watching the...

tkf · April 19, 2020, 5:48am

IIUC the compiler may optimize the inline assembly if it can. GCC and Clang has asm volatile for the programmer to explicitly tell them to not touch it. At least that’s what learnt from watching the talk explaining clobber() and escape():

This used is in google/benchmark, too:

github.com

google/benchmark/blob/c078337494086f9372a46b4ed31a3ae7b3f1a6a2/include/benchmark/benchmark.h#L307-L315


      
          // The DoNotOptimize(...) function can be used to prevent a value or
          // expression from being optimized away by the compiler. This function is
          // intended to add little to no overhead.
          // See: https://youtu.be/nXaxk27zwlk?t=2441
          #ifndef BENCHMARK_HAS_NO_INLINE_ASSEMBLY
          template <class Tp>
          inline BENCHMARK_ALWAYS_INLINE void DoNotOptimize(Tp const& value) {
            asm volatile("" : : "r,m"(value) : "memory");
          }

Topic		Replies	Views
Impact of Spectre on Julia? Internals & Design	19	2490	January 14, 2018
DarkIntegers, a pure Julia library for modulo integer and polynomial arithmetic Package Announcements package , announcement	4	816	March 9, 2020
Rust vs Julia Offtopic rust	11	33649	May 19, 2022
Julia on embedded devices & validation thereof General Usage	36	3057	July 16, 2022
Julia written in Julia? Internals & Design question	37	6321	July 6, 2019

Constant-time arithmetic operations

Related topics