Kraionix
Kraionix
Hello. I am learning low-level optimizations using SIMD on x86-64. I wrote the code from scratch for the sake of practice. I optimized the SSE and AVX2 implementations of ImMemchr...
I also tested `ImGui::TextUnformatted` with a 16 MB `std::string` buffer, in which every 131 characters is `\n`, the performance on AVX2 ImMemchr increased by about 60%, compared to regular memchr....
I added an optimized ImStrlen with implementations on SSE and AVX2.