Incrementing Vectors (2019)

Re:Incrementing Vectors (2019) by 684577567345 on Friday September 23, 2022 @02:17AM (#32948644)

I could not reproduce the authors results on clang 8.0 or gcc 8.1 at O3. The uint8_t version performed roughly 6 times faster than the uint32_t version. The post was published in 2019 so perhaps something has changed since then.

https://quick-bench.com/q/93vPCntI82FHbwAMv1ODUd_FjPc

[ Parent | on HN ]

Re:Incrementing Vectors (2019) by TinkersW on Friday September 23, 2022 @02:29AM (#32948709)

Your vectors are local to the function, so not the same thing, trivial for compiler to know they aren't aliased.

[ Parent | on HN ]

Re:Incrementing Vectors (2019) by yarrrrrrrrrrrr on Friday September 23, 2022 @05:49AM (#32949833)

https://quick-bench.com/q/x9kqob0PB5FsWQ65tFgpOlpaWOA

Repro. As the other commenter said, the internal data pointers need to be of unknown provenance (e.g. non-local).

[ Parent | on HN ]