Feature/foa fast iteration

[joaquintides]

Changed iterator increment code for "regular layout" group implementations (basically, those for SSE2 and Neon). Iteration speed improves dramatically, look for "Running erasure" in:

https://github.com/boostorg/boost_unordered_benchmarks/commit/46129697b05fab67c42cfa2b918a6e40af26d6e6

[cppalliance-bot]

An automated preview of the documentation is available at https://187.unordered.prtest.cppalliance.org/libs/unordered/doc/html/unordered.html

[pdimov]

Looks good to me, although I'm not sure we should be merging to develop now. Maybe wait for the beta to go out in case there are critical fixes we need to apply.

[joaquintides]

The release notes added in the PR target 1.83, I just submitted it because it's ready but we can wait of course.

[cppalliance-bot]

An automated preview of the documentation is available at https://187.unordered.prtest.cppalliance.org/libs/unordered/doc/html/unordered.html

[cppalliance-bot]

An automated preview of the documentation is available at https://187.unordered.prtest.cppalliance.org/libs/unordered/doc/html/unordered.html

[codecov[bot]]

Codecov Report

Merging #187 (1c5640c) into develop (9aedb95) will increase coverage by 0.00%. The diff coverage is 100.00%.

Additional details and impacted files

[](https://codecov.io/gh/boostorg/unordered/pull/187?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=boostorg) ```diff @@ Coverage Diff @@ ## develop #187 +/- ## ======================================== Coverage 97.85% 97.85% ======================================== Files 91 91 Lines 13849 13875 +26 ======================================== + Hits 13552 13578 +26 Misses 297 297 ``` | [Impacted Files](https://codecov.io/gh/boostorg/unordered/pull/187?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=boostorg) | Coverage Δ | | |---|---|---| | [include/boost/unordered/detail/foa.hpp](https://codecov.io/gh/boostorg/unordered/pull/187?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=boostorg#diff-aW5jbHVkZS9ib29zdC91bm9yZGVyZWQvZGV0YWlsL2ZvYS5ocHA=) | `99.84% <100.00%> (+<0.01%)` | :arrow_up: | ------ [Continue to review full report at Codecov](https://codecov.io/gh/boostorg/unordered/pull/187?src=pr&el=continue&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=boostorg). > **Legend** - [Click here to learn more](https://docs.codecov.io/docs/codecov-delta?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=boostorg) > `Δ = absolute (impact)`, `ø = not affected`, `? = missing data` > Powered by [Codecov](https://codecov.io/gh/boostorg/unordered/pull/187?src=pr&el=footer&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=boostorg). Last update [9aedb95...1c5640c](https://codecov.io/gh/boostorg/unordered/pull/187?src=pr&el=lastupdated&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=boostorg). Read the [comment docs](https://docs.codecov.io/docs/pull-request-comments?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=boostorg).

[joaquintides]

Latest commit (1c5640cfbe84213b9d2a7eff68e2db744cfa3817) didn't affect performance: https://github.com/boostorg/boost_unordered_benchmarks/commit/90e178376ffefe73d3e6fc7fa24ef70a26d429a6

[cmazakas]

Latest commit (1c5640c) didn't affect performance: boostorg/boost_unordered_benchmarks@90e1783

Well, my logic was more about the code self-documenting the assumptions.

It wasn't clear to me why we needed the null pointer check before calling ++it.

[cmazakas]

@pdimov @joaquintides

Now that the Boost beta is out, should we feel free to go and merge this PR?

[joaquintides]

FWIW, this feature is already integrated in feature/cfoa, so we can either drop this PR or merge and rebase that branch.

[pdimov]

We can merge it to develop, I suppose, just not to master.