Re: [PATCH 3/4] t/perf: avoid testing bitmaps without lookup table

From: Junio C Hamano <gitster@pobox.com>
Date: Apr 18, 2025, 04:24 UTC
Message-ID: <xmqqzfgeaygh.fsf@gitster.g>
In-Reply-To: <xmqqcydae8fg.fsf@gitster.g>

Junio C Hamano <gitster@pobox.com> writes:

> I somehow have a feeling that removal of these "performance" tests
> is less worrysome than removing correctness tests, but as long as we
> claim to support both configurations (i.e. with and without lookup
> tables), it feels a bit premature to remove tests for one of them.

In case the implication was missed, I was hinting that in the longer term, once one variant proves to be better than the other variant(s) in any and all aspects, it would be a great move to remove the other one(s). It is exactly what is happening on the recursive-ort front.

Once we become so confident about correctness and performance with the configuration with lookup tables that we are willing to lose an escape hatch to operate without them, we can obviously remove these tests for configuration without lookup tables. If we are not there yet, and still rely on the "escape hatch" value of the configuration that does not use the lookup tables, we want to make sure that the escape hatch still functions, right ;-)?

Thanks.