From: Junio C Hamano <gitster@pobox.com>
Date: Fri, 14 Nov 2025 20:06:07 GMT
Subject: Re: [PATCH v3 06/10] xdiff: split xrecord_t.ha into line_hash and minimal_perfect_hash
Message-ID: <xmqq346gidlc.fsf@gitster.g>
In-Reply-To: <CAH=ZcbCJ4MXnHpspuT+KkeR6LRTQrzh-7v5ep9S8WPRjdteR8g@mail.gmail.com>

Ezekiel Newren <ezekielnewren@gmail.com> writes:

>> How much extra memory pressure does this change cause?  In a single
>> instance of xrecord_t, we used to have a single ulong plus a pointer
>> and a size_t; now we replaced the single ulong with two 8-byte words,
>> so 33% more memory per record, which is not so huge a deal?
>
> This was asked and answered earlier in this patch series [1].

In short, this step does bloat, but the memory usage will shrink
when the members are moved elsewhere in future patches?

> [1] https://lore.kernel.org/git/CAH=ZcbD7FeRHtYvN_4=qHApB-AwK18=KRU2SGWNg8ADkrFM-Fw@mail.gmail.com/