Re: [PATCH v4] write-tree: integrate with sparse index

From: Junio C Hamano <gitster@pobox.com>
To: Shuqi Liang <cheskaqiqi@gmail.com>, Victoria Dye <vdye@github.com>
Cc: git@vger.kernel.org, derrickstolee@github.com
Subject: Re: [PATCH v4] write-tree: integrate with sparse index
Date: Mon, 24 Apr 2023 08:14:21 -0700	[thread overview]
Message-ID: <xmqq7cu1ia7m.fsf@gitster.g> (raw)
In-Reply-To: <8b2a754c-6162-54d9-e9ba-fd994058066c@github.com> (Victoria Dye's message of "Fri, 21 Apr 2023 14:42:34 -0700")

Victoria Dye <vdye@github.com> writes:

> Shuqi Liang wrote:
>> Update 'git write-tree' to allow using the sparse-index in memory
>> without expanding to a full one.
>> 
>> The recursive algorithm for update_one() was already updated in 2de37c5
>> (cache-tree: integrate with sparse directory entries, 2021-03-03) to
>> handle sparse directory entries in the index. Hence we can just set the
>> requires-full-index to false for "write-tree".
>> 
>> The `p2000` tests demonstrate a ~96% execution time reduction for 'git
>> write-tree' using a sparse index:
>> 
>> Test                                           before  after
>> -----------------------------------------------------------------
>> 2000.78: git write-tree (full-v3)              0.34    0.33 -2.9%
>> 2000.79: git write-tree (full-v4)              0.32    0.30 -6.3%
>> 2000.80: git write-tree (sparse-v3)            0.47    0.02 -95.8%
>> 2000.81: git write-tree (sparse-v4)            0.45    0.02 -95.6%
>
> Please update your commit message to explain only the incremental updates on
> top of 1a65b41b38a (write-tree: integrate with sparse index, 2023-04-03);
> that patch's message (what you have here) does not accurately describe what
> _this_ patch is doing.

Good point.

In addition, as this is the first iteration of a follow-up topic,
"v4" on the subject line is a bit misleading.  Let's treat it as a
new and separate topic that build on top of the previous achievement.

Thanks for working on this topic and mentoring a new contributor.