Re: What's cooking in git.git (Sep 2021, #08; Mon, 27)

From: Junio C Hamano <gitster@pobox.com>
To: Neeraj Singh <nksingh85@gmail.com>, Elijah Newren <newren@gmail.com>
Cc: "Ævar Arnfjörð Bjarmason" <avarab@gmail.com>,
	"Git Mailing List" <git@vger.kernel.org>,
	"Jeff King" <peff@peff.net>,
	"Jonathan Nieder" <jrnieder@gmail.com>,
	"Neeraj Singh" <neerajsi@microsoft.com>
Subject: Re: What's cooking in git.git (Sep 2021, #08; Mon, 27)
Date: Thu, 07 Oct 2021 15:01:42 -0700	[thread overview]
Message-ID: <xmqq4k9spk8p.fsf@gitster.g> (raw)
In-Reply-To: <CANQDOddZ-KYTB4q0nYNDinis8aKktm6Ek6F+mJouTV-yRtTpUw@mail.gmail.com> (Neeraj Singh's message of "Tue, 28 Sep 2021 16:53:13 -0700")

Neeraj Singh <nksingh85@gmail.com> writes:

> Elijah,
>
> Here's a branch of your changes based on the amalgamated tmp-objdir code:
> https://github.com/neerajsi-msft/git/commits/neerajsi/remerge-diff
>
> This commit adapts your code to use the amalgamated API:
> https://github.com/neerajsi-msft/git/commit/725328fe1d8be8326d2ddef78e164ca21450b100

It seems that the discussion petered out at this point.

Right now I have a version of ns/remerge-diff before this adjustment
in 'seen', and Neeraj's latest version is kept out of 'seen' as they
do not play well together without an adjustment like that.

What's the good way forward?  I do not deeply care which one goes
first, but I have a feeling that the need by remerge-diff that wants
to discard temporary objects would involve more work to make it safe
than the need by batched fsync where newly created objects will not
be discarded but merely moved to the primary store before the end of
the operation, so from that point of view, it seems simpler and
safer to queue ns/batched-fsync topic first (especially given that
it is a no-op until the end-user opts into the experiment), and have
a remerge-diff that uses the infrastructure from Neeraj's topic.

What's your take on the rebase Neeraj made, Elijah (at the URL
above)?

Thanks.