Re: [PATCH v5] pack-objects mru

From: Junio C Hamano <gitster@pobox.com>
Date: Aug 11, 2016, 15:11 UTC
Message-ID: <xmqqmvkj2vdm.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <20160811095710.p2bffympjlwmv3gc@sigill.intra.peff.net>

Jeff King <peff@peff.net> writes:

> So considering "--depth" as a space-saving measure for --aggressive does
> not seem that effective. But it feels weird to quietly drop actions
> people might have done with previous aggressive runs.

That argument cuts both ways, doesn't it?

If the user explicitly asks to use lower "--depth" from the command line when the second repack runs, the intention is clear: the existing pack may use delta chains that are too long and is detrimental to the run-time performance, and the user wants to correct it by repacking with shorter delta chain.

Should the act of letting "gc --auto" use lower "--depth", by not configuring to always use deeper chain, be interpreted the same way? I am not sure. The old packing with large --depth is something the user did long time ago, and the decision the user made not to use large depth always is also something the user did long time ago. I do not think it is so cut-and-dried which one of the two conflicting wishes we should honor when running the second repack, especially when it is run unattended like "gc --auto" does.