Re: [PATCH] is_promisor_object(): walk promisor packs in pack-order

From: Derrick Stolee <derrickstolee@github.com>
To: Jeff King <peff@peff.net>
Cc: Richard Oliver <roliver@roku.com>, Taylor Blau <me@ttaylorr.com>,
	git@vger.kernel.org, jonathantanmy@google.com
Subject: Re: [PATCH] is_promisor_object(): walk promisor packs in pack-order
Date: Thu, 16 Jun 2022 10:00:43 -0400	[thread overview]
Message-ID: <1682f396-018a-8751-1907-7c244fe06ea9@github.com> (raw)
In-Reply-To: <YqrTsbXbEjx0Pabn@coredump.intra.peff.net>

On 6/16/2022 2:54 AM, Jeff King wrote:
> On Thu, Jun 16, 2022 at 02:07:41AM -0400, Jeff King wrote:
> 
>> Those rev-lists run in 1.7s and 224s respectively. Ouch!
> 
> Even though I expected the second one to be slow, I was shocked at just
> how slow it was. The patch below speeds it up by a factor of 2, and I
> think is worth applying separately, regardless of anything else being
> discussed in this thread.

I agree. It's extremely simple for the value it provides.

> diff --git a/packfile.c b/packfile.c
> index 8e812a84a3..ed69fe457b 100644
> --- a/packfile.c
> +++ b/packfile.c
> @@ -2275,7 +2275,8 @@ int is_promisor_object(const struct object_id *oid)
>  		if (has_promisor_remote()) {
>  			for_each_packed_object(add_promisor_object,
>  					       &promisor_objects,
> -					       FOR_EACH_OBJECT_PROMISOR_ONLY);
> +					       FOR_EACH_OBJECT_PROMISOR_ONLY |
> +					       FOR_EACH_OBJECT_PACK_ORDER);
>  		}

Thanks,
-Stolee