git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: "Robert Coup via GitGitGadget" <gitgitgadget@gmail.com>
Cc: git@vger.kernel.org, Jonathan Tan <jonathantanmy@google.com>,
	John Cai <johncai86@gmail.com>,
	Jeff Hostetler <git@jeffhostetler.com>,
	Derrick Stolee <derrickstolee@github.com>,
	Robert Coup <robert@coup.net.nz>
Subject: Re: [PATCH v2 2/8] fetch-pack: add repairing
Date: Thu, 24 Feb 2022 22:46:01 -0800	[thread overview]
Message-ID: <xmqqwnhjfnpy.fsf@gitster.g> (raw)
In-Reply-To: <2d817a65db55750d27986c2e2ab5b1723d17081e.1645719218.git.gitgitgadget@gmail.com> (Robert Coup via GitGitGadget's message of "Thu, 24 Feb 2022 16:13:32 +0000")

"Robert Coup via GitGitGadget" <gitgitgadget@gmail.com> writes:

> @@ -694,6 +696,9 @@ static void mark_complete_and_common_ref(struct fetch_negotiator *negotiator,
>  
>  	save_commit_buffer = 0;
>  
> +	if (args->repair)
> +		return;
> +

Reading how the original value of save_commit_buffer is saved away,
the variable gets cleared and then gets restored before the function
returns in the normal codepath, this new code looks wrong.  Hitting
this early return after clearing the variable means nobody will
restore the saved value of the variable, no?

> @@ -1027,9 +1032,6 @@ static struct ref *do_fetch_pack(struct fetch_pack_args *args,
>  	struct fetch_negotiator negotiator_alloc;
>  	struct fetch_negotiator *negotiator;
>  
> -	negotiator = &negotiator_alloc;
> -	fetch_negotiator_init(r, negotiator);

I know why you want to force the "noop" negotiator while repairing,
but it is unclear why you need to move this down in the function.

>  	sort_ref_list(&ref, ref_compare_name);
>  	QSORT(sought, nr_sought, cmp_ref_by_name);
>  
> @@ -1119,9 +1121,16 @@ static struct ref *do_fetch_pack(struct fetch_pack_args *args,
>  	if (!server_supports_hash(the_hash_algo->name, NULL))
>  		die(_("Server does not support this repository's object format"));
>  
> +	negotiator = &negotiator_alloc;
> +	if (args->repair) {
> +		fetch_negotiator_init_noop(negotiator);
> +	} else {
> +		fetch_negotiator_init(r, negotiator);
> +	}

Hmph.  I am debating myself if hardcoding the implementation detail
of "when repairing, the noop negitiator is the only useful one" like
this code does is a sensible thing to do.  If we later need to tweak
the choice of negotiator used depending on the caller's needs,
perhaps fetch_negotiator_init() should gain a new flags word, i.e.

	fetch_negotiator_init(struct repository *,
			      struct fetch_negotiator *,
			      unsigned flags)

where "Use negotiator suitable for the repairing fetch" could be a
single bit in the flags word, making this caller more like:

	negotiator = &negotiator_alloc;
	flags = 0;
	if (args->repair)
		flags |= FETCH_NEGOTIATOR_REPAIRING;
	fetch_negotiator_init(r, negotiator, flags);

perhaps.  That way, [1/8] becomes unnecessary.

>  	mark_complete_and_common_ref(negotiator, args, &ref);
>  	filter_refs(args, &ref, sought, nr_sought);
> -	if (everything_local(args, &ref)) {
> +	if (!args->repair && everything_local(args, &ref)) {
>  		packet_flush(fd[1]);
>  		goto all_done;
>  	}
> @@ -1587,7 +1596,10 @@ static struct ref *do_fetch_pack_v2(struct fetch_pack_args *args,
>  	struct strvec index_pack_args = STRVEC_INIT;
>  
>  	negotiator = &negotiator_alloc;
> -	fetch_negotiator_init(r, negotiator);
> +	if (args->repair)
> +		fetch_negotiator_init_noop(negotiator);
> +	else
> +		fetch_negotiator_init(r, negotiator);

Likewise.

>  	packet_reader_init(&reader, fd[0], NULL, 0,
>  			   PACKET_READ_CHOMP_NEWLINE |
> @@ -1613,7 +1625,7 @@ static struct ref *do_fetch_pack_v2(struct fetch_pack_args *args,
>  			/* Filter 'ref' by 'sought' and those that aren't local */
>  			mark_complete_and_common_ref(negotiator, args, &ref);
>  			filter_refs(args, &ref, sought, nr_sought);
> -			if (everything_local(args, &ref))
> +			if (!args->repair && everything_local(args, &ref))
>  				state = FETCH_DONE;
>  			else
>  				state = FETCH_SEND_REQUEST;
> diff --git a/fetch-pack.h b/fetch-pack.h
> index 7f94a2a5831..bbb663edda8 100644
> --- a/fetch-pack.h
> +++ b/fetch-pack.h
> @@ -42,6 +42,7 @@ struct fetch_pack_args {
>  	unsigned update_shallow:1;
>  	unsigned reject_shallow_remote:1;
>  	unsigned deepen:1;
> +	unsigned repair:1;
>  
>  	/*
>  	 * Indicate that the remote of this request is a promisor remote. The

  reply	other threads:[~2022-02-25  6:46 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-01 15:49 [PATCH 0/6] [RFC] partial-clone: add ability to refetch with expanded filter Robert Coup via GitGitGadget
2022-02-01 15:49 ` [PATCH 1/6] fetch-negotiator: add specific noop initializor Robert Coup via GitGitGadget
2022-02-01 15:49 ` [PATCH 2/6] fetch-pack: add partial clone refiltering Robert Coup via GitGitGadget
2022-02-04 18:02   ` Jonathan Tan
2022-02-11 14:56     ` Robert Coup
2022-02-17  0:05       ` Jonathan Tan
2022-02-01 15:49 ` [PATCH 3/6] builtin/fetch-pack: add --refilter option Robert Coup via GitGitGadget
2022-02-01 15:49 ` [PATCH 4/6] fetch: " Robert Coup via GitGitGadget
2022-02-01 15:49 ` [PATCH 5/6] t5615-partial-clone: add test for --refilter Robert Coup via GitGitGadget
2022-02-01 15:49 ` [PATCH 6/6] doc/partial-clone: mention --refilter option Robert Coup via GitGitGadget
2022-02-01 20:13 ` [PATCH 0/6] [RFC] partial-clone: add ability to refetch with expanded filter Junio C Hamano
2022-02-02 15:02   ` Robert Coup
2022-02-16 13:24     ` Robert Coup
2022-02-02 18:59 ` Jonathan Tan
2022-02-02 21:58   ` Robert Coup
2022-02-02 21:59     ` Robert Coup
2022-02-07 19:37 ` Jeff Hostetler
2022-02-24 16:13 ` [PATCH v2 0/8] fetch: add repair: full refetch without negotiation (was: "refiltering") Robert Coup via GitGitGadget
2022-02-24 16:13   ` [PATCH v2 1/8] fetch-negotiator: add specific noop initializor Robert Coup via GitGitGadget
2022-02-25  6:19     ` Junio C Hamano
2022-02-28 12:22       ` Robert Coup
2022-02-24 16:13   ` [PATCH v2 2/8] fetch-pack: add repairing Robert Coup via GitGitGadget
2022-02-25  6:46     ` Junio C Hamano [this message]
2022-02-28 12:14       ` Robert Coup
2022-02-24 16:13   ` [PATCH v2 3/8] builtin/fetch-pack: add --repair option Robert Coup via GitGitGadget
2022-02-24 16:13   ` [PATCH v2 4/8] fetch: " Robert Coup via GitGitGadget
2022-02-24 16:13   ` [PATCH v2 5/8] t5615-partial-clone: add test for fetch --repair Robert Coup via GitGitGadget
2022-02-24 16:13   ` [PATCH v2 6/8] maintenance: add ability to pass config options Robert Coup via GitGitGadget
2022-02-25  6:57     ` Junio C Hamano
2022-02-28 12:02       ` Robert Coup
2022-02-28 17:07         ` Junio C Hamano
2022-02-25 10:29     ` Ævar Arnfjörð Bjarmason
2022-02-28 11:51       ` Robert Coup
2022-02-24 16:13   ` [PATCH v2 7/8] fetch: after repair, encourage auto gc repacking Robert Coup via GitGitGadget
2022-02-28 16:40     ` Ævar Arnfjörð Bjarmason
2022-02-24 16:13   ` [PATCH v2 8/8] doc/partial-clone: mention --repair fetch option Robert Coup via GitGitGadget
2022-02-28 16:43   ` [PATCH v2 0/8] fetch: add repair: full refetch without negotiation (was: "refiltering") Ævar Arnfjörð Bjarmason
2022-02-28 17:27     ` Robert Coup
2022-02-28 18:54       ` [PATCH v2 0/8] fetch: add repair: full refetch without negotiation Junio C Hamano
2022-02-28 22:20       ` [PATCH v2 0/8] fetch: add repair: full refetch without negotiation (was: "refiltering") Ævar Arnfjörð Bjarmason
2022-03-04 15:04   ` [PATCH v3 0/7] " Robert Coup via GitGitGadget
2022-03-04 15:04     ` [PATCH v3 1/7] fetch-negotiator: add specific noop initializer Robert Coup via GitGitGadget
2022-03-04 15:04     ` [PATCH v3 2/7] fetch-pack: add refetch Robert Coup via GitGitGadget
2022-03-04 15:04     ` [PATCH v3 3/7] builtin/fetch-pack: add --refetch option Robert Coup via GitGitGadget
2022-03-04 15:04     ` [PATCH v3 4/7] fetch: " Robert Coup via GitGitGadget
2022-03-04 21:19       ` Junio C Hamano
2022-03-07 11:31         ` Robert Coup
2022-03-07 17:27           ` Junio C Hamano
2022-03-09 10:00             ` Robert Coup
2022-03-04 15:04     ` [PATCH v3 5/7] t5615-partial-clone: add test for fetch --refetch Robert Coup via GitGitGadget
2022-03-04 15:04     ` [PATCH v3 6/7] fetch: after refetch, encourage auto gc repacking Robert Coup via GitGitGadget
2022-03-04 15:04     ` [PATCH v3 7/7] doc/partial-clone: mention --refetch fetch option Robert Coup via GitGitGadget
2022-03-09  0:27     ` [PATCH v3 0/7] fetch: add repair: full refetch without negotiation (was: "refiltering") Calvin Wan
2022-03-09  9:57       ` Robert Coup
2022-03-09 21:32         ` [PATCH v3 0/7] fetch: add repair: full refetch without negotiation Junio C Hamano
2022-03-10  1:07           ` Calvin Wan
2022-03-10 14:29           ` Robert Coup
2022-03-21 17:58             ` Calvin Wan
2022-03-21 21:34               ` Robert Coup
2022-03-28 14:02     ` [PATCH v4 0/7] fetch: add repair: full refetch without negotiation (was: "refiltering") Robert Coup via GitGitGadget
2022-03-28 14:02       ` [PATCH v4 1/7] fetch-negotiator: add specific noop initializer Robert Coup via GitGitGadget
2022-03-28 14:02       ` [PATCH v4 2/7] fetch-pack: add refetch Robert Coup via GitGitGadget
2022-03-31 15:09         ` Ævar Arnfjörð Bjarmason
2022-04-01 10:26           ` Robert Coup
2022-03-28 14:02       ` [PATCH v4 3/7] builtin/fetch-pack: add --refetch option Robert Coup via GitGitGadget
2022-03-28 14:02       ` [PATCH v4 4/7] fetch: " Robert Coup via GitGitGadget
2022-03-31 15:18         ` Ævar Arnfjörð Bjarmason
2022-04-01 10:31           ` Robert Coup
2022-03-28 14:02       ` [PATCH v4 5/7] t5615-partial-clone: add test for fetch --refetch Robert Coup via GitGitGadget
2022-03-31 15:20         ` Ævar Arnfjörð Bjarmason
2022-04-01 10:36           ` Robert Coup
2022-03-28 14:02       ` [PATCH v4 6/7] fetch: after refetch, encourage auto gc repacking Robert Coup via GitGitGadget
2022-03-31 15:22         ` Ævar Arnfjörð Bjarmason
2022-04-01 10:51           ` Robert Coup
2022-03-28 14:02       ` [PATCH v4 7/7] docs: mention --refetch fetch option Robert Coup via GitGitGadget
2022-03-28 17:38         ` Junio C Hamano

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=xmqqwnhjfnpy.fsf@gitster.g \
    --to=gitster@pobox.com \
    --cc=derrickstolee@github.com \
    --cc=git@jeffhostetler.com \
    --cc=git@vger.kernel.org \
    --cc=gitgitgadget@gmail.com \
    --cc=johncai86@gmail.com \
    --cc=jonathantanmy@google.com \
    --cc=robert@coup.net.nz \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).