git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jonathan Tan <jonathantanmy@google.com>
To: gitgitgadget@gmail.com
Cc: git@vger.kernel.org, hariom18599@gmail.com,
	Jonathan Tan <jonathantanmy@google.com>
Subject: Re: [PATCH 1/2] fetch-pack: remove fetch_if_missing=0
Date: Thu,  7 May 2020 12:43:54 -0700	[thread overview]
Message-ID: <20200507194354.33347-1-jonathantanmy@google.com> (raw)
In-Reply-To: <eb0cbeeeed080596c130f657186894999ae6121b.1587412477.git.gitgitgadget@gmail.com>

> From: Hariom Verma <hariom18599@gmail.com>
> 
> Commit 6462d5e ("fetch: remove fetch_if_missing=0", 2019-11-08)
> strove to remove the need for fetch_if_missing=0 from the fetching
> mechanism, so it is plausible to attempt removing fetch_if_missing=0
> from fetch-pack as well.
> 
> Signed-off-by: Hariom Verma <hariom18599@gmail.com>

As Christian said [1], please include tests like in the commit you
mentioned. For a change like this, I think that the test is the most
important part.

Also include a justification for why it's safe to remove
fetch_if_missing=0. You can probably cite the aforementioned commit to
say that it covers the fetch_pack() method, and then go through the rest
of the code to see if any may inadvertently fetch an object.

Also, the fetch-pack and index-pack parts can be sent in separate patch
sets, so you might want to concentrate on one command first.

[1] https://lore.kernel.org/git/CAP8UFD2SNnpKWtYUztZ76OU7zBsrXyYhG_Zds1wi+NqBKCv+Qw@mail.gmail.com/

> diff --git a/fetch-pack.c b/fetch-pack.c
> index 1734a573b01..1ca643f6491 100644
> --- a/fetch-pack.c
> +++ b/fetch-pack.c
> @@ -1649,7 +1649,7 @@ static void update_shallow(struct fetch_pack_args *args,
>  		struct oid_array extra = OID_ARRAY_INIT;
>  		struct object_id *oid = si->shallow->oid;
>  		for (i = 0; i < si->shallow->nr; i++)
> -			if (has_object_file(&oid[i]))
> +			if (has_object_file_with_flags(&oid[i], OBJECT_INFO_SKIP_FETCH_OBJECT))
>  				oid_array_append(&extra, &oid[i]);
>  		if (extra.nr) {
>  			setup_alternate_shallow(&shallow_lock,

Hmm...this triggers when the user requests a clone that is both partial
and shallow, and the server reports a shallow object that it didn't send
back as a packfile; and it causes another fetch to be sent. This is a
separate issue, but Hariom, if you'd like to take a look at this, that
would work out too. You'll need to figure out how to make the server
send back shallow lines referencing objects that are not in the packfile
- one way to do it is to use one-time-perl. (Search the codebase to see
how it is used.) This is probably more complex, though.

  parent reply	other threads:[~2020-05-07 19:44 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-20 19:54 [PATCH 0/2] [WIP] removed fetch_if_missing global Hariom Verma via GitGitGadget
2020-04-20 19:54 ` [PATCH 1/2] fetch-pack: remove fetch_if_missing=0 Hariom Verma via GitGitGadget
2020-05-07 13:17   ` Christian Couder
2020-05-07 15:02     ` Junio C Hamano
2020-05-09 17:48       ` Hariom verma
2020-05-09 17:40     ` Hariom verma
2020-05-07 19:43   ` Jonathan Tan [this message]
2020-05-09 18:00     ` Hariom verma
2023-02-20 18:13     ` Kousik Sanagavarapu
2023-02-22 18:19       ` Jonathan Tan
2023-02-22 17:45     ` Kousik Sanagavarapu
2020-04-20 19:54 ` [PATCH 2/2] index-pack: " Hariom Verma via GitGitGadget
2020-05-07 13:10 ` [PATCH 0/2] [WIP] removed fetch_if_missing global Christian Couder
2020-05-09 17:35   ` Hariom verma
2023-02-17 17:20 ` Kousik Sanagavarapu
2023-02-18 17:00   ` Christian Couder
2023-02-19  3:57     ` Kousik Sanagavarapu
2023-02-19  7:37       ` Christian Couder
2023-02-19 10:40   ` Hariom verma

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=20200507194354.33347-1-jonathantanmy@google.com \
    --to=jonathantanmy@google.com \
    --cc=git@vger.kernel.org \
    --cc=gitgitgadget@gmail.com \
    --cc=hariom18599@gmail.com \
    /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).