All of lore.kernel.org
 help / color / mirror / Atom feed
From: Matheus Tavares Bernardino <matheus.bernardino@usp.br>
To: Jonathan Tan <jonathantanmy@google.com>
Cc: git <git@vger.kernel.org>,
	Emily Shaffer <emilyshaffer@google.com>,
	Junio C Hamano <gitster@pobox.com>,
	Ramsay Jones <ramsay@ramsayjones.plus.com>,
	Josh Steadmon <steadmon@google.com>
Subject: Re: [PATCH v2 7/8] submodule-config: pass repo upon blob config read
Date: Mon, 16 Aug 2021 16:57:05 -0300	[thread overview]
Message-ID: <CAHd-oW6o2nSd15NPE3H8zjTtTehVWLsxF6oR5cjYw8uKUEVhUQ@mail.gmail.com> (raw)
In-Reply-To: <CAHd-oW7nv-Y_QmhA2gcX5GoWxus88inmbcXQ5kPiSRLBdhtoUw@mail.gmail.com>

On Mon, Aug 16, 2021 at 11:32 AM Matheus Tavares Bernardino
<matheus.bernardino@usp.br> wrote:
>
> On Fri, Aug 13, 2021 at 6:05 PM Jonathan Tan <jonathantanmy@google.com> wrote:
> >
> > When reading the config of a submodule, if reading from a blob, read
> > using an explicitly specified repository instead of by adding the
> > submodule's ODB as an alternate and then reading an object from
> > the_repository.
>
> Great!
>
> At first, I thought this would also allow us to remove another
> NEEDSWORK comment in grep_submodule(), together with a lock
> protection:
>
> /*
>  * NEEDSWORK: repo_read_gitmodules() might call
>  * add_to_alternates_memory() via config_from_gitmodules(). This
>  * operation causes a race condition with concurrent object readings
>  * performed by the worker threads. That's why we need obj_read_lock()
>  * here. It should be removed once it's no longer necessary to add the
>  * subrepo's odbs to the in-memory alternates list.
>  */
> obj_read_lock();
> repo_read_gitmodules(subrepo, 0);
>
> Back when I wrote this comment, my conclusion was that the alternates
> mechanics were the only thread-unsafe object-reading operations in
> repo_read_gitmodules()'s call chains. So once the add-to-alternates
> mechanics were gone, we could also remove the lock.
>
> But with further inspection now, I see that this is not really the
> case. For example, we have a few global variables in packfile.c
> collecting some statistics (pack_mmap_calls, pack_open_windows, etc.)
> which are updated on obj readings from both the_repository *and*
> submodules.

Sorry, this is incorrect. I forgot that repo_read_object_file() (which
is part of repo_read_gitmodules()'s call chain) also acquires the
obj_read_mutex before accessing those global variables. So the
NEEDSWORK might be right.

Nevertheless, I think it might be better to look into
repo_read_gitmodules() more carefully before removing this lock. And
this is something for another series. Sorry about the noise.

  reply	other threads:[~2021-08-16 19:57 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-10 18:28 [PATCH 0/7] In grep, no adding submodule ODB as alternates Jonathan Tan
2021-08-10 18:28 ` [PATCH 1/7] submodule: lazily add submodule ODBs " Jonathan Tan
2021-08-10 21:13   ` Junio C Hamano
2021-08-13 16:53     ` Jonathan Tan
2021-08-11 21:33   ` Emily Shaffer
2021-08-13 16:23     ` Jonathan Tan
2021-08-10 18:28 ` [PATCH 2/7] grep: use submodule-ODB-as-alternate lazy-addition Jonathan Tan
2021-08-11 21:36   ` Emily Shaffer
2021-08-13 16:31     ` Jonathan Tan
2021-08-10 18:28 ` [PATCH 3/7] grep: typesafe versions of grep_source_init Jonathan Tan
2021-08-10 21:38   ` Junio C Hamano
2021-08-11 21:42   ` Emily Shaffer
2021-08-11 23:07     ` Ramsay Jones
2021-08-13 16:32       ` Jonathan Tan
2021-08-11 22:45   ` Matheus Tavares Bernardino
2021-08-12 16:49     ` Junio C Hamano
2021-08-13 16:33       ` Jonathan Tan
2021-08-10 18:28 ` [PATCH 4/7] grep: read submodule entry with explicit repo Jonathan Tan
2021-08-11 21:44   ` Emily Shaffer
2021-08-13 16:39     ` Jonathan Tan
2021-08-10 18:28 ` [PATCH 5/7] grep: allocate subrepos on heap Jonathan Tan
2021-08-11 21:50   ` Emily Shaffer
2021-08-13 16:42     ` Jonathan Tan
2021-08-10 18:28 ` [PATCH 6/7] grep: add repository to OID grep sources Jonathan Tan
2021-08-11 21:52   ` Emily Shaffer
2021-08-13 16:44     ` Jonathan Tan
2021-08-11 23:28   ` Matheus Tavares Bernardino
2021-08-13 16:47     ` Jonathan Tan
2021-08-10 18:28 ` [PATCH 7/7] t7814: show lack of alternate ODB-adding Jonathan Tan
2021-08-11 21:55   ` Emily Shaffer
2021-08-11 22:22     ` Matheus Tavares Bernardino
2021-08-13 16:50       ` Jonathan Tan
2021-08-11 21:29 ` [PATCH 0/7] In grep, no adding submodule ODB as alternates Emily Shaffer
2021-08-11 22:49 ` Josh Steadmon
2021-08-13 21:05 ` [PATCH v2 0/8] " Jonathan Tan
2021-08-13 21:05   ` [PATCH v2 1/8] submodule: lazily add submodule ODBs " Jonathan Tan
2021-08-13 21:05   ` [PATCH v2 2/8] grep: use submodule-ODB-as-alternate lazy-addition Jonathan Tan
2021-08-13 21:05   ` [PATCH v2 3/8] grep: typesafe versions of grep_source_init Jonathan Tan
2021-08-16 15:06     ` Matheus Tavares Bernardino
2021-08-13 21:05   ` [PATCH v2 4/8] grep: read submodule entry with explicit repo Jonathan Tan
2021-08-13 21:05   ` [PATCH v2 5/8] grep: allocate subrepos on heap Jonathan Tan
2021-08-13 21:44     ` Junio C Hamano
2021-08-16 19:42       ` Jonathan Tan
2021-08-13 21:05   ` [PATCH v2 6/8] grep: add repository to OID grep sources Jonathan Tan
2021-08-16 14:48     ` Matheus Tavares Bernardino
2021-08-16 19:44       ` Jonathan Tan
2021-08-13 21:05   ` [PATCH v2 7/8] submodule-config: pass repo upon blob config read Jonathan Tan
2021-08-16 14:32     ` Matheus Tavares Bernardino
2021-08-16 19:57       ` Matheus Tavares Bernardino [this message]
2021-08-16 20:02       ` Jonathan Tan
2021-08-16 15:48     ` Matheus Tavares Bernardino
2021-08-16 20:09       ` Jonathan Tan
2021-08-16 20:57         ` Jonathan Tan
2021-08-13 21:05   ` [PATCH v2 8/8] t7814: show lack of alternate ODB-adding Jonathan Tan
2021-08-16 15:14   ` [PATCH v2 0/8] In grep, no adding submodule ODB as alternates Matheus Tavares Bernardino
2021-08-16 21:09 ` [PATCH v3 " Jonathan Tan
2021-08-16 21:09   ` [PATCH v3 1/8] submodule: lazily add submodule ODBs " Jonathan Tan
2021-08-16 21:09   ` [PATCH v3 2/8] grep: use submodule-ODB-as-alternate lazy-addition Jonathan Tan
2021-08-16 21:09   ` [PATCH v3 3/8] grep: typesafe versions of grep_source_init Jonathan Tan
2021-08-16 21:09   ` [PATCH v3 4/8] grep: read submodule entry with explicit repo Jonathan Tan
2021-08-16 21:09   ` [PATCH v3 5/8] grep: allocate subrepos on heap Jonathan Tan
2021-08-16 21:09   ` [PATCH v3 6/8] grep: add repository to OID grep sources Jonathan Tan
2021-09-27 12:08     ` Ævar Arnfjörð Bjarmason
2021-09-27 16:45       ` [RFC PATCH 0/3] grep: don'\''t add subrepos to in-memory alternates Matheus Tavares
2021-09-27 17:30         ` Ævar Arnfjörð Bjarmason
2021-08-16 21:09   ` [PATCH v3 7/8] submodule-config: pass repo upon blob config read Jonathan Tan
2021-08-16 21:09   ` [PATCH v3 8/8] t7814: show lack of alternate ODB-adding Jonathan Tan
2021-08-17 19:29   ` [PATCH v3 0/8] In grep, no adding submodule ODB as alternates Matheus Tavares Bernardino
2021-09-08  0:26   ` Junio C Hamano
2021-09-08 15:31     ` Matheus Tavares Bernardino
2021-09-08 18:45       ` 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=CAHd-oW6o2nSd15NPE3H8zjTtTehVWLsxF6oR5cjYw8uKUEVhUQ@mail.gmail.com \
    --to=matheus.bernardino@usp.br \
    --cc=emilyshaffer@google.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jonathantanmy@google.com \
    --cc=ramsay@ramsayjones.plus.com \
    --cc=steadmon@google.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.