Git Mailing List Archive on lore.kernel.org
 help / color / Atom feed
From: Johannes Schindelin <Johannes.Schindelin@gmx.de>
To: Emily Shaffer <emilyshaffer@google.com>
Cc: git@vger.kernel.org, Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH v3] fetch: emphasize failure during submodule fetch
Date: Fri, 17 Jan 2020 11:54:29 +0100 (CET)
Message-ID: <nycvar.QRO.7.76.6.2001171152410.46@tvgsbejvaqbjf.bet> (raw)
In-Reply-To: <20200116222012.162642-1-emilyshaffer@google.com>

Hi Emily,

On Thu, 16 Jan 2020, Emily Shaffer wrote:

> In cases when a submodule fetch fails when there are many submodules, the error
> from the lone failing submodule fetch is buried under activity on the other
> submodules if more than one fetch fell back on fetch-by-oid. Call out a failure
> late so the user is aware that something went wrong, and where.
>
> Because fetch_finish() is only called synchronously by
> run_processes_parallel, mutexing is not required around
> submodules_with_errors.
>
> Signed-off-by: Emily Shaffer <emilyshaffer@google.com>
> ---
> Since v2, removed mutexing as run_processes_parallel() creates
> subprocesses, not threads, and is in fact synchronous. Also added
> translation marker for the error message.

Excellent. For Git for Windows' ever-green branch based on `pu`, I had to
introduce this:

-- snip --
Subject: [PATCH] fixup??? fetch: emphasize failure during submodule fetch

This would be the first user of `PTHREAD_MUTEX_INITIALIZER`, but as we
have a hard time to emulate this with critical sections in Windows,
let's not do that.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
---
 submodule.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/submodule.c b/submodule.c
index a65db54cece..ec54f4db2cd 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1286,7 +1286,7 @@ struct submodule_parallel_fetch {
 };
 #define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
 		  STRING_LIST_INIT_DUP, \
-		  NULL, 0, 0, STRBUF_INIT, PTHREAD_MUTEX_INITIALIZER}
+		  NULL, 0, 0, STRBUF_INIT}

 static int get_fetch_recurse_config(const struct submodule *submodule,
 				    struct submodule_parallel_fetch *spf)
@@ -1614,6 +1614,7 @@ int fetch_populated_submodules(struct repository *r,
 	spf.default_option = default_option;
 	spf.quiet = quiet;
 	spf.prefix = prefix;
+	pthread_mutex_init(&spf.submodule_errors_mutex, NULL);

 	if (!r->worktree)
 		goto out;
-- snap --

I look forward to v3 hitting `pu` and making this change obsolete.

Thanks,
Dscho

>
>  - Emily
>
>  submodule.c | 19 +++++++++++++++----
>  1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/submodule.c b/submodule.c
> index 9da7181321..ee74acee27 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -1280,10 +1280,12 @@ struct submodule_parallel_fetch {
>  	/* Pending fetches by OIDs */
>  	struct fetch_task **oid_fetch_tasks;
>  	int oid_fetch_tasks_nr, oid_fetch_tasks_alloc;
> +
> +	struct strbuf submodules_with_errors;
>  };
>  #define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
>  		  STRING_LIST_INIT_DUP, \
> -		  NULL, 0, 0}
> +		  NULL, 0, 0, STRBUF_INIT}
>
>  static int get_fetch_recurse_config(const struct submodule *submodule,
>  				    struct submodule_parallel_fetch *spf)
> @@ -1547,7 +1549,10 @@ static int fetch_finish(int retvalue, struct strbuf *err,
>  	struct string_list_item *it;
>  	struct oid_array *commits;
>
> -	if (retvalue)
> +	if (!task || !task->sub)
> +		BUG("callback cookie bogus");
> +
> +	if (retvalue) {
>  		/*
>  		 * NEEDSWORK: This indicates that the overall fetch
>  		 * failed, even though there may be a subsequent fetch
> @@ -1557,8 +1562,9 @@ static int fetch_finish(int retvalue, struct strbuf *err,
>  		 */
>  		spf->result = 1;
>
> -	if (!task || !task->sub)
> -		BUG("callback cookie bogus");
> +		strbuf_addf(&spf->submodules_with_errors, "\t%s\n",
> +			    task->sub->name);
> +	}
>
>  	/* Is this the second time we process this submodule? */
>  	if (task->commits)
> @@ -1627,6 +1633,11 @@ int fetch_populated_submodules(struct repository *r,
>  				   &spf,
>  				   "submodule", "parallel/fetch");
>
> +	if (spf.submodules_with_errors.len > 0)
> +		fprintf(stderr, _("Errors during submodule fetch:\n%s"),
> +			spf.submodules_with_errors.buf);
> +
> +
>  	argv_array_clear(&spf.args);
>  out:
>  	free_submodules_oids(&spf.changed_submodule_names);
> --
> 2.25.0.rc1.283.g88dfdc4193-goog
>
>

      reply index

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-10 19:55 [PATCH] " Emily Shaffer
2020-01-10 20:18 ` Junio C Hamano
2020-01-10 23:01   ` Emily Shaffer
2020-01-16  2:59 ` [PATCH v2] " Emily Shaffer
2020-01-16 18:23   ` Junio C Hamano
2020-01-16 21:55     ` Emily Shaffer
2020-01-16 22:04       ` Emily Shaffer
2020-01-16 22:20   ` [PATCH v3] " Emily Shaffer
2020-01-17 10:54     ` Johannes Schindelin [this message]

Reply instructions:

You may reply publically 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=nycvar.QRO.7.76.6.2001171152410.46@tvgsbejvaqbjf.bet \
    --to=johannes.schindelin@gmx.de \
    --cc=emilyshaffer@google.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.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

Git Mailing List Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/git/0 git/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 git git/ https://lore.kernel.org/git \
		git@vger.kernel.org
	public-inbox-index git

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.git


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git