git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: "Carlo Marcelo Arenas Belón" <carenas@gmail.com>
Cc: git@vger.kernel.org, lehmacdj@gmail.com
Subject: Re: [PATCH 2/3] credential-cache: check for windows specific errors
Date: Sun, 12 Sep 2021 18:10:44 -0700	[thread overview]
Message-ID: <xmqqwnnlgu0b.fsf@gitster.g> (raw)
In-Reply-To: <20210912202830.25720-3-carenas@gmail.com> ("Carlo Marcelo Arenas Belón"'s message of "Sun, 12 Sep 2021 13:28:29 -0700")

Carlo Marcelo Arenas Belón  <carenas@gmail.com> writes:

> Connect and reset errors aren't what will be expected by POSIX but
> are compatible with the ones used by WinSock.
>
> Alternatively a translation layer for read could be added (similar
> to the one provided by mingw-write()) to translate the odd EINVAL,
> into a more reasonable EPIPE, but this is more localized and still
> unlikely to cause confusion in other systems.

Are we sure EINVAL or ENETDOWN from outside the Windows world is
tolerable after read_in_full() failure in this context?

I'd rather see something like

	if (!r || (r < 0 && connection_closed(errno)))
		break;

with a local helper function connection_closed() allowing EINVAL
only with "#ifdef WINDOWS" to be safe.  The same for "the socket is
not yet open" codepath with another local helper function.

Thanks.

>
> Signed-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>
> ---
>  builtin/credential-cache.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/builtin/credential-cache.c b/builtin/credential-cache.c
> index 76a6ba3722..b12a79ae01 100644
> --- a/builtin/credential-cache.c
> +++ b/builtin/credential-cache.c
> @@ -28,7 +28,8 @@ static int send_request(const char *socket, const struct strbuf *out)
>  		int r;
>  
>  		r = read_in_full(fd, in, sizeof(in));
> -		if (r == 0 || (r < 0 && errno == ECONNRESET))
> +		if (r == 0 ||
> +			(r < 0 && (errno == ECONNRESET || errno == EINVAL)))
>  			break;
>  		if (r < 0)
>  			die_errno("read error from cache daemon");
> @@ -75,7 +76,7 @@ static void do_cache(const char *socket, const char *action, int timeout,
>  	}
>  
>  	if (send_request(socket, &buf) < 0) {
> -		if (errno != ENOENT && errno != ECONNREFUSED)
> +		if (errno != ENOENT && errno != ECONNREFUSED && errno != ENETDOWN)
>  			die_errno("unable to connect to cache daemon");
>  		if (flags & FLAG_SPAWN) {
>  			spawn_daemon(socket);

  reply	other threads:[~2021-09-13  1:10 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-12 20:28 [PATCH 0/3] windows: allow building without NO_UNIX_SOCKETS Carlo Marcelo Arenas Belón
2021-09-12 20:28 ` [PATCH 1/3] t0301: fixes for windows compatibility Carlo Marcelo Arenas Belón
2021-09-13  1:04   ` Junio C Hamano
2021-09-13  5:34   ` Bagas Sanjaya
2021-09-13  7:13     ` Carlo Arenas
2021-09-13 18:01       ` Junio C Hamano
2021-09-12 20:28 ` [PATCH 2/3] credential-cache: check for windows specific errors Carlo Marcelo Arenas Belón
2021-09-13  1:10   ` Junio C Hamano [this message]
2021-09-12 20:28 ` [PATCH 3/3] git-compat-util: include declaration for unix sockets Carlo Marcelo Arenas Belón
2021-09-13  8:55 ` [PATCH v2 0/3] windows: allow building without NO_UNIX_SOCKETS Carlo Marcelo Arenas Belón
2021-09-13  8:55   ` [PATCH v2 1/3] t0301: fixes for windows compatibility Carlo Marcelo Arenas Belón
2021-09-13 11:50     ` Johannes Schindelin
2021-09-13 18:09       ` Junio C Hamano
2021-09-13  8:55   ` [PATCH v2 2/3] credential-cache: check for windows specific errors Carlo Marcelo Arenas Belón
2021-09-13 11:58     ` Johannes Schindelin
2021-09-13  8:56   ` [PATCH v2 3/3] git-compat-util: include declaration for unix sockets Carlo Marcelo Arenas Belón
2021-09-13 11:59     ` Johannes Schindelin
2021-09-13 11:42   ` [PATCH v2 0/3] windows: allow building without NO_UNIX_SOCKETS Johannes Schindelin
2021-09-14  7:25   ` [PATCH v3 " Carlo Marcelo Arenas Belón
2021-09-14  7:25     ` [PATCH v3 1/3] t0301: fixes for windows compatibility Carlo Marcelo Arenas Belón
2021-11-02  0:37       ` Ævar Arnfjörð Bjarmason
2021-09-14  7:25     ` [PATCH v3 2/3] credential-cache: check for windows specific errors Carlo Marcelo Arenas Belón
2021-09-14 16:43       ` Junio C Hamano
2021-09-14 19:09       ` What should happen in credential-cache on recoverable error without SPAWN option? Junio C Hamano
2021-09-14 19:33         ` Jeff King
2021-09-14  7:26     ` [PATCH v3 3/3] git-compat-util: include declaration for unix sockets in windows Carlo Marcelo Arenas Belón
2021-09-14 16:37     ` [PATCH v3 0/3] windows: allow building without NO_UNIX_SOCKETS 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=xmqqwnnlgu0b.fsf@gitster.g \
    --to=gitster@pobox.com \
    --cc=carenas@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=lehmacdj@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).