git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff King <peff@peff.net>
To: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Cc: "Junio C Hamano" <gitster@pobox.com>,
	"SZEDER Gábor" <szeder.dev@gmail.com>,
	"Git mailing list" <git@vger.kernel.org>,
	"Clemens Buchacher" <drizzd@gmx.net>
Subject: [PATCH 1/2] fetch: avoid calling write_or_die()
Date: Sun, 3 Mar 2019 11:55:37 -0500	[thread overview]
Message-ID: <20190303165537.GA23755@sigill.intra.peff.net> (raw)
In-Reply-To: <20190303165447.GA31116@sigill.intra.peff.net>

The write_or_die() function has one quirk that a caller might not
expect: when it sees EPIPE from the write() call, it translates that
into a death by SIGPIPE. This doesn't change the overall behavior (the
program exits either way), but it does potentially confuse test scripts
looking for a non-signal exit code.

Let's switch away from using write_or_die() in a few code paths, which
will give us more consistent exit codes. It also gives us the
opportunity to write more descriptive error messages, since we have
context that write_or_die() does not.

Note that this won't do much by itself, since we'd typically be killed
by SIGPIPE before write_or_die() even gets a chance to do its thing.
That will be addressed in the next patch.

Signed-off-by: Jeff King <peff@peff.net>
---
 fetch-pack.c | 9 ++++++---
 pkt-line.c   | 6 ++++--
 2 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/fetch-pack.c b/fetch-pack.c
index 812be15d7e..dca249e9d7 100644
--- a/fetch-pack.c
+++ b/fetch-pack.c
@@ -191,8 +191,10 @@ static void send_request(struct fetch_pack_args *args,
 	if (args->stateless_rpc) {
 		send_sideband(fd, -1, buf->buf, buf->len, LARGE_PACKET_MAX);
 		packet_flush(fd);
-	} else
-		write_or_die(fd, buf->buf, buf->len);
+	} else {
+		if (write_in_full(fd, buf->buf, buf->len) < 0)
+			die_errno("unable to write to remote");
+	}
 }
 
 static void insert_one_alternate_object(struct fetch_negotiator *negotiator,
@@ -1163,7 +1165,8 @@ static int send_fetch_request(struct fetch_negotiator *negotiator, int fd_out,
 
 	/* Send request */
 	packet_buf_flush(&req_buf);
-	write_or_die(fd_out, req_buf.buf, req_buf.len);
+	if (write_in_full(fd_out, req_buf.buf, req_buf.len) < 0)
+		die_errno("unable to write request to remote");
 
 	strbuf_release(&req_buf);
 	return ret;
diff --git a/pkt-line.c b/pkt-line.c
index d4b71d3e82..093b2f3976 100644
--- a/pkt-line.c
+++ b/pkt-line.c
@@ -88,13 +88,15 @@ static void packet_trace(const char *buf, unsigned int len, int write)
 void packet_flush(int fd)
 {
 	packet_trace("0000", 4, 1);
-	write_or_die(fd, "0000", 4);
+	if (write_in_full(fd, "0000", 4) < 0)
+		die_errno("unable to write flush packet");
 }
 
 void packet_delim(int fd)
 {
 	packet_trace("0001", 4, 1);
-	write_or_die(fd, "0001", 4);
+	if (write_in_full(fd, "0001", 4) < 0)
+		die_errno("unable to write delim packet");
 }
 
 int packet_flush_gently(int fd)
-- 
2.21.0.675.g01c085a870


  reply	other threads:[~2019-03-03 16:55 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-06 15:11 t5570-git-daemon fails with SIGPIPE on OSX SZEDER Gábor
2018-08-06 15:31 ` SZEDER Gábor
2019-02-08  8:32   ` Johannes Schindelin
2019-02-08 12:54     ` SZEDER Gábor
2018-08-14 22:32 ` Jeff King
2018-08-14 22:37   ` Jeff King
2019-02-08  9:02   ` Johannes Schindelin
2019-02-08  9:28     ` Johannes Schindelin
2019-02-08 19:54       ` Jeff King
2019-03-01 15:02         ` Johannes Schindelin
2019-03-01 19:00           ` Jeff King
2019-03-02 21:21             ` Johannes Schindelin
2019-03-03 16:54               ` Jeff King
2019-03-03 16:55                 ` Jeff King [this message]
2019-03-04 13:42                   ` [PATCH 1/2] fetch: avoid calling write_or_die() Duy Nguyen
2019-03-05  4:11                     ` Jeff King
2019-03-03 16:58                 ` [PATCH 2/2] fetch: ignore SIGPIPE during network operation Jeff King
2019-03-04  1:11                   ` Junio C Hamano
2019-03-05  4:11                     ` Jeff King
2019-03-03  1:21             ` t5570-git-daemon fails with SIGPIPE on OSX Junio C Hamano
2019-03-03 14:56               ` Johannes Schindelin

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=20190303165537.GA23755@sigill.intra.peff.net \
    --to=peff@peff.net \
    --cc=Johannes.Schindelin@gmx.de \
    --cc=drizzd@gmx.net \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=szeder.dev@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).