git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Adam Spiers <git@adamspiers.org>
To: git list <git@vger.kernel.org>
Subject: [PATCH 3/5] Documentation: add caveats about I/O buffering for check-{attr,ignore}
Date: Thu, 11 Apr 2013 02:59:33 +0100	[thread overview]
Message-ID: <1365645575-11428-3-git-send-email-git@adamspiers.org> (raw)
In-Reply-To: <1365645575-11428-1-git-send-email-git@adamspiers.org>

check-attr and check-ignore have the potential to deadlock callers
which do not read back the output in real-time.  For example, if a
caller writes N paths out and then reads N lines back in, it risks
becoming blocked on write() to check-*, and check-* is blocked on
write back to the caller.  Somebody has to buffer; the pipe buffers
provide some leeway, but they are limited.

Thanks to Peff for pointing this out:

    http://article.gmane.org/gmane.comp.version-control.git/220534

Signed-off-by: Adam Spiers <git@adamspiers.org>
---
 Documentation/git-check-attr.txt   |  5 +++++
 Documentation/git-check-ignore.txt |  5 +++++
 Documentation/git.txt              | 16 +++++++++-------
 3 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/Documentation/git-check-attr.txt b/Documentation/git-check-attr.txt
index 5abdbaa..a7be80d 100644
--- a/Documentation/git-check-attr.txt
+++ b/Documentation/git-check-attr.txt
@@ -56,6 +56,11 @@ being queried and <info> can be either:
 'set';;		when the attribute is defined as true.
 <value>;;	when a value has been assigned to the attribute.
 
+Buffering happens as documented under the `GIT_FLUSH` option in
+linkgit:git[1].  The caller is responsible for avoiding deadlocks
+caused by overfilling an input buffer or reading from an empty output
+buffer.
+
 EXAMPLES
 --------
 
diff --git a/Documentation/git-check-ignore.txt b/Documentation/git-check-ignore.txt
index 854e4d0..4014d28 100644
--- a/Documentation/git-check-ignore.txt
+++ b/Documentation/git-check-ignore.txt
@@ -66,6 +66,11 @@ are also used instead of colons and hard tabs:
 <source> <NULL> <linenum> <NULL> <pattern> <NULL> <pathname> <NULL>
 
 
+Buffering happens as documented under the `GIT_FLUSH` option in
+linkgit:git[1].  The caller is responsible for avoiding deadlocks
+caused by overfilling an input buffer or reading from an empty output
+buffer.
+
 EXIT STATUS
 -----------
 
diff --git a/Documentation/git.txt b/Documentation/git.txt
index 6a875f2..eecdb15 100644
--- a/Documentation/git.txt
+++ b/Documentation/git.txt
@@ -808,13 +808,15 @@ for further details.
 
 'GIT_FLUSH'::
 	If this environment variable is set to "1", then commands such
-	as 'git blame' (in incremental mode), 'git rev-list', 'git log',
-	and 'git whatchanged' will force a flush of the output stream
-	after each commit-oriented record have been flushed.   If this
-	variable is set to "0", the output of these commands will be done
-	using completely buffered I/O.   If this environment variable is
-	not set, Git will choose buffered or record-oriented flushing
-	based on whether stdout appears to be redirected to a file or not.
+	as 'git blame' (in incremental mode), 'git rev-list', 'git
+	log', 'git check-attr', 'git check-ignore', and 'git
+	whatchanged' will force a flush of the output stream after
+	each commit-oriented record have been flushed.  If this
+	variable is set to "0", the output of these commands will be
+	done using completely buffered I/O.  If this environment
+	variable is not set, Git will choose buffered or
+	record-oriented flushing based on whether stdout appears to be
+	redirected to a file or not.
 
 'GIT_TRACE'::
 	If this variable is set to "1", "2" or "true" (comparison
-- 
1.8.2.1.347.g37e0606

  parent reply	other threads:[~2013-04-11  1:59 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-08 18:13 RFC: two minor tweaks to check-ignore to help git-annex assistant Adam Spiers
2013-04-08 21:56 ` Junio C Hamano
2013-04-08 22:20 ` Jeff King
2013-04-11  1:59 ` [PATCH 1/5] check-ignore: move setup into cmd_check_ignore() Adam Spiers
2013-04-11  1:59   ` [PATCH 2/5] check-ignore: allow incremental streaming of queries via --stdin Adam Spiers
2013-04-11  5:31     ` Jeff King
2013-04-11 10:55       ` Adam Spiers
2013-04-11 11:20     ` Adam Spiers
2013-04-11 18:33       ` Jeff King
2013-04-11  1:59   ` Adam Spiers [this message]
2013-04-11  5:31     ` [PATCH 3/5] Documentation: add caveats about I/O buffering for check-{attr,ignore} Jeff King
2013-04-11  1:59   ` [PATCH 4/5] t0008: remove duplicated test fixture data Adam Spiers
2013-04-11  1:59   ` [PATCH 5/5] check-ignore: add -n / --non-matching option Adam Spiers
2013-04-11  5:25   ` [PATCH 1/5] check-ignore: move setup into cmd_check_ignore() Jeff King
2013-04-11 11:05     ` Adam Spiers
2013-04-11 12:05       ` [PATCH v2 1/5] t0008: remove duplicated test fixture data Adam Spiers
2013-04-11 12:05         ` [PATCH v2 2/5] check-ignore: add -n / --non-matching option Adam Spiers
2013-04-11 12:05         ` [PATCH v2 3/5] check-ignore: move setup into cmd_check_ignore() Adam Spiers
2013-04-11 12:05         ` [PATCH v2 4/5] check-ignore: allow incremental streaming of queries via --stdin Adam Spiers
2013-04-11 19:11           ` Jeff King
2013-04-11 20:31             ` Adam Spiers
2013-04-11 20:40               ` Jeff King
2013-04-22 18:03               ` Junio C Hamano
2013-04-24  8:02                 ` Adam Spiers
2013-04-29 22:55                   ` [PATCH] t0008: use named pipe (FIFO) to test check-ignore streaming Adam Spiers
2013-04-11 21:04           ` [PATCH v2 4/5] check-ignore: allow incremental streaming of queries via --stdin Aaron Schrab
2013-04-11 22:55             ` Adam Spiers
2013-04-11 12:05         ` [PATCH v2 5/5] Documentation: add caveats about I/O buffering for check-{attr,ignore} Adam Spiers
2013-04-11 18:09           ` Junio C Hamano
2013-04-11 20:12             ` [PATCH v3 " Adam Spiers
2013-04-12  2:12               ` Junio C Hamano
2013-04-12 11:00                 ` Adam Spiers
2013-04-11 18:35       ` [PATCH 1/5] check-ignore: move setup into cmd_check_ignore() Jeff King

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=1365645575-11428-3-git-send-email-git@adamspiers.org \
    --to=git@adamspiers.org \
    --cc=git@vger.kernel.org \
    /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).