git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "SZEDER Gábor" <szeder.dev@gmail.com>
To: Andrew Baumann <Andrew.Baumann@microsoft.com>
Cc: "Johannes Schindelin" <johannes.schindelin@gmx.de>,
	git@vger.kernel.org, "SZEDER Gábor" <szeder.dev@gmail.com>
Subject: [PATCH] setup.c: don't try to access '//HEAD' during repo discovery
Date: Fri,  3 Nov 2017 02:06:45 +0100	[thread overview]
Message-ID: <20171103010645.25499-1-szeder.dev@gmail.com> (raw)
In-Reply-To: <MWHPR21MB0159089DB91A30E5E5CFCB479E5C0@MWHPR21MB0159.namprd21.prod.outlook.com>

Commit ce9b8aab5 (setup_git_directory_1(): avoid changing global state,
2017-03-13) changed how the git directory is discovered, and as a side
effect when the discovery reaches the root directory Git tries to
access paths like '//HEAD' and '//objects'.  This interacts badly with
Cygwin, which interprets it as a UNC file share, and so demand-loads a
bunch of extra DLLs and attempts to resolve/contact the server named
HEAD.  This obviously doesn't work too well, especially over a slow
network link.

Special case the root directory in is_git_directory() to prevent
accessing paths with leading double slashes.

Reported-by: Andrew Baumann <Andrew.Baumann@microsoft.com>
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
---

I'm not quite sure whether this is the right or complete fix.  I can't
test it on Cygwin, and I don't know whether Git for Windows is
affected with it's drive prefixes in paths.  Anyway, at least strace
output on Linux looks good to me.

 setup.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/setup.c b/setup.c
index 03f51e056..0cfc5e676 100644
--- a/setup.c
+++ b/setup.c
@@ -311,6 +311,10 @@ int is_git_directory(const char *suspect)
 	int ret = 0;
 	size_t len;
 
+	/* To avoid accessing '//HEAD' & co when checking the root directory */
+	if (!strcmp(suspect, "/"))
+		suspect = "";
+
 	/* Check worktree-related signatures */
 	strbuf_addf(&path, "%s/HEAD", suspect);
 	if (validate_headref(path.buf))
-- 
2.15.0.67.gb67a46776


  parent reply	other threads:[~2017-11-03  1:06 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-02 23:45 git tries to stat //HEAD when searching for a repo, leading to huge delays on Cygwin Andrew Baumann
2017-11-03  1:03 ` Jeff King
2017-11-03 12:32   ` Johannes Schindelin
2017-11-03 18:29     ` Jeff King
2017-11-03  1:06 ` SZEDER Gábor [this message]
2017-11-03 12:36   ` [PATCH] setup.c: don't try to access '//HEAD' during repo discovery 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=20171103010645.25499-1-szeder.dev@gmail.com \
    --to=szeder.dev@gmail.com \
    --cc=Andrew.Baumann@microsoft.com \
    --cc=git@vger.kernel.org \
    --cc=johannes.schindelin@gmx.de \
    /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).