linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
To: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v2] kmod: Avoid deadlock by recursive kmod call.
Date: Wed, 4 Jan 2012 22:08:09 +0900	[thread overview]
Message-ID: <201201042208.JHG52695.OLStHFQOFFMVJO@I-love.SAKURA.ne.jp> (raw)
In-Reply-To: <201201040743.q047h9OD055670@www262.sakura.ne.jp>

Tetsuo Handa wrote:
> I noticed that the system deadlocks if do_execve() request by kmod() triggered
> recursive do_execve() request.
> 
> An example operation for triggering this deadlock:
> 
>   # : > /tmp/dummy
>   # chmod 755 /tmp/dummy
>   # echo /tmp/dummy > /proc/sys/kernel/hotplug
>   # modprobe whatever
> 
> Although this patch works for me, I'm not sure that this is a correct fix.
> Also, my analysis in the patch description may be wrong. Please check.

Well, we should not unconditionally prevent all vfork()ed threads from using
request_module() at search_binary_handler(). We should prevent only kernel
threads created by __call_usermodehelper(). Updated to filter only kmod threads.
--------------------
[PATCH v2] kmod: Avoid deadlock by recursive kmod call.

call_usermodehelper(UMH_WAIT_EXEC or UMH_WAIT_PROC) request depends on an
assumption that do_execve() will not trigger recursive
call_usermodehelper(UMH_WAIT_EXEC or UMH_WAIT_PROC) request, for
"khelper cannot start processing recursive call_usermodehelper() request until
do_execve() of original call_usermodehelper() request completes" but
"do_execve() of original call_usermodehelper() cannot be completed until
recursive call_usermodehelper() request completes".

There are several callers that break this assumption and lead to deadlock.
We need to make sure that do_execve() request by kmod thread does not trigger
recursive call_usermodehelper(UMH_WAIT_PROC or UMH_WAIT_EXEC).

Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
---
 include/linux/sched.h |    3 +++
 kernel/kmod.c         |   18 +++++++++++++++++-
 2 files changed, 20 insertions(+), 1 deletion(-)

--- linux-3.1.7.orig/include/linux/sched.h
+++ linux-3.1.7/include/linux/sched.h
@@ -1305,6 +1305,9 @@ struct task_struct {
 	unsigned sched_reset_on_fork:1;
 	unsigned sched_contributes_to_load:1;
 
+	/* Prevent recursive kmod request. */
+	unsigned kmod_thread:1;
+
 	pid_t pid;
 	pid_t tgid;
 
--- linux-3.1.7.orig/kernel/kmod.c
+++ linux-3.1.7/kernel/kmod.c
@@ -189,6 +189,13 @@ fail:
 	do_exit(0);
 }
 
+static int call_helper(void *data)
+{
+	/* Do not trigger recursive kmod call. */
+	current->kmod_thread = 1;
+	return ____call_usermodehelper(data);
+}
+
 void call_usermodehelper_freeinfo(struct subprocess_info *info)
 {
 	if (info->cleanup)
@@ -252,7 +259,7 @@ static void __call_usermodehelper(struct
 		pid = kernel_thread(wait_for_helper, sub_info,
 				    CLONE_FS | CLONE_FILES | SIGCHLD);
 	else
-		pid = kernel_thread(____call_usermodehelper, sub_info,
+		pid = kernel_thread(call_helper, sub_info,
 				    CLONE_VFORK | SIGCHLD);
 
 	switch (wait) {
@@ -428,6 +435,15 @@ int call_usermodehelper_exec(struct subp
 		retval = -EBUSY;
 		goto out;
 	}
+	/*
+	 * We can't call wait_for_completion() if current thread was created by
+	 * call_helper(), or we will deadlock when current thread calls
+	 * request_module() from search_binary_handler().
+	 */
+	if (wait != UMH_NO_WAIT && current->kmod_thread) {
+		retval = -EBUSY;
+		goto out;
+	}
 
 	sub_info->complete = &done;
 	sub_info->wait = wait;

  reply	other threads:[~2012-01-04 13:08 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-12-27  7:01 [3.1.6] Deadlock upon recursive call_usermodehelper_exec() penguin-kernel
2011-12-27  8:14 ` Tetsuo Handa
2012-01-04  7:43 ` [PATCH] kmod: Avoid deadlock by recursive kmod call Tetsuo Handa
2012-01-04 13:08   ` Tetsuo Handa [this message]
     [not found]     ` <201201062139.JFC73472.SMQFOFOFHLVOtJ@I-love.SAKURA.ne.jp>
2012-01-19  0:45       ` [PATCH v3] " Andrew Morton
2012-01-19  2:07         ` Tetsuo Handa
2012-01-24  4:32           ` [PATCH v4] " Tetsuo Handa
2012-01-26  0:40             ` Andrew Morton
2012-01-26  1:32               ` Tetsuo Handa

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=201201042208.JHG52695.OLStHFQOFFMVJO@I-love.SAKURA.ne.jp \
    --to=penguin-kernel@i-love.sakura.ne.jp \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@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).