mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Schspa Shi <schspa@gmail.com>
To: mcgrof@kernel.org
Cc: linux-kernel@vger.kernel.org, Schspa Shi <schspa@gmail.com>,
	syzbot+10d19d528d9755d9af22@syzkaller.appspotmail.com,
	syzbot+70d5d5d83d03db2c813d@syzkaller.appspotmail.com,
	syzbot+83cb0411d0fcf0a30fc1@syzkaller.appspotmail.com
Subject: [PATCH] umh: fix UAF when the process is being killed
Date: Tue, 15 Nov 2022 22:02:33 +0800	[thread overview]
Message-ID: <20221115140233.21981-1-schspa@gmail.com> (raw)

When the process is killed, wait_for_completion_state will return with
-ERESTARTSYS, and the completion variable in the stack will be freed.
If the user-mode thread is complete at the same time, there will be a UAF.

Please refer to the following scenarios.
            T1                                  T2
------------------------------------------------------------------
call_usermodehelper_exec
                                   call_usermodehelper_exec_async
                                   << do something >>
                                   umh_complete(sub_info);
                                   comp = xchg(&sub_info->complete, NULL);
                                   /* we got the completion */
                                   << context switch >>

    << Being killed >>
	retval = wait_for_completion_state(sub_info->complete, state);
	if (!retval)
		goto wait_done;

	if (wait & UMH_KILLABLE) {
		/* umh_complete() will see NULL and free sub_info */
		if (xchg(&sub_info->complete, NULL))
			goto unlock;
        << we can't got the completion >>
	}
	....
unlock:
	helper_unlock();
	return retval;
}

/**
 * the completion variable in stack is end of life cycle.
 * and maybe freed due to process is recycled.
 */
                                   --------UAF here----------
                                   if (comp)
                                       complete(comp);

To fix it, we can put the completion variable in the subprocess_info
variable.

Reported-by: syzbot+10d19d528d9755d9af22@syzkaller.appspotmail.com
Reported-by: syzbot+70d5d5d83d03db2c813d@syzkaller.appspotmail.com
Reported-by: syzbot+83cb0411d0fcf0a30fc1@syzkaller.appspotmail.com

Signed-off-by: Schspa Shi <schspa@gmail.com>
---
 include/linux/umh.h | 1 +
 kernel/umh.c        | 6 +++---
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/include/linux/umh.h b/include/linux/umh.h
index 5d1f6129b847..801f7efbc825 100644
--- a/include/linux/umh.h
+++ b/include/linux/umh.h
@@ -20,6 +20,7 @@ struct file;
 struct subprocess_info {
 	struct work_struct work;
 	struct completion *complete;
+	struct completion done;
 	const char *path;
 	char **argv;
 	char **envp;
diff --git a/kernel/umh.c b/kernel/umh.c
index 850631518665..3ed39956c777 100644
--- a/kernel/umh.c
+++ b/kernel/umh.c
@@ -380,6 +380,7 @@ struct subprocess_info *call_usermodehelper_setup(const char *path, char **argv,
 	sub_info->cleanup = cleanup;
 	sub_info->init = init;
 	sub_info->data = data;
+	init_completion(&sub_info->done);
   out:
 	return sub_info;
 }
@@ -405,7 +406,6 @@ EXPORT_SYMBOL(call_usermodehelper_setup);
 int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait)
 {
 	unsigned int state = TASK_UNINTERRUPTIBLE;
-	DECLARE_COMPLETION_ONSTACK(done);
 	int retval = 0;
 
 	if (!sub_info->path) {
@@ -431,7 +431,7 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait)
 	 * This makes it possible to use umh_complete to free
 	 * the data structure in case of UMH_NO_WAIT.
 	 */
-	sub_info->complete = (wait == UMH_NO_WAIT) ? NULL : &done;
+	sub_info->complete = (wait == UMH_NO_WAIT) ? NULL : &sub_info->done;
 	sub_info->wait = wait;
 
 	queue_work(system_unbound_wq, &sub_info->work);
@@ -444,7 +444,7 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait)
 	if (wait & UMH_FREEZABLE)
 		state |= TASK_FREEZABLE;
 
-	retval = wait_for_completion_state(&done, state);
+	retval = wait_for_completion_state(sub_info->complete, state);
 	if (!retval)
 		goto wait_done;
 
-- 
2.37.3


             reply	other threads:[~2022-11-15 14:02 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-15 14:02 Schspa Shi [this message]
2022-12-05 11:38 ` Schspa Shi
2022-12-12  5:10   ` Luis Chamberlain
2022-12-12 11:04     ` Schspa Shi
2022-12-12 13:38       ` Schspa Shi
2022-12-13 23:03         ` Luis Chamberlain
2022-12-14  2:28           ` Schspa Shi
2022-12-14 19:57           ` Luis Chamberlain
2022-12-15  6:16             ` Schspa Shi
2022-12-22  5:45               ` Schspa Shi
2022-12-22  6:16                 ` Luis Chamberlain
2022-12-22  6:50                   ` Schspa Shi
2022-12-22 11:56                     ` Schspa Shi
2022-12-22 12:09                       ` Schspa Shi
2022-12-23 15:01                         ` Luis Chamberlain
2023-01-13  5:42                           ` Schspa Shi
2023-01-24 17:39                             ` Luis Chamberlain

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=20221115140233.21981-1-schspa@gmail.com \
    --to=schspa@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mcgrof@kernel.org \
    --cc=syzbot+10d19d528d9755d9af22@syzkaller.appspotmail.com \
    --cc=syzbot+70d5d5d83d03db2c813d@syzkaller.appspotmail.com \
    --cc=syzbot+83cb0411d0fcf0a30fc1@syzkaller.appspotmail.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

all inboxes | Powered by JetHome®