linux-audit.redhat.com archive mirror
 help / color / mirror / Atom feed
From: Peter Hurley <peter@hurleysoftware.com>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: linux-audit@redhat.com, Jiri Slaby <jslaby@suse.cz>,
	Peter Hurley <peter@hurleysoftware.com>
Subject: [PATCH 06/15] tty: audit: Ignore current association for audit push
Date: Tue, 10 Nov 2015 21:05:51 -0500	[thread overview]
Message-ID: <1447207560-16410-7-git-send-email-peter@hurleysoftware.com> (raw)
In-Reply-To: <1447207560-16410-1-git-send-email-peter@hurleysoftware.com>

In canonical read mode, each line read and logged is pushed separately
with tty_audit_push(). For all single-threaded processes and multi-threaded
processes reading from only one tty, this patch has no effect; the last line
read will still be the entry pushed to the audit log because the tty
association cannot have changed between tty_audit_add_data() and
tty_audit_push().

For multi-threaded processes reading from different ttys concurrently,
the audit log will have mixed log entries anyway. Consider two ttys
audited concurrently:

CPU0                           CPU1
----------                     ------------
tty_audit_add_data(ttyA)
                               tty_audit_add_data(ttyB)
tty_audit_push()
                               tty_audit_add_data(ttyB)
                               tty_audit_push()

This patch will now cause the ttyB output to be split into separate
audit log entries.

However, this possibility is equally likely without this patch:

CPU0                           CPU1
----------                     ------------
                               tty_audit_add_data(ttyB)
tty_audit_add_data(ttyA)
tty_audit_push()
                               tty_audit_add_data(ttyB)
                               tty_audit_push()

Mixed canonical and non-canonical reads have similar races.

Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
---
 drivers/tty/n_tty.c     |  2 +-
 drivers/tty/tty_audit.c | 11 +++--------
 include/linux/tty.h     |  2 +-
 3 files changed, 5 insertions(+), 10 deletions(-)

diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c
index a3ad312..93f85a6 100644
--- a/drivers/tty/n_tty.c
+++ b/drivers/tty/n_tty.c
@@ -2104,7 +2104,7 @@ static int canon_copy_from_read_buf(struct tty_struct *tty,
 			ldata->line_start = ldata->read_tail;
 		else
 			ldata->push = 0;
-		tty_audit_push(tty);
+		tty_audit_push();
 	}
 	return eof_push ? -EAGAIN : 0;
 }
diff --git a/drivers/tty/tty_audit.c b/drivers/tty/tty_audit.c
index 5f65653..5ae4839 100644
--- a/drivers/tty/tty_audit.c
+++ b/drivers/tty/tty_audit.c
@@ -313,9 +313,9 @@ void tty_audit_add_data(struct tty_struct *tty, const void *data, size_t size)
 /**
  *	tty_audit_push	-	Push buffered data out
  *
- *	Make sure no audit data is pending for @tty on the current process.
+ *	Make sure no audit data is pending on the current process.
  */
-void tty_audit_push(struct tty_struct *tty)
+void tty_audit_push(void)
 {
 	struct tty_audit_buf *buf;
 	unsigned long flags;
@@ -331,13 +331,8 @@ void tty_audit_push(struct tty_struct *tty)
 	spin_unlock_irqrestore(&current->sighand->siglock, flags);
 
 	if (buf) {
-		int major, minor;
-
-		major = tty->driver->major;
-		minor = tty->driver->minor_start + tty->index;
 		mutex_lock(&buf->mutex);
-		if (buf->major == major && buf->minor == minor)
-			tty_audit_buf_push(buf);
+		tty_audit_buf_push(buf);
 		mutex_unlock(&buf->mutex);
 		tty_audit_buf_put(buf);
 	}
diff --git a/include/linux/tty.h b/include/linux/tty.h
index f8a20a8..8a73d84 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -607,7 +607,7 @@ extern void tty_audit_add_data(struct tty_struct *tty, const void *data,
 extern void tty_audit_exit(void);
 extern void tty_audit_fork(struct signal_struct *sig);
 extern void tty_audit_tiocsti(struct tty_struct *tty, char ch);
-extern void tty_audit_push(struct tty_struct *tty);
+extern void tty_audit_push(void);
 extern int tty_audit_push_current(void);
 #else
 static inline void tty_audit_add_data(struct tty_struct *tty, const void *data,
-- 
2.6.3

  parent reply	other threads:[~2015-11-11  2:06 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-11  2:05 [PATCH 00/15] Rework tty audit Peter Hurley
2015-11-11  2:05 ` [PATCH 01/15] tty: audit: Early-out pty master reads earlier Peter Hurley
2015-11-11  2:05 ` [PATCH 02/15] tty: audit: Never audit packet mode Peter Hurley
2015-11-11  2:05 ` [PATCH 03/15] tty: audit: Remove icanon mode from call chain Peter Hurley
2015-11-12 19:10   ` Richard Guy Briggs
2015-11-12 19:58     ` Peter Hurley
2015-11-13  2:15       ` Richard Guy Briggs
2015-11-13  2:27         ` Peter Hurley
2015-11-13  3:28           ` Richard Guy Briggs
2015-11-16 13:25             ` Peter Hurley
2015-11-11  2:05 ` [PATCH 04/15] tty: audit: Defer audit buffer association Peter Hurley
2015-11-11  2:05 ` [PATCH 05/15] tty: audit: Take siglock directly Peter Hurley
2015-11-11  2:05 ` Peter Hurley [this message]
2015-11-11  2:05 ` [PATCH 07/15] tty: audit: Combine push functions Peter Hurley
2015-11-11  2:05 ` [PATCH 08/15] tty: audit: Track tty association with dev_t Peter Hurley
2015-11-11  2:05 ` [PATCH 09/15] tty: audit: Handle tty audit enable atomically Peter Hurley
2015-11-11  2:05 ` [PATCH 10/15] tty: audit: Remove false memory optimization Peter Hurley
2015-11-11  2:05 ` [PATCH 11/15] tty: audit: Remove tty_audit_buf reference counting Peter Hurley
2015-11-11  2:05 ` [PATCH 12/15] tty: audit: Simplify first-use allocation Peter Hurley
2015-11-11  2:05 ` [PATCH 13/15] tty: audit: Check audit enable first Peter Hurley
2015-11-11  2:05 ` [PATCH 14/15] tty: audit: Always push audit buffer before TIOCSTI Peter Hurley
2015-11-11  2:06 ` [PATCH 15/15] tty: audit: Poison tty_audit_buf while process exits Peter Hurley
2015-11-13  2:31 ` [PATCH 00/15] Rework tty audit Peter Hurley
2015-12-21  0:39 ` Paul Moore
2016-01-10  4:58 ` [RESEND][PATCH " Peter Hurley
2016-01-10  4:58   ` [RESEND][PATCH 01/15] tty: audit: Early-out pty master reads earlier Peter Hurley
2016-01-10  4:58   ` [RESEND][PATCH 02/15] tty: audit: Never audit packet mode Peter Hurley
2016-01-10  4:58   ` [RESEND][PATCH 03/15] tty: audit: Remove icanon mode from call chain Peter Hurley
2016-01-10  4:58   ` [RESEND][PATCH 04/15] tty: audit: Defer audit buffer association Peter Hurley
2016-01-10  4:58   ` [RESEND][PATCH 05/15] tty: audit: Take siglock directly Peter Hurley
2016-01-10  4:58   ` [RESEND][PATCH 06/15] tty: audit: Ignore current association for audit push Peter Hurley
2016-01-10  5:36     ` kbuild test robot
2016-01-10  7:00       ` Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 07/15] tty: audit: Combine push functions Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 08/15] tty: audit: Track tty association with dev_t Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 09/15] tty: audit: Handle tty audit enable atomically Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 10/15] tty: audit: Remove false memory optimization Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 11/15] tty: audit: Remove tty_audit_buf reference counting Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 12/15] tty: audit: Simplify first-use allocation Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 13/15] tty: audit: Check audit enable first Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 14/15] tty: audit: Always push audit buffer before TIOCSTI Peter Hurley
2016-01-10  4:59   ` [RESEND][PATCH 15/15] tty: audit: Poison tty_audit_buf while process exits Peter Hurley

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=1447207560-16410-7-git-send-email-peter@hurleysoftware.com \
    --to=peter@hurleysoftware.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=jslaby@suse.cz \
    --cc=linux-audit@redhat.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).