lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [day] [month] [year] [list]
Message-Id: <20240927084056.7193-1-chenqiuji666@gmail.com>
Date: Fri, 27 Sep 2024 16:40:56 +0800
From: Qiu-ji Chen <chenqiuji666@...il.com>
To: amit@...nel.org,
	arnd@...db.de,
	gregkh@...uxfoundation.org
Cc: virtualization@...ts.linux.dev,
	linux-kernel@...r.kernel.org,
	baijiaju1990@...il.com,
	Qiu-ji Chen <chenqiuji666@...il.com>,
	stable@...r.kernel.org
Subject: [PATCH] virtio: console: Fix atomicity violation in fill_readbuf()

The atomicity violation issue is due to the invalidation of the function 
port_has_data()'s check caused by concurrency. Imagine a scenario where a 
port that contains data passes the validity check but is simultaneously 
assigned a value with no data. This could result in an empty port passing 
the validity check, potentially leading to a null pointer dereference 
error later in the program, which is inconsistent.

To address this issue, we believe there is a problem with the original 
logic. Since the validity check and the read operation were separated, it 
could lead to inconsistencies between the data that passes the check and 
the data being read. Therefore, we moved the main logic of the 
port_has_data function into this function and placed the read operation 
within a lock, ensuring that the validity check and read operation are 
not separated, thus resolving the problem.

This possible bug is found by an experimental static analysis tool
developed by our team. This tool analyzes the locking APIs
to extract function pairs that can be concurrently executed, and then
analyzes the instructions in the paired functions to identify possible
concurrency bugs including data races and atomicity violations.

Fixes: 203baab8ba31 ("virtio: console: Introduce function to hand off data from host to readers")
Cc: stable@...r.kernel.org
Signed-off-by: Qiu-ji Chen <chenqiuji666@...il.com>
---
 drivers/char/virtio_console.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index de7d720d99fa..5aaf07f71a4c 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -656,10 +656,14 @@ static ssize_t fill_readbuf(struct port *port, u8 __user *out_buf,
 	struct port_buffer *buf;
 	unsigned long flags;
 
-	if (!out_count || !port_has_data(port))
+	if (!out_count)
 		return 0;
 
-	buf = port->inbuf;
+	spin_lock_irqsave(&port->inbuf_lock, flags);
+	buf = port->inbuf = get_inbuf(port);
+	spin_unlock_irqrestore(&port->inbuf_lock, flags);
+	if (!buf)
+		return 0;
 	out_count = min(out_count, buf->len - buf->offset);
 
 	if (to_user) {
-- 
2.34.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ