linux-kernel - [PATCH v2] media: vidtv: Fix use-after-free in vidtv_bridge_dvb

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: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <20221108070630.211084-1-chenzhongjin@huawei.com>
Date:   Tue, 8 Nov 2022 15:06:30 +0800
From:   Chen Zhongjin <chenzhongjin@...wei.com>
To:     <linux-kernel@...r.kernel.org>, <linux-media@...r.kernel.org>
CC:     <dwlsalmeida@...il.com>, <mchehab@...nel.org>,
        <chenzhongjin@...wei.com>
Subject: [PATCH v2] media: vidtv: Fix use-after-free in vidtv_bridge_dvb_init()

KASAN reports a use-after-free:
BUG: KASAN: use-after-free in dvb_dmxdev_release+0x4d5/0x5d0 [dvb_core]
Call Trace:
 ...
 dvb_dmxdev_release+0x4d5/0x5d0 [dvb_core]
 vidtv_bridge_probe+0x7bf/0xa40 [dvb_vidtv_bridge]
 platform_probe+0xb6/0x170
 ...
Allocated by task 1238:
 ...
 dvb_register_device+0x1a7/0xa70 [dvb_core]
 dvb_dmxdev_init+0x2af/0x4a0 [dvb_core]
 vidtv_bridge_probe+0x766/0xa40 [dvb_vidtv_bridge]
 ...
Freed by task 1238:
 dvb_register_device+0x6d2/0xa70 [dvb_core]
 dvb_dmxdev_init+0x2af/0x4a0 [dvb_core]
 vidtv_bridge_probe+0x766/0xa40 [dvb_vidtv_bridge]
 ...

It is because the error handling in vidtv_bridge_dvb_init() is wrong.

First, vidtv_bridge_dmx(dev)_init() will clean themselves when fail, but
goto fail_dmx(_dev): calls release functions again, which causes
use-after-free.

Also, in fail_fe, fail_tuner_probe and fail_demod_probe, j = i will cause
out-of-bound when i finished its loop (i == NUM_FE). And the loop
releasing is wrong, although now NUM_FE is 1 so it won't cause problem.

Fix this by correctly releasing everything.

Fixes: f90cf6079bf6 ("media: vidtv: add a bridge driver")
Signed-off-by: Chen Zhongjin <chenzhongjin@...wei.com>
---
v1 -> v2
After some testing I found we can't call dvb_frontend_detach() directly
because it is not safe and can cause double free now.

Obviously single dvb_unregister_frontend() can't handle all resources for
dvb_register_frontend(), because the fe->refcount is initialized to two
but unregister only put one of it.

So there are still memory leaked for frontend. These code should be
refactored to completely remove dvb_attach and dettach.

Anyway it is fault of frontend part. Now this patch only fix fail_dmx
and fail_dmx_dev cases which makes kenel panic.
---
 .../media/test-drivers/vidtv/vidtv_bridge.c   | 22 +++++++------------
 1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/drivers/media/test-drivers/vidtv/vidtv_bridge.c b/drivers/media/test-drivers/vidtv/vidtv_bridge.c
index 82620613d56b..dff7265a42ca 100644
--- a/drivers/media/test-drivers/vidtv/vidtv_bridge.c
+++ b/drivers/media/test-drivers/vidtv/vidtv_bridge.c
@@ -459,26 +459,20 @@ static int vidtv_bridge_dvb_init(struct vidtv_dvb *dvb)
 	for (j = j - 1; j >= 0; --j)
 		dvb->demux.dmx.remove_frontend(&dvb->demux.dmx,
 					       &dvb->dmx_fe[j]);
-fail_dmx_dev:
 	dvb_dmxdev_release(&dvb->dmx_dev);
-fail_dmx:
+fail_dmx_dev:
 	dvb_dmx_release(&dvb->demux);
+fail_dmx:
+fail_demod_probe:
+	for (i = i - 1; i >= 0; --i) {
+		dvb_unregister_frontend(dvb->fe[i]);
 fail_fe:
-	for (j = i; j >= 0; --j)
-		dvb_unregister_frontend(dvb->fe[j]);
+		dvb_module_release(dvb->i2c_client_tuner[i]);
 fail_tuner_probe:
-	for (j = i; j >= 0; --j)
-		if (dvb->i2c_client_tuner[j])
-			dvb_module_release(dvb->i2c_client_tuner[j]);
-
-fail_demod_probe:
-	for (j = i; j >= 0; --j)
-		if (dvb->i2c_client_demod[j])
-			dvb_module_release(dvb->i2c_client_demod[j]);
-
+		dvb_module_release(dvb->i2c_client_demod[i]);
+	}
 fail_adapter:
 	dvb_unregister_adapter(&dvb->adapter);
-
 fail_i2c:
 	i2c_del_adapter(&dvb->i2c_adapter);

-- 
2.17.1