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: <20200107202950.782951-1-arnd@arndb.de>
Date:   Tue,  7 Jan 2020 21:29:40 +0100
From:   Arnd Bergmann <arnd@...db.de>
To:     David Kershner <david.kershner@...sys.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Ken Cox <jkc@...hat.com>
Cc:     Oleksandr Natalenko <oleksandr@...hat.com>,
        Arnd Bergmann <arnd@...db.de>,
        Ben Romer <sparmaintainer@...sys.com>,
        linux-kernel@...r.kernel.org
Subject: [PATCH] visorbus: fix uninitialized variable access

The setup_crash_devices_work_queue function only partially initializes
the message it sends to chipset_init, leading to undefined behavior:

drivers/visorbus/visorchipset.c: In function 'setup_crash_devices_work_queue':
drivers/visorbus/visorchipset.c:333:6: error: '((unsigned char*)&msg.hdr.flags)[0]' is used uninitialized in this function [-Werror=uninitialized]
  if (inmsg->hdr.flags.response_expected)

Set up the entire structure, zero-initializing the 'response_expected'
flag.

This was apparently found by the patch that added the -O3 build option
in Kconfig.

Fixes: 12e364b9f08a ("staging: visorchipset driver to provide registration and other services")
Signed-off-by: Arnd Bergmann <arnd@...db.de>
---
 drivers/visorbus/visorchipset.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/visorbus/visorchipset.c b/drivers/visorbus/visorchipset.c
index ca752b8f495f..cb1eb7e05f87 100644
--- a/drivers/visorbus/visorchipset.c
+++ b/drivers/visorbus/visorchipset.c
@@ -1210,14 +1210,17 @@ static void setup_crash_devices_work_queue(struct work_struct *work)
 {
 	struct controlvm_message local_crash_bus_msg;
 	struct controlvm_message local_crash_dev_msg;
-	struct controlvm_message msg;
+	struct controlvm_message msg = {
+		.hdr.id = CONTROLVM_CHIPSET_INIT,
+		.cmd.init_chipset = {
+			.bus_count = 23,
+			.switch_count = 0,
+		},
+	};
 	u32 local_crash_msg_offset;
 	u16 local_crash_msg_count;
 
 	/* send init chipset msg */
-	msg.hdr.id = CONTROLVM_CHIPSET_INIT;
-	msg.cmd.init_chipset.bus_count = 23;
-	msg.cmd.init_chipset.switch_count = 0;
 	chipset_init(&msg);
 	/* get saved message count */
 	if (visorchannel_read(chipset_dev->controlvm_channel,
-- 
2.20.0

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ