linux-kernel - Re: [v4l-dvb-maintainer] [2.6.23 regression fix] usbvision: don't return an uninitialized value

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>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <Pine.LNX.4.58.0707230254390.5535@shell2.speakeasy.net>
Date:	Mon, 23 Jul 2007 03:02:29 -0700 (PDT)
From:	Trent Piepho <xyzzy@...akeasy.org>
To:	Adrian Bunk <bunk@...sta.de>
cc:	Thierry MERLE <thierry.merle@...e.fr>,
	"Dwaine P. Garden" <dwainegarden@...ers.com>,
	Mauro Carvalho Chehab <mchehab@...radead.org>,
	v4l-dvb maintainer list <v4l-dvb-maintainer@...uxtv.org>,
	Linux Kernel Mailing list <linux-kernel@...r.kernel.org>,
	Michal Piotrowski <michal.k.k.piotrowski@...il.com>
Subject: Re: [v4l-dvb-maintainer] [2.6.23 regression fix] usbvision: don't
 return an uninitialized	value

On Mon, 23 Jul 2007, Adrian Bunk wrote:
> Commit c5f48367fe54c46805774eeea8e828de54a5ad7b introduced this return
> of an uninitialized variable.
>
> Spotted by the Coverity checker.

It spotted that, but it seems it missed the bigger bug, reg->val is
unsigned.  Since reg->val can't be less than 0, it will never actually
return the unitialized variable.

This patch should fix it correctly.  The vidioc_g_register function was
broken too, it wasn't actually setting reg->val to the register value it
read.

--------------------------------------------------------------------
usbvision: fix bugs in vidioc_[sg]_register functions

From: Trent Piepho <xyzzy@...akeasy.org>

s_register was assigning the return code to (unsigned)reg->val, rather than
errCode, which it what it would return.  Except reg->val can't be < 0, so it
would never actually return an error.

g_register never actually put the value it read into reg->val.

Signed-off-by: Trent Piepho <xyzzy@...akeasy.org>

diff -r 8e72fe5675b3 -r d5c4fffbbad3 linux/drivers/media/video/usbvision/usbvision-video.c
--- a/linux/drivers/media/video/usbvision/usbvision-video.c	Sat Jul 21 17:26:40 2007 -0700
+++ b/linux/drivers/media/video/usbvision/usbvision-video.c	Mon Jul 23 02:58:31 2007 -0700
@@ -550,6 +550,7 @@ static int vidioc_g_register (struct fil
 		    __FUNCTION__, errCode);
 		return errCode;
 	}
+	reg->val = errCode;
 	return 0;
 }

@@ -564,8 +565,8 @@ static int vidioc_s_register (struct fil
 	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
 		return -EINVAL;
 	/* NT100x has a 8-bit register space */
-	reg->val = (u8)usbvision_write_reg(usbvision, reg->reg&0xff, reg->val);
-	if (reg->val < 0) {
+	errCode = usbvision_write_reg(usbvision, reg->reg&0xff, reg->val);
+	if (errCode < 0) {
 		err("%s: VIDIOC_DBG_S_REGISTER failed: error %d",
 		    __FUNCTION__, errCode);
 		return errCode;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/