linux-kernel - Re: [PATCH] scsi: fix potential integer signedness problem

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: <1453148090.2363.13.camel@HansenPartnership.com>
Date:	Mon, 18 Jan 2016 12:14:50 -0800
From:	James Bottomley <James.Bottomley@...senPartnership.com>
To:	Insu Yun <wuninsu@...il.com>, martin.petersen@...cle.com,
	linux-scsi@...r.kernel.org, linux-kernel@...r.kernel.org
Cc:	taesoo@...ech.edu, yeongjin.jang@...ech.edu, insu@...ech.edu,
	changwoo@...ech.edu
Subject: Re: [PATCH] scsi: fix potential integer signedness problem

On Mon, 2016-01-18 at 14:55 -0500, Insu Yun wrote:
> Since len is signed integer variable, it is better to check
> whether len is non-negative or not.
> If non-negative value len gives, it can cause heap overflow.

Hey, this patch is full of a ton of undocumented (and mostly spurious)
whitespace changes, which make it very hard to see the actual patch

> Signed-off-by: Insu Yun <wuninsu@...il.com>
> ---
>  drivers/scsi/scsi_lib.c | 33 ++++++++++++++++++---------------
>  1 file changed, 18 insertions(+), 15 deletions(-)
[...]
Which is this, I think:

> @@ -2349,6 +2349,9 @@ scsi_mode_select(struct scsi_device *sdev, int
> pf, int sp, int modepage,
>  	unsigned char *real_buffer;
>  	int ret;
>  
> +	if (len < 0)
> +		return -EINVAL;
> +
>  	memset(cmd, 0, sizeof(cmd));
>  	cmd[1] = (pf ? 0x10 : 0) | (sp ? 0x01 : 0);

What's the actual problem here?  There's only one user of
scsi_mode_select: an internal one which always sends in a positive
length, so this check is entirely unnecessary.

James