Re: [PATCH] scsi: int overflow in st_ioctl()

Re: [PATCH] scsi: int overflow in st_ioctl()

["Kai Mäkisara (Kolumbus)"]

On 17.12.2013, at 16.43, Yongjian Xu xuyongjiande@gmail.com wrote:

> From: Yongjian Xu <xuyongjiande@gmail.com>
> 
> mtc.mt_count comes from user-space.
> int overflow may occur:
> mtc.mt_count++;
> mtc.mt_count—;

I agree that this is a problem. However, it seems that there is also another problem: the SPACE command uses a 24-bit count field and this overflows far below INT_MAX. Should the count in MTFSF etc. be limited to (2<<24-2)? This would make the checks you suggest unnecessary. (-2 so that the behavior does not depend on whether we are at file mark or not.)

I am a bit hesitant with this suggestion because someone may use count INT_MAX to space to the end of the tape. This probably succeeds now but after the fix it would return an error.

Thanks,
Kai

--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] scsi: int overflow in st_ioctl()

[Yu Chen]

Because the local variable of st_int_ioctl function, such as long ltmp
be assigned by arg.
I think  we should limit the parameter arg of st_int_ioctl function to
the range of 24-bit signed integer.
Yong Jianxu, please update your patch.

2013/12/18 "Kai Mäkisara (Kolumbus)" <kai.makisara@kolumbus.fi>:
> On 17.12.2013, at 16.43, Yongjian Xu xuyongjiande@gmail.com wrote:
>
>> From: Yongjian Xu <xuyongjiande@gmail.com>
>>
>> mtc.mt_count comes from user-space.
>> int overflow may occur:
>> mtc.mt_count++;
>> mtc.mt_count—;
>
> I agree that this is a problem. However, it seems that there is also another problem: the SPACE command uses a 24-bit count field and this overflows far below INT_MAX. Should the count in MTFSF etc. be limited to (2<<24-2)? This would make the checks you suggest unnecessary. (-2 so that the behavior does not depend on whether we are at file mark or not.)
>
> I am a bit hesitant with this suggestion because someone may use count INT_MAX to space to the end of the tape. This probably succeeds now but after the fix it would return an error.
>
> Thanks,
> Kai
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/



-- 
Best Regards
==============================================
Yu Chen
Ph.D.  Associate Professor
System Software&Software Engineering Group,
Dept. of Computer Science and Technology
Tsinghua University, Beijing 100084, P.R. China
E-Mail: mailto:yuchen@tsinghua.edu.cn  chyyuu@gmail.com
==============================================