linux-kernel - Re: [PATCH] video: fbdev: metronomefb: Fix buffer overflow in load

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: <333658df-a767-437f-9566-857e8ff5867f@gmail.com>
Date: Fri, 15 Nov 2024 18:42:27 +0530
From: Suraj Sonawane <surajsonawane0215@...il.com>
To: Helge Deller <deller@....de>
Cc: linux-fbdev@...r.kernel.org, dri-devel@...ts.freedesktop.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH] video: fbdev: metronomefb: Fix buffer overflow in
 load_waveform()

On 14/11/24 20:13, Helge Deller wrote:
> On 11/12/24 21:28, Suraj Sonawane wrote:
>> Fix an error detected by the Smatch tool:
>>
>> drivers/video/fbdev/metronomefb.c:220 load_waveform() error:
>> buffer overflow 'wfm_hdr->stuff2a' 2 <= 4
>> drivers/video/fbdev/metronomefb.c:220 load_waveform() error:
>> buffer overflow 'wfm_hdr->stuff2a' 2 <= 4
>>
>> The access to wfm_hdr->stuff2a in the loop can lead to a buffer
>> overflow if stuff2a is not large enough. To fix this, a check was
>> added to ensure that stuff2a has sufficient space before accessing
>> it. This prevents the overflow and improves the safety of the code.
>>
>> Signed-off-by: Suraj Sonawane <surajsonawane0215@...il.com>
>> ---
>>   drivers/video/fbdev/metronomefb.c | 6 ++++++
>>   1 file changed, 6 insertions(+)
>>
>> diff --git a/drivers/video/fbdev/metronomefb.c 
>> b/drivers/video/fbdev/metronomefb.c
>> index 6f0942c6e..9da55cef2 100644
>> --- a/drivers/video/fbdev/metronomefb.c
>> +++ b/drivers/video/fbdev/metronomefb.c
>> @@ -210,6 +210,12 @@ static int load_waveform(u8 *mem, size_t size, 
>> int m, int t,
>>       }
>>       wfm_hdr->mc += 1;
>>       wfm_hdr->trc += 1;
>> +
>> +    if (sizeof(wfm_hdr->stuff2a) < 5) {
>> +        dev_err(dev, "Error: insufficient space in stuff2a\n");
>> +        return -EINVAL;
>> +    }
>> +
>>       for (i = 0; i < 5; i++) {
>>           if (*(wfm_hdr->stuff2a + i) != 0) {
>>               dev_err(dev, "Error: unexpected value in padding\n");
> 
> That patch is completely wrong.
> There is
> /* the waveform structure that is coming from userspace firmware */
> struct waveform_hdr {
>          ....
>          u8 stuff2a[2];
>          u8 stuff2b[3];
> 
> So, I *believe* the for-next loop wants to walk acrosss stuff2a and 
> stuff2b,
> which have 5 entries together. So, basically the original code isn't nice
> but still correct.
> Your "sizeof()" check will always be false and is the wrong patch.
> 
> If at all, I think the stuff2a and stuff 2b arrays should be joined.
> Something like
>          u8 stuff2[5]; /* this is actually 2-entry stuff2a and 3-entry 
> stuff2b */
> But again, I don't know much about this driver.
> 
> Helge

Thank you for the brief feedback. I see your point regarding stuff2a and 
stuff2b. I’ll study this approach and revise the patch if I find it to 
be the correct solution.

Best regards,
Suraj Sonawane