netdev - Re: [PATCH v3] net: asix: fix uninit value bugs

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] [thread-next>] [day] [month] [year] [list]

Date:   Sat, 14 Aug 2021 19:20:21 +0300
From:   Pavel Skripkin <paskripkin@...il.com>
To:     Andrew Lunn <andrew@...n.ch>
Cc:     davem@...emloft.net, kuba@...nel.org, linux@...pel-privat.de,
        robert.foss@...labora.com, linux-usb@...r.kernel.org,
        netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
        syzbot+a631ec9e717fb0423053@...kaller.appspotmail.com
Subject: Re: [PATCH v3] net: asix: fix uninit value bugs

On 8/14/21 6:36 PM, Andrew Lunn wrote:
> On Sat, Aug 14, 2021 at 04:55:05PM +0300, Pavel Skripkin wrote:
>> Syzbot reported uninit-value in asix_mdio_read(). The problem was in
>> missing error handling. asix_read_cmd() should initialize passed stack
>> variable smsr, but it can fail in some cases. Then while condidition
>> checks possibly uninit smsr variable.
>> 
>> Since smsr is uninitialized stack variable, driver can misbehave,
>> because smsr will be random in case of asix_read_cmd() failure.
>> Fix it by adding error handling and just continue the loop instead of
>> checking uninit value.
>> 
>> Also, same loop was used in 3 other functions. Fixed uninit value bug
>> in them too.
> 
> Hi Pavel
> 
> Which suggests it might make sense to refactor the code to make a
> helper? I will leave you to decide if you want to do that.
> 
> The code does looks correct now.
> 
> 	Andrew
> 

I noticed strange thing. For example: driver looped 30 times, there 
wasn't any errors with usb transfer, but Host_En bit is not set. 
Datasheet says, that if Host_En is not set, that means software access 
will be ignored. Driver code doesn't handle this situation. We only 
check if ret is -ENODEV or -ETIMEOUT.

I guess, next register access will fail, but anyway, does it make sense 
to return when Host_En bit is not set?


With regards,
Pavel Skripkin