linux-kernel - Re: [PATCH RFC v2 0/6] staging: r8188eu: avoid 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]

Message-ID: <e9ad740e-d970-1a1e-fee9-b26d8c371506@gmail.com>
Date:   Sun, 22 Aug 2021 23:19:26 +0300
From:   Pavel Skripkin <paskripkin@...il.com>
To:     "Fabio M. De Francesco" <fmdefrancesco@...il.com>,
        Larry.Finger@...inger.net, phil@...lpotter.co.uk,
        gregkh@...uxfoundation.org, straube.linux@...il.com
Cc:     linux-staging@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: Re: [PATCH RFC v2 0/6] staging: r8188eu: avoid uninit value bugs

On 8/22/21 11:06 PM, Fabio M. De Francesco wrote:
> On Sunday, August 22, 2021 7:38:11 PM CEST Pavel Skripkin wrote:
>> On 8/22/21 8:36 PM, Fabio M. De Francesco wrote:
>> > On Sunday, August 22, 2021 4:35:05 PM CEST Pavel Skripkin wrote:
>> >> Hi, Greg, Larry and Phillip!
>> >> 
>> >> I noticed, that new staging driver was added like 3 weeks ago and I decided
>> >> to look at the code, because drivers in staging directory are always buggy.
>> >> 
>> >> The first thing I noticed is *no one* was checking read operations result, but
>> >> it can fail and driver may start writing random stack values into registers. It
>> >> can cause driver misbehavior or device misbehavior.
>> >> 
>> >> To avoid this type of bugs, I've changed rtw_read* API. Now all rtw_read
>> >> funtions return an error, when something went wrong with usb transfer.
>> >> 
>> >> It helps callers to break/return earlier and don't write random values to
>> >> registers or to rely on random values.
>> >> 
>> >> Why is this pacth series RFC?
>> >>   1. I don't have this device and I cannot test these changes.
>> >>   2. I don't know how to handle errors in each particular case. For now, function
>> >>      just returns or returns an error. That's all. I hope, driver maintainers will
>> >>      help with these bits.
>> >>   3. I guess, I handled not all uninit value bugs here. I hope, I fixed
>> >>      at least half of them
>> >> 
>> >> v1 -> v2:
>> >>   1. Make rtw_read*() return an error instead of initializing pointer to error
>> >>   2. Split one huge patch to smaller ones for each rtw_read{8,16,32} function
>> >>      changes
>> >>   3. Add new macro for printing register values (It helps to not copy-paste error
>> >>      handling)
>> >>   4. Removed {read,write}_macreg (Suggested by Phillip)
>> >>   5. Rebased on top of staging-next
>> >>   6. Cleaned checkpatch errors and warnings
>> >> 
>> >> Only build-tested, since I don't have device with r8118eu chip
>> >> 
>> >> Pavel Skripkin (6):
>> >>   staging: r8188eu: remove {read,write}_macreg
>> >>   staging: r8188eu: add helper macro for printing registers
>> >>   staging: r8188eu: add error handling of rtw_read8
>> >>   staging: r8188eu: add error handling of rtw_read16
>> >>   staging: r8188eu: add error handling of rtw_read32
>> >>   staging: r8188eu: make ReadEFuse return an int
>> > 
>> > Hi Pavel,
>> > 
>> > I've just read your v2 of the series. I had no time to read each and every line,
>> > however, I suppose that I saw enough to say that I think they are a huge
>> > improvement over v1. I really like your patches and if I were you, I'd drop
>> > that RFC tag.
>> > 
>> 
>> Thank you, Fabio! I appreciate it :)
>> 
>> 
>> With regards,
>> Pavel Skripkin
> 
> Hi Pavel,
> 
> I've read more code of your series and I'm ready to give a formal ack. However,
> I'm not sure about the rules: can it be also given to RFC or only to "real" patches?
> 
> As I've already said, they look good and I like them. So, the entire series is...
> 

AFAIK, RFC patches can be treated as normal patches, if 
reviewers/maintainers don't have objections to code. I am not sure about 
it, since it's my first experience with RFCs :)

Anyway, thank you for ACKing. Let's see what Larry thinks about it. I 
believe, he can find some bugs in my code since it's not tested at all 
:) I hope, my r8118eu device will come soon and I will be able to test 
this series...

> Acked-by: Fabio M. De Francesco <fmdefrancesco@...il.com>
> 
> If the rules don't allow to formally ack RFC, I will be happy to ack again the final product.
> 
> I also want to say that I enjoyed discussing this work with you on this long thread. :-)
> 

Me too, thank you. Technical discussions are the best part of linux 
kernel development process, IMO :)



With regards,
Pavel Skripkin