netdev - Re: [PATCH net-next 0/1] Introducing OpenVPN Data Channel Offload

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: Mon, 8 Jan 2024 00:32:23 +0100
From: Antonio Quartulli <antonio@...nvpn.net>
To: Sergey Ryazanov <ryazanov.s.a@...il.com>
Cc: netdev@...r.kernel.org, Jakub Kicinski <kuba@...nel.org>
Subject: Re: [PATCH net-next 0/1] Introducing OpenVPN Data Channel Offload

Hi Sergey,

Thanks for jumping in

On 06/01/2024 23:29, Sergey Ryazanov wrote:
> Hi Antonio,
> 
> On 06.01.2024 23:57, Antonio Quartulli wrote:
>> I tend to agree that a unique large patch is harder to review, but
>> splitting the code into several paches proved to be quite cumbersome,
>> therefore I prefered to not do it. I believe the code can still be
>> reviewed file by file, despite in the same patch.
> 
> I am happy to know that project is ongoing. But I had stopped the review 
> after reading these lines. You need AI to review at once "35 files 
> changed, 5914 insertions(+)". Last time I checked, I was human. Sorry.
> 
> Or you can see it like this: if submitter does not care, then why anyone 
> else should?

I am sorry - I did not mean to be careless/sloppy.

I totally understand, but I truly burnt so much time on finding a 
reasonable way to split this patch that I had to give up at some point.

I get your input, but do you think that turning it into 35 patches of 1 
file each (just as a random example), will make it easier to digest?

Anyway, I will give it another try (the test robot complained about 
something, so it seems I need to resend the patch anyway) and I'll see 
where I land.

Cheers!

> 
>> ** KNOWN ISSUE:
>> Upon module unloading something is not torn down correctly and sometimes
>> new packets hit dangling netdev pointers. This problem did not exist
>> when the RTNL API was implemented (before interface handling was moved
>> to Netlink). I was hoping to get some feedback from the netdev community
>> on anything that may look wrong.
> 
> A small hint, if the series is not going to be merged, then it is better 
> to mark it as RFC.
> 
> -- 
> Sergey

-- 
Antonio Quartulli
OpenVPN Inc.