netdev - Re: [PATCH v2 net-next 2/2] tcp: add more DROP REASONs in receive process

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: <CAL+tcoDUtUp+qJ1zxDyLUi_Y19+-68V4pTJhf0S0g51Y-bt=bg@mail.gmail.com>
Date: Fri, 9 Feb 2024 19:39:24 +0800
From: Jason Xing <kerneljasonxing@...il.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: davem@...emloft.net, kuba@...nel.org, pabeni@...hat.com, 
	dsahern@...nel.org, netdev@...r.kernel.org, 
	Jason Xing <kernelxing@...cent.com>
Subject: Re: [PATCH v2 net-next 2/2] tcp: add more DROP REASONs in receive process

[...]
>
> Your patch is too large/risky, not that I am trying to be gentle here...
>
> You should know better that we are not going to accept it as is.
> Please split your patches into smaller ones.
>
> Eg, a single patch to change tcp_rcv_synsent_state_process() return value.
> It used to return -1, 0, or 1.
> Take the time to document for tcp_rcv_synsent_state_process() what are
> the new possible return values.
>
> Then other changes, one at a time, in a logical way.
>
> Smaller patches are easier to review, even if it forces the author to
> think a bit more about how to
> make his series a work of art. Everyone wins, because we spend less
> time and we learn faster.

Now I understand. I'll separate the current patch.

One more question about whether I should split the 1/2 patch into some
smaller patches, say:
1) first, add some definitions into include/net/dropreason-core.h
2) then use those definitions in cookie_v4_check()
or
1) first, add some definitions.
2) second, add kfree_skb_reason() into cookie_v4_check() but only
NOT_SPECIFIED fake reason; get rid of 'goto discard;' in
tcp_v4_do_rcv() if @nsk is NULL.
3) third, use those definitions.
something like that...this way could make each patch more precise and clean.

What would you recommend about 1/2 patch?

Thanks,
Jason

>
> Thank you.