netdev - Re: [PATCH net-next 1/1] net/sched: Disambiguate verdict from return code

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: <ncblypbr7vshhgeuuxxvs6k7lapu4ooxyiecqn3yitlhnmw5bl@tdjlc4s4tgvj>
Date: Fri, 6 Oct 2023 11:59:46 -0600
From: Daniel Xu <dxu@...uu.xyz>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Daniel Borkmann <daniel@...earbox.net>, 
	Jamal Hadi Salim <jhs@...atatu.com>, Victor Nogueira <victor@...atatu.com>, xiyou.wangcong@...il.com, 
	jiri@...nulli.us, davem@...emloft.net, edumazet@...gle.com, pabeni@...hat.com, 
	paulb@...dia.com, netdev@...r.kernel.org, kernel@...atatu.com, 
	martin.lau@...ux.dev, bpf@...r.kernel.org
Subject: Re: [PATCH net-next 1/1] net/sched: Disambiguate verdict from return
 code

On Fri, Oct 06, 2023 at 07:12:15AM -0700, Jakub Kicinski wrote:
> On Fri, 6 Oct 2023 15:49:18 +0200 Daniel Borkmann wrote:
> > > Which will no longer work with the "pack multiple values into
> > > the reason" scheme of subsys-specific values :(  
> > 
> > Too bad, do you happen to know why it won't work? 
> 
> I'm just guessing but the reason is enum skb_drop_reason
> and the values of subsystem specific reasons won't be part
> of that enum.

Yeah, looks like the subsystem reasons are different enums right?
There's probably a way to still support it in bpftrace but it might take
some minor changes and/or ugly conditionals in scripts.

But I also wonder: why are the subsystem error codes not included into
`enum skb_drop_reason`? It looks like the enum space is partitioned
already. And the modules have already registered themselves into core
kernel (in `enum skb_drop_reason_subsys`). So even if modules are
compiled out there are still hints of it laying around vmlinux.

Thanks,
Daniel

[..]