netdev - Re: [PATCH net-next v5 06/15] net-timestamp: prepare for isolating two modes of SO

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: <CAL+tcoARXLJCvE1hjACH5B_rbxM-B4yGrROaVamp=11N2mnoKw@mail.gmail.com>
Date: Thu, 16 Jan 2025 07:50:45 +0800
From: Jason Xing <kerneljasonxing@...il.com>
To: Martin KaFai Lau <martin.lau@...ux.dev>
Cc: davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org, 
	pabeni@...hat.com, dsahern@...nel.org, willemdebruijn.kernel@...il.com, 
	willemb@...gle.com, ast@...nel.org, daniel@...earbox.net, andrii@...nel.org, 
	eddyz87@...il.com, song@...nel.org, yonghong.song@...ux.dev, 
	john.fastabend@...il.com, kpsingh@...nel.org, sdf@...ichev.me, 
	haoluo@...gle.com, jolsa@...nel.org, horms@...nel.org, bpf@...r.kernel.org, 
	netdev@...r.kernel.org
Subject: Re: [PATCH net-next v5 06/15] net-timestamp: prepare for isolating
 two modes of SO_TIMESTAMPING

On Thu, Jan 16, 2025 at 6:11 AM Martin KaFai Lau <martin.lau@...ux.dev> wrote:
>
> On 1/12/25 3:37 AM, Jason Xing wrote:
> >   void __skb_tstamp_tx(struct sk_buff *orig_skb,
> >                    const struct sk_buff *ack_skb,
> >                    struct skb_shared_hwtstamps *hwtstamps,
> > -                  struct sock *sk, int tstype)
> > +                  struct sock *sk, bool sw, int tstype)
>
> Instead of adding a new "bool sw" and changing all callers, is it the same as
> testing "!hwtstamps" ?

Actually, I had a version using the hwtstamps, then I realized that
hardware or driver may go wrong and pass a NULL hwstamps. It's indeed
unlikely to happen. If so, timestamping code will consider it as a
software timestamp.

I don't expect that thing happening, ensuring our code is robust
enough. The original timestamping code seems not deal with this case
as we cannot see some particular test in __skb_tstamp_tx(). The worst
thing is if it happens, we would never know and treat it as a software
SCM_TSTAMP_SND case.

Does it make any sense to you?

Thanks,
Jason