netdev - Re: [PATCH] tcp: SO_TIMESTAMP implementation for TCP

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1272693648.2230.78.camel@edumazet-laptop>
Date:	Sat, 01 May 2010 08:00:48 +0200
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Tom Herbert <therbert@...gle.com>
Cc:	Bill Fink <billfink@...dspring.com>,
	David Miller <davem@...emloft.net>, netdev@...r.kernel.org
Subject: Re: [PATCH] tcp: SO_TIMESTAMP implementation for TCP

Le vendredi 30 avril 2010 à 22:40 -0700, Tom Herbert a écrit :
> > Not being a kernel hacker, I will naively ask if the kernel tracing
> > facility could somehow be used to provide the desired info (or could
> > be modified to provide it).
> >
> 
> We did consider kernel tracing (more in the context of implementing
> RFC 4898).  In the case of trying get per packet timestamps,
> correlating a ktrace event with an application message is probably too
> high to make it practical.  If it weren't for the cost of
> timestamp'ing every single skb being received, we'd probably have
> SO_TIMESTAMP turned on permanently for many connections.  For now
> we're settling for a percentage of messages for sampling.

Tom, did you tried to reuse existing skb or sk tstamps ?



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html