netdev - RE: [PATCH net-next v2 1/6] skbuff: Add skb_list

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: <063D6719AE5E284EB5DD2968C1650D6D1CAD6218@AcuExch.aculab.com>
Date:	Fri, 30 Jan 2015 12:25:57 +0000
From:	David Laight <David.Laight@...LAB.COM>
To:	'Pravin Shelar' <pshelar@...ira.com>,
	Alexander Duyck <alexander.duyck@...il.com>
CC:	David Miller <davem@...emloft.net>, netdev <netdev@...r.kernel.org>
Subject: RE: [PATCH net-next v2 1/6] skbuff: Add skb_list_linearize()

From: Pravin Shelar
...
> > I'm not completely convinced this is even necessary.  Seems like you are
> > wasting cycles copying the frames around when you could probably just
> > pull the header of the first frame and then use page frages to fill in
> > the rest.
> >
> 
> This is what is done when possible. But skb merging is not always
> possible, for example when  MAX_SKB_FRAGS limit is reached. You can
> have a look at STT implementation.

If MAX_SKB_FRAGS is reached why not just generate two frames?
Except in pathological cases with a log of small fragments
the linearize is likely to be more expensive than the processing.

	David