netdev - Re: changing dev->needed_headroom/needed

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: <20130820092939.7d4cad35@nehalam.linuxnetplumber.net>
Date:	Tue, 20 Aug 2013 09:29:39 -0700
From:	Stephen Hemminger <stephen@...workplumber.org>
To:	Johannes Berg <johannes@...solutions.net>
Cc:	Florian Fainelli <f.fainelli@...il.com>,
	Eric Dumazet <eric.dumazet@...il.com>,
	Ben Hutchings <bhutchings@...arflare.com>,
	netdev <netdev@...r.kernel.org>, linux-wireless@...r.kernel.org
Subject: Re: changing dev->needed_headroom/needed_tailroom?

On Tue, 20 Aug 2013 18:24:57 +0200
Johannes Berg <johannes@...solutions.net> wrote:

> On Tue, 2013-08-20 at 09:20 -0700, Stephen Hemminger wrote:
> 
> > All code must check for needed headroom first, and copy packet
> > if space is not available. Since excess headroom is always safe,
> > most devices just always use the same worst case headroom.
> > 
> > Even with your changes this will still be necessary since packets will
> > be still in flight while features change.
> 
> I agree, it will always be necessary. I guess the question boils down to
> whether/what N bytes headroom more or less actually make a difference.
> 
> The wireless stack currently sets needed_headroom to 34 or so I think,
> but I will need to increase by 8 which was why I had this question to
> start with. However, those 34 are an absolute worst case - in most cases
> it's much less...
> 
> So I suppose the other question we should ask is will increasing from
> ~34 to ~42 make a significant different that it's worth thinking about
> avoiding it in the common cases at all?
> 
> johannes
> 

The only thing that might matter is alignment. but +/- 8 bytes isn't going
to change that.
--
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