From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [PATCH net-next] net: increase LL_MAX_HEADER if HYPERV_NET is
 enabled
Date: Sun, 01 Nov 2015 14:36:20 -0800
Message-ID: <1446417380.6254.93.camel@edumazet-glaptop2.roam.corp.google.com>
References: <BN1PR0301MB07705F68B22ED84061E37337CA2F0@BN1PR0301MB0770.namprd03.prod.outlook.com>
	 <1446249142.6254.47.camel@edumazet-glaptop2.roam.corp.google.com>
	 <1446398459.6254.90.camel@edumazet-glaptop2.roam.corp.google.com>
	 <20151101.155859.1206189858863918018.davem@davemloft.net>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Cc: haiyangz@microsoft.com, edumazet@google.com,
	netdev@vger.kernel.org, kys@microsoft.com
To: David Miller <davem@davemloft.net>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-pa0-f46.google.com ([209.85.220.46]:35733 "EHLO
	mail-pa0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753087AbbKAWgV (ORCPT
	<rfc822;netdev@vger.kernel.org>); Sun, 1 Nov 2015 17:36:21 -0500
Received: by pasz6 with SMTP id z6so127332080pas.2
        for <netdev@vger.kernel.org>; Sun, 01 Nov 2015 14:36:21 -0800 (PST)
In-Reply-To: <20151101.155859.1206189858863918018.davem@davemloft.net>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Sun, 2015-11-01 at 15:58 -0500, David Miller wrote:
> From: Eric Dumazet <eric.dumazet@gmail.com>
> Date: Sun, 01 Nov 2015 09:20:59 -0800
> 
> > From: Eric Dumazet <edumazet@google.com>
> > 
> > My recent commit, attaching SYNACK messages to request sockets
> > exposed a too small LL_MAX_HEADER when netvsc_drv.c is in use,
> > because this driver sets a needed_headroom of 220 bytes.
> > 
> > Increase LL_MAX_HEADER in this case, to avoid a realloc of all
> > TCP frames.
> > 
> > In another patch, I'll make skb_set_owner_w() more robust.
> > 
> > Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener")
> > Signed-off-by: Eric Dumazet <edumazet@google.com>
> > Bisected-by: Haiyang Zhang <haiyangz@microsoft.com>
> 
> Using a value of 256 just because HYPER-V is crazy imposes a huge
> unnecessary burdon upon the rest of the stack.
> 
> I rejected a previous attempt to use such a huge value for
> LL_MAX_HEADER, and I will do so again here.  We need a different fix
> for this issue, one that doesn't hurt everyone.

Sure, I was planning to send the skb_set_owner_w() fix as well.

I was not aware you already objected to LL_MAX_HEADER increase.