From owner-freebsd-current@FreeBSD.ORG Fri Sep 17 21:43:25 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A4A2116A4CE for ; Fri, 17 Sep 2004 21:43:25 +0000 (GMT) Received: from aposerv.p-i-n.com (aposerv.p-i-n.com [145.253.185.5]) by mx1.FreeBSD.org (Postfix) with ESMTP id D18A843D1D for ; Fri, 17 Sep 2004 21:43:24 +0000 (GMT) (envelope-from rabe@p-i-n.com) Received: from p-i-n.com (inside.p-i-n.com [129.10.9.21]) by aposerv.p-i-n.com (8.12.11/8.12.11) with ESMTP id i8HLffRU086266 for ; Fri, 17 Sep 2004 23:41:41 +0200 (CEST) (envelope-from rabe@p-i-n.com) Received: (from rabe@localhost) by p-i-n.com (8.11.6/8.11.6) id i8HLfeK26141 for freebsd-current@freebsd.org; Fri, 17 Sep 2004 23:41:40 +0200 (CEST) (envelope-from rabe) Date: Fri, 17 Sep 2004 23:41:40 +0200 From: "Raphael H. Becker" To: freebsd-current@freebsd.org Message-ID: <20040917234140.H55054@p-i-n.com> References: <20040917203851.F55054@p-i-n.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: ; from rwatson@freebsd.org on Fri, Sep 17, 2004 at 02:42:39PM -0400 Organization: PHOENIX Pharmahandel AG & Co KG, Mannheim, Deutschland Subject: Re: de0 doze off X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Sep 2004 21:43:25 -0000 On Fri, Sep 17, 2004 at 02:42:39PM -0400, Robert Watson wrote: > > > Another possible testing avenue (which might be slightly unstable but > > > would illustrate the point) would be to leave debug.mpsafenet=1, done > > > disable preemption, done > > > and remove the IFF_NEEDSGIANT flag from the if_de driver flags. [snip: diff] done > Looks good. This may result in reduced stability, but it may also tell us > specifically if there's a problem with the IFF_NEEDSGIANT mechanism. two parallel wget saturate the link to an average of ~12MByte/sec (netstat 1) for some hours. seems to work perfect now, constant full speed. I transferred around 180 copies of the miniinst.iso and ran a md5 on lots of the file to detect xfer-errors, nothing failed, no more errors. HTH Raphael Becker