From owner-cvs-all@FreeBSD.ORG Thu Apr 1 08:19:15 2004 Return-Path: Delivered-To: cvs-all@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E760516A4CE; Thu, 1 Apr 2004 08:19:15 -0800 (PST) Received: from prg.traveller.cz (prg.traveller.cz [193.85.2.77]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3249D43D3F; Thu, 1 Apr 2004 08:19:15 -0800 (PST) (envelope-from mime@traveller.cz) Received: from prg.traveller.cz (localhost [127.0.0.1]) i31GJDKh090514; Thu, 1 Apr 2004 18:19:13 +0200 (CEST) Received: from localhost (mime@localhost)i31GJDYD090511; Thu, 1 Apr 2004 18:19:13 +0200 (CEST) Date: Thu, 1 Apr 2004 18:19:13 +0200 (CEST) From: Michal Mertl To: Alexey Dokuchaev In-Reply-To: <20040401105857.GA72284@regency.nsu.ru> Message-ID: <20040401180334.Y72588@prg.traveller.cz> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: Ruslan Ermilov cc: cvs-all@freebsd.org cc: Nate Lawson Subject: Re: cvs commit: src/sys/pci if_ste.c X-BeenThere: cvs-all@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: CVS commit messages for the entire tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 01 Apr 2004 16:19:16 -0000 On Thu, 1 Apr 2004, Alexey Dokuchaev wrote: > On Thu, Apr 01, 2004 at 09:42:41AM +0300, Ruslan Ermilov wrote: > > On Wed, Mar 31, 2004 at 01:47:16PM -0800, Nate Lawson wrote: > > > On Wed, 31 Mar 2004, Ruslan Ermilov wrote: > > > > Modified files: > > > > sys/pci if_ste.c > > > > Log: > > > > Under a heavy RX load, at least with D-Link DFE-550TX adapters, > > > > the driver's RX ring head may fall behind the chip, causing the > > > > stuck traffic, disordered packets, etc. Work around this by > > > > adopting the technique of resyncing RX head used in dc(4) and > > > > xl(4) drivers, but do it in a slightly different place to reduce > > > > the number of resyncs needed. > > > > > > > > Also, set the NIC's RX polling period to a more meaningful value, > > > > to stop overloading the PCI bus (this also reduces the number of > > > > resyncs by a factor of 3 or more in a long run; the actual number > > > > is very dependent on a nature of the traffic). > > > > > > > > Maintain the statistics counter as the hw.ste_rxsyncs sysctl. > > > > > > This pollutes the toplevel hw sysctl space. Please move it to > > > hw.ste.rxsyncs or better, debug.*. > > > > > Seems I followed the bad practice from dc(4) and fxp(4). I will > > move it to hw.ste, thanks. > > Care to fix this "bad practice" in dc(4) and fxp(4) as well? :) Or > should we wait for their maintainers, or possibly filling the PR ? > > ./danfe Wouldn't it make some sense to include unit number too? -- Michal Mertl