From owner-freebsd-net@FreeBSD.ORG Thu Apr 19 13:03:48 2007 Return-Path: X-Original-To: freebsd-net@freebsd.org Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id C3C6216A403 for ; Thu, 19 Apr 2007 13:03:48 +0000 (UTC) (envelope-from proks@logos.uptel.net) Received: from logos.uptel.net (logos.uptel.net [195.138.170.125]) by mx1.freebsd.org (Postfix) with ESMTP id 6BFF913C480 for ; Thu, 19 Apr 2007 13:03:48 +0000 (UTC) (envelope-from proks@logos.uptel.net) Received: from logos.uptel.net (logos.uptel.net [195.138.170.125]) by logos.uptel.net (Postfix) with ESMTP id 181F433C83 for ; Thu, 19 Apr 2007 16:03:47 +0300 (EEST) Date: Thu, 19 Apr 2007 16:03:47 +0300 (EEST) From: "Prokofiev S.P." To: freebsd-net@freebsd.org Message-ID: <20070419160040.X17270@logos.uptel.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Subject: Re: Hang up/down re0: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 19 Apr 2007 13:03:48 -0000 Thanks for commit ! Problem with vlanhwtag is not only with ssh... I experimented also with ftp: upload - 1-3 KB/s download - 10 MB/s And here experiment with iperf: freebsd - server with re0 logos - server with fxp0 freebsd>[13:18:22]#>iperf -s ------------------------------------------------------------ Server listening on TCP port 5001 TCP window size: 64.0 KByte (default) ------------------------------------------------------------ [ 4] local 195.138.170.121 port 5001 connected with logos port 62951 [ 4] 0.0-10.0 sec 111 MBytes 93.0 Mbits/sec freebsd>[13:19:30]#>iperf -c logos ------------------------------------------------------------ Client connecting to logos, TCP port 5001 TCP window size: 65.0 KByte (default) ------------------------------------------------------------ [ 3] local 195.138.170.121 port 50422 connected with logos port 5001 [ 3] 0.0-12.8 sec 128 KBytes 81.6 Kbits/sec Thus we have the assymetric traffic (u/d~1/1000) at re0 with vlanhwtag... The same situation was observed with the driver rl (u/d~1/2) several years ago. On Tue, 17 Apr 2007, Pyun YongHyeon wrote: > On Tue, Apr 17, 2007 at 01:35:57PM +0300, Prokofiev S.P. wrote: > > Hello Pyun YongHyeon! > > > > I have updated src on April, 13th and versions of these files differ > > from yours, therefore I have corrected your patch (attached). > > After rebuilding kernel the bug has disappeared. > > Very good job! Thank you! > > Commit please in source. > > > > Thanks for testing. It seems that was a long standing bug. > I've burned all mana. I'll commit it tomorrow. > > > But even earlier I have found one more bug on this NIC. > > If to not switch off "vlanhwtag" (ifconfig re0 -vlanhwtag) ssh connection > > is lost through time and can not establish new. > > > > Do you mean VLAN h/w tagging didn't work at all for ssh connection? > What about other connections with VLAN? > > -- > Regards, > Pyun YongHyeon >