From mint-bounce@lists.fishpool.fi Tue Jan 22 17:28:01 2008 X-Original-To: fnaumann@mail.boerde.de Delivered-To: fnaumann@mail.boerde.de Message-ID: <473BC7EE.7040009@atari-source.com> Date: Wed, 14 Nov 2007 23:15:42 -0500 From: Mark Duckworth User-Agent: Thunderbird 2.0.0.6 (Macintosh/20070728) MIME-Version: 1.0 To: Peter Slegg Cc: mint@fishpool.com Subject: Re: [MiNT] Mintnet success ? References: <473B68F2.2020001@atari-source.com> <0002f323.017c6ae8327a@smtp.freeola.net> In-Reply-To: <0002f323.017c6ae8327a@smtp.freeola.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ecartis-version: Ecartis v1.0.0 Sender: mint-bounce@lists.fishpool.fi Errors-To: mint-bounce@lists.fishpool.fi X-original-sender: mduckworth@atari-source.com Precedence: bulk List-help: List-unsubscribe: List-Id: X-List-ID: List-subscribe: List-owner: List-post: X-Virus-Scanned: by amavisd-new at relay.boerde.de X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on relay.boerde.de X-Spam-Status: No, hits=-1.0 tagged_above=-50.5 required=7.0 tests=BAYES_00 X-Spam-Level: Status: X-Status: X-Keywords: Peter Slegg wrote: > On Wed, 14 Nov 2007 16:30:26 -0500, Mark Duckworth wrote: > >> The MTU of ethernet is 1500. Under will definitely work fine, but you >> will incur more overhead. Sounds to me like you're ahead of the game >> though seeing as how it's actually working. I don't think this is >> negotiable when it comes to setting it higher though :-P The fact that >> it's working indicates an undersized/corrupted buffer somewhere (or >> perhaps a defect with the card). >> >> Thanks, >> Mark >> >> > > I've tried several different drivers and 3 different cards so we > can eliminate those possibilities. > > Why does lo0 have mtu bigger than 1500 ? > > lo0: flags=0x4b > inet 127.0.0.1 netmask 255.255.255.0 broadcast 127.0.0.255 > metric 0 mtu 16384 > > Is 1500 a Mint limitation ? On a linux box I could ping packets > much bigger than 1500 > > Peter > > > > You're thinking of different things. An ethernet frame is 1500 bytes. ping is ICMP protocol which is higher level than the ethernet packet and can do things like send several 1500 byte packets in order to make up whatever sized ICMP packet you are sending. In fact even though ethernet has a 1500 byte packet size, you can actually send less due to overhead. BUT if you tell your system your underlying device can handle larger packet sizes and the buffers/protocol cannot the end of every packet would be truncated. So you have a packet [---] You have a large chunk of data (which is many packets) [---][---][---][---][---] And if you set your MTU too high the end of the packets will get cut off giving you packets with the ends cut off so you get [---[---[---[---[--- If you can understand what I am saying. In your case, there's probably some pause between each packet send. The lower MTU probably adds enough overhead and slows things down just enough that the corruption disappears. I've seen things like this with serial on my older machines. You've tried different cards but did you try different ISA slots (milan) or ethernecs? I'm willing to bet it would be hard to eliminate all of the possibilities (swap machine with same boot disk) due to lack of hardware. Thanks, Mark