From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mout.gmx.net ([212.227.17.20]:62391 "EHLO mout.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752920AbaJJKJ0 (ORCPT ); Fri, 10 Oct 2014 06:09:26 -0400 Message-ID: <5437B048.4090100@gmx.net> Date: Fri, 10 Oct 2014 12:09:12 +0200 From: JPT MIME-Version: 1.0 To: Patrick Boettcher CC: linux-media@vger.kernel.org Subject: Re: technisat-usb2: i2c-error References: <5434226B.3010804@gmx.net> <20141009172614.5e16f240@vdr> In-Reply-To: <20141009172614.5e16f240@vdr> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Sender: linux-media-owner@vger.kernel.org List-ID: Hi Patrick, Am 09.10.2014 um 17:26 schrieb Patrick Boettcher: > Hi Jan, > >> What exactly do the i2c-errors mean? > > I can't tell you exactly what happens in the device, but I can tell you > that I have the same problem with my device on my PC sometimes. > > In addition to this I2c-failures from time to time the box is quite > sensitive regarding: repowering, replugging and host-rebooting. This is > a USB-device-firmware problem which makes the device crash and all > subsequent USB-transfers are failed. Reloading the module or replugging > the device will make it work again. That's bad news. Could you please add this info on the Wikipage? http://www.linuxtv.org/wiki/index.php/Technisat_SkyStar_USB_HD > How many days without interruption did you use the device? Well, this was about 2 days and 1.5 ;) recordings It's now another 3 days. and there were only a few i2c errors, but nothing serious. I did only two SD recordings in the meantime. No problems yet. I changed the watchdog from 60 secs to 10 mins. I added a delay of 10 secs between unload and reload of the modules in the runvdr script. I hope the machine will be able to recover automatically next time. > I was following quietly you're discussion with Antti. Has someone taken > care of the your changes regarding the transfer-size? No. Since I don't understand anything about what I did ;) And I don't know the process of officially working on the kernel... (Just wondered if I will ever learn that much about kernel programming;) > I think it should included. Then someone with more insight should change the buffers and create a patch. o.O Jan