From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from vsmtp14.tin.it ([212.216.176.118]:56705 "EHLO vsmtp14.tin.it" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753559Ab3LEKtz (ORCPT ); Thu, 5 Dec 2013 05:49:55 -0500 Received: from [192.168.1.26] (79.32.49.233) by vsmtp14.tin.it (8.6.060.28) (authenticated as andrea.venturi@tin.it) id 5210101911734EDA for linux-media@vger.kernel.org; Thu, 5 Dec 2013 11:44:03 +0100 Message-ID: <52A058EF.9000701@avalpa.com> Date: Thu, 05 Dec 2013 11:43:59 +0100 From: Andrea Venturi MIME-Version: 1.0 To: linux-media@vger.kernel.org Subject: advice on Easycap dongles and VBI interface.. Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-media-owner@vger.kernel.org List-ID: i'm working on a project for legacy support of a teletext server with composite output interface. i supposed to "ingest" teletext using the /dev/vbi support on linux so i've got an Easycap dongle Bus 001 Device 008: ID 05e1:0408 Syntek Semiconductor Co., Ltd STK1160 Video Capture Device AFAIK (i didn't open) it's using a video decoder SC7113 supposed to be a clone of SAA7113. this device is supposed to work on an Ubuntu 12.04 with old kernel 3.2 so i had to blacklist the stock "RE" kernel module easycap and install the "retrofit" from here: https://github.com/ezequielgarcia/stk1160-standalone i see the saa7115 is loaded and i can also see audio and video through mplayer.. $ lsmod Module Size Used by saa7115 18447 1 stk1160 27732 0 videobuf2_core 28148 1 stk1160 snd_ac97_codec 110213 1 stk1160 .... but in the dev fs there's no sign of /dev/vbi interface.. so my questions are: - are these SAA7113 clones really copycat with all the features supported? - has the SAA711x driver ever been used for /dev/vbi feature? i used to work with Bt878 cards for that feature. - is the stock 3.x mainline kernel of the stk1160 really improved/different/VBI enabled? - is there a way to extract VBI lines on /dev/video0? - is there an easy way to enable or at least test (directly on the USB interface) the VBI interface? - finally which approach do you suggest for supporting this ancient feature, if feasibiliy tests are ok: - a libusb quick hack? - an implementation of the bindings between user level /dev/vbi and underlying SAA711x routines? bests andrea venturi