Hi Mauro, On Tuesday 26 January 2016 07:09:55 Mauro Carvalho Chehab wrote: > Em Mon, 25 Jan 2016 18:01:21 -0200 Mauro Carvalho Chehab escreveu: > > Em Mon, 25 Jan 2016 21:32:21 +0200 Laurent Pinchart escreveu: > > > Hi Mauro, > > > > > > Let's see, I can't test em28xx, could you try remove the CONF_SHARED_PIN > > > change and replacing the write in s_stream with a read-modify-write that > > > disables the output (bits 3, 2 and 0) ? If that works I'll test it with > > > the omap3 isp when I'll be back home. > > > > Didn't work. I'll do more tests later (or tomorrow). > > The root cause weren't at tvp5150 side, but, instead, at em28xx that > were only calling s_stream() to disable the stream, but never enabling it. > > I'm wander why it was doing such thing. Well, s_stream() came years > after the em28xx driver, so I suspect it was a hack to fix some issue > with a particular device. Let's hope that the change won't cause any > regressions on such hardware. Let's blame history :-) > The good news is that em28xx doesn't need MISC_CTL to be filled with > 0x6f to stream. Just 0x09 is enough. So, the same initialization > needed by OMAP3 will work there. > > I'm posting the patch in a few. Great, thanks. -- Regards, Laurent Pinchart -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
