On Mon, Jul 13, 2009 at 4:08 PM, Devin Heitmueller<dheitmueller@xxxxxxxxxxxxxx> wrote: > On Mon, Jul 13, 2009 at 9:51 AM, Markus Rechberger<mrechberger@xxxxxxxxx> wrote: >> On Mon, Jul 13, 2009 at 5:23 AM, Devin >> Heitmueller<dheitmueller@xxxxxxxxxxxxxx> wrote: >>> Hello Mauro, >>> >>> Please pull from >>> http://kernellabs.com/hg/~dheitmueller/pinnacle_hybrid_2881 for the >>> following: >>> >>> em28xx: set GPIO properly for Pinnacle Hybrid Pro analog support >>> em28xx: make support work for the Pinnacle Hybrid Pro (eb1a:2881) >>> >> >> You might add a message to dmesg, that this setting can burn other >> devices which use the eb1a:2881 vendor/productid. Especially Pinnacle >> they didn't add Svideo/Composite to their device and it works slightly >> different than the others. >> >> Best Regards, >> Markus > > Hello Markus, > > Thank you for taking the time to review the patch and provide > feedback. Two questions: > > 1. Are you suggesting that a warning be shown if someone manually > uses "card=52", or do you mean that we should show a warning if they > actually have the Pinnacle Hybrid Pro 320e? I am using an eeprom > checksum to ensure that not all devices with eb1a:2881 use the device > profile, but rather only devices with the matching hash, so in theory > the profile should only be associated with that one card. > > 2. Are there any additional details you can provide about the > differences in the Pinnacle device you know of which is wired up > differently than all the others? > I am not entirely sure about which device you have, but there's one which does not support Svideo/Composite - there's no connector for it and internally it's not routed properly (although the hardware would be able to support it). I will check those devices this week. > Personally, I would like to get rid of the "card=" modprobe option > completely, as it presents a considerable risk of users who do not > know any better damaging their hardware. > > I am obviously very concerned about any cases where devices could be > damaged, and any additional information you could provide to help > mitigate that risk is certainly welcome. > I think there's no official way to solve this properly as this information is disclosed, so it will be up to the user if he burns his device or not when he uses a device with a generic ID. I still have those 2 Terratec devices which have a slightly melted case from the good old days where reverse engineered values have been used. Best Regards, Markus -- 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
