Hi Sab, On Friday 11 January 2008, RUMI Szabolcs wrote: > Hi! > > On Thu, 10 Jan 2008 21:34:13 +0100 > > Laurent Pinchart <[EMAIL PROTECTED]> wrote: > > > What makes this difference that it works while other apps don't? > > > > Skype might be buggy ;-) I know that Skype tries to change the frame > > rate while streaming. The driver currently allows that but doesn't > > handle the request properly. It might make the webcam crash. > > > > I plan to fix it, probably by disallowing frame rate changes while > > streaming. > > I'd be more than happy to test that fix you are planning... ;-)
Then grab the latest SVN version. I've just committed a patch that disallows changing the frame format or the frame size while streaming. > I've read somewhere that there have been some people who tried to > create various patches in order to get this camera to work but you > refused to integrate them because you didn't wanted to taint your > nice code with such ugly and/or occasionally working things which > sounds reasonable. The early UVC Logitech webcams were all buggy. Under some particular timing conditions the firmware misses an interrupt which results in errors. Those timing conditions, while very uncommon under Windows, happen quite often for some Linux users. A few people tried various workarounds but no definitive fix has been found. > I ain't got no clue about no nothing but I thought maybe there could > be made an established interface for that like a "workarounds=0x0000" > module parameter where 0x0000 is a bitmask that can be used to enable > or disable the various kludges which iron out such bugs like the ones > that the QC Fusion has. It might happen with various cameras even in > the future that they've got a bug that is ugly to work around or it > is no good for some reason to always work around them in the same way. > > Of course the default would be 0x0000 as in no workarounds enabled > for the main driver to be nice and solid but any of them could be > easily enabled and tested by those who need them to get their camera > working. This way they could also get evolved within the broader > linux-uvc community including people who are not able or willing > to patch/hack their kernel sources on their own for example. > > What do you think? The driver already includes workarounds for various broken devices. Have a look at the UVC_QUIRK_* macros in uvcvideo.h and how they are associated with devices in uvc_driver.c. For early Logitech devices, however, no satisfying workaround has been find so far. Best regards, Laurent Pinchart _______________________________________________ Linux-uvc-devel mailing list [email protected] https://lists.berlios.de/mailman/listinfo/linux-uvc-devel
