>> There's some dubious timeout code in cx24123_pll_writereg(), I guess >> this should go. Is this copied from the Windoze driver? >> > > I didn't receive an answer to this. Do you care to comment? > Can we just drop the timeout cruft? > Windows code, no. What's dubious about it? It's supposed to avoid situations where the pll isn't working correctly. It returns an error and avoids hanging situations. Which I've personally witnessed. If you don't like safety, it can be removed. In reality, similar testing should also be added to the same handler in the 24108(?) module. Steve
