On Fri, Mar 18, 2005 at 11:11:48PM +0100, Kenneth Aafl?y wrote: > On Friday 18 March 2005 22:28, Rainer Schubert wrote: > > On Fri, 18 Mar 2005, [iso-8859-15] Fran?ois-Xavier Payet wrote: > > > My card is finally working quite well, but I still have got a problem > > > The image is perfect for a few seconds, but then it become badder and badder > > > > With my card (300i) it's working for half an hour or more without errors, > > but if I zap too fast from one channel to the next or even to a random > > channel, I get the same read errors that you have seen. > > > > There are more registers than just reg=1 presenting problems with reading, > > buit as I don't really know what they are doing, I can't be of any use to > > you. > > > > > From the moment the image begins to get bad, dmesg gets me this message : > > > mt352_read_register: readreg error (reg=1, ret==-5) > > > > > > Does anyone knows something about this? > > > > Maybe kraxel@xxxxxxxxxxx (Gerd Knorr) can tell you what to do. If he does, > > please let me know what he suggested. > > There is a comment about the STATUS_* registers in mt352.c that states that > if they are read to fast the pinnacle might loose lock. Would you please > care to try the attached patch, which limits the read to two times a sec? But it doesn't cause i2c errors. Maybe the bug is in the i2c adapter driver (saa7134 I think)? Johannes
