Nicolas Will wrote: > You should really stick to the 1.10 firmware. 03-pre1 was an earlier > test and has more issues. Well, I've figured out what was going on. Seems if you run 'make' against the source tree with one kernel, it will always build modules for said kernel until you run 'make distclean'. I started by building for 2.6.24-8-generic, then upgraded to -10-generic, then to -11-generic, and only did a 'make clean; make; sudo make install' when I rebuilt v4l-dvb, as this worked for Madwifi. So to summarise, if you're going to reuse the same source tree for multiple kernels, make distclean before making the drivers, or it'll build for the last kernel you built for... Not sure if it installs to the running kernel, but it certainly doesn't use the headers for the running kernel... But at least the card seems to be behaving now. Up 10 hours with: options dvb-usb-dib0700 debug=15 force_lna_activation=1 options dvb_usb disable_rc_polling=1 options usbcore autosuspend=-1 I've also blacklisted dvb-usb-dib0700 and modprobe'd it in an rc-script, so my HVR-3000 ends up as device 0 and the two T-500 tuners end up as devices 1 and 2; said shell script also sets up symlinks for the 1st and 2nd front-ends on the HVR to devices 10 and 11, because MythTV doesn't like the idea of a card having multiple front-ends... The clock is once again running... -- Phil. | (\_/) This is Bunny. Copy and paste Bunny lists@xxxxxxxxxxxxx | (='.'=) into your signature to help him gain http://www.philpem.me.uk/ | (")_(") world domination. _______________________________________________ linux-dvb mailing list linux-dvb@xxxxxxxxxxx http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
