Your fix worked. Great job! We're finally getting CVS back to good shape. I'll keep working on the PCI drivers in kernel/bussses. Things should go much faster when there aren't problems on the i2c side... Mark Studebaker wrote: > It sounds like you got it, I'll test it tonight and let you know. > thanks > mds > > "Mark M. Hoffman" wrote: > >>* Mark M. Hoffman <mhoffman at lightlink.com> [2003-06-03 23:07:38 -0400]: >> >>>* Mark D. Studebaker <mds at paradyne.com> [2003-06-03 22:41:20 -0400]: >>> >>>>Works fine on ISA. >>>> >>>>If you modprobe an i2c bus driver after w83781d is modprobed, the bus >>>>driver modprobe hangs, >>>>presumably while w83781d is scanning the bus. >>>>If you modprobe w83781d after an i2c bus driver is modprobed, the w83781d >>>>modprobe hangs. >>>> >>>>Any ideas let us know. >>> >>>I assume you're talking about lk2-4, not the 2.5 stuff that recently >>>went to lkml. >>> >>>I haven't ever been able to load lk2-4 w83781d.o. The symptoms are >>>similar: it hangs as soon as I modprobe it (with I2C adapter). I'll >>>try to debug that one so you can hack on the other chips I don't >>>have. :) >> >>The function i2c_check_addr() in i2c-core.c was causing deadlock. >>I'm surprised if this didn't happen with any other chips. I fixed >>it and *very* lightly tested it (need sleep!). Anyway, w83781d >>seems to work now on I2C lk2-4 branch. >> >>Regards, >> >>-- >>Mark M. Hoffman >>mhoffman at lightlink.com > >
