On Fri, Oct 01, 2004 at 01:22:45PM +0100, Adrian Cox wrote: > On Fri, 2004-10-01 at 07:52, Greg KH wrote: > > On Fri, Sep 24, 2004 at 10:21:08PM +0200, Michael Hunold wrote: > > > > If we have a PCI card where we exactly know what we are doing, we can > > > use the NO_PROBE flag to effectively block any probing and can use the > > > proposed interface to manually connect the clients. > > > > But why? The .class feature can accomplish this too. Just create a new > > class for this type of adapter and device. Then only that device will > > be able to be connected to that adapter, just like you want to have > > happen, right? > > Either the i2c devices need to be able to support a list of permitted > adapters, or the i2c adapters need a list of permitted clients. A single > class isn't adequate. Consider the following scenario: > > The FooTV123 has multiplexor MX3R0K3 and frontend XYZZY, the TVMatic3000 > has frontend XYZZY and multiplexor MX31337, and the FooTV124 has > multiplexor MX31337 and frontend FR012. All three cards are installed in > the same machine. In the worst case the probe code for MX31337 puts > MX3R0K3 into a state that requires a hard reset. > > Manual connection of clients makes it easier to develop a driver outside > the kernel tree, then merge it when ready, without having to allocate a > number from a central authority. Ok, I now understand better, thanks for putting up with me :) So, got a patch to do this? thanks, greg k-h
