Hello!
> If lanmedia advertises a tulip PCI id and a unique subvendor/device id,
> then both tulip and lanmedia drivers will try to grab the same device.
>
> Currently the Tulip driver needs to either list all existing
> subvendor/device ids (not feasible), or manually exclude specific
> subvendor/device ids in the xxx_init_one method.
>
> The more I think about the more I like adding an 'exclude_table' field
> to struct pci_driver.
>
> Does the attached patch look ok to everyone?
I very dislike adding of such complex extensions to the originally very simple
PCI probing API we have, especially when it's done to fight one single case
of brain-dead devices. I think the right place for this hack is the tulip
driver itself which should handle it in its probe function (either by testing
the lanmedia subsystem ID manually or calling pci_match_device on its local
blacklist).
About the tulip/lanmedia vs. modutils problem: I think it would be better
to just teach the tool reading the ID lists that it should prefer the most
specific match (think classless routing :) ). In case both driver modules
are compiled, everything works right, if they are not, the worst thing that
can happen is that modprobe will load the wrong driver (which will ignore
the device) instead of saying that there exists no driver. Not much harm,
I think.
Have a nice fortnight
-- Martin `MJ' Mares <mj@ucw.cz> <mj@suse.cz> http://atrey.karlin.mff.cuni.cz/~mj/ "Immanuel doesn't pun, he Kant."- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Apr 30 2000 - 21:00:10 EST