On Mon, May 24, 2010 at 16:45, Scott James Remnant <scott@xxxxxxxxxx> wrote:
> On Mon, 2010-05-24 at 16:38 +0200, Kay Sievers wrote:
>
>> On Mon, 2010-05-24 at 15:34 +0100, Scott James Remnant wrote:
>> > On Mon, 2010-05-24 at 16:15 +0200, Kay Sievers wrote:
>> >
>> > > To be able to let udev support kernel module on-demand loading, the next
>> > > kernel will export:
>> > > alias: devname:<name>
>> > > for some well-defined common kernel modules. This will let the kernel call
>> > > modprobe when the device node is accessed, the subsystem/driver will be
>> > > initialized and the open() of the device node will magically succeed.
>> > >
>> > Isn't this already covered by {char,block}-major-* aliases?
>>
>> No. We need the device name, which isn't available from the module
>> itself, and in many cases different from the module name. the *-major-*
>> alias is used to get the numbers to export though.
>>
>> And this must be limited to explicitly annotated modules too, and not
>> for all sorts of broken stuff, which must not exist in /dev just because
>> the module is available.
>>
> Ah, so this is basically just a bunch of static mknod() calls that udevd
> must do when it starts?
Yes, it does this already, will be part of udev 155.
> Why isn't devtmpfs doing those?
How would devtmpfs know that there is a tun.ko ppp_generic.ko laying
around somewhere on the disk, and could be loaded? :)
Kay
--
To unsubscribe from this list: send the line "unsubscribe linux-modules" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
