I've been talking to Linus further about this. Probably we will stick in
just an isapnp enabler and the sysctl api bit which allows you to set cards
up and then load modules, not the callbacks or notional support for other
PnP interfaces
In 2.3.x we will switch all the existing resource allocators to be wrappers
for a single generic resource object managing arbitary resource types and
supporting relocation and hot swapping etc. That'll need input for all sorts
of people - especially the PCMCIA folks
My major remaining concenns are the totally broken VM layer, the fact
the scsi error handler thread stuff appears somewhat dubious and the lack
of adequate kernel resource tracking where it hasnt scaled up with the
scaling of things like descriptor sizes and maximum buffers.
It would be very bad to shit a 2.2 any user can trivially take down.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu