Re: cvs commit: src/sys/i386/i386 nexus.c src/sys/i386/include atomic.h src/sys/kern kern_poll.c lwkt_serialize.c src/sys/net if.c if_var.h rtsock.c src/sbin/ifconfig ifconfig.c src/sys/dev/netif/dc if_dc.c src/sys/dev/netif/em if_em.c if_em.h ...

[Joerg Sonnenberger <joerg@xxxxxxxxxxxxxxxxx>]

On Wed, May 25, 2005 at 11:18:51AM -0700, Matthew Dillon wrote:
>     My main desire is to simplify the APIs.  e.g. Having the driver's
>     *interrupt* routine try to register the poll really complicates the
>     poll registration code.

I said before that I want to move most of the common nic_ioctl
into ether_ioctl and deal with appropiate callbacks. I didn't do
that in the first patch, because it would have made much larger.
Given that most of the conditional can go away, the API point is
weak. Most importantly, the current code doesn't allow a device
"poll only" operation, which is useful for the no-interrupt case.
When we push the IFCAP handling for the default cases down into
ether_ioctl, the effective per-driver change is almost 0.

Joerg