On Sat, 17 Mar 2007, Quentin Garnier wrote:

> The thing you're missing is that attributes are not something the user
> always sees.  E.g., ep* at isa? brings in attribute ep_isa and ep.  If
> you want debug for the ISA attachment, it's different from debug for the
> rest of the driver.  Yes, it's far fetched, but there are probably
> better examples of the issue.

No, thats true - its one of the details that may need to be glossed over
in some way (though, since this is a DEBUG mechanism, I don't see why a
few sharp corners can not be tolerated)

> Well, I don't really understand why you're reluctant to change files.*.

Well I was on a different track and didn't see a need..  taking your
example for instance, we have:

attach	ep at isa with ep_isa: elink
file	dev/isa/if_ep_isa.c		ep_isa

attach	ep at isapnp with ep_isapnp
file	dev/isapnp/if_ep_isapnp.c	ep_isapnp

attach	ep at eisa with ep_eisa
file	dev/eisa/if_ep_eisa.c		ep_eisa

what would you propose we need to add to that, in order to make fine
grained debug output of ep(4) easily accessible?

While I think that improving 'options DEBUG' is a worthy goal (possibly
better than my initial proposal :), I'm still not seeing any need to
twiddle the files.* defs to enable it.. if the knobs were to be sysctl
knobs, then we could already use any of the following namespaces:

	debug.dev.isa.if_ep_isa
	debug.dev.isapnp.if_ep_isapnp
	debug.dev.eisa.if_ep_eisa

(or, because file names are unique)

	debug.if_ep_isa
	debug.if_ep_isapnp
	debug.if_ep_eisa

(or simply)

	debug.ep

(or just using the attribute)

	debug.ep_isa
	debug.ep_isapnp
	debug.ep_eisa

without adding anything but some config(1) magic to generate a source
file containing the knobs?

iain