spelling
feedback and ok tb@ jmc@ ok ratchov@

Use ieee80211_has_addr4() inline function where possible.

ok stsp@

No need to include <net/if_arp.h>

This header is only needed because <netinet/if_ether.h> declares a
structure that needs it. But it turns out that <net/if.h> already
includes it as workaround.

A proper solution would be to stop declarting "struct ether_arp"
there. But no driver should need this header.

unifdef INET

include bpfilter.h so the softc will be the same size in all files

Replace all ovbcopy with memmove; swap the src and dst arguments too
ok otto

Implememnt AP side power saving for acx(4). Tested on acx11.
With some input and OK from kettenis@ and OK sthen@

remove proc.h include from uvm_map.h. This has far reaching effects, as
sysctl.h was reliant on this particular include, and many drivers included
sysctl.h unnecessarily. remove sysctl.h or add proc.h as needed.
ok deraadt

unbreak build.

pointed out by mglocker@

replace the ieee80211_wepkey structure with a more generic ieee80211_key
one that can be used with other ciphers than WEP.

Another round of makro killing. Kill _ACX_CONF_FUNC() and replace all
calls with acx_set_conf()/acx_get_conf(). OK mglocker@

Kill some more macros. This time ACX_NOARG_FUNC and ACX_INIT_TMPLT_FUNC
bite the dust and make the code easier to read. Help and OK mglocker@

First part of fixing broken beacon frames in acx(4) AP mode based on a
diff from Sepherosa Ziehau (DragonFly); The firmware TIM element template
was initialized wrong.

Though, our ieee80211_alloc_beacon() routine calculates it's own TIM
element, which now results in having two TIMs in an acx(4) beacon.

Would it be an idea to introduce a ieee80211_alloc_beacon() flag to
turn the TIM element off? Because fiddling out the TIM element in the
driver for cases in which the firmware calculates its own TIM is
a bit ugly ...

ok claudio@

First round of evil macro removal. _acx_set_##name##_tmplt bites the dust.
OK mglocker@

avoid extra inclusions; ok jsg

Make use of ieee80211_std_rateset

add missing sys/device.h header.

Replace existing rate adaption code with the ieee80211_amrr framework.

ok jsg@

spacing

Turn off debug messages by default.
Reminded by deraadt@

Enable WEP.

ok jsg@

Add bpf hooks. Based on slightly modified diff from jsg.

ok jsg@

Some KNF.

Adjust copyright.

Fix more cvs Id tag tweaks in copyright notes.

Spotted by <gwyllion@ulyssis.org>

Add/Fix copyright notes.

Change to OpenBSD cvs tag.

Inital import for the acx(4) driver.

ok deraadt@ jsg@

Use ieee80211_has_addr4() inline function where possible.

ok stsp@

No need to include <net/if_arp.h>

This header is only needed because <netinet/if_ether.h> declares a
structure that needs it. But it turns out that <net/if.h> already
includes it as workaround.

A proper solution would be to stop declarting "struct ether_arp"
there. But no driver should need this header.

unifdef INET

include bpfilter.h so the softc will be the same size in all files

Replace all ovbcopy with memmove; swap the src and dst arguments too
ok otto

Implememnt AP side power saving for acx(4). Tested on acx11.
With some input and OK from kettenis@ and OK sthen@

remove proc.h include from uvm_map.h. This has far reaching effects, as
sysctl.h was reliant on this particular include, and many drivers included
sysctl.h unnecessarily. remove sysctl.h or add proc.h as needed.
ok deraadt

unbreak build.

pointed out by mglocker@

replace the ieee80211_wepkey structure with a more generic ieee80211_key
one that can be used with other ciphers than WEP.

Another round of makro killing. Kill _ACX_CONF_FUNC() and replace all
calls with acx_set_conf()/acx_get_conf(). OK mglocker@

Kill some more macros. This time ACX_NOARG_FUNC and ACX_INIT_TMPLT_FUNC
bite the dust and make the code easier to read. Help and OK mglocker@

First part of fixing broken beacon frames in acx(4) AP mode based on a
diff from Sepherosa Ziehau (DragonFly); The firmware TIM element template
was initialized wrong.

Though, our ieee80211_alloc_beacon() routine calculates it's own TIM
element, which now results in having two TIMs in an acx(4) beacon.

Would it be an idea to introduce a ieee80211_alloc_beacon() flag to
turn the TIM element off? Because fiddling out the TIM element in the
driver for cases in which the firmware calculates its own TIM is
a bit ugly ...

ok claudio@

First round of evil macro removal. _acx_set_##name##_tmplt bites the dust.
OK mglocker@

avoid extra inclusions; ok jsg

Make use of ieee80211_std_rateset

add missing sys/device.h header.

Replace existing rate adaption code with the ieee80211_amrr framework.

ok jsg@

spacing

Turn off debug messages by default.
Reminded by deraadt@

Enable WEP.

ok jsg@

Add bpf hooks. Based on slightly modified diff from jsg.

ok jsg@

Some KNF.

Adjust copyright.

Fix more cvs Id tag tweaks in copyright notes.

Spotted by <gwyllion@ulyssis.org>

Add/Fix copyright notes.

Change to OpenBSD cvs tag.

Inital import for the acx(4) driver.

ok deraadt@ jsg@