| Age | Commit message (Collapse) | Author |
|---|
|
- use the new queue macros.
- use IFQ_POLL() to peek at the next packet.
- use IFQ_IS_EMPTY() for empty check.
- drivers should always check if (m == NULL) after IFQ_DEQUEUE(),
since it could return NULL even when IFQ_IS_EMPTY() is FALSE
under rate-limiting.
- drivers are supposed to call if_start from tx complete interrupts
(in order to trigger the next dequeue under rate-limiting).
|
|
|
|
Cleanup as suggested by millert@
|
|
ok, millert@
|
|
|
|
card. This is useful for PCI adapters where the card is not actually
present. Also check some more wi_read_record() calls and try to
deal sanely (avoid using uninitialized data).
|
|
is info on the PCI -> PCMCIA dumb bridge and not the card itself.
|
|
|
|
o Endian fixes
o Missing WI_RID_AUTH_CNTL and WI_RID_MICROWAVE_OVEN bits
o Move promisc mode setting to be before prism2 detect since
some prism2 firmware revs need to have promisc to support WEP
and it is useless to set promisc and then clear it again.
o Add missing clear of ifp->if_timer when IFF_OACTIVE is cleared
|
|
two lines for readability. In practice this doesn't take an extra
line on the screen.
|
|
|
|
|
|
|
|
|
|
an(4) -- no real code changes at this point.
Rename if_wavelan_ieee.h to if_wi_ieee.h while I'm at it and
create an if_wivar.h similar to NetBSD.
PCI adaptors exist that are not true PCMCIA controllers so
wi(4) @ PCI is possible (though the code is not working yet ;-).
|
