summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2002-04-05Add another check for sc->wi_gone; from FreeBSDTodd C. Miller
2002-04-05Actually set wi_gone when we detach, based on what the an driver does.Todd C. Miller
2002-04-05remove un-needed stuff. I leave NFSCLIENT for hughTheo de Raadt
2002-04-05Use the OF set-colors operation instead of the color! operation, thisDale Rahn
interface seems to work on more machines, also is more efficient when XF4/wsfb is setting the palette. ok deraadt@
2002-04-04permit maxusers down to 2Theo de Raadt
2002-04-04permit compiling without NFSTheo de Raadt
2002-04-04Some PCI machines do not implement the uperf registers. Those that doJason Wright
provide an 'sc at ebus'. Use that instead of a direct attachment of uperf to psycho.
2002-04-04lithuanian kbdmap, made from the script at http://open.bsd.ltMichael Shalayeff
2002-04-04wi_init() now takes a struct wi_softc * not a void * since an sc pointerTodd C. Miller
is available everywhere wi_init() is called. deraadt@ OK
2002-04-04Catch broken devices on sparc64 (where broken == devices that don't haveJason Wright
a function 0). Many thanks to Joey Coleman <joeycoleman@acm.org> for fixing this.
2002-04-04s/0/NULL for pointersJason Wright
2002-04-04Fetch the "idprom" property from the root and if that fails get itJason Wright
from the nvram.
2002-04-03fix possible alignment problem; with markus@, angelos@ deraadt@ ok.Federico G. Schwindt
2002-04-03The PLX version of the NDC NCP130 PCI card w/ embedded Prism2 doesTodd C. Miller
not map the PLX registers so the driver cannot enable interrupts on the PLX chip (this is OK since the card's EEPROM does it for us). To work around this we check that what we expect to be the PLX registers are a) I/O mapped and b) have a non-zero size. OK mickey@
2002-04-03In Host-AP mode, set IBSS SSID to the normal network id. This makesTodd C. Miller
"ifconfig nwid FOO" work correctly in Host-AP mode.
2002-04-03Don't process WCCPv1-style GRE packets, unless allowed (via sysctlAngelos D. Keromytis
variable); GRE processing must also be allowed. From joeycoleman@acm.org, ok deraadt@ niklas@
2002-04-03WCCP sysctl variable -- ok deraadt@ niklas@Angelos D. Keromytis
2002-04-03move the "ukbd_enable: %s: bad call on=%d" from DIAGNOSTIC to UKBD_DEBUGJason Wright
2002-04-03Replace the implementation of microtime with one written in C that doesn't ↵Jason Wright
go backwards on machines w/out counter-timer Remove a bunch of uncessary code (#if 0, #undef, etc) [still exhibits bad behavior on counter-timer machines, but it's the same behavior as the assembler version]
2002-04-03in PCITAG_SETNODE s/t/x to match usage; from netbsdJason Wright
2002-04-03deal with ccb shortage correctly, ok mickey@Niklas Hallqvist
2002-04-03- read the rx completion register exactly once.Jason Wright
- don't complain about still owned descriptors tested by myself and Joey Coleman <joeycoleman@acm.org>
2002-04-03back out ALTQ diff which breaks it; see PR 2517Theo de Raadt
2002-04-03gx doesn't actually do bitbangingNathan Binkert
2002-04-03disable userland crypto for 3.1Theo de Raadt
2002-04-02gx doesn't use vtophysNathan Binkert
2002-04-02keylen is either 8 (40 bit secret ) or 16 (104 bit)Markus Friedl
wrong index used for RC4SWAP during key setup IV is stored in little endian format; works for millert@
2002-04-02time and time+4 might not be the same in the %hi (copycat jason)Michael Shalayeff
2002-04-02match on BCM5421S. ok deraadtDale Rahn
2002-04-02syncDale Rahn
2002-04-02add xxBROADCOM 0x000e as BCM5421S, ok deraadt@Dale Rahn
2002-04-02move to gxTheo de Raadt
2002-04-02bit defns for %tick (basically the ticker bits and the npt bit)Jason Wright
2002-04-02uncomment lxtphy(4) and enable eephy(4) in the RAMDISK kernel;lebel
asked by miod@, OK'ed by drahn@
2002-04-02new PowerBook G4 has a Marvell 88E1000 Gigabit PHY, model 6,lebel
so activate the eephy(4) device for it
2002-04-02add support for Marvell 88E1000 Gigabit PHY, model 6;lebel
from nate@, with one forgotten closing ) and a typo. :)
2002-04-02synclebel
2002-04-02Marvell 88E1000 Gigabit PHY, model 6; from nate@lebel
2002-04-02gx works on alpha and i386 add a commented line for the driverNathan Binkert
2002-04-02Add some new IDs for the Marvell Gigabit PHYNathan Binkert
2002-04-02syncNathan Binkert
2002-04-02Add gx(4)Nathan Binkert
This driver is for the Intel gigabit cards including those based on 82542, 82543 and 82544 chips. Base driver from FreeBSD Support for 82544 from me.
2002-04-02syncNathan Binkert
2002-04-02Fix the names for the Intel Gigabit cardsNathan Binkert
2002-04-02fix keyid encoding wi_tx_keyMarkus Friedl
2002-04-02Prism2 firmware should be reported as version a.b.c not a.b variant c.Todd C. Miller
Only Lucent uses the 'variant' form. Also, prism2 firmware 0.8.2 does WEP w/o requiring promisc mode (from NetBSD). OK mickey@
2002-04-02regenTodd C. Miller
2002-04-023Com AirConnect 3CRWE777A is really just a 3CRWE737A in disguise.Todd C. Miller
They use 3CRWE777A because that is the part number for the PCI adapter it comes with but the card can be used w/o it.
2002-04-02always reset promisc in hostap mode (a comment added), a couple of ↵Michael Shalayeff
formatiing and byte-swapping fixens as well
2002-04-01freem mbuf on input pkt check failureMichael Shalayeff