summaryrefslogtreecommitdiff
path: root/sys/arch/mvme68k/dev
AgeCommit message (Collapse)Author
1996-08-20init sc_mem; cleanupTheo de Raadt
1996-06-11tty_attach()Theo de Raadt
1996-06-11use void * for pa/va instead of caddr_tTheo de Raadt
1996-06-11use intr_findvec()Theo de Raadt
1996-06-11call am7990_config()Theo de Raadt
1996-05-29update for new generic bootchuck
1996-05-27import from sun3chuck
1996-05-10Add some code bulletproofing, I have had this around for some timeDale S. Rahn
I actually thought it was checked in. This should stop any extra transmit interupts occuring and a possible crash if another system talks to the tty port before it is completely configured. Also allows debugging when no processes have the port open for writing. If abort is pressed before getty starts.
1996-05-10split le driver, if_name/if_unit -> if_xname/if_softcTheo de Raadt
1996-05-10fix sync scsiTheo de Raadt
1996-05-07delete impossible to reach code, use more defined constantsTheo de Raadt
1996-05-06correct my interrupt bugsTheo de Raadt
1996-05-06fix russian typoTheo de Raadt
1996-05-05add general ether_ioctl call in net/if_ethersubr.c,Michael Shalayeff
NS,IPX,X.25 special processing is now handled in there. reflect this amazing addition in all the ether ifaces. ppl, pls check the stuff.
1996-04-28wdsc driver hacked from amiga driver by Steve WoodfordTheo de Raadt
1996-04-28cfattach/cfdriver split; we hate caddr_tTheo de Raadt
1996-04-28cfattach/cfdriver splitTheo de Raadt
1996-04-28add OpenBSD headerTheo de Raadt
1996-02-03Ensure siop_acb alignment via malloc rather than let alignment dependchuck
on the size of MI data structures that come before it in the softc. Chip will not function properly if alignment is wrong. Detected and fixed by: Chuck Cranor <chuck@ccrc.wustl.edu> and Michael L Hitch <osymh@gemini.oscs.montana.edu>
1995-12-27use splnet instead of splimpTheo de Raadt
1995-12-16Two changes are made in this checkin.Dale S. Rahn
The first is a change to force the console to stay 8N1, the configuration of the bug/prom during the first open to the device by init. The second is to not enable interupts until the port is opened. Previously receive and CD change interrutps were enabled at config time. If a character arrived before the port was opened for the first time, it would crash the system.
1995-12-01we cannot jump into ddb until the interrupt has been acknowledged to theTheo de Raadt
chip. set a flag and defer calling ddb till end of interrupt routine.
1995-12-01previous verbosity patch was wrongTheo de Raadt
1995-12-01add CONSOLEBREAKDDB support. also be less verbose at boot..Theo de Raadt
1995-12-01fixed ddb, improved fifo handling, new still-not-working dma supportTheo de Raadt
from drahn@pacific.urbana.mcd.mot.com
1995-12-01for now, assume no flash on 167 (is this true?)Theo de Raadt
1995-11-30do not drop into debugger on simple recoverable errorsTheo de Raadt
1995-11-19in clcninit: we need to use IIOV to convert the physical address ofchuck
the console to a virtual address. otherwise we can not print anything because our register pointer is wrong (this is on the 167).
1995-11-17add a neat hack for write, but it still does not workTheo de Raadt
1995-11-13comment a bit betterTheo de Raadt
1995-11-07$Id$ throughoutTheo de Raadt
update many copyrights
1995-10-29fixes from daleTheo de Raadt
1995-10-20kill duplicateTheo de Raadt
1995-10-18old filesTheo de Raadt
1995-10-18mvme68k port by me. some parts by dale rahn.Theo de Raadt
1995-10-18initial import of NetBSD treeTheo de Raadt