summaryrefslogtreecommitdiff
path: root/sys/arch/arc
AgeCommit message (Collapse)Author
1997-04-10Attempt to clean up local devices.Per Fogelstrom
Added support for new QED RT52x0 processors.
1997-04-10additions & fixes from niklas & iTheo de Raadt
1997-04-06correct labelsector confusionTheo de Raadt
1997-04-05correct DEV_BSIZE vs lp->d_secsize confusion; spotted by hte late night icb ↵Theo de Raadt
gang. Other ports need fixing still
1997-04-04Define in_addr_t and in_port_t everywhere but use u_intXX_t forTodd C. Miller
htonl and friends.
1997-03-26Add ast, tel and IPSECNiklas Hallqvist
1997-03-26Replace "isa_io" with "bus". Not isa specific.Per Fogelstrom
1997-03-23Adding more arches require more structuring and cleanup. More to come...Per Fogelstrom
1997-03-23Moved from pica since they are not Pica bus specificPer Fogelstrom
1997-03-17Serial console trouble again...Per Fogelstrom
1997-03-12Addition of support for a PCI based Vr4300 board from Algorithmics, the P-4032.Per Fogelstrom
Changes to io macros were done to handle sparse bus addressing dynamically. This is a first cut (rough).
1997-02-28Add missing _MACHINE & _MACHINE_ARCH defines + some RCS IdsNiklas Hallqvist
1997-02-25This file was never used so... ZAP!Per Fogelstrom
1997-02-15add kernel_text symbol (just like other archs)Michael Shalayeff
1997-02-06need a frame.h; empty for now; PR#74, gwen@rtmx.comTheo de Raadt
1997-02-04send VM_PROT_* code for si_trapnoTheo de Raadt
1997-02-03More siginfo implementations (alpha and mips might even work)Theo de Raadt
move "siginfo_t *" to 2nd arg of signal handler as 1003.1b requires. I really wish I had 1003.1b documentation.
1997-02-02attempt at siginfo.. Per must look this overTheo de Raadt
1997-02-01Implement BUFCACHEPERCENT. Allow bufpages/nbuf larger than 1 (still max outTheo de Raadt
at all buffers be MAXBSIZE). Remove faulty limit test of buffer pages allocation. The check is now done at the bufpages+nbuf calculations.
1997-01-28add addr arg to sendsig/trapsignal; no siginfo_t support yetTheo de Raadt
1997-01-24do not build fake disklabel partitions for MBR partitions that have ↵Theo de Raadt
out-of-range start/size fields
1997-01-22pmap enter fixes for I/O mapping.Per Fogelstrom
1997-01-22Fixed some pmap problems when porting XFree86 (mmap of i/o etc).Per Fogelstrom
1997-01-19Ooops. This should have been committed along with pccons.cPer Fogelstrom
1997-01-17mmap for TYNE. mods in ps/2 mouse driver to be more generic...Per Fogelstrom
1997-01-14RAW_PARTTheo de Raadt
1997-01-10Use "locore.S" versions of block io operations.Per Fogelstrom
They handle unaligned memory references.
1997-01-04Go to Buslogic and SCSINiklas Hallqvist
1997-01-04Print warning when not recognizing the boot device.Niklas Hallqvist
Also get the partition number from the BIOS when building bootdev.
1996-12-23i think this fixes COMPAT_ULTRIXTheo de Raadt
1996-12-23use _{NLIST,KERN}_DO_{AOUT,ELF,ECOFF} to select binary typesTheo de Raadt
1996-12-11isadma belongs in files.isaNiklas Hallqvist
1996-12-11MCLSHIFT=11, and calc MCLBYTES in such terms. found by pefoTheo de Raadt
1996-12-11Sync headers to ld.so for gdb/arc supportimp
1996-12-09arc port can use the same disksubr code, whee!Theo de Raadt
1996-12-09For the moment, use the same code that we use in the pmax port to generateimp
core files. This may change again as the gdb port progresses, but is my best current guess. Core files generated by OpenBSD/arc with this change are now known by the file command as being core files.
1996-11-30Com driver moved, still somewhat ISA-dependent though, but it willNiklas Hallqvist
improve. Pica and PCMCIA-dependent stuff factored out.
1996-11-30Cleaned up. lpt is MI now.Niklas Hallqvist
1996-11-28Adaption to new bus.h from NetBSD. Various driver has been updated toNiklas Hallqvist
NetBSD 961107 state. Other has not, partly because sometimes the bus.h changes was coupled with other changes.
1996-11-28Add special case code for the rPC44. The DMA buffers on this machineimp
are not cache coherent, so for the moment use uncached addresses for all buffer access. In the future, more specific cache flushing should be done. rPC44 now boots into multi user off SCSI disk or the network, builds its own kernel and seems to be making its way toward finishing a make build. Correct -Wall complaint when BTDEBUG defined while I'm here.
1996-11-25Fix comment typoNiklas Hallqvist
1996-11-25htons et al. works on explicit 16- and 32-bit quantities and not theNiklas Hallqvist
machine dependent "short" and "long" integer. Correct and enhance manpage. Change all short and longs to u_int16_t and u_int32_t, respectively. OpenBSD RCSIds
1996-11-24Fix typoNiklas Hallqvist
1996-11-23remrq -> remrunqueuekstailey
1996-11-23added const to second parameter of cfprint_t routineskstailey
1996-11-12s/bus.h/bus.old.h/ to make the transit to the bus_space bus.h an easy roadNiklas Hallqvist
possibly taken in small steps
1996-11-12Sync to realityNiklas Hallqvist
1996-11-12Add PCCONS_FORCE_WORD mode to limit the moves to 16bit in the VGA memNiklas Hallqvist
area. This is needed for stupid VLB VGA cards that accept larger accesses but ignores everything but the 16 low bits.
1996-11-12Use table to map the BIOS bootdev to our name instead of picking outNiklas Hallqvist
certain characters out of the BIOS string
1996-11-06use dk_mountrootTheo de Raadt