summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Expand)Author
2005-09-26Return EIO if reading fails.Mark Kettenis
2005-09-26Match Apple Shasta PCI ID as found on G5 systems.Mark Kettenis
2005-09-26Match "u3" as found in G5 systems.Mark Kettenis
2005-09-26Consider nodes named "mpic" as potential open-pic candidates.Mark Kettenis
2005-09-26- correct framebuffer calculation offset by marking off garbage bits inMartin Reindl
2005-09-26Delete wrong comments, originally came from Makefile.mvme88k.Kenji Aoyama
2005-09-25Define symbolic constants for the processor identification register fieldsMiod Vallat
2005-09-25TyposMiod Vallat
2005-09-25Define the symbolic register indexes in trapframe in <m68k/reg.h> onlyMiod Vallat
2005-09-25Remove unused defines and macros.Miod Vallat
2005-09-25Change the size parameter of cmmu_flush_tlb() from bytes to pages. This makesMiod Vallat
2005-09-25Turn CPU_INFO_FOREACH into a real construct, like all queue(3) iterators,Miod Vallat
2005-09-25Do not consider userland trap #496 (BUG system call on mvme88k) as specialMiod Vallat
2005-09-25Do not try increasing NKMEMPAGE_MAX on sun4m at all, until we makeMiod Vallat
2005-09-25On Sun4m, be sure to steal the overlapping part of the iommu dvma spaceMiod Vallat
2005-09-23- fix PowerNow flags, cosmetic, only affects dmesg output.Brad Smith
2005-09-22NEWINTR is not an option on OpenBSD.Dale Rahn
2005-09-21isbad() is only used on SMD disks on sparc; remove it on other platforms.Miod Vallat
2005-09-21Reduce NKMEMPAGES_MAX_DEFAULT_SUN4M, or we may allocate too much of kernel_mapMiod Vallat
2005-09-20unbreak tree, how did you do that miodTheo de Raadt
2005-09-20Turn CPU_INFO_FOREACH into a real construct, like all queue(3) iterators,Miod Vallat
2005-09-20remove unused pci_intr_evcnt() macro.Brad Smith
2005-09-19Make idle a global variable so that gprof can see it and the time canDale Rahn
2005-09-19floppy support; form@openbsd.ruTheo de Raadt
2005-09-17boot uses abstract 'hd' device names, and 'set tty' worksUwe Stuehler
2005-09-15Remove the (unused) syscall() indirection on arm.Miod Vallat
2005-09-15In syscall(), do not recompute the struct proc * after the syscall hasMiod Vallat
2005-09-15Change child_return() to record a proper ktrace record for vfork childMiod Vallat
2005-09-15Define PG_M_U as PG_M | PG_U, and use it where appropriate; no functionalMiod Vallat
2005-09-15Fix typo.Mark Kettenis
2005-09-15Stop compiling the texte console rotation code #ifdef __zaurus__, but useMiod Vallat
2005-09-15get rid of MD m68k_trunc_page and m68k_round_page and replace by MIMartin Reindl
2005-09-14Remove unused label.Marco Peereboom
2005-09-13Oops, fix ivec[] prototype.Miod Vallat
2005-09-13merge the macglobals one-liner into locoreMartin Reindl
2005-09-12Change the NKMEMPAGES range to 4-64MB for 32bit arches, and 8-128MB for 64bitMiod Vallat
2005-09-12cleanup commentsMartin Reindl
2005-09-12- Add byte-swap and stride options to bus_space.Martin Reindl
2005-09-12provide m68k_page_offset()Martin Reindl
2005-09-11ivec[] needs to be one entry larger on 188.Miod Vallat
2005-09-11cope with the ___moddi3 issue, same as mvme68kTheo de Raadt
2005-09-10beef up and correct the example sectionMartin Reindl
2005-09-08make comments match the fact that we are on sun4u, plus minor cleanupMartin Reindl
2005-09-06fix two conf space related hw erratas; dino-moh-hummMichael Shalayeff
2005-09-06regenMichael Shalayeff
2005-09-06add proj namesand those more modern cpus we do not have ids for yetMichael Shalayeff
2005-09-06Initialize max_cpus for 88110-based systems (i.e. 197 boards). Makes pmapMiod Vallat
2005-09-06Do not depend on MAX_CPUS being 4.Miod Vallat
2005-09-06Correct control registers descriptive names for 88110.Miod Vallat
2005-09-06MVME88K -> M88K in include file guard symbols. No functional changes.Miod Vallat