summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2001-01-13use md in_cksumMichael Shalayeff
2001-01-13Driver for Cirrus Logic CS4281 sound chips. Originally written for NetBSD byAaron Campbell
Tatoku Ogaito, banged into shape for OpenBSD by me. Recording has not been tested yet, but mpg123 works fine. Sound still works after suspend/resume. XXX: On the IBM ThinkPad X20, must disable pcibios else the cs4281 won't generate interrupts. Cause yet unknown but probably not the driver's fault. YMMV on other laptops with this chip.
2001-01-13Booting kernel with MACHINE_NEW_NONCONTIG. UVM code added but not working.Steve Murphree
New stand config. Lots of header fixes. Can now cross-compile i386->m88k.
2001-01-13s!UFS!FFSMiod Vallat
2001-01-13own in_cksum versionMichael Shalayeff
2001-01-12no pmap_changebit no more; better barrier thingMichael Shalayeff
2001-01-12fix PTE_BITSMichael Shalayeff
2001-01-12move pmap_physseg back to vmparam.h where it belongs.Michael Shalayeff
define PMAP_PREFER in pmap.h, as afar as i understand it works. no more pmap_changebit() grow USRIOSIZE in vmparam.h
2001-01-12do not print pidr3,pidr4 which we do not use anywayMichael Shalayeff
2001-01-12this is even uglier hack, but it also works; theo did not help (;Michael Shalayeff
2001-01-12Update vm interface to MACHIN_NEW_NONCONTIG. Fix compile warning in pcctwo.cSteve Murphree
2001-01-11Add maestroMarc Espie
2001-01-11read rom through the rom handleMichael Shalayeff
2001-01-11there is no pseudo-device crypto either; sleep theoMichael Shalayeff
2001-01-11ich/auichMichael Shalayeff
2001-01-11move ich to auich at mickey's requestTheo de Raadt
2001-01-08Compare the end of symbol table as passed from boot with the addressHugh Graham
of _end in the kernel, and base Sysmap on whichever is higher. This permits stripped kernels to load despite unexpected values from libsa.
2001-01-07- add EMODD emulation (EMODF) -- this fixes modf() and some libm codeBrandon Creighton
on systems which don't have native EMODD (i.e., most of them) - big cleanup of the getval_* stuff - have one routine doing most of the work instead of four doing the same thing - add some miscellaneous routines -- count sig. bits - add more operand addressing modes (still not all fully implemented) both the existing POLYD and EMODD can use them - add me to copyright
2001-01-05Add i386 specific sysctl to modify halt -p processing, required to makeMarco S Hyman
powerdown work correctly on some processors. Idea from mickey@ and smcho@tsp.korea.ac.kr. Tested my me and angelos@. Note: If you want to try these out you'll need to cd /sys/arch/i386/include && sudo install -m 444 -o root -g root cpu.h /usr/include/i386 cd /usr/src/sbin/sysctl && make obj && make && sudo make install to get a sysctl that understands the new variable (machdep.apmhalt).
2001-01-04Sync with genassym.cf symbol name changes.Miod Vallat
2001-01-04Convert the old genassym.c to .cf style. Remove unneeded definitions,Miod Vallat
change some names to be not different from other m68k arches.
2001-01-04Harmonization with other m68k arch Makefiles. Stricter compilationMiod Vallat
flags, compile C code with -m68020. Switch the generation of assym.h from a standalone genassym.c to a regular genassym.cf.
2001-01-04KNF, add prototypes to compile with -Wmissing-prototypes.Miod Vallat
2001-01-04Remove netintr().Miod Vallat
2001-01-04Y2K fix for TODC - Submitted by Bob Harbour. Thanks Bob!Steve Murphree
2001-01-04Y2K fix for TODC - Submitted by Bob Harbour. Thanks Bob!Steve Murphree
2001-01-03Sync with GENERIC + cleanupMiod Vallat
2001-01-03Sync with GENERIC, minor cleanups.Miod Vallat
2001-01-03Remove hardcoded SunOS-compatible drive name assignments. No other archMiod Vallat
does that, and it's more confusing than useful. Suggested by fries@
2001-01-03Remove bogus comment. Also fix ramdisk kernel detection.Miod Vallat
2001-01-03netintr() doesn't exist anymoreMiod Vallat
2001-01-03Use netisr_dispatch.Miod Vallat
2001-01-02handle no-event as the end of event fetch loop; my fix for pr-1588 by Sam ↵Michael Shalayeff
Isaacson <sbi@mobius.colorado.edu>
2000-12-29kiss in the rng; all those evil voices kept me pushing for it...Michael Shalayeff
2000-12-28Define UADDR not as fixed address, but rather as what it really is:Miod Vallat
a negative offset relative to the top of the adressing space. This fixes the miscompilation problems in locore_r2000.S with binutils 2.10. Tested by maja@ and myself.
2000-12-28Alter compilation flags, since binutils have changed some of their defaultMiod Vallat
settings between 2.7 and 2.10. Inspired by a similar NetBSD change. Also, while I'm at it, add more -W stuff to CDIAGFLAGS
2000-12-28mvme88k updates to -current. finally!Steve Murphree
2000-12-27better english and fix some spelling mistakes, Ok'd by aaron@Brad Smith
adjucent -> adjacent usefull -> useful
2000-12-25always read new byte from rng, then decide wheather to start a new cycleMichael Shalayeff
2000-12-21People have difficulty spelling 'initial' and derivatives thereof (too manyAaron Campbell
"i's" I guess).
2000-12-19This fixes keyboard lock-ups I was getting wiggling the mouse inConstantine Sapuntzakis
console mode while using my omnicube KVM switch. The keyboard lock-ups were caused by the keyboard/mouse interface controller's input buffer filling up. This was caused because the mouse was sending data to the computer and the pms driver wasn't draining it. The driver used to rely on the fact that it could disable the mouse device when the mouse driver was closed. Disabling the mouse device prevents it from sending data to the computer. At the same time pms disabled the mouse device, pms disabled interrupt notification of the arrival of new data. This regime doesn't work so well with KVM switches which re-enable the mouse. In this case, the mouse will send data to the computer but the keyboard/moust interface controller won't raise an interrupt (because pms disabled interrupts). pms will never read from the input buffer.
2000-12-17Compaq SMART Array RAID controllers.Michael Shalayeff
based on netbsd driver. testing helps from brad@ eisa untested, but should work (;
2000-12-17Add a new possible value for the machdep.allowaperture sysctl: `2'Matthieu Herrb
allows access to the whole 1st megabyte of memory, for use by XFree86 4.0.x int10 code. (ok, deraadt@) CVS: ---------------------------------------------------------------------- sys/arch/i386/i386/mem.c share/man/man4/man4.i386/xf86.4 CVS: ----------------------------------------------------------------------
2000-12-15Define BUS_SPACE_BARRIER_{READ,WRITE} as synonmys for BUS_BARRIER_{READ,WRITE}Artur Grabowski
2000-12-15ITE_INGRF checks needed, from NetBSD.Marc Espie
Reported by Lars Hecking.
2000-12-15Grow the number of reserved pt pages.Artur Grabowski
Solves the 64MB problem. Somewhat based on NetBSD.
2000-12-14native ELF and ELF for nlistArtur Grabowski
2000-12-14Enable loading of ELF kernels.Artur Grabowski
2000-12-14First shot at making ELF compile.Artur Grabowski
2000-12-12Add an extra apm_periodic_check() call before the one we check forAngelos D. Keromytis
errors; that way, Theo's laptop can still turn off APM when there's an error, and the IBM 770Z won't gratuitously disconnect APM (since the first event is always a bogus one).
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-29 14:29:56 +0000