summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2006-01-15Do not put HZ in the kernel configuration file anymore, and let the kernelMiod Vallat
decide by itself: Quadra-style machines with A/UX style interrupts and clock at level 6 will use 100Hz now, while other machines will remain at 60Hz.
2006-01-14Revert 1.12 and go back to 32K MAXPHYS; 64k MAXPHYS exposes subtle bugsMiod Vallat
and is not reliable enough.
2006-01-13support i915. adapted from freebsd by Dimitry Andric. does not breakTheo de Raadt
at least i810 on x40, but made x41 work. ok matthieu
2006-01-13list other maxds modelsTheo de Raadt
2006-01-13Bring a few improvements from NetBSD:Miod Vallat
- recognize a few more models. - better memory range detection, especially if they are in direct translation windows on 68040. - better internal video address detection.
2006-01-13KNF and remove dead parts.Miod Vallat
2006-01-13Sync OSS code with NetBSD; supports poweroff now.Miod Vallat
2006-01-13Remove unused bad{,b,w,l}addr.Miod Vallat
2006-01-13Remove the Mac Rom Glue code completely. With the ADB ``direct'' code beingMiod Vallat
used by default, and since all PRAM accesses are either directly fiddling with VIA registers or through ADB commands, the MRG code has no reason to stay. This means the kernel is now not running unknown PROM code anymore.
2006-01-13Allow proper detaching of wdc, so that detaching a mediabay device properlyMiod Vallat
releases resources and works as expected. Thanks to Thomas Maschutznig for testing.
2006-01-13Fix printf format string due to types changeTodd C. Miller
2006-01-13Make size_t, ssize_t, intptr_t and uintptr_t consistent amongst ourTodd C. Miller
architectures. They are now defined as unsigned long, long, long and unsigned long respectively.
2006-01-13white space vs tabs; no binary changeMichael Shalayeff
2006-01-12Move to using gdt only (no more ldt in general case) but with a variableTobias Weingartner
limit selector, so that the w^x line can float much more dynamically. Much work done by tom. Tested by various people. Addresses concerns of (Julien Tinnes) <julien ATHOST cr0.org>
2006-01-12oopsTheo de Raadt
2006-01-12after range-checking save nkpde back; toby@ ok and discussion w/ hshoexer@Michael Shalayeff
2006-01-12- add a missing htons() to the receive checksum routine to fixBrad Smith
checksumming on little endian systems. - only set the Ok flag in the checksum routine. Checksumming fix from: Pyun YongHyeon <pyunyh at gmail dot com>
2006-01-12Fix up interrupt routing on SiS 963.Jonathan Gray
From Ian McWilliam ian at dodo.com.au. ok mickey@
2006-01-11Use bounded string functions while constructing the disassembly lines; plusMiod Vallat
this compiles into smaller code!
2006-01-11Let's move the orphans to the Attic.Miod Vallat
2006-01-11hotplug(4) and hotplugd(8) hereMartin Reindl
2006-01-11Switch to __HAVE_DEVICE_REGISTER to find the device we're booted from.Miod Vallat
Simpler, better.
2006-01-10Simplify external macfb interfaces, kill an intermediate attachment structure.Miod Vallat
2006-01-10The __CONCAT macro here collides with the one in cdefs.h pulled in viaMartin Reindl
errno.h. Don't include errno.h and export EFAULT therefore, similiar to hppa.
2006-01-10missing cast to vaddr_tMartin Reindl
2006-01-10struct cfdriver shall not be declared as const (oops).Miod Vallat
2006-01-10need sys/types.h for vaddr_tJolan Luff
2006-01-10Really pass -fno-stack-protector in CFLAGS as intended, and define theMiod Vallat
load addresses in one single place; from mvme88k
2006-01-10cull the deadTheo de Raadt
2006-01-10comment amdpmTheo de Raadt
2006-01-10use ${INSTALL_STRIP} ok miodTheo de Raadt
2006-01-10The __CONCAT macro here collides with the one in cdefs.h pulled in viaMartin Reindl
errno.h. Don't include errno.h and export EFAULT therefore. looks good miod@
2006-01-10Remove duplicate _C_LABEL define. OK miod@Todd C. Miller
2006-01-09Fix last change; we need machine/reg.h for struct sigcontext toTodd C. Miller
be fully defined.
2006-01-09Spank the clock: remove relics of old GPROF code, compute tickfix forMiod Vallat
better clock adjustment, consider clocks past year 2010 as possible, and do not display the PRAM clock if it consistent.
2006-01-09Introduce a wsdisplay type for mac68k frame buffers; we might need to becomeMiod Vallat
less generic in the future, though.
2006-01-09Disable an(4) on BE archs until it works.Jonathan Gray
2006-01-09Remove clock tick adjustment computation inherited from luna68k where someMiod Vallat
models run with hz=60.
2006-01-09Remove completely obsolete comment from the pre-wscons sparc days inMiod Vallat
zs_getc().
2006-01-09Ditch the grf frame buffer abstraction, and directly attach macfb toMiod Vallat
either obio or nubus. No functional change, shaves a few more KB...
2006-01-09Remove duplicate defines with if_ether.hMiod Vallat
2006-01-09amdpm also exists as a 2nd iic bus on amd8111 machinesTheo de Raadt
2006-01-09Restore console bell functionnality lost in the wscons merge.Miod Vallat
2006-01-09Use protected version of register_t in struct sigcontext for theTodd C. Miller
case where _XOPEN_SOURCE was defined but BSD-specific types are not visible.
2006-01-09Enable hotplug(4); Tested by Michael Knudsen and me;Robert Nagy
ok brad@ deraadt@
2006-01-09skip the darn cereal devices, which i think are just markers for the host ↵Theo de Raadt
chip i think
2006-01-08Remove unused grf_softc members; do not keep temporary data in grf_softcMiod Vallat
either, instead keep it on the stack or directly in attachment structures. No functional change.
2006-01-08Don't provide bass and treble mixer controls if the underlying hardware doesn'tMark Kettenis
provide these. Put them in the outputs class. ok joris@
2006-01-08Drop native MacOS support from the adb direct code, and skip callbackMiod Vallat
wrappers used to convert between MacOS pascal calling convention and the C calling convention, unless option MRG_ADB. No functional change.
2006-01-08Display unhandled adb devices as "whatever at adb0 addr X" like all buses do;Miod Vallat
while there, make the detailed descriptions dependent on ADBVERBOSE (enabled in GENERIC) instead of DIAGNOSTIC.