summaryrefslogtreecommitdiff
path: root/sys/arch/i386/include
AgeCommit message (Collapse)Author
2001-06-05License clarification from David Mazieres, ok deraadt@Peter Valchev
2001-05-08From netbsd: implementations of _bus_dmamap_load_mbuf() andJason Wright
_bus_dmamap_load_uio().
2001-05-06Update some comments wrt. the CLSIZE changes.Artur Grabowski
And remove that memory price comment from 1981. It is amusing, but also confusing because the math in there is only correct on vax.
2001-05-05PMAP_NEW and UVM are no longer optional on i386.Artur Grabowski
2001-05-05Rename configure() to cpu_configure().Artur Grabowski
Move it from cpu_startup() to main().
2001-05-05Get rid of CLSIZE and all related stuff.Artur Grabowski
CLSIZE -> 1 CLBYTES -> PAGE_SIZE OLOFSET -> PAGE_MASK etc. At the same time some archs needed some cleaning in vmparam.h so that goes in at the same time.
2001-04-30Move proc0's TSS and LDT initialization to a separate functionArtur Grabowski
that's called at the end of configure instead of cpu_startup. Inspired by NetBSD.
2001-04-30Move gdt_init prototype to gdt.hArtur Grabowski
2001-04-07Additional console ioctl definitionsThorsten Lockert
2001-04-07OK, PMAP_NEW by default once again. This time it should work better forNiklas Hallqvist
userland since art removed the vm_pmap field that made UVM and old VM differ. An option PMAP_OLD is provided should someone need to configure with the old pmap. This is suspected to go away soon so I will not document it.
2001-04-03If you cannot test your work, please do not do commits and then break userlandTheo de Raadt
2001-04-02PMAP_NEW is now default behaviour, get old pmap with option PMAP_OLD.Niklas Hallqvist
However PMAP_OLD is just there for emergencies, and will go away in the future
2001-03-28Replaced with bt8xx.h; byebye.Federico G. Schwindt
2001-03-24Remove an #ifdef PMAP_NEW so userland (specifically ps) doesn'tTodd C. Miller
have to define PMAP_NEW all over the place. This means there is an unused entry in struct pcb for pmap.old but that's going away (and it doesn't hurt anything).
2001-03-22Merge in NetBSD's PMAP_NEW, still disabledNiklas Hallqvist
2001-02-28scan the isa hole for `optional rom's.Michael Shalayeff
found proms are excluded from the iomem_ex such that devices cannot map on the same memory. next step would be for pcic and similar devices to choose memory windows from unused spots in the iomem_ex. currently prom checksum test is not enforced due to broken stinkpad bioses, which do not pass the checksum test. testing from aaron@, fgsch@, fries@, millert@
2001-02-19Avoid losing rtc after suspend/resume on some laptops. Ok provos@Hakan Olsson
2001-02-06missed ) in if()Michael Shalayeff
2001-02-06i can be stooooopidMichael Shalayeff
2001-02-04Fix an off-by-{1,2,4} error in i386_space_copy(). Console is now working muchAaron Campbell
more nicely, but not perfect yet.
2001-02-04Minor indent botch.Aaron Campbell
2001-02-03make overlaping bus space copy operations work; solves vga's scrolling problemsMichael Shalayeff
2001-01-30Support for the wscons machine-independent console driver on the i386. ToAaron Campbell
become default in a few moments once deraadt commits new kernel config files. IMPORTANT NOTE: A few important pieces are still required to be worked out over the next few days. Users (and developers) relying on X should probably stay away from -current until further notice. This should not take long, but here is your first heads up. If you decide to upgrade your kernel anyway, be sure to update your /etc/ttys file to use vt100 instead of vt220. The rest of the fallout from this should be minor.
2001-01-26more special regs definitions; use names from pctr.h for those it usesMichael Shalayeff
2001-01-26define large page frame mask; also remove trailing spacesMichael Shalayeff
2001-01-25constify cpuid tablesMichael Shalayeff
2001-01-25detect Transmeta cpusTheo de Raadt
2001-01-24KGDB for i386 from NetBSD via niklas, with tweaks and accommodations.Hugh Graham
Tested by various, vetted by mickey & theo.
2001-01-23recognize P4; testing by techsupport@lokmail.net, the unfortunate owner of ↵Theo de Raadt
such a thing
2001-01-22Define ARCH_ELFSIZE that is the default elf size on this arch.Artur Grabowski
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).
2000-12-06use __x__ formats for __attribute__ arguments; guenther@gac.eduTheo de Raadt
2000-11-10seperate -> separate, okay aaron@Niels Provos
2000-08-17pcibios tells us everything, no bios_pciinfoMichael Shalayeff
2000-08-17pcibios0 at bios0 flags 0x0000Michael Shalayeff
this is now a real device, which looks in flags for disabling address, interrupt or bus fixup. bios0's flags are to disable bios32 and pcibios probing. since boot_config support editing flags, it's posible to adjust pcibios functionality in the generic kernel upon boot time. man is coming.
2000-08-05$OpenBSD$Niklas Hallqvist
2000-08-05$OpenBSD$Niklas Hallqvist
2000-08-05pastoNiklas Hallqvist
2000-08-02permit build with HZ user-defined; <wd@ics.nara-wu.ac.jpTheo de Raadt
2000-07-31Add defines for ULLONG_MAX, LLONG_MAX, and LLONG_MIN in terms ofTodd C. Miller
UQUAD_MAX, QUAD_MAX, and QUAD_MIN respectively.
2000-07-27fix espie@'s screwup; frenchies can't type! noticed by aaron@Michael Shalayeff
2000-07-26correct constraints for __{in,out}c(); from Hiroki Nakano <cas@trans-nt.com>Michael Shalayeff
2000-07-06Change splsoftclock() to raise the spl, instead of lowering it.Hakan Olsson
Add spllowersoftclock() for the former behaviour. Update comment. (art@ ok)
2000-07-05Add a breakpoint() inline for use with KGDB.Hugh Graham
Ok'd by mickey.
2000-06-07look into config flags for version enforcement and other flagsMichael Shalayeff
2000-06-05Argh. forgot this when committing the reaper stuff.Artur Grabowski
2000-05-24Add an extra byte to the end of struct pcb and make sure that it is set toBrandon Creighton
0xff. Intel (vol1 section 9.5.2) says that there must be a byte inside the TSS after the iomap because it always reads two bytes when checking permissions for io accesses. before this, bits 1016-1023 were ignored. This means that the entire pcb_iomap (and i386_*_ioperm) are accurate; pr#1190 fixed
2000-04-25A fix to the dreaded isadmaattach panic which hunts people playing withNiklas Hallqvist
large memory machines. This time I really hope we can continue quite a bit away over the Gig.
2000-04-08Initial check-in for support of 32-bit CardBus PC Cards; from NetBSD. On manyAaron Campbell
machines, this code needs the new PCIBIOS* options enabled in the kernel config file to work, but your mileage may vary. Included is a working 3c575 driver for 3Com 10/100 CardBus PC Card NICs (tested only with the 'C' revision). The 3c575 is the pccard version of the PCI EtherLink XL cards, and thus the xl driver has been split into /sys/dev/ic.
2000-04-07Protect MAXPHYS against redefinition- this allows you to make a configmjacob
option for increasing it to something reasonable for large I/O subsystems (you get 22MB/s on some of the DataDirect RAID arrays at 64K, about 37MB/s when you get go to 256KB transfers).
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-28 13:41:14 +0000