summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2002-09-20flush cpu state on read/write fpregs, force reload on writeMichael Shalayeff
2002-09-20gonna need cr10, aka ccr from the trapframeMichael Shalayeff
2002-09-20Support to run the 24 bit fb in 8 bit mode or cg12 in 1 bit modeFederico G. Schwindt
via flags, using 0x0001. miod@ ok.
2002-09-18Don't forget to map the prom executable. Otherwise we'll have troubleArtur Grabowski
starting X after the prom has been bumped from the iTLB. matthieu@ jason@ ok.
2002-09-18From NetBSD and in line with recent local changes:Hugh Graham
: date: 2002/09/13 03:00:18; author: chuck; state: Exp; lines: +3 -3 : increase sc_maxxfer from 63K to 64K. otherwise fsck fails for : file systems with larger block sizes.
2002-09-18Add lofn (working), and nofn (not working and commented out)Jason Wright
2002-09-17It turns out that disabling selection timeout is not a good idea, onMiod Vallat
MVME177 at least. Restore previous behaviour.
2002-09-17print out the ccr and rctrMichael Shalayeff
2002-09-176 year old typo in _KERN_DO_ELFMiod Vallat
2002-09-17Work around a vax-specific problem where mmap()ing large amount of memoryMiod Vallat
would cause a panic. To do so, instead of panic'ing in pmap_protect() if the pages which are supposed to be used for the mapping are out of the affordable pmap limits, just return, and handle this in pmap_enter, by either returning EFAULT if pmap_enter has been invoked with PMAP_CANFAIL, or panicing there if there is no escape. ok art@
2002-09-17I sense too much cut 'n paste.Artur Grabowski
Allow this to build with DEBUG.
2002-09-17handle fpu exceptions properly, might use a regress, i guessMichael Shalayeff
2002-09-17Until vgafb on macppc properly support non-console operation, doDale Rahn
not allow it to execute code which could panic the otherwise booting system.
2002-09-17Switch device register routine from last match to first match. FixesHugh Graham
problems one of my machines was having with phantom scsi luns and no way to discern which one was the boot device.
2002-09-16Limit buffer cache size to 35% kvm, not 70.Artur Grabowski
drahn@ ok
2002-09-16It's more convinient to spell convinient as ``convenient'', actually.Miod Vallat
2002-09-15Begone.Miod Vallat
2002-09-15Fix color choices to match wsemul_sun requirements.Miod Vallat
Tested by and ok matthieu@
2002-09-15check for errors on tod pdc ops. say that bad time is before 82Michael Shalayeff
2002-09-15better alias checking, verified w/ the regressMichael Shalayeff
2002-09-15be more precise on what we save on traps. flush fpu regs in pcb since they ↵Michael Shalayeff
are possibly accessed through non-coherent mappings
2002-09-15fix for an interrupt happening in the 2 insn window during a syscall, ↵Michael Shalayeff
comment added; from fredette@
2002-09-15kill dangling spacesMichael Shalayeff
2002-09-15backout prematureTheo de Raadt
2002-09-15KNFTheo de Raadt
2002-09-15KNFTheo de Raadt
2002-09-15nice printingTheo de Raadt
2002-09-15pcmcia works somewhatTheo de Raadt
2002-09-14recognize AppleKiwiTheo de Raadt
2002-09-14spacesTheo de Raadt
2002-09-13why was BOOT_CONFIG left out for so long?Theo de Raadt
2002-09-12Support serial console on RAMDISK.Dale Rahn
2002-09-12Change the PMAP_PAGEIDLEZERO api to take the struct vm_page instead of the pa.Artur Grabowski
2002-09-12Change the PMAP_{MAP,UNMAP}_POOLPAGE api to take a vm_page as argumentArtur Grabowski
and return a VM_PAGE. This is to allow sparc64 to cheaply record the VAC color for those pages.
2002-09-12reset want_resched earlier, set curproc after we are done w/ switching, ↵Michael Shalayeff
rearrange syscall savements
2002-09-12give a condition trap own treatment, per millicode; and, oh horror, swap ↵Michael Shalayeff
return of quad_t, sort of, sometimes
2002-09-11Some early callers of pmap_growkernel call it before vm_page structuresArtur Grabowski
are initialized. So we can't to PHYS_TO_VM_PAGE becuase there are no vm_pages. Reintroduce the old pmap_zero_page renamed to pmap_zero_phys that can zero pages without struct vm_page.
2002-09-11Include zstty.h, so that the NZSTTY check will be correct.Dale Rahn
2002-09-11When freeing the extent, include the page offset, not just the pageDale Rahn
base. This is important when allocating/freeing sub page size mappings.
2002-09-11correct argsize for ktrsyscallMichael Shalayeff
2002-09-11provide a convinient reset labelMichael Shalayeff
2002-09-11stack grows the other way, but the frame is the other way to that. heheMichael Shalayeff
2002-09-11update from sparc64 version:Jason Wright
-fix several stupid bugs -get encoding setup correct -add support for capture
2002-09-10copy syscall args in the syscall() based on theMichael Shalayeff
amount given in the sysent. deal w/ the words swappage phenomena due to reverse long word storage on the stack and that being reverse copied by words yielding long word being word swapped. do it only for syscall(2) and __syscall(2) since that's how those w/ long word args syscalls get used (for proper alignment). discussed and ideas from: deraadt@, drahn@ and fredette@ note: this might be a good point for a regress.
2002-09-10randomize space ids, plus we are not gonna run out now (;Michael Shalayeff
2002-09-10comment out a bit of unused code nowMichael Shalayeff
2002-09-10move to 3.2-betaTheo de Raadt
2002-09-10rework the fpu csw a bit better and move non-trap related stuff to laterMichael Shalayeff
2002-09-10Change the pmap_zero_page and pmap_copy_page API to take the struct vm_page *Artur Grabowski
instead of the pa. Most callers already had it handy and those who didn't only called it for managed pages and were outside time-critical code. This will allow us to make those functions clean and fast on sparc and sparc64 letting us to avoid unnecessary cache flushes. deraadt@ miod@ drahn@ ok.
2002-09-10Remove some debugging code, KNF, and fix encoding selection.Jason Wright
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-26 11:46:35 +0000