summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2001-12-06move rx handling over to bus_dma... that just leaves one vtophys() in tx.Jason Wright
2001-12-06update the radio chipset supportMichael Shalayeff
2001-12-06Keep track of how many pages a vnode hold with vhold and vholdreleArtur Grabowski
so that we can get back the old behavior where a vnode with cached data is less likely to be recycled than a vnode without cached data. XXX - This is a brute-force solution - we do it where uvmexp.vnodepages are changed, I am not really sure it is correct but people have been very happy with the diff so far and want this in the tree.
2001-12-06clear all inrterrupts separately and before enabling any.Michael Shalayeff
this prevents stray interrupts and sometimes hangs, caused by looping on stray unmapped interrupt.
2001-12-06move the setup buffer and pad into a bus_dma alloced areaJason Wright
(current score: 2 vtophys calls to go...)
2001-12-06Checkpoint bus_dma work in progress:Jason Wright
- descriptor lists are now bus_dma allocated and manipulated (for those keeping score: 6 vtophys dead, 4 to go)
2001-12-06be more careful with spaces and tabs, guysTheo de Raadt
2001-12-06unbreak alpha kernel.Eric Jackson
2001-12-06remove kame IPSEC code within #ifdef.Jun-ichiro itojun Hagino
2001-12-06put __attribute__ to header decls to avoid align strangeness. sync with kameJun-ichiro itojun Hagino
2001-12-06kill register decls. sync better with kame.Jun-ichiro itojun Hagino
2001-12-06remove obsolete comment; we don't support IPv4 mapped addrJun-ichiro itojun Hagino
2001-12-06remove #if 0'ed portion (for KAME IPSEC - not needed)Jun-ichiro itojun Hagino
2001-12-06Fix typos, from NetBSDMiod Vallat
2001-12-06MAPPEDCOPY no more.Miod Vallat
2001-12-06Reality updated: no nedd to speak about "old" machines and "current" machinesMiod Vallat
as the hp300 series are extinct, remove more dead options, clarify other, etc.
2001-12-05Enable rootdev auto-configuration by device drivers during boot.Thierry Deval
Add support for "raid" devices (prepare future raidframe changes). OK deraadt@, Reviewed by drahn@, jason@, mickey@ and miod@
2001-12-05remove some ancient crud from the "modern machine" floppyTheo de Raadt
2001-12-05match on Symbol's Spectrum24 pcmcia/prism2, form netbsdMichael Shalayeff
2001-12-05add simpletech spectrum24, from netbsdMichael Shalayeff
2001-12-05I don't know who is to blame, heko@ for bOrking this, or myself to have notMiod Vallat
noticed for so long...
2001-12-05syncChristian Weisgerber
2001-12-05Add AdLib MSC 16 PnP.Christian Weisgerber
From Georg Schwarz <georg.schwarz@iname.com>; ok deraadt@
2001-12-05change KMEMPAGES back to 64M, 128M does not work on large memory machines.Dale Rahn
ok art@
2001-12-05If we assume (just pure speculation) that there will be a pmap sometime inArtur Grabowski
the future that wants to allocate pv entries for every pmap_enter and wants to allocate those pv entries from kmem_map, it might be a good idea to init the kmem_map before initializing the kmemusage struct (because kmemusage allocates memory).
2001-12-05Put back to the old KMEM size, until we can increase this and have it workDale Rahn
on large memory machines.
2001-12-05calculate and use nkmempages, not NKMEMCLUSTERS.Artur Grabowski
2001-12-05Remove NKMEMCLUSTERS.Artur Grabowski
2001-12-05VM_KMEM_SIZE no longer needed.Artur Grabowski
2001-12-05Change pcache_flush_line to pcache_flush_page since this is the only wayArtur Grabowski
we use the function. At the same time fix the pte pool backend functions to allocate and map the memory themselves.
2001-12-05kill ugly spacesMichael Shalayeff
2001-12-05unsmoke radiumMichael Shalayeff
2001-12-05add radiotack defsMichael Shalayeff
2001-12-05update for radio drivers from tm@oganer.net und jumbo@narod.ruMichael Shalayeff
2001-12-05KNF whackTheo de Raadt
2001-12-05choose one AC97 name and stick with it a bit more oftenTheo de Raadt
2001-12-05Same stuff as cgthree:Jason Wright
- fix initialization so that the cursor is initialized to be in the same place the prom left it. - remove some trial and error #if 0 stuff - fix copyo's in a2int() - ditch *_romcursoraddr for a direct call to romgetcursoraddr() - allow this to attach (without panic()) if it is not the console.
2001-12-05- fix initialization so that the cursor is initialized to be in the sameJason Wright
place the prom left it. - remove some trial and error #if 0 stuff - fix copyo's in a2int() - ditch *_romcursoraddr for a direct call to romgetcursoraddr() - allow this to attach (without panic()) if is is not the console.
2001-12-05Enable MSCP disk support, since although untested due to lack ofHugh Graham
hardware, it's possible this is working...
2001-12-05sync with netbsd. untested.Hugh Graham
2001-12-05nuke rominterpret and update romgetcursoraddr() to work on sparc64; based on ↵Jason Wright
NetBSD work.
2001-12-05Make OF_interpret() work correctly; based on NetBSD fix, but maintainingJason Wright
old OF_interpret(cmd, nreturns, ...) prototype (why/where/when would you need nargs anyway?)
2001-12-05make nkmempages dynamic based on memory. okay art@ from netbsd:Niels Provos
date: 2000/02/11 19:22:52; author: thorpej; Add some very simple code to auto-size the kmem_map. We take the amount of physical memory, divide it by 4, and then allow machine dependent code to place upper and lower bounds on the size. Export the computed value to userspace via the new "vm.nkmempages" sysctl. NKMEMCLUSTERS is now deprecated and will generate an error if you attempt to use it. The new option, should you choose to use it, is called NKMEMPAGES, and two new options NKMEMPAGES_MIN and NKMEMPAGES_MAX allow the user to configure the bounds in the kernel config file.
2001-12-05Make DEBUG not default, and make the default compileNiklas Hallqvist
2001-12-05Oops forgot to regen.Artur Grabowski
2001-12-05Break out the part that lowers v_holdcnt in brelvp into an own functionArtur Grabowski
and make it and vhold into public interfaces.
2001-12-05Update pmap_update macro for arches Art missed. Still just a noop.Todd C. Miller
2001-12-04Yet another sync to NetBSD uvm.Artur Grabowski
Today we add a pmap argument to pmap_update() and allocate map entries for kernel_map from kmem_map instead of using the static entries. This should get rid of MAX_KMAPENT panics. Also some uvm_loan problems are fixed.
2001-12-04Readd VOP_MMAP, will be used soon. Right now it's just a question toArtur Grabowski
the filesystem if we're allowed to mmap the file.
2001-12-04Add declaration of "state" to _simple_lock_assert(). Since it isTodd C. Miller
an int this can safely be omitted but it is bad style to do so. Closes PR 2223.
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-01 17:55:55 +0000