CVS log for src/sys/arch/vax/vax/pmap.c

Up to [cvs.NetBSD.org] / src / sys / arch / vax / vax

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

Pull up following revision(s) (requested by mrg in ticket #516):

	sys/arch/vax/vax/pmap.c: revision 1.199

fix locking botch: use mutex_spin_exit(), not _exit(), for PMAP_UNLOCK.
fixes LOCKDEBUG issue in first attempt to unlock it.

also call PMAP_UNLOCK in one non-fatal out of memory case.

Pull up following revision(s) (requested by thorpej in ticket #515):

	sys/arch/vax/vax/pmap.c: revision 1.200

rmspace(): when a PT page is put back onto the free list, invalidate
it's VA in the TB.

rmptep(): Add a comment that the no TIBS is needed for newly-freed PT
pages here because all of the callers perform a TBIA.

pmap_is_modified(): If we fall into the slow path and encounter a modified
page during PV list walk, set PG_M in the pvh_attrs to avoid falling into
the slow path again.

rmspace(): when a PT page is put back onto the free list, invalidate
it's VA in the TB.

rmptep(): Add a comment that the no TIBS is needed for newly-freed PT
pages here because all of the callers perform a TBIA.

fix locking botch: use mutex_spin_exit(), not _exit(), for PMAP_UNLOCK.

fixes LOCKDEBUG issue in first attempt to unlock it.

also call PMAP_UNLOCK in one non-fatal out of memory case.

XXX: pullup-10

vax/pmap.c: Sort includes.  Nix traliing whitespace.

No functional change intended.

This now needs <sys/syncobj.h> to compile.

Pull up following revision(s) (requested by abs in ticket #1829):

	sys/arch/vax/vax/pmap.c: revision 1.196
	sys/arch/vax/include/trap.h: revision 1.25

Change CASMAGIC to 0xFEDABABE so that it cannot accidentally end up in
valid kernel memory.  Due to the VARM accesses above S0 should always
give a ptelen trap.

Bug found by Kalvis Duckmanton.


Ensure that the kernel do not try to allocate a S0 segment larger than 1G,
since the hardware prohibits that.

Pull up following revision(s) (requested by abs in ticket #1648):

	sys/arch/vax/vax/pmap.c: revision 1.196
	sys/arch/vax/include/trap.h: revision 1.25

Change CASMAGIC to 0xFEDABABE so that it cannot accidentally end up in
valid kernel memory.  Due to the VARM accesses above S0 should always
give a ptelen trap.

Bug found by Kalvis Duckmanton.


Ensure that the kernel do not try to allocate a S0 segment larger than 1G,
since the hardware prohibits that.

Pull up following revision(s) (requested by abs in ticket #201):

	sys/arch/vax/vax/pmap.c: revision 1.196
	sys/arch/vax/include/trap.h: revision 1.25

Change CASMAGIC to 0xFEDABABE so that it cannot accidentally end up in
valid kernel memory.  Due to the VARM accesses above S0 should always
give a ptelen trap.

Bug found by Kalvis Duckmanton.


Ensure that the kernel do not try to allocate a S0 segment larger than 1G,
since the hardware prohibits that.

Ensure that the kernel do not try to allocate a S0 segment larger than 1G,
since the hardware prohibits that.

Support save/restore of AST levels in the PCB for context switching.

Code written by ragge@ , tested by oster@.

vax: Remove a vestige from before the device_t/softc split.

Verified MicroVAX 4000/96 still boots.

cpu0 at mainbus0: KA49, NVAX, 10KB L1 cache, 256KB L2 cache

fix various typos, mainly in comments.

fix various typos in comments and log messages.

Move proc_lock into the data segment.  It was dynamically allocated because
at the time we had mutex_obj_alloc() but not __cacheline_aligned.

Mostly merge changes from HEAD upto 20200411

Merge changes from current as of 20200406

PR port-vax/55094: vax pmap needs locking adjustments

Make the adjustments noted in the PR and don't call uvm_wait() or do
WAITOK ever - UVM takes care of that.

Sync with head.

- Move the LW_RUNNING flag back into l_pflag: updating l_flag without lock
  in softint_dispatch() is risky.  May help with the "softint screwup"
  panic.

- Correct the memory barriers around zombies switching into oblivion.

Sync with head.

Hopefully fix some problems seen with MP support on non-x86, in particular
where curcpu() is defined as curlwp->l_cpu:

- mi_switch(): undo the ~2007ish optimisation to unlock curlwp before
  calling cpu_switchto().  It's not safe to let other actors mess with the
  LWP (in particular l->l_cpu) while it's still context switching.  This
  removes l->l_ctxswtch.

- Move the LP_RUNNING flag into l->l_flag and rename to LW_RUNNING since
  it's now covered by the LWP's lock.

- Ditch lwp_exit_switchaway() and just call mi_switch() instead.  Everything
  is in cache anyway so it wasn't buying much by trying to avoid saving old
  state.  This means cpu_switchto() will never be called with prevlwp ==
  NULL.

- Remove some KERNEL_LOCK handling which hasn't been needed for years.

in many device attach paths, allocate memory with M_WAITOK instead of M_NOWAIT
and remove code to handle failures that can no longer happen.

Sync with HEAD.  77 conflicts resolved - all of them $NetBSD$

Allow allocation of PTE space to fail in pmap_enter if PMAP_CANFAIL is set.
This fixes the panic part of PR port-vax/28379.

update from HEAD

Sync with HEAD

Update copyright notice for Ludd (remove clause 3 & 4).

Sync with HEAD

Sync with HEAD.  (Note that most of these changes are simply $NetBSD$
tag issues.)

"Make NetBSD great again!"

Introduce uvm_hotplug(9) to the kernel.

Many thanks, in no particular order to:

TNF, for funding the project.

Chuck Silvers - for multiple API reviews and feedback.
Nick Hudson - for testing on multiple architectures and bugfix patches.
Everyone who helped with boot testing.

KeK (http://www.kek.org.in) for hosting the primary developers.

switch all ports to use uvm_init.c:uvm_md_init()

uvm_setpagesize() is now subsumed within this funciton.

sync with head

merge to -current.

- bringing kmeminit_nkmempages back and revert pmaps that called this early
- use nkmempages to scale the kmem_arena
- reducing diff to pre kmem/vmem change
   (NKMEMPAGES_MAX_DEFAULT will need adjusting on some archs)

Fix the initial sizing of the system page table / address space.

Sync with HEAD.

sync with head

Remove pmap's simple_lock; switch to atomic ops
Change pvtable simple_lock to mutex
Switch to kmem (goodbye malloc).

sync with head

Cleanups includes.  Change <machine/XXX.h> to <sys/XXX.h>
Remove unneeded includes.

Use more VM_PHYSMEM_*() accessors.  No functional changes.

Machine dependent code is considered as part of UVM.  Include
internal API header.

Implement pmap_physload_device(9) to replace xmd(4) MD backend.
Implement pmap_mmap(9) and use it from mem(4) and xmd(4).

Sync with HEAD.

sync with head.

sync with head

A big bunch of bugfixes from Johnny Billquist. Highlights:
- Makes the VAX8600 work as expected (500kg of hardware :-)
- Fix a hard-tracked bug causing VAXen to hang at splhigh.

Fix build.

Always use struct vm_physseg *vm_physmem_ptrs[] in MD code.

sync with head

- Use uvm_lwp_setuarea() instead of directly setting address to lwp_t::l_addr.
- Replace most remaining uses of l_addr with uvm_lwp_getuarea() or lwp_getpcb().
- Amend assembly in ports where it accesses PCB via struct user.
- Rename L_ADDR to L_PCB in few places.  Reduce sys/user.h inclusions.

Kill proc0paddr.  Use lwp0.l_addr instead.

- Use lwp_getpcb() on VAX, clean from struct user usage.
- Include sys/user.h in MD proc.h .

Add a flags argument to pmap_kenter_pa(9).
Patch showed on tech-kern@ http://mail-index.netbsd.org/tech-kern/2009/11/04/msg006434.html
No objections.

kill extra whitespaces
reviewed by tsutsui@

Correct the previous by adding an obviously overlooked
"struct pmap *pm" argument declaration to pmap_vax_swappable().

Remove uarea swap-out functionality:

- Addresses the issue described in PR/38828.
- Some simplification in threading and sleepq subsystems.
- Eliminates pmap_collect() and, as a side note, allows pmap optimisations.
- Eliminates XS_CTL_DATA_ONSTACK in scsipi code.
- Avoids few scans on LWP list and thus potentially long holds of proc_lock.
- Cuts ~1.5k lines of code.  Reduces amd64 kernel size by ~4k.
- Removes __SWAP_BROKEN cases.

Tested on x86, mips, acorn32 (thanks <mpumford>) and partly tested on
acorn26 (thanks to <bjh21>).

Discussed on <tech-kern>, reviewed by <ad>.

Sync with HEAD.

Commit is split, to avoid a "too many arguments" protocol error.

sync with head.

Sync with HEAD.

change pmap flags argument from int to u_int.
discussed with christos@ on source-changes-d@

bcopy -> memcpy

bzero -> memset

Sync with HEAD.

Use "defopt MODULAR" in sys/conf/files, and #include "opt_modular.h"
in all kernel sources that use the MODULAR option.
Proposed in tech-kern on 18 Jan 2009.

Sync with HEAD.

Sync with HEAD.

kill MALLOC and FREE macros.

Update haad-dm branch to haad-dm-base2.

Make kernel_pmap_ptr a const.  Requested by steve_martin.

Make pmap_kernel() a MI macro for struct pmap *kernel_pmap_ptr,
which is now the "API" provided by the pmap module.  pmap_kernel()
remains as the syntactic sugar.

Bonus cosmetics round: move all the pmap_t pointer typedefs into
uvm_pmap.h.

Thanks to Greg Oster for providing cpu muscle for doing test builds.

Remove LKMs and switch to the module framework, pass 1.

Proposed on tech-kern@.

Sync with HEAD.

sync with head.

Network protocol interrupts can now block on locks, so merge the globals
proclist_mutex and proclist_lock into a single adaptive mutex (proc_lock).
Implications:

- Inspecting process state requires thread context, so signals can no longer
  be sent from a hardware interrupt handler. Signal activity must be
  deferred to a soft interrupt or kthread.

- As the proc state locking is simplified, it's now safe to take exit()
  and wait() out from under kernel_lock.

- The system spends less time at IPL_SCHED, and there is less lock activity.

Sync with HEAD.

sync with head.

sync with HEAD

sync with head.

After a process in swapped in, refresh the pcb's paddr.

Rototill the vax code.  Switch to devicet/PRIV_ALLOC.  Cleanup vax autoconf
code.  Move to prototype definitions.  staticfy, constify, avoid casting.
Use device_* accessors.

sync with head.

A number of performance changes:
Change internal register SSP to contain curlwp insread curcpu().  This
shrinks the GENERIC by 2KB.
Add rwlock stubs (rw_enter, rw_exit, rw_tryenter).
Simplify simple lock code for uniprocessors.
Do SPL raise/lower in cpu_handle_ipi instead of simple_locks.

Fix a bug in the pmap pcb tracking code.  While here, rewrite it to simplier.
Use a field in the pcb itself (since it's basically free) and keep track of
what pmap "owns" a pcb (for consistency checking).  use M_ZERO as appropriate.

Make NetBSD/vax work again.  Add support for __HAVE_FAST_SOFTINTS.
Add a few small optimizations.

ci_curlwp must be set to &lwp0 very early in the boot.  Now we get further
into the kernel before dying.

Sync with HEAD.

sync with head.

misc type cleanups

Sync with HEAD.

Sync with HEAD

sync with HEAD

Sync with HEAD.

sync with head.

defparam PAGER_MAP_SIZE.

Merge the ppcoea-renovation branch to HEAD.

This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree.  Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches.  The work
for this branch was done by a variety of people, too long to list here.

TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.

NOTES:
pmppc was removed as an arch, and moved to a evbppc target.

sync with head.

Sync with head.

Sync with head.

Update to HEAD.

merge yamt-idlelwp branch.  asked by core@.  some ports still needs work.

from doc/BRANCHES:

	idle lwp, and some changes depending on it.

	1. separate context switching and thread scheduling.
	   (cf. gmcgarry_ctxsw)
	2. implement idle lwp.
	3. clean up related MD/MI interfaces.
	4. make scheduler(s) modular.

sync with head.

Pull up following revision(s) (requested by mhitch in ticket #569):
	sys/arch/vax/vax/pmap.c: revision 1.149
Workaround a gcc 4.1 issue.  don't use s.m = s2.m = s3.m = ...
assign them individually.

Workaround a gcc 4.1 issue.  don't use s.m = s2.m = s3.m = ...
assign them individually.

Remove ci_exit, remove idlepcb and exitstack.

Sync with head.

Sync with HEAD.

Shrink VAX kmutex from 12 bytes to 8.  Fix various LOCKDEBUG/DIAGNOSTIC
problems.

- Make the proclist_lock a mutex. The write:read ratio is unfavourable,
  and mutexes are cheaper use than RW locks.
- LOCK_ASSERT -> KASSERT in some places.
- Hold proclist_lock/kernel_lock longer in a couple of places.

Fix caddr_t fallout.

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

- sync with head.
- move sched_changepri back to kern_synch.c as it doesn't know PPQ anymore.

sync with head.

TRUE -> true, FALSE -> false

Replace the Mach-derived boolean_t type with the C99 bool type.  A
future commit will replace use of TRUE and FALSE with true and false.

Change the process/lwp flags seen by userland via sysctl back to the
P_*/L_* naming convention, and rename the in-kernel flags to avoid
conflict. (P_ -> PK_, L_ -> LW_ ). Add back the (now unused) LSDEAD
constant.

Restores source compatibility with pre-newlock2 tools like ps or top.

Reviewed by Andrew Doran.

file pmap.c was added on branch yamt-idlelwp on 2007-02-27 16:53:22 +0000

Add an implementation for krwlock_t.  Initialize ci_mtx_count to 1 (biased).
Fix a few buglets in the kmutex_t implementation.

Partially adapt the VAX port to the newlock2 changes.  These are untested
but they do at least compile.

sync with head.

Sync with head.

sync with head

remove details of the kernel malloc() implementation from header files:
 - change MALLOC() and FREE() to just call their function equivalents.
 - remove references to other malloc()-related constants.

sync with head

sync with head

Merge from HEAD.

Don't bcopy to 0 because gcc4 doesn't like.  Instead use a structure copy
to 0 to do the equivalent thing.  XXX this is evil.

sync with head.

Sync with head.

sync with head - hopefully this will work

sync with head.

Fix typo in lhs cast removal.

sync with head.

Fix various places where assignment happen to casted lvalues.
(t)v = foo; is not legal C.

merge ktrace-lwp.

Fix the sync with head I botched.

Sync with HEAD.

Sync with HEAD

Add CPU_INFO_INTERATOR/FOREACH support.

Uppercase CPU, plural is CPUs.

Replace the traditional buffer memory management -- based on fixed per buffer
virtual memory reservation and a private pool of memory pages -- by a scheme
based on memory pools.

This allows better utilization of memory because buffers can now be allocated
with a granularity finer than the system's native page size (useful for
filesystems with e.g. 1k or 2k fragment sizes).  It also avoids fragmentation
of virtual to physical memory mappings (due to the former fixed virtual
address reservation) resulting in better utilization of MMU resources on some
platforms.  Finally, the scheme is more flexible by allowing run-time decisions
on the amount of memory to be used for buffers.

On the other hand, the effectiveness of the LRU queue for buffer recycling
may be somewhat reduced compared to the traditional method since, due to the
nature of the pool based memory allocation, the actual least recently used
buffer may release its memory to a pool different from the one needed by a
newly allocated buffer. However, this effect will kick in only if the
system is under memory pressure.

TOPDOWN_VM is now mandatory on vax.

Add a default: branch with a panic() to the switch statement to
handle "impossible" segment types, so that the compiler can know
that 'pt' has been initialized after the switch statement.

Avoid the pvtable when devices are mapped. Fixes panics when unmapping
framebuffers.  Set aside space in kernel map for framebuffer on VS4k.

__KERNEL_RCSID()

Back out the following chagne:
    http://mail-index.netbsd.org/source-changes/2003/05/08/0068.html

There were some side-effects that I didn't anticipate, and fixing them
is proving to be more difficult than I thought, do just eject for now.
Maybe one day we can look at this again.

Fixes PR kern/21517.

Simplify the way the bounds of the managed kernel virtual address
space is advertised to UVM by making virtual_avail and virtual_end
first-class exported variables by UVM.  Machine-dependent code is
responsible for initializing them before main() is called.  Anything
that steals KVA must adjust these variables accordingly.

This reduces the number of instances of this info from 3 to 1, and
simplifies the pmap(9) interface by removing the pmap_virtual_space()
function call, and removing two arguments from pmap_steal_memory().

This also eliminates some kludges such as having to burn kernel_map
entries on space used by the kernel and stolen KVA.

This also eliminates use of VM_{MIN,MAX}_KERNEL_ADDRESS from MI code,
this giving MD code greater flexibility over the bounds of the managed
kernel virtual address space if a given port's specific platforms can
vary in this regard (this is especially true of the evb* ports).

Use PAGE_SIZE rather than NBPG.
XXX Except in one place, which will require more work.

Change the pmap to use 512-byte pages as user page table pages instead of
PAGE_SIZE pages (4k). An average small program uses ~6 4k pages (24k),
while the same program only uses ~20 512 byte pages (10k), so it will be
a small memory usage improvement. The large improvement will be the upcoming
ability to share page table pages between processes for shared libraries.

Remaining: should be able to give back ptp pages to the system.

Enable USE_TOPDOWN_VM. This also makes it possible to use large address
spaces, so bump MAXDSIZ to 1GB.

D'oh!  Remove unused local variable too.

Remove references to nswbuf.  It is entirely unused.

Use M_VMPMAP instead of M_HTABLE.

Merge the nathanw_sa branch.

Make this compile again.

Sync with HEAD.

Fix inconsistencies in various definitions of avail_*, virtual_*, and
proc0paddr.

LWPify pmap_rmproc().  The comment says it is based on uvm_swapout_threads().
Brought the same LWP changes across from there.

sync kqueue branch with HEAD

catch up with -current.

Catch up to -current.

Make nbuf, nswbuf, and bufpages unsigned.  Make all operations on these
variables unsigned, and update places where their values are printed.

Catch up to -current.

Pull up revision 1.116 (requested by ragge in ticket #542):
Increase the interrupt stack, to avoid problems reported on port-vax by
Olaf Seibert. (IPSec takes much stack space).

Increase the interrupt stack, to avoid problems reported on port-vax by
Olaf Seibert. (IPSec takes much stack space).

catch up with -current.

Curproc->curlwp renaming.

Change uses of "curproc->l_proc" back to "curproc", which is more like the
original use. Bare uses of "curproc" are now "curlwp".

"curproc" is now #defined in proc.h as ((curlwp) ? (curlwp)->l_proc) : NULL)
so that it is always safe to reference curproc (*de*referencing curproc
is another story, but that's always been true).

catch up with -current on kqueue branch

Catch up to -current.

Pull up revision 1.115 (requested by ragge in ticket #229):
Must multiply the pte count with its size when copying ptes when the P1
region is expanded. Fixes bug reported by Matt Thomas on port-vax.

Must multiply the pte count with its size when copying ptes when the P1
region is expanded. Fixes bug reported by Matt Thomas on port-vax.

pullup-1-6 ticket #192:

syssrc/sys/arch/vax/vax/pmap.c 1.114

Original log message:

Pipes and swap disks takes a not insignificant amount of KVA, so therefore
it must be used in KVA calculation.

Pipes and swap disks takes a not insignificant amount of KVA, so therefore
it must be used in KVA calculation.

Catch up to -current.

Fix crash problems reported by mrg and bouyer in separate mails.

Check that each region is inuse before freeing their related extent.
This fixes (hopefully) PR#16164.

Catch up to -current.
(CVS: It's not just a program. It's an adventure!)

Initial nathanw_sa kernel support for vax.

Inline functions that mostly do not need to be called.  Partly
rearrange/change the code to be able to avoid some inefficient functions.
Profiling shows that the time spent in pmap now is decreased by 20%. (!)

Catch up with -current.

Major update of the vax pmap:
	- Reinstall the "dynamic page table length" that was removed some
	  years ago.
	- Limit the user page table submap to max 5% of available memory.
	- Free the page table space when a process is swapped out.
	- If the UPT submap runs out of space, throw away pmap mappings
	  using the same algorithm as for swapping processes.

As a result of this, 4MB machines are useable again and it's even possible
to compile a kernel for 2MB machines (but it will be slow... :-)

Still to do:
	- Multiprocessor fixes.
	- More profiling.

usrptsize was not correct calculated, fixed.
Add space to system page table for the UVM kernel area.
This fixes the KVM usage problem that Manuel Bouyer reported a while ago.

Sync kqueue branch with -current.

Apply patch (requested by ragge):
  Handle remapping of pages entered by pmap_kenter_pa() properly.
  Fixes PR#12231.

Catch up with -current.

file pmap.c was added on branch nathanw_sa on 2002-03-29 23:22:44 +0000

Check if page to be remapped in pmap_enter() is entered by pmap_kenter_pa(),
to decide whether to remove the previous mapping from pv_table or not.
Should fix PR#12231.

implement pmap_wired_count().

Update the kqueue branch to HEAD.

Update pmap_update to now take the updated pmap as an argument.
This will allow improvements to the pmaps so that they can more easily defer expensive operations, eg tlb/cache flush, til the last possible moment.

Currently this is a no-op on most platforms, so they should see no difference.

Reviewed by Jason.

Merge Aug 24 -current into the kqueue branch.

Don't substract KERNBASE, xor it to clear.   The former causes GNU ld to
bitch about a signed overflow and this code is only done once, incurring
an extra instruction isn't going to matter much in the long run.

Do more intelligent calculation of KVM size. This reduces the amount of
pre-allocated physical memory significantly.
A 11/750, for example, get 2MB more free (of 14MB), about 12%.

Catch up to -current.

implement pmap_clear_modify() correctly.

make this compile without DEBUG.

The beginning of pmap locks. While here, some cleaning and KNF.

Sprinkle pmap_update() calls after calls to:
- pmap_enter()
- pmap_remove()
- pmap_protect()
- pmap_kenter_pa()
- pmap_kremove()
as described in pmap(9).

These calls are relatively conservative.  It may be possible to
optimize these a little more.

Garbage-collect pmap_bootstrap_alloc(); nothing uses it.

Sync with HEAD.

Remove pmap_kenter_pgs().  It was never really adopted by
anything, and the interface itself wasn't as flexible as
callers would have probably liked.

Make pmap_virtual_space() a required pmap function, even on platforms
which have pmap_steal_memory().  This is to reduce the API differences
between pmaps that implement pmap_steal_memory() and pmaps which do
not.

Note that pmap_steal_memory() needs to adjust *vstartp and/or
*vendp only if it used addresses within the range provided to UVM
via the pmap_virtual_space() call.  I.e. it is not necessary to do
so in any current pmap_steal_memory() implementation.

Sync with HEAD

splimp -> splvm

Catch up with -current.

Be sure that all kernel stack pages are valid after swapin. Fixes PR#12520.

Sync with HEAD.

eliminate the KERN_* error codes in favor of the traditional E* codes.
the mapping is:

KERN_SUCCESS			0
KERN_INVALID_ADDRESS		EFAULT
KERN_PROTECTION_FAILURE		EACCES
KERN_NO_SPACE			ENOMEM
KERN_INVALID_ARGUMENT		EINVAL
KERN_FAILURE			various, mostly turn into KASSERTs
KERN_RESOURCE_SHORTAGE		ENOMEM
KERN_NOT_RECEIVER		<unused>
KERN_NO_ACCESS			<unused>
KERN_PAGES_LOCKED		<unused>

Sync with HEAD.

Check page reference in pmap_clear_reference(). Fixes vax hang problem
with ubc, tested by Chuck Silvers.

Sync with HEAD

Do not touch pv_table when mapping/unmapping I/O registers. This caused
mmap() of framebuffers to give unpredictable faults.

Sync with HEAD.

two changes to pmap_enter():
 - allow entry of the same mapping that's already there.
   this can happen with UBC.
 - if PMAP_CANFAIL, return failure rather than sleeping
   when we fail to allocate a ptp.

eliminate TRUNC_PAGE() and ROUND_PAGE() in favor of their
lowercase counterparts.  also, a little misc cleanup.

Update thorpej_scsipi to -current as of a month ago
A i386 GENERIC kernel compiles without the siop, ahc and bha drivers
(will be updated later). i386 IDE/ATAPI and ncr work, as well as
sparc/esp_sbus. alpha should work as well (untested yet).
siop, ahc and bha will be updated once I've updated the branch to current
-current, as well as machine-dependant code.

Pullup 1.89 by patch [ragge]:
If a process is locked into memory with mlockall() (like ntpd is)
pmap_enter() will get called trying to wire an already wired page again
if sysctl() is called. To avoid a panic the "wired" check is moved after
the double-mapping check.

If a process is locked into memory with mlockall() (like ntpd is)
pmap_enter() will get called trying to wire an already wired page again
if sysctl() is called. To avoid a panic the "wired" check is moved after
the double-mapping check.

Make the pmap: enter on myself panic conditional under DIAGNOSTIC

Pull up revision 1.87 (requested by ragge):

Wiring of a page were done at the wrong place, so pages in kernel space
never got the wired bit set. This caused panics if a swapped out process
was swapped in again and the kernel stack had not yet been unmapped.
While here, add a forgotten lock initializer.

Wiring of a page were done at the wrong place, so pages in kernel space
never got the wired bit set. This caused panics if a swapped out process
was swapped in again and the kernel stack had not yet been unmapped.
While here, add a forgotten lock initializer.

Pull up revisions (requested by ragge):
pmap.c 1.85-1.86
pmap.h 1.40

Remove some unneccessary TBIA's. Keep track of wired pages.
Fix PR#8503 (refcount error).

Remove some unneccessary TBIA's. Keep track of wired pages.
Fix PR#8503 (refcount error).

Do mtpr(0, PR_TBIA), not mtpr(1, PR_TBIA). This has worked on all CPUs
except for KA88.

remove include of <vm/vm.h>. <vm/vm.h> -> <uvm/uvm_extern.h>

remove/move more mach vm header files:

	<vm/pglist.h> -> <uvm/uvm_pglist.h>
	<vm/vm_inherit.h> -> <uvm/uvm_inherit.h>
	<vm/vm_kern.h> -> into <uvm/uvm_extern.h>
	<vm/vm_object.h> -> nothing
	<vm/vm_pager.h> -> into <uvm/uvm_pager.h>

also includes a bunch of <vm/vm_page.h> include removals (due to redudancy
with <vm/vm.h>), and a scattering of other similar headers.

Sync w/ netbsd-1-5-base.

Allow printf's from slave CPUs by stealing the v_putc function.
Easier than expected because the printout lock is hold in the higher levels.

Code to spin up other CPUs on a VAX 8200 system.
Haven't solved the printf() problem yet, though.

Use the cpu_info struct to store cpu-specific data in the same way in
both uniprocessor and multiprocessor environments. Use the otherwise
unused internal CPU register SSP to store the cpu_info pointer.
The macros curcpu(), curproc, cpu_number() and need_resched() are now the
same in both uniprocessor and multiprocessor environments.

Changes to use the RPB for different tasks. Much simpler now to find the
boot device.

Be a little bit more generous with TBIA.

Instead of checking vm_physmem[<physseg>].pgs to determine if
uvm_page_init() has completed, add a boolean uvm.page_init_done,
and test against that.  Use this same boolean (rather than
pmap_initialized) in pmap_growkernel() to determine if we are
being called via uvm_page_init() to grow the kernel address space.

This fixes a problem on some i386 configurations where pmap_init()
itself was needing to have the kernel page table grown, and since
pmap_initialized was not yet set to TRUE, pmap_growkernel() was
choosing the wrong code path.

Fix tested by Havard Eidnes.

Merge parts of chs-ubc2 into the trunk:
* Remove the casts to vaddr_t from the round_page() and trunc_page() macros to
  make them type-generic, which is necessary i.e. to operate on file offsets
  without truncating them.
* In due course, cast pointer arguments to these macros to an appropriate
  integral type (paddr_t, vaddr_t).

Originally done by Chuck Silvers, updated by myself.

Apply patch (requested by toddpw):
  Fix bugs in support for VaxStation 2000 and 3100, vax pmap, and
  the NCR scsi driver.

Apply patch (requested by ragge):
  Fix a problem when the system goes out of free pages.

Pull up to last week's -current.

CL* discarding.
Move physmem setting to pmap_bootstrap().
Fix security problem in process_write_regs(), found by Klaus Klein.

bring in latest KAME (as of 19991130, KAME/NetBSD141) into kame branch
just for reference purposes.
This commit includes 1.4 -> 1.4.1 sync for kame branch.

The branch does not compile at all (due to the lack of ALTQ and some other
source code).  Please do not try to modify the branch, this is just for
referenre purposes.

synchronization to latest KAME will take place on HEAD branch soon.

Sync with -current

adapt to jason new pmap_enter change

Update for pmap_enter() API change.  No functional difference.

Avoid calling cninit() too early.

eliminate the PMAP_NEW option by making it required for all ports.
ports which previously had no support for PMAP_NEW now implement
the pmap_k* interfaces as wrappers around the non-k versions.

Update from trunk.

A bunch of changes:
- Free pte pages not in use anymore.
- Inline pmap_extract().
- Fix annoying page reference/modify bug. Fixes PR#7858 & PR#7859.

pullup via patch (ragge): fix critical paging/swapping problems

Rewrite the page table entry routines. Don't take a pte invalid fault for
missing pte's, instead map in pte entries in pmap_enter(). The user ptes
is no more handled by the VM system. All this made swapping start working
on VAX again.
Still to do:
- Keep refcount per pte page, so that those pages get free'd when the
  process is swapped out. Right now they are only free'd when the pmap
  is destroyed.

Many thanks to Chuck Silvers for all help finding the deadlock problems.

Change the pmap_extract() interface to:
	boolean_t pmap_extract(pmap_t, vaddr_t, paddr_t *);
This makes it possible for the pmap to map physical address 0.

Sync w/ -current.

Use pmap_steal_memory() for early memory allocation.

Sync w/ -current.

Page tables must be pageable (where did this disappear?)
Fix a PGOFSET -> VAX_PGOFSET miss.
Stylistic cleanup.

Change the vm_map's "entries_pageable" member to a r/o flags member, which
has PAGEABLE and INTRSAFE flags.  PAGEABLE now really means "pageable",
not "allocate vm_map_entry's from non-static pool", so update all map
creations to reflect that.  INTRSAFE maps are maps that are used in
interrupt context (e.g. kmem_map, mb_map), and thus use the static
map entry pool (XXX as does kernel_map, for now).  This will eventually
change now these maps are locked, as well.

Rewrite the "emulate page reference bits" code to avoid a bunch of
pv_lists traversal and unneccessary page faults.

Only allocate QDSS memory if there is a QDSS as console.
QDSS should also work on KA650 now. (untested)

Use "access_type" in pmap_enter() more intelligent.
Allocate device register space bigger than the logical page size
from the kernel map, thus decreasing the pre-allocated page table memory.

Changes for modified pmap_enter() API:
* Map the message buffer with access_type = VM_PROT_READ|VM_PROT_WRITE `just
  because'.
* Map the file system buffers with access_type = VM_PROT_READ|VM_PROT_WRITE to
  avoid possible problems with pagemove().
* Do not use VM_PROT_EXEC with either of the above.
* Map pages for /dev/mem with access_type = prot.  Also, DO NOT use
  pmap_kenter() for this, as we DO NOT want to lose modification information.
* Map pages in dumpsys() with VM_PROT_READ.
* Map pages in m68k mappedcopyin()/mappedcopyout() and writeback() with
  access_type = prot.
* For now, bus_dma*(), pmap_map(), vmapbuf(), and similar functions still use
  access_type = 0.  This should probably be revisited.

completely remove Mach VM support.  all that is left is the all the
header files as UVM still uses (most of) these.

Another giant change: Allocate register space dynamic instead of compiled-in.
This is done on a physical page size basis, instead of virtual (as the
(on vax yet non-existing) bus_* routines does). This is similar to the
way uba allocation is done.

Fix some void pointers.

Allocate (almost) all interrupt vectors dynamically. Simplifies much
work when adding support for new machines and devices.

Giant change: NBPG now set to 4k and CLSIZE == 1 for vax. This change
made a whole bunch of annoying bugs disappear; mostly depending on
bad use of NBPG in non-MD code. The VAX port was the only port that
used this historical "feature".

The CL* macros should probably go away totally, there is no reason
at all to keep them.

A "return 1" should be "return 0". Caused machine to crash as soon as
it started paging.

- Save R/M bits after a page is unmapped.
- Keep track of mapping count (for statistics).
- vm_offset_t -> vaddr_t/paddr_t.
- Move away counting of available memory.

VAX logical page size increased to 4k.

Fix PMAP_NEW and make it the default.

Define one page free list, and put all pages on it.

Remove a forgotten debug printout.

It is no longer necessary for pmap_pinit() and pmap_release() to be
pmap interface functions, as NetBSD no longer uses statically allocated
pmaps (except for the kernel pmap, which is special-cased anyhow).

Remove the ancient kernel-stack area at top of P1 region. Make PMAP_NEW
almost work. Also Wall cleaning.

Add support for QDSS graphic console. Code originated from 4.4BSD,
ported to NetBSD by Boris Gjenero <bgjenero@undergrad.math.uwaterloo.ca>

Support for UVM on VAXen.

Re-implement page reference bit emulation by using the (otherwise unused)
valid bit. This is faster than the "unmap all" solution that were described
in that Mach paper _and_ it eliminates the need for checking the wired bit.
As a result of this; swapping started working again on vax :-)

Fix to make vfork work on machines with separated PCB/mtpr PTE registers.

Fix page size initialisation.

- Adjust so that access to the PCB is not required in most pmap functions.
  Store the page {0,1} base and length registers in the pmap structure,
  and implement pmap_activate(), which stores them in the PCB (for the
  hardware's use).
- Re-implement pmap_pinit(); allocate the PTEs here, not in cpu_fork().

Move some interrupt routines out to assembler.
Optimize (slightly) pmap_clear_modify.

Major rewriting, optimization and simplifying of the pmap code:

- Map in all physical memory first in system space. This reduces
  pmap_copy_page() and pmap_zero_page to 3 resp. 2 instructions.
- Have fized position user page tables. Makes the pv_table smaller
  and all reverse references faster (and simpler).
- Remove the wiring code. Nobody doesn't even know what a DR32 is anymore.
- Simulate page reference bit by setting page invalid, as suggested by
  Rich Draves in a paper for 1991 Mach Usenix Symposium.

This reduced the time spent in the pmap module to between 70-75% of
the previous; and made process startup _much_ faster.

Add a sometimes needed round_page().
Always flush TLB when messing around with system mapping.

Update marc-pcmcia branch from trunk.

Implement the kernel part of pr-1891. This allows for a more flexible sized
msgbuf. Note that old 'dmesg' and 'syslogd' binaries will continue running,
though old 'dmesg' binaries will output a few bytes of junk at the start of
the buffer, and will miss a few bytes at the end of the buffer.

Fixed bug causing machines with a memory size not a multiple of 16k
to crash. This is true on many MicroVAXen.

Remove __VM_PMAP_HACK.

Clean up DZ console routines. Use memory size from boot (if provided).

Clean up console autoconfiguration so that the VAXstation DZ11-like
console work.

Merge in changes from Trunk

Patches to make VS2000 to work. From Ari Suutari.

Basic change to cpu-config code. This will make support of new
cpu types much easier.

Walk a step closer to more intelligent User PTE handling.

backout previous kprintf change

printf -> kprintf, sprintf -> ksprintf

Locore changes for VS and VAX8200.

Fixed all (proto)type errors. Fixes PR 2377.

Added prototypes to everything. Made all files compile with -Wall.

Do not have debugging enabled by default.

Support for VAX 8600/8650 added. Works with lots of Unibus adapters,
and will probably work with Massbus adapters as well. (Not tested,
but it's the same code as for 11/780). Ubareset's may cause crashes
on 8600 also like 11/780, but they are more uncommon. No support
for console RL02 yet, but it's likely to come.

Support for VAX 11/780 - 11/785 system types added.
There are no support (yet) for the RX01 console floppy,
even if it is actually needed to create boot floppies.
(The arff utility is also missing, so...)

Fix things that -O2 broke. Cleanup.

Fix type errors.

Fix #includes.

CPU dependent code moved out of cpu independent modules.

Major rewriting of pmap_bootstrap; CPU dependent allocations moved
to related file.

Fixed USRPTSIZE trouble. Cleaned up.

Bug fix: Must have user readability of kernel text on Microvax
even when using DDB.

DDB needs writable kernel text.

rudimentary DDB support.

Enabled auto reboot. Fixed bug in conf.c. Started use of rpb.
Detecting kernel stack overflow. Removed local mapping of
sigtramp code. Frob kernel stack pages in cpu_swapin().
Cleaned up some code and changed to KNF.

Kernel message buffer implemented.

Removed kernel stack relocation and double-mapping.
Speeded up fork() significantly.

Bring back pmap_kernel(), for now always inlined as a pointer to
kernel_pmap_store.

v_cmap --> vmmap

Speed-up of pmap and trap.
Fixes due to uVAX support.

Support for uVAXII, new tmscp driver added.

Zillions of bugfixes. Obsolete files taken away. New style config
fixed. Inline assembler commonly used.

Lots of fixes and updates.

new RCS ID format.

Lots of bug fixes; clock and mem files new.

Lots of bug fixes. Generic console support added. Major changes in machdep.c.
pmap.c now supports user process mapping. Page faults handles correctly.

file pmap.c was added on branch netbsd-1-0 on 1994-08-16 23:47:36 +0000

Lots of bug fixes. Generic console support added. Major changes in machdep.c.
pmap.c now supports user process mapping. Page faults handles correctly.

Initial VAX port merging.