1
0
mirror of https://git.FreeBSD.org/src.git synced 2024-12-25 11:37:56 +00:00
Commit Graph

829 Commits

Author SHA1 Message Date
Peter Wemm
f8ddc1e209 Print correct function name in panics 1997-09-13 15:04:52 +00:00
Jonathan Lemon
987b847efc Do not consider VM_PROT_OVERRIDE_WRITE to be part of the protection
entry when handling a fault.  This is set by procfs whenever it wants
to write to a page, as a means of overriding `r-x COW' entries, but
causes failures in the `rwx' case.

Submitted by:	 bde
1997-09-12 15:58:47 +00:00
Bruce Evans
41fadeeb28 Removed yet more vestiges of config-time swap configuration and/or
cleaned up nearby cruft.
1997-09-07 16:21:11 +00:00
Bruce Evans
79624e2147 Removed unused #includes. 1997-09-01 03:17:34 +00:00
Bruce Evans
4de628dec4 Some staticized variables were still declared to be extern. 1997-09-01 02:55:50 +00:00
Bruce Evans
dfeca1b8ae Print a device number in hex instead of decimal. 1997-09-01 02:28:32 +00:00
Poul-Henning Kamp
a051452ae2 Change the 0xdeadb hack to a flag called VDOOMED.
Introduce VFREE which indicates that vnode is on freelist.
Rename vholdrele() to vdrop().
Create vfree() and vbusy() to add/delete vnode from freelist.
Add vfree()/vbusy() to keep (v_holdcnt != 0 || v_usecount != 0)
  vnodes off the freelist.
Generalize vhold()/v_holdcnt to mean "do not recycle".
Fix reassignbuf()s lack of use of vhold().
Use vhold() instead of checking v_cache_src list.
Remove vtouch(), the vnodes are always vget'ed soon enough
  after for it to have any measuable effect.
Add sysctl debug.freevnodes to keep track of things.
Move cache_purge() up in getnewvnodes to avoid race.
Decrement v_usecount after VOP_INACTIVE(), put a vhold() on
  it during VOP_INACTIVE()
Unmacroize vhold()/vdrop()
Print out VDOOMED and VFREE flags (XXX: should use %b)

Reviewed by:		dyson
1997-08-31 07:32:39 +00:00
Peter Wemm
54f42e4ba0 Allow non-page aligned file offset mmap's, providing that the system is
allowed to choose the address, or that the MAP_FIXED address has the same
remainder when modulo PAGE_SIZE as the file offset.  Apparently this is
posix1003.1b specified behavior.  SVR4 and the other *BSD's allow it too.
It costs us nothing to support and means we don't get EINVAL on some mmap
code that works perfectly elsewhere.

Obtained from: NetBSD
1997-08-30 18:50:06 +00:00
Bruce Evans
b9dcd593ff Fixed type mismatches for functions with args of type vm_prot_t and/or
vm_inherit_t.  These types are smaller than ints, so the prototypes
should have used the promoted type (int) to match the old-style function
definitions.  They use just vm_prot_t and/or vm_inherit_t.  This depends
on gcc features to work.  I fixed the definitions since this is easiest.
The correct fix may be to change the small types to u_int, to optimize
for time instead of space.
1997-08-25 22:15:31 +00:00
John Dyson
89721f6f1a This is a trial improvement for the vnode reference count while on the vnode
free list problem.  Also, the vnode age flag is no longer used by the
vnode pager.  (It is actually incorrect to use then.)  Constructive
feedback welcome -- just be kind.
1997-08-22 03:56:37 +00:00
Bruce Evans
b1037dcd53 #include <machine/limits.h> explicitly in the few places that it is required. 1997-08-21 20:33:42 +00:00
Steve Passe
7cbfd031b6 Added includes of smp.h for SMP.
This eliminates a bazillion warnings about implicit s_lock & friends.
1997-08-18 03:29:21 +00:00
John Dyson
03e9c6c101 Fix kern_lock so that it will work. Additionally, clean-up some of the
VM systems usage of the kernel lock (lockmgr) code.  This is a first
pass implementation, and is expected to evolve as needed.  The API
for the lock manager code has not changed, but the underlying implementation
has changed significantly.  This change should not materially affect
our current SMP or UP code without non-standard parameters being used.
1997-08-18 02:06:35 +00:00
John Dyson
1c5ff0a712 The "cutsie" register parameter passing that I had mistakenly used breaks
profiling.  Since it doesn't really improve perf much, I have backed it
out.
1997-08-10 00:12:13 +00:00
John Dyson
f1c1c5b5a4 More vm_zone cleanup. The sysctl now accounts for items better, and
counts the number of allocations.
1997-08-07 03:52:55 +00:00
John Dyson
507b10b48c Add exposure of some vm_zone allocation stats by sysctl. Also, change
the initialization parameters of some zones in VM map.  This contains
only optimizations and not bugfixes.
1997-08-06 04:58:05 +00:00
John Dyson
ba9be04c72 Fixed the commit botch that was causing crashes soon after system
startup.  Due to the error, the initialization of the zone for
pv_entries was missing.  The system should be usable again.
1997-08-05 23:03:24 +00:00
John Dyson
0d65e566b9 Another attempt at cleaning up the new memory allocator. 1997-08-05 22:24:31 +00:00
John Dyson
b79933ebfa Fix some bugs, document vm_zone better. Add copyright to vm_zone.h. Use
the new zone code in pmap.c so that we can get rid of the ugly ad-hoc
allocations in pmap.c.
1997-08-05 22:07:27 +00:00
John Dyson
f2adc8bb27 Modify pmap to use our new memory allocator. Also, change the vm_map_entry
allocations to be interrupt safe.
1997-08-05 01:32:52 +00:00
John Dyson
565bca977d A very simple zone allocator. 1997-08-05 00:07:31 +00:00
John Dyson
3075778b63 Get rid of the ad-hoc memory allocator for vm_map_entries, in lieu of
a simple, clean zone type allocator.  This new allocator will also be
used for machine dependent pmap PV entries.
1997-08-05 00:02:08 +00:00
Bruce Evans
1fd0b0588f Removed unused #includes. 1997-08-02 14:33:27 +00:00
John Dyson
dc2efb2766 Add the ability for the pageout daemon to measure stats on memory usage before
the system is out of memory.  The daemon does a minimal amount of work that
increases as the system becomes more likely to run out of memory and page in/out.

The default tuning is fairly low in background CPU usage, and sysctl variables
have been added to enable flexable operation.  This is an experimental feature
that will likely be changed and improved over time.
1997-07-27 04:49:19 +00:00
John Dyson
11cccda1de Fix a very subtile problem that causes unnessary numbers of objects backing
a single logical object.
Submitted by:	Alan Cox <alc@cs.rice.edu>
1997-07-27 04:44:12 +00:00
John Dyson
0a0a85b3e0 Add support for 4MB pages. This includes the .text, .data, .data parts
of the kernel, and also most of the dynamic parts of the kernel.  Additionally,
4MB pages will be allocated for display buffers as appropriate (only.)

The 4MB support for SMP isn't complete, but doesn't interfere with operation
either.
1997-07-17 04:34:03 +00:00
Tor Egge
208d433777 Don't try upgrading an existing exclusive lock in vm_map_user_pageable.
This should close PR kern/3180.
Also remove a bogus unconditional call to vm_map_unlock_read in
vm_map_lookup.
1997-06-23 21:51:03 +00:00
Peter Wemm
3b18caba29 Kill some stale leftovers from the earlier attempts at SMP per-cpu pages 1997-06-22 15:47:16 +00:00
John Dyson
3c631446d3 Remove a window during running down a file vnode. Also, the OBJ_DEAD
flag wasn't being respected during vref(), et. al.  Note that this
isn't the eventual fix for the locking problem.  Fine grained SMP
in the VM and VFS code will require (lots) more work.
1997-06-22 03:00:24 +00:00
John Dyson
4a40e3d42f Correct the return code for the mlock system call. Also add the stubs
for mlockall and munlockall.
1997-06-15 23:35:32 +00:00
John Dyson
dbc806e731 Fix a reference problem with maps. Only appears to manifest itself when
sharing address spaces.
1997-06-15 23:33:52 +00:00
Peter Wemm
0228905ae4 Update the #include "opt_smpxxx.h" includes - opt_smp.h isn't needed
very much in the generic parts of the kernel now.
1997-05-29 02:57:22 +00:00
Doug Rabson
32ad9cb531 Fix a few bugs with NFS and mmap caused by NFS' use of b_validoff
and b_validend.  The changes to vfs_bio.c are a bit ugly but hopefully
can be tidied up later by a slight redesign.

PR:		kern/2573, kern/2754, kern/3046 (possibly)
Reviewed by:	dyson
1997-05-19 14:36:56 +00:00
John Dyson
6160099735 Check the correct queue for waking up the pageout daemon. Specifically,
the pageout daemon wasn't always being waken up appropriately when the
(cache + free) queues were depleted.
Submitted by:	David S. Miller <davem@jenolan.rutgers.edu>
1997-05-01 14:36:01 +00:00
Peter Wemm
477a642cee Man the liferafts! Here comes the long awaited SMP -> -current merge!
There are various options documented in i386/conf/LINT, there is more to
come over the next few days.

The kernel should run pretty much "as before" without the options to
activate SMP mode.

There are a handful of known "loose ends" that need to be fixed, but
have been put off since the SMP kernel is in a moderately good condition
at the moment.

This commit is the result of the tinkering and testing over the last 14
months by many people.  A special thanks to Steve Passe for implementing
the APIC code!
1997-04-26 11:46:25 +00:00
Peter Wemm
92da7e012d Send this to the Attic so there's no mixups over which kern_lock.c is in
use in -current.
1997-04-21 13:39:56 +00:00
Peter Wemm
e108835bbc Unused variable (upobj is now purely handled within pmap) 1997-04-14 03:40:42 +00:00
John Dyson
5856e12e69 Fully implement vfork. Vfork is now much much faster than even our
fork. (On my machine, fork is about 240usecs, vfork is 78usecs.)

Implement rfork(!RFPROC !RFMEM), which allows a thread to divorce its memory
	from the other threads of a group.

Implement rfork(!RFPROC RFCFDG), which closes all file descriptors, eliminating
	possible existing shares with other threads/processes.

Implement rfork(!RFPROC RFFDG), which divorces the file descriptors for a
	thread from the rest of the group.

Fix the case where a thread does an exec.  It is almost nonsense for a thread
	to modify the other threads address space by an exec, so we
	now automatically divorce the address space before modifying it.
1997-04-13 01:48:35 +00:00
Peter Wemm
a2a1c95c10 The biggie: Get rid of the UPAGES from the top of the per-process address
space. (!)

Have each process use the kernel stack and pcb in the kvm space.  Since
the stacks are at a different address, we cannot copy the stack at fork()
and allow the child to return up through the function call tree to return
to user mode - create a new execution context and have the new process
begin executing from cpu_switch() and go to user mode directly.
In theory this should speed up fork a bit.

Context switch the tss_esp0 pointer in the common tss.  This is a lot
simpler since than swithching the gdt[GPROC0_SEL].sd.sd_base pointer
to each process's tss since the esp0 pointer is a 32 bit pointer, and the
sd_base setting is split into three different bit sections at non-aligned
boundaries and requires a lot of twiddling to reset.

The 8K of memory at the top of the process space is now empty, and unmapped
(and unmappable, it's higher than VM_MAXUSER_ADDRESS).

Simplity the pmap code to manage process contexts, we no longer have to
double map the UPAGES, this simplifies and should measuably speed up fork().

The following parts came from John Dyson:

Set PG_G on the UPAGES that are now in kernel context, and invalidate
them when swapping them out.

Move the upages object (upobj) from the vmspace to the proc structure.

Now that the UPAGES (pcb and kernel stack) are out of user space, make
rfork(..RFMEM..) do what was intended by sharing the vmspace
entirely via reference counting rather than simply inheriting the mappings.
1997-04-07 07:16:06 +00:00
Peter Wemm
2100d64585 Commit a typo fix that's been sitting in my tree for ages, quite forgotten.
The typo was detected once apon a time with the -Wunused compile option.
The result was that a block of code for implementing
madvise(.. MADV_SEQUENTIAL..) behavior was "dead" and unused, probably
negating the effect of activating the option.

Reviewed by: dyson
1997-04-06 16:16:11 +00:00
John Dyson
7d78abc9d9 Make vm_map_protect be more complete about map simplification. This
is useful when a process changes it's page range protections very
much.
Submitted by:	Alan Cox <alc@cs.rice.edu>
1997-04-06 03:04:31 +00:00
John Dyson
15cb98465f Correction to the prototype for vm_fault. 1997-04-06 02:30:56 +00:00
John Dyson
a04c970a7a Fix the gdb executable modify problem. Thanks to the detective work
by Alan Cox <alc@cs.rice.edu>, and his description of the problem.

The bug was primarily in procfs_mem, but the mistake likely happened
due to the lack of vm system support for the operation.  I added
better support for selective marking of page dirty flags so that
vm_map_pageable(wiring) will not cause this problem again.

The code in procfs_mem is now less bogus (but maybe still a little
so.)
1997-04-06 02:29:45 +00:00
Bruce Evans
3f39dbc52d Removed potentially harmful garbage <vm/lock.h> and fixed bogus
use of it.  It was actually harmless because the use was null due
to fortuitous include orders and identical (wrong) idempotency
macros.
1997-04-01 08:39:07 +00:00
David Greenman
9caaadb63a Changed the way that the exec image header is read to be filesystem-
centric rather than VM-centric to fix a problem with errors not being
detectable when the header is read.
Killed exech_map as a result of these changes.
There appears to be no performance difference with this change.
1997-03-31 11:11:26 +00:00
Bruce Evans
3ac4d1ef0c Don't #include <sys/fcntl.h> in <sys/file.h> if KERNEL is defined.
Fixed everything that depended on getting fcntl.h stuff from the wrong
place.  Most things don't depend on file.h stuff at all.
1997-03-23 03:37:54 +00:00
John Dyson
c5d593ae63 Fix a significant error in the accounting for pre-zeroed pages. This
is a candidate for RELENG_2_2...
1997-03-23 02:44:54 +00:00
John Dyson
eb2c768ebb When removing IN_RECURSE support during the Lite/2 merge, read/write
to/from mmaped regions was broken.  This commit fixes the breakage, and
uses the new Lite/2 locking mechanisms.
1997-03-08 04:33:47 +00:00
Bruce Evans
2f558c3e59 Removed a wrong LK_INTERLOCK flag. 1997-02-27 15:38:41 +00:00
Peter Wemm
6875d25465 Back out part 1 of the MCFH that changed $Id$ to $FreeBSD$. We are not
ready for it yet.
1997-02-22 09:48:43 +00:00
Bruce Evans
697030ed3d Removed vestiges of Mach lock types.
vm_map.h:
Removed #include of <sys/proc.h>.  curproc is only used in some macros
and users of the macros already include <sys/proc.h>.
1997-02-18 14:07:03 +00:00
Garrett Wollman
10825343b0 Provide an alternative interface to contigmalloc() which allows a specific
map to be used when allocating the kernel va (e.g., mb_map).  The VM
gurus may want to look this over.
1997-02-13 19:37:40 +00:00
John Dyson
996c772f58 This is the kernel Lite/2 commit. There are some requisite userland
changes, so don't expect to be able to run the kernel as-is (very well)
without the appropriate Lite/2 userland changes.

The system boots and can mount UFS filesystems.

Untested: ext2fs, msdosfs, NFS
Known problems: Incorrect Berkeley ID strings in some files.
		Mount_std mounts will not work until the getfsent
		library routine is changed.

Reviewed by:	various people
Submitted by:	Jeffery Hsu <hsu@freebsd.org>
1997-02-10 02:22:35 +00:00
John Dyson
5069bf5747 Another fix to inheriting shared segments. Do the copy on write
thing if needed.
Submitted by:	Alan Cox <alc@cs.rice.edu>
1997-01-31 04:10:41 +00:00
David Greenman
098415100b Added a check/panic for v_usecount being 0 (no vnode reference) in
vnode_pager_alloc().
1997-01-24 22:20:23 +00:00
John Dyson
fed9a9032e Fix two problems where a NULL object is dereferenced. One problem
was in the VM_INHERIT_SHARE case of vmspace_fork, and also in vm_map_madvise.
Submitted by:	Alan Cox <alc@cs.rice.edu>
1997-01-22 01:34:48 +00:00
John Dyson
6e20a16589 Make MADV_FREE work better. Specifically, it did not wait for
the page to be unbusy, and it caused some algorithmic problems
as a result.  There were some other problems with it also, so
this is a general cleanup of the code.
Submitted by:	Douglas Crosher <dtc@scrooge.ee.swin.oz.au> and myself.
1997-01-20 02:25:14 +00:00
John Dyson
afa07f7e83 Change the map entry flags from bitfields to bitmasks. Allows
for some code simplification.
1997-01-16 04:16:22 +00:00
David Greenman
649c409d03 Fix bug related to map entry allocations where a sleep might be attempted
when allocating memory for network buffers at interrupt time. This is due
to inadequate checking for the new mcl_map. Fixed by merging mb_map and
mcl_map into a single mb_map.

Reviewed by:	wollman
1997-01-15 20:46:02 +00:00
Bruce Evans
16a02c1105 Removed redundant spl0()'s from kernel processes. They were work-arounds
for a bug in fork().
1997-01-15 19:05:08 +00:00
Jordan K. Hubbard
1130b656e5 Make the long-awaited change from $Id$ to $FreeBSD$
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.

Boy, I'm glad we're not using sup anymore.  This update would have been
insane otherwise.
1997-01-14 07:20:47 +00:00
John Dyson
d4a272db61 Slightly correct the code that moves pages from the active to the
inactive queue.  This is only a minor performance improvement, but will
not affect perf on machines that don't have ref bits.
1997-01-11 07:22:24 +00:00
John Dyson
9b5a5d81be Prepare better for multi-platform by eliminating another required
pmap routine (pmap_is_referenced.)  Upper level recoded to use
pmap_ts_referenced.
1997-01-11 07:19:02 +00:00
John Dyson
106031ef73 Undo the collapse breakage (swap space usage problem.) 1997-01-03 17:02:28 +00:00
John Dyson
3c018e7214 Guess what? We left alot of the old collapse code that is not needed
anymore with the "full" collapse fix that we added about 1yr ago!!!  The
code has been removed by optioning it out for now, so we can put it back
in ASAP if any problems are found.
1997-01-01 04:45:05 +00:00
John Dyson
8cc7e047a3 A very significant improvement in the management of process maps
and objects.  Previously, "fancy" memory management techniques
such as that used by the M3 RTS would have the tendancy of chopping
up processes allocated memory into lots of little objects.  Alan
has come up with some improvements to migtigate the sitution to
the point where even the M3 RTS only has one object for bss and
it's managed memory (when running CVSUP.)  (There are still cases where the
situation isn't improved when the system pages -- but this is much much
better for the vast majority of cases.)  The system will now be able
to much more effectively merge map entries.

Submitted by:	Alan Cox <alc@cs.rice.edu>
1996-12-31 16:23:38 +00:00
John Dyson
d0aea04fe0 Let the VM system know that on certain arch's that VM_PROT_READ
also implies VM_PROT_EXEC.  We support it that way for now,
since the break system call by default gives VM_PROT_ALL.  Now
we have a better chance of coalesing map entries when mixing
mmap/break type operations.  This was contributing to excessive
numbers of map entries on the modula-3 runtime system.  The
problem is still not "solved", but the situation makes more
sense.

Eventually, when we work on architectures where VM_PROT_READ
is orthogonal to VM_PROT_EXEC, we will have to visit this
issue carefully (esp. regarding security issues.)
1996-12-30 05:31:21 +00:00
John Dyson
bc0d333478 EEEK!!! useracc and kernacc didn't lock their respective
maps.  Additionally, eliminate the map->hint distortion
associated with useracc.  That may/may-not be the "right"
thing to do -- but time will tell.
Submitted by:	Partially by Alan Cox <alc@cs.rice.edu>
1996-12-30 03:56:11 +00:00
John Dyson
595236df9b Superficial cleanup of comment. 1996-12-29 02:33:12 +00:00
John Dyson
b7b2aac2b6 Eliminate the redundancy due to the similarity between the routines
vm_map_simplify and vm_map_simplify_entry.  Make vm_map_simplify_entry
handle wired maps so that we can get rid of vm_map_simplify.  Modify
the callers of vm_map_simplify to properly use vm_map_simplify_entry.
Submitted by:	Alan Cox <alc@cs.rice.edu>
1996-12-28 23:07:49 +00:00
John Dyson
94328e9057 The code unnecessarily created an object with no handle up-front, which
has the negative effect of disabling some map optimizations.  This
patch defers the creation of the object until it needs to be at fault time.
Submitted by:	Alan Cox <alc@cs.rice.edu>
1996-12-28 22:40:44 +00:00
Joerg Wunsch
e9822d926c Make DFLDSIZ and MAXDSIZ fully-supported options.
"Don't forget to do a ``make depend''" :-)
1996-12-22 23:17:09 +00:00
John Dyson
7aaaa4fd5d Implement closer-to POSIX mlock semantics. The major difference is
that we do allow mlock to span unallocated regions (of course, not
mlocking them.)  We also allow mlocking of RO regions (which the old
code couldn't.)  The restriction there is that once a RO region is
wired (mlocked), it cannot be debugged (or EVER written to.)

Under normal usage, the new mlock code will be a significant improvement
over our old stuff.
1996-12-14 17:54:17 +00:00
John Dyson
0362d7d737 Expunge inlines... 1996-12-07 07:44:05 +00:00
John Dyson
62487bb4db Fix a map entry leak problem found by DG. Also, de-inline a function
vm_map_entry_dispose, because it won't help being inlined.
1996-12-07 06:19:37 +00:00
John Dyson
cdc2c29161 Make vm_map_insert much more intelligent in the MAP_NOFAULT case so
that map entries are coalesced when appropriate.  Also, conditionalize
some code that is currently not used in vm_map_insert.  This mod
has been added to eliminate unnecessary map entries in buffer map.

Additionally, there were some cases where map coalescing could be done
when it shouldn't.  That problem has been resolved.
1996-12-07 00:03:43 +00:00
John Dyson
09e0c6ccdd Implement a new totally dynamic (up to MAXPHYS) buffer kva allocation
scheme.  Additionally, add the capability for checking for unexpected
kernel page faults.  The maximum amount of kva space for buffers hasn't
been decreased from where it is, but it will now be possible to do so.

This scheme manages the kva space similar to the buffers themselves.  If
there isn't enough kva space because of usage or fragementation, buffers
will be reclaimed until a buffer allocation is successful.  This scheme
should be very resistant to fragmentation problems until/if the LFS code
is fixed and uses the bogus buffer locking scheme -- but a 'fixed' LFS
is not likely to use such a scheme.

Now there should be NO problem allocating buffers up to MAXPHYS.
1996-11-30 22:41:49 +00:00
John Dyson
e0c5a895f1 Make the kernel smaller with at worst a neutral effect on perf by
de-inlining some VM calls.  (Actually, I measured a small improvement.)
1996-11-28 23:15:07 +00:00
John Dyson
5b0a74089d Improve the locality of reference for variables in vm_page and
vm_kern by moving them from .bss to .data.  With this change,
there is a measurable perf improvement in fork/exec.
1996-11-17 02:38:31 +00:00
John Dyson
db2c0faa4c Vastly improved contigmalloc routine. It does not solve the
problem of allocating contiguous buffer memory in general, but
make it much more likely to work at boot-up time.  The best
chance for an LKM-type load of a sound driver is immediately
after the mount of the root filesystem.

This appears to work for a 64K allocation on an 8MB system.
1996-11-05 04:19:08 +00:00
John Dyson
851c12ff1d Change mmap to use OBJT_DEFAULT instead of OBJT_SWAP by default
for anonymous objects.  The system will automatically change the
type to SWAP if needed (for size or pageout reasons.)
1996-10-29 22:07:11 +00:00
Poul-Henning Kamp
281cd9b020 The way we get a vnode for swapdev is not quite kosher. In particular
it breaks in the DEVFS_ROOT case.  replicate a bit too much of bdevvp()
in here to circumvent the problem.  The real problem is the magic that
lives in bdevsw[1].
1996-10-27 22:31:00 +00:00
John Dyson
fcae040bc0 Remove a bogus optimization in the mmap code. It is superfluous,
and at best is the same speed as the unoptimized code.  At worst, it
slows down trivial programs.
1996-10-24 02:56:23 +00:00
John Dyson
a669a6e9a9 Make processes waken up eligible for immediate swap-in. 1996-10-17 02:58:20 +00:00
John Dyson
ad98052216 Clean up the rundown of the object backing a vnode. This should fix
NFS problems associated with forcible dismounts.
1996-10-17 02:49:35 +00:00
Bruce Evans
aad9af2ba3 Removed nested include of <sys/proc.h> from <vm/vm_object.h> and fixed
the one place that depended on it.  wakeup() is now prototyped in
<sys/systm.h> so that it is normally visible.

Added nested include of <sys/queue.h> in <vm/vm_object.h>.  The queue
macros are a more fundamental prerequisite for <vm/vm_object.h> than
the wakeup prototype and previously happened to be included by
namespace pollution from <sys/proc.h> or elsewhere.
1996-10-15 18:24:34 +00:00
John Dyson
675878e732 Move much of the machine dependent code from vm_glue.c into
pmap.c.  Along with the improved organization, small proc fork
performance is now about 5%-10% faster.
1996-10-15 03:16:45 +00:00
Poul-Henning Kamp
1111860c29 Remove a stale comment. 1996-10-13 07:16:50 +00:00
Bruce Evans
8ba0c490a9 Removed __pure's and __pure2's. __pure is a no-op for recent versions
of gcc by definition, and __pure2 is a no-op in effect (presumably the
compiler can see when an inline function has no side effects).
1996-10-12 20:09:48 +00:00
John Dyson
853a7bc893 Make the default cache size optim to be 256K, the old default was
64K.  The change has essentially neutral effect on those machines with
little or no cache, and has a positive effect on "normal" machines
with 256K or more cache.
1996-10-06 22:26:13 +00:00
John Dyson
f7d6dab2fd Fix a problem with the page coloring code that the system will not always
be able to use all of the free pages.  This can manifest as a panic
using DIAGNOSTIC, or as a panic on an indirect memory reference.
1996-10-06 18:27:39 +00:00
Bruce Evans
322dfc2bd5 Fixed undeclared variables for the !(PQ_L2_SIZE > 1) case.
Removed redundant #include.
1996-09-28 17:53:18 +00:00
John Dyson
a2f4a84696 Reviewed by:
Submitted by:
Obtained from:
1996-09-28 03:33:40 +00:00
David Greenman
cd6eea255f Fixed bug with reversed trunc/round_page() in madvise...start must be
trunced, end must be rounded.
1996-09-19 10:12:41 +00:00
Bruce Evans
dd106ca742 Removed iprintf(). It was copied to db_iprintf() in ddb. 1996-09-15 11:24:21 +00:00
Bruce Evans
c7c34a24a3 Attached vm ddb commands show map', show vmochk', `show object',
`show vmopag', `show page' and `show pageq'.  Moved all vm ddb stuff
to the ends of the vm source files.

Changed printf() to db_printf(), `indent' to db_indent, and iprintf()
to db_iprintf() in ddb commands.  Moved db_indent and db_iprintf()
from vm to ddb.

vm_page.c:
Don't use __pure.  Staticized.

db_output.c:
Reduced page width from 80 to 79 to inhibit double spacing for long
lines (there are still some problems if words are printed across
column 79).
1996-09-14 11:54:59 +00:00
John Dyson
9fea9a6f5b The whole issue of not support VOP_LOCK for VBLK devices should be
rethought.  This fixes YET another problem with unmounting filesystems.
The root cause is not fixed here, but at least the problem has gone
away.
1996-09-10 05:28:23 +00:00
John Dyson
4334b0d815 Fixed the use of the wrong variable in vm_map_madvise. 1996-09-08 23:49:47 +00:00
John Dyson
5070c7f8c5 Addition of page coloring support. Various levels of coloring are afforded.
The default level works with minimal overhead, but one can also enable
full, efficient use of a 512K cache.  (Parameters can be generated
to support arbitrary cache sizes also.)
1996-09-08 20:44:49 +00:00
John Dyson
b8e251a56d Improve the scalability of certain pmap operations. 1996-09-08 16:57:53 +00:00
John Dyson
6476c0d204 Even though this looks like it, this is not a complex code change.
The interface into the "VMIO" system has changed to be more consistant
and robust.  Essentially, it is now no longer necessary to call vn_open
to get merged VM/Buffer cache operation, and exceptional conditions
such as merged operation of VBLK devices is simpler and more correct.

This code corrects a potentially large set of problems including the
problems with ktrace output and loaded systems, file create/deletes,
etc.

Most of the changes to NFS are cosmetic and name changes, eliminating
a layer of subroutine calls.  The direct calls to vput/vrele have
been re-instituted for better cross platform compatibility.

Reviewed by: davidg
1996-08-21 21:56:23 +00:00
John Dyson
67bf686897 Backed out the recent changes/enhancements to the VM code. The
problem with the 'shell scripts' was found, but there was a 'strange'
problem found with a 486 laptop that we could not find.  This commit
backs the code back to 25-jul, and will be re-entered after the snapshot
in smaller (more easily tested) chunks.
1996-07-30 03:08:57 +00:00
David Greenman
0f281c28fa Slight performance tweak for previous commit. 1996-07-28 02:54:09 +00:00
John Dyson
f230c45cbe Undo part of the scalability commit. Many of the changes
in vm_fault had some performance enhancements not ready
for prime time.  This commit backs out some of the changes.
1996-07-28 01:14:01 +00:00
John Dyson
bf6dfc7b35 Allow sequentially created mmap'ed anonymous regions to coalesce. There
is little or no reason to create a swap pager for small mmap's.  The
vm_map_insert code will automatically create a swap pager if the object
becomes too large.  This fix, per a request from phk.
1996-07-27 17:21:41 +00:00
John Dyson
3b297e93b8 Clean up some lint. 1996-07-27 04:22:12 +00:00
John Dyson
feb32a8fa9 Remove experimental header file. My test-build must have picked it
up in an unexpected place.
Submitted by:	jkh
1996-07-27 04:06:11 +00:00
John Dyson
819c1c6f43 Missing (prototype) change from the previous commit. 1996-07-27 03:47:35 +00:00
John Dyson
4f4d35edf0 This commit is meant to solve a couple of VM system problems or
performance issues.

	1) The pmap module has had too many inlines, and so the
	   object file is simply bigger than it needs to be.
	   Some common code is also merged into subroutines.
	2) Removal of some *evil* PHYS_TO_VM_PAGE macro calls.
	   Unfortunately, a few have needed to be added also.
	   The removal caused the need for more vm_page_lookups.
	   I added lookup hints to minimize the need for the
	   page table lookup operations.
	3) Removal of some bogus performance improvements, that
	   mostly made the code more complex (tracking individual
	   page table page updates unnecessarily).  Those improvements
	   actually hurt 386 processors perf (not that people who
	   worry about perf use 386 processors anymore :-)).
	4) Changed pv queue manipulations/structures to be TAILQ's.
	5) The pv queue code has had some performance problems since
	   day one.  Some significant scalability issues are resolved
	   by threading the pv entries from the pmap AND the physical
	   address instead of just the physical address.  This makes
	   certain pmap operations run much faster.  This does
	   not affect most micro-benchmarks, but should help loaded system
	   performance *significantly*.  DG helped and came up with most
	   of the solution for this one.
	6) Most if not all pmap bit operations follow the pattern:
		pmap_test_bit();
		pmap_clear_bit();
	   That made for twice the necessary pv list traversal.   The
	   pmap interface now supports only pmap_tc_bit type operations:
	   pmap_[test/clear]_modified, pmap_[test/clear]_referenced.
	   Additionally, the modified routine now takes a vm_page_t arg
	   instead of a phys address.  This eliminates a PHYS_TO_VM_PAGE
	   operation.
	7) Several rewrites of routines that contain redundant code to
	   use common routines, so that there is a greater likelihood of
	   keeping the cache footprint smaller.
1996-07-27 03:24:10 +00:00
Bruce Evans
6ab46d52a5 Don't use NULL in non-pointer contexts. 1996-07-12 04:12:25 +00:00
John Dyson
502ba6e4a8 Back-off on the previous commit, specifically remove the look-ahead
optimization on the active queue scan.  I will do this correctly later.
1996-07-08 03:22:55 +00:00
John Dyson
c8c4b40cca Fix a problem with the pageout daemon RSS limiting, where it degrades
performance to LRU or worse when RSS limiting takes effect.  Also,
make an end condition in the active queue scan more efficient in the
case where pages are removed from the active queue as a side effect
of a pmap operation.
1996-07-08 02:25:53 +00:00
David Greenman
9579ee641a In all special cases for spl or page_alloc where kmem_map is check for,
mb_map (a submap of kmem_map) must also be checked.
Thanks to wcarchive (err...sort of) for demonstrating this bug.
1996-07-07 03:27:41 +00:00
John Dyson
a6e6bcc5f4 Properly set the PG_MAPPED and PG_WRITEABLE flags. This fixes some potential
problems with vm_map_remove/vm_map_delete.
1996-07-02 02:08:02 +00:00
John Dyson
877329e059 Make -current consistant with -stable regarding time that a process
sleeps before being swapped out.  The time is increased from 4 secs to
10 secs.  Originally I had decreased it from 20 to 4, but that is a bit
severe.  20 is too long though.
1996-06-30 21:16:18 +00:00
David Greenman
01155bd720 Make sure we have an object in the map entry before trying to trim pages
from it.
1996-06-29 09:17:17 +00:00
John Dyson
38efa82b23 This commit does a couple of things:
Re-enables the RSS limiting, and the routine is now tail-recursive,
	making it much more safe (eliminates the possiblity of kernel stack
	overflow.) Also, the RSS limiting is a little more intelligent about
	finding the likely objects that are pushing the process over the limit.

	Added some sysctls that help with VM system tuning.

New sysctl features:
	1)	Enable/disable lru pageout algorithm.
		vm.pageout_algorithm = 0, default algorithm that works
			well, especially using X windows and heavy
			memory loading.  Can have adverse effects,
			sometimes slowing down program loading.

		vm.pageout_algorithm = 1, close to true LRU.  Works much
			better than clock, etc.  Does not work as well as
			the default algorithm in general.  Certain memory
			"malloc" type benchmarks work a little better with
			this setting.

		Please give me feedback on the performance results
		associated with these.

	2)	Enable/disable swapping.
		vm.swapping_enabled = 1, default.

		vm.swapping_enabled = 0, useful for cases where swapping
			degrades performance.

		The config option "NO_SWAPPING" is still operative, and
		takes precedence over the sysctl.  If "NO_SWAPPING" is
		specified, the sysctl still exists, but "vm.swapping_enabled"
		is hard-wired to "0".

Each of these can be changed "on the fly."
1996-06-26 05:39:27 +00:00
John Dyson
f0e2953e5e Fix some serious problems with limits checking in the sbrk(2)/brk(2)
code.
Reviewed by:	bde
1996-06-25 00:36:46 +00:00
John Dyson
a001376dc3 Remove RSS limiting until I rewrite the code to be non-recursive. The
code can overrun the kernel stack under very stressful conditions.
1996-06-24 04:30:24 +00:00
John Dyson
2a4eb04bfd Improve algorithm for page hash queue. It was previously about
as bad as it could be.  This algorithm appears to improve fork
performance (barely) measurably.
1996-06-21 05:39:22 +00:00
John Dyson
ef743ce6ed Several bugfixes/improvements:
1) Make it much less likely to miss a wakeup in vm_page_free_wakeup
	2) Create a new entry point into pmap: pmap_ts_referenced, eliminates
	   the need to scan the pv lists twice in many cases.  Perhaps there
	   is alot more to do here to work on minimizing pv list manipulation
	3) Minor improvements to vm_pageout including the use of pmap_ts_ref.
	4) Major changes and code improvement to pmap.  This code has had
	   several serious bugs in page table page manipulation.  In order
	   to simplify the problem, and hopefully solve it for once and all,
	   page table pages are no longer "managed" with the pv list stuff.
	   Page table pages are only (mapped and held/wired) or
	   (free and unused) now.  Page table pages are never inactive,
	   active or cached.  These changes have probably fixed the
	   hold count problems, but if they haven't, then the code is
	   simpler anyway for future bugfixing.
	5) The pmap code has been sorely in need of re-organization, and I
	   have taken a first (of probably many) steps.  Please tell me
	   if you have any ideas.
1996-06-17 03:35:40 +00:00
John Dyson
b5b40fa62b Various bugfixes/cleanups from me and others:
1) Remove potential race conditions on waking up in vm_page_free_wakeup
   by making sure that it is at splvm().
2) Fix another bug in vm_map_simplify_entry.
3) Be more complete about converting from default to swap pager
   when an object grows to be large enough that there can be
   a problem with data structure allocation under low memory
   conditions.
4) Make some madvise code more efficient.
5) Added some comments.
1996-06-16 20:37:31 +00:00
David Greenman
664275648a Move a case of PG_MAPPED being set before a pmap_enter(). This will likely
make no difference, but it will make it consistent with other uses of
PG_MAPPED.
1996-06-14 23:26:40 +00:00
John Dyson
419702a468 Fix a very significant cnt.v_wire_count leak in vm_page.c, and some
minor leaks in pmap.c.  Bruce Evans made me aware of this problem.
1996-06-12 06:52:12 +00:00
John Dyson
5fcf66debe Fix some serious errors in vm_map_simplify_entries. 1996-06-12 04:03:21 +00:00
John Dyson
3091ee0955 Mostly superficial code improvements, add a diagnostic. The
code improvements include significant simplification of the reservation
of the swap pager control blocks for reads.  Add a panic for an inconsistent
swap pager control block count.
1996-06-10 04:58:48 +00:00
John Dyson
c82b01813e Keep the vm_fault/vm_pageout from getting into an "infinite paging loop", by
reserving "cached" pages before waking up the pageout daemon.  This will reserve
the faulted page, and keep the system from thrashing itself to death given
this condition.
1996-06-10 00:25:40 +00:00
John Dyson
886d3e1150 Adjust the threshold for blocking on movement of pages from the cache
queue in vm_fault.

Move the PG_BUSY in vm_fault to the correct place.

Remove redundant/unnecessary code in pmap.c.

Properly block on rundown of page table pages, if they are busy.

I think that the VM system is in pretty good shape now, and the following
individuals (among others, in no particular order) have helped with this
recent bunch of bugs, thanks!  If I left anyone out, I apologize!

Stephen McKay, Stephen Hocking, Eric J. Chet, Dan O'Brien, James Raynard,
Marc Fournier.
1996-06-08 06:48:35 +00:00
John Dyson
6b6f000870 Keep page-table pages from ever being sensed as dirty. This should fix
some problems with the page-table page management code, since it can't
deal with the notion of page-table pages being paged out or in transit.
Also, clean up some stylistic issues per some suggestions from
Stephen McKay.
1996-06-05 03:31:49 +00:00
John Dyson
ff97964a2e Disable madvise optimizations for device pager objects (some of the
operations don't work with FICTITIOUS pages.)  Also, close a window
between PG_MANAGED and pmap_enter that can mess up the accounting of
the managed flag.  This problem could likely cause a hold_count error
for page table pages.
1996-06-01 20:50:57 +00:00
John Dyson
f35329ac0f This commit is dual-purpose, to fix more of the pageout daemon
queue corruption problems, and to apply Gary Palmer's code cleanups.
David Greenman helped with these problems also.  There is still
a hang problem using X in small memory machines.
1996-05-31 00:38:04 +00:00
John Dyson
545901f794 Correct some unfortunately chosen constants, otherwise, not enough
pages are calculated for deferred allocation of swap pager data structures.
This is a follow-on to the previous commit to this file.
1996-05-29 06:33:30 +00:00
John Dyson
b182ec9eb4 After careful review by David Greenman and myself, David had found a
case where blocking can occur, thereby giving other process's a chance
to modify the queue where a page resides.  This could cause numerous
process and system failures.
1996-05-29 05:15:33 +00:00
John Dyson
a5b6fd29a3 Make sure that pageout deadlocks cannot occur. There is a problem
that the datastructures needed to support the swap pager can take
enough space to fully deplete system memory, and cause a deadlock.
This change keeps large objects from being filled with dirty pages
without the appropriate swap pager datastructures.  Right now,
default objects greater than 1/4 the size of available system memory
are converted to swap objects, thereby eliminating the risk of deadlock.
1996-05-29 05:12:23 +00:00
John Dyson
85a376eb93 Fix a couple of problems in the pageout_scan routine. First, there is
a condition when blocking can occur, and the daemon did not check properly
for a page remaining on the expected queue.  Additionally, the inactive
target was being set much too large for small memory machines.  It is now
being calculated based upon the amount of user memory available on every
pageout daemon run.  Another problem was that if memory was very low, the
pageout daemon could fail repeatedly to traverse the inactive queue.
1996-05-26 07:52:09 +00:00
John Dyson
0ed4376231 I think this covers (fixes) the last batch of freeing active/held/busy page
problem.  BY MISTAKE, the vm_page_unqueue (or equiv) was removed from the
vm_fault code.  Really bad things appear to happen if a page is on a queue
while it is being faulted.
1996-05-26 05:30:33 +00:00
John Dyson
f777ab7b8b Add an assert to vm_page_cache. We should never cache a dirty page. 1996-05-24 05:20:15 +00:00
John Dyson
1eeaa1e31f Add apparently needed splvm protection to the active queue, and eliminate
an unnecessary test for dirty pages if it is already known to be dirty.
1996-05-24 05:19:15 +00:00
John Dyson
3077a9c2f4 Eliminate inefficient check for dirty pages for pages in the PQ_CACHE
queue.  Also, modify the MADV_FREE policy (it probably still isn't the final
version.)
1996-05-24 05:17:21 +00:00
John Dyson
a9d4727439 Make the conversion from the default pager to swap pager more robust
in the face of low memory conditions.
1996-05-24 05:14:44 +00:00
John Dyson
99ea1af0a6 Eliminate a vm_page_free, busy panic, in kern_malloc. 1996-05-23 02:24:55 +00:00
John Dyson
0a47b48b9f Initial support for MADV_FREE, support for pages that we don't care
about the contents anymore.  This gives us alot of the advantage of
freeing individual pages through munmap, but with almost none of the
overhead.
1996-05-23 00:45:58 +00:00
John Dyson
4a62209c07 After reviewing the previous commit to vm_object, the page protection
is never necessary, not just for PG_FICTICIOUS.
1996-05-21 17:13:31 +00:00
John Dyson
07c647c528 Don't protect non-managed pages off during object rundown. This fixes
a hang that occurs under certain circumstances when exiting X.
1996-05-21 05:26:27 +00:00
John Dyson
867a482d66 Initial support for mincore and madvise. Both are almost fully
supported, except madvise does not page in with MADV_WILLNEED, and
MADV_DONTNEED doesn't force dirty pages out.
1996-05-19 07:36:50 +00:00
John Dyson
7f5fe93fc7 One more file missing from the mega-commit. This inlines some very
simple routines in vm_page.c, so that an unnecessary subroutine call
is removed.
1996-05-18 04:00:18 +00:00
John Dyson
1b4435b8ce File mistakenly left out of the previous mega-commit. This provides
a global defn for 'exech_map.'
1996-05-18 03:52:13 +00:00
John Dyson
b18bfc3da7 This set of commits to the VM system does the following, and contain
contributions or ideas from Stephen McKay <syssgm@devetir.qld.gov.au>,
Alan Cox <alc@cs.rice.edu>, David Greenman <davidg@freebsd.org> and me:

	More usage of the TAILQ macros.  Additional minor fix to queue.h.
	Performance enhancements to the pageout daemon.
		Addition of a wait in the case that the pageout daemon
		has to run immediately.
		Slightly modify the pageout algorithm.
	Significant revamp of the pmap/fork code:
		1) PTE's and UPAGES's are NO LONGER in the process's map.
		2) PTE's and UPAGES's reside in their own objects.
		3) TOTAL elimination of recursive page table pagefaults.
		4) The page directory now resides in the PTE object.
		5) Implemented pmap_copy, thereby speeding up fork time.
		6) Changed the pv entries so that the head is a pointer
		   and not an entire entry.
		7) Significant cleanup of pmap_protect, and pmap_remove.
		8) Removed significant amounts of machine dependent
		   fork code from vm_glue.  Pushed much of that code into
		   the machine dependent pmap module.
		9) Support more completely the reuse of already zeroed
		   pages (Page table pages and page directories) as being
		   already zeroed.
	Performance and code cleanups in vm_map:
		1) Improved and simplified allocation of map entries.
		2) Improved vm_map_copy code.
		3) Corrected some minor problems in the simplify code.
	Implemented splvm (combo of splbio and splimp.)  The VM code now
		seldom uses splhigh.
	Improved the speed of and simplified kmem_malloc.
	Minor mod to vm_fault to avoid using pre-zeroed pages in the case
		of objects with backing objects along with the already
		existant condition of having a vnode.  (If there is a backing
		object, there will likely be a COW...  With a COW, it isn't
		necessary to start with a pre-zeroed page.)
	Minor reorg of source to perhaps improve locality of ref.
1996-05-18 03:38:05 +00:00
Garrett Wollman
cb7545a995 Allocate mbufs from a separate submap so that NMBCLUSTERS works as
expected.
1996-05-10 19:28:55 +00:00
Poul-Henning Kamp
aa8de40ae5 Another sweep over the pmap/vm macros, this time with more focus on
the usage.  I'm not satisfied with the naming, but now at least there is
less bogus stuff around.
1996-05-03 21:01:54 +00:00
Poul-Henning Kamp
e911eafcba removed:
CLBYTES PD_SHIFT PGSHIFT NBPG PGOFSET CLSIZELOG2 CLSIZE pdei()
        ptei() kvtopte() ptetov() ispt() ptetoav() &c &c
new:
        NPDEPG

Major macro cleanup.
1996-05-02 14:21:14 +00:00
Poul-Henning Kamp
a8c5fef5e6 KGDB is dead. It may come back one day if somebody does it. 1996-05-02 09:34:51 +00:00
John Dyson
3ea2f344e0 Move the map entry allocations from the kmem_map to the kernel_map. As
a side effect, correct the associated object offset.
1996-04-29 22:04:57 +00:00
John Dyson
0891ef4c9a This fixes kmem_malloc/kmem_free (and malloc/free of objects of > 8K).
A page index was calculated incorrectly in vm_kern, and vm_object_page_remove
removed pages that should not have been.
1996-04-24 04:16:45 +00:00
Bruce Evans
bd105bb750 Fixed a spl hog. The vmdaemon process ran entirely at splhigh. It
sometimes disabled clock interrupts for 60 msec or more on a P133.
Clock interrupts were lost ...

Reviewed by:	dyson
1996-04-11 21:05:25 +00:00
John Dyson
d3a3498598 Reinstitute the map lock for processes being swapped out. This
is needed because of the vm_fault used to bring the page table page
for the kernel stack (UPAGES) back in.  The consequence of the
previous incorrect change was a system hang.
1996-04-09 04:36:58 +00:00
John Dyson
b5cfb15fad Map lock checks not needed anymore for swapping out. We don't use
map operations for it anymore.  Certain deadlocks should never happen
anymore.
1996-04-08 03:42:01 +00:00
Bruce Evans
6ffde942bf Removed never-used #includes of <machine/cpu.h>. Many were apparently
copied from bad examples.
1996-04-07 17:39:28 +00:00
John Dyson
030ad08012 Fixed a problem that the UPAGES of a process were being run down
in a suboptimal manner.  I had also noticed some panics that appeared
to be at least superficially caused by this problem.  Also, included
are some minor mods to support more general handling of page table page
faulting.  More details in a future commit.
1996-04-03 05:23:44 +00:00
David Greenman
46268a606f Revert to previous calculation of vm_object_cache_max: it simply works
better in most real-world cases.
1996-03-29 06:28:48 +00:00
Bruce Evans
8375baabed Undid last revision. It duplicated part of second last revision. 1996-03-28 15:40:17 +00:00
Marc G. Fournier
7f49be143c devfs_add_devsw() -> devfs_add_devswf modifications
Reviewed by:	julian@freebsd.org
1996-03-28 14:36:48 +00:00
John Dyson
bb35ebd6cc Add a function prototype for pmap_prefault. 1996-03-28 04:54:50 +00:00
John Dyson
30dcfc09f2 VM performance improvements, and reorder some operations in VM fault
in anticipation of a fix in pmap that will allow the mlock system call to work
without panicing the system.
1996-03-28 04:53:28 +00:00
John Dyson
f32dbbeeed More map_simplify fixes from Alan Cox. This very significanly improves the
performance when the map has been chopped up.  The map simplify operations
really work now.
Reviewed by: dyson
Submitted by:	Alan Cox <alc@cs.rice.edu>
1996-03-28 04:22:17 +00:00
Bruce Evans
5ea390eff5 Added drum device.
Submitted by:	partly by "Marc G. Fournier" <scrappy@ki.net>
1996-03-27 20:09:26 +00:00
John Dyson
ad5dd2341c Fix the problem that unmounting filesystems that are backed by a VMIO
device have reference count problems.  We mark the underlying object
ono-persistent, and account for the reference count that the VM system
maintainsfor the special device close.  This should fix the removable
device problem.
1996-03-19 05:13:22 +00:00
David Greenman
8f2ec877b8 Force device mappings to always be shared. It doesn't make sense for them
to ever be COW and we need the mappings to be shared for backward
compatibilty.

Reviewed by:	dyson
1996-03-16 15:00:05 +00:00
John Dyson
308c24ba5e This commit is as a result of a comment by Alan Cox (alc@cs.rice.edu)
regarding the "real" problem with maps that we have been having
over the last few weeks.  He noted that the first_free pointer was
left dangling in certain circumstances -- and he was right!!!   This
should fix the map problems that we were having, and also give us the
advantage of being able to simplify maps more aggressively.
1996-03-13 01:18:14 +00:00
John Dyson
2fc2c638d5 Fix the map corruption problem that appears as a u_map allocation
error.
1996-03-12 13:46:13 +00:00
John Dyson
5850152d95 Allow mmap'ed devices to work correctly across forks. The sanest
solution appeared to be to allow the child to maintain the same mapping as
the parent.
1996-03-12 02:27:20 +00:00
Jeffrey Hsu
1b67ec6de9 For Lite2: proc LIST changes.
Reviewed by:	 davidg & bde
1996-03-11 06:11:43 +00:00
John Dyson
9ea857084d Delay forking a process until there are more pages available. It was
possible to deadlock with the low threshold that we had used.
1996-03-09 06:57:53 +00:00
John Dyson
9ee58740bc Modify a threshold for waking up the pageout daemon. Also, add a consistancy
check for making sure that held pages aren't freed (DG).
1996-03-09 06:56:39 +00:00
John Dyson
c68f9c929b Add a missing initialization of the hold_count for device pager ficticiouse
pages.
1996-03-09 06:54:41 +00:00
John Dyson
6ac5bfdb3a Fix a calculation for a paging parameter. 1996-03-09 06:53:27 +00:00
John Dyson
67cc64f4c7 Fix two problems:
The pmap_remove in vm_map_clean incorrectly unmapped the entire
	map entry.
	The new vm_map_simplify_entry code had an error (the offset
	of the combined map entry was not set correctly.)
Submitted by:	Alan Cox <alc@cs.rice.edu>
1996-03-09 06:52:05 +00:00
John Dyson
65bc79b85f Set the page valid bits in fewer places, as opposed to being scattered
in various places.
1996-03-09 06:48:26 +00:00
John Dyson
45952afcc7 Fix a problem in the swap pager that caused some of the pages that
were paged in under low swap space conditions to both loose their
backing store and their dirty bits.  This would cause pages to
be demand zeroed under certain conditions in low VM space conditions
and consequential sig-11's or sig-10's.  This situation was made
worse lately when the level for swap space reclaim threshold was
increased.
1996-03-06 04:31:46 +00:00
John Dyson
8a02c104f9 Fix a problem that pages in a mapped region were not always
properly invalidated.  Now we traverse the object shadow chain
properly.
1996-03-04 02:04:24 +00:00
John Dyson
836e5d1360 In order to fix some concurrency problems with the swap pager early
on in the FreeBSD development, I had made a global lock around the
rlist code.  This was bogus, and now the lock is maintained on a
per resource list basis.  This now allows the rlist code to be used for
almost any non-interrupt level application.
1996-03-03 21:11:08 +00:00
Peter Wemm
5e004bea6f Remove the #ifdef notyet from the prototype of vm_map_simplify. John
re-enabled the function but missed the prototype, causing a warning.
1996-03-03 18:53:10 +00:00
Peter Wemm
9154ee6aec Oops.. I nearly forgot the actual core of the length/rounding/etc fixes
that Bruce asked for.

These still are not quite perfect, and in particular, it can get
upset on extreme boundary cases (addr = 0xfff, len = 0xffffffff,
which would end up mapping a single page rather than failing), but
this is better code that I committed before.

(note, the VM system does not (apparently) support single mmap segment
sizes above 0x80000000 anyway)
1996-03-02 17:14:09 +00:00
John Dyson
de5f6a7765 1) Eliminate unnecessary bzero of UPAGES.
2) Eliminate unnecessary copying of pages during/after forks.
3) Add user map simplification.
1996-03-02 02:54:24 +00:00
Peter Wemm
dabee6fecc kern_descrip.c: add fdshare()/fdcopy()
kern_fork.c: add the tiny bit of code for rfork operation.
kern/sysv_*: shmfork() takes one less arg, it was never used.
sys/shm.h: drop "isvfork" arg from shmfork() prototype
sys/param.h: declare rfork args.. (this is where OpenBSD put it..)
sys/filedesc.h: protos for fdshare/fdcopy.
vm/vm_mmap.c: add minherit code, add rounding to mmap() type args where
it makes sense.
vm/*: drop unused isvfork arg.

Note: this rfork() implementation copies the address space mappings,
it does not connect the mappings together.  ie: once the two processes
have split, the pages may be shared, but the address space is not. If one
does a mmap() etc, it does not appear in the other.  This makes it not
useful for pthreads, but it is useful in it's own right for having
light-weight threads in a static shared address space.

Obtained from: Original by Ron Minnich, extended by OpenBSD
1996-02-23 18:49:25 +00:00
David Greenman
5afce28270 Add a "NO_SWAPPING" option to disable swapping. This was originally done
to help diagnose a problem on wcarchive (where the kernel stack was
sometimes not present), but is useful in its own right since swapping
actually reduces performance on some systems (such as wcarchive).
Note: swapping in this context means making the U pages pageable and has
nothing to do with generic VM paging, which is unaffected by this option.

Reviewed by:	 <dyson>
1996-02-22 10:57:37 +00:00
John Dyson
a02051c37a Fixed a really bogus problem with msync ripping pages away from
objects before they were written.  Also, don't allow processes
without write access to remove pages from vm_objects.
1996-02-11 22:03:49 +00:00
John Dyson
dca5129987 Changed vm_fault_quick in vm_machdep.c to be global. Needed for
new pipe code.
1996-02-04 22:09:12 +00:00
David Greenman
1af87c9263 "out of space" -> "out of swap space". 1996-01-31 13:14:21 +00:00
David Greenman
729b1e5149 Improved killproc() log message and made it and the other similar message
tolerant of p_ucred being invalid. Starting using killproc() where
appropriate.
1996-01-31 12:44:33 +00:00
David Greenman
8c73da1e15 Print a more descriptive message when the mb_map is filled (out of mbuf
clusters), and tell the operator what to do about it (increase maxusers).
1996-01-31 12:05:52 +00:00
Mike Pritchard
6c5e9bbdf5 Fix a bunch of spelling errors in the comment fields of
a bunch of system include files.
1996-01-30 23:02:38 +00:00
David Greenman
2c68345ab4 Added a check/panic for vm_map_find failing to find space for the page
tables/u-pages when forking. This is a "can't happen" case. :-)
1996-01-29 12:10:30 +00:00
Bruce Evans
324e9ed2a4 Added a `boundary' arg to vm_alloc_page_contig(). Previously the only
way to avoid crossing a 64K DMA boundary was to specify an alignment
greater than the size even when the alignment didn't matter, and for
sizes larger than a page, this reduced the chance of finding enough
contiguous pages.  E.g., allocations of 8K not crossing a 64K boundary
previously had to be allocated on 8K boundaries; now they can be
allocated on any 4K boundary except (64 * n + 60)K.

Fixed bugs in vm_alloc_page_contig():
- the last page wasn't allocated for sizes smaller than a page.
- failures of kmem_alloc_pageable() weren't handled.

Mutated vm_page_alloc_contig() to create a more convenient interface
named contigmalloc().  This is the same as the one in 1.1.5 except
it has `low' and `high' args, and the `alignment' and `boundary'
args are multipliers instead of masks.
1996-01-27 00:13:33 +00:00
Poul-Henning Kamp
f782b11a04 Don't use %r, we havn't got it anymore.
Submitted by: bde
1996-01-25 07:15:40 +00:00
John Dyson
bd7e5f992e Eliminated many redundant vm_map_lookup operations for vm_mmap.
Speed up for vfs_bio -- addition of a routine bqrelse to greatly diminish
	overhead for merged cache.
Efficiency improvement for vfs_cluster.  It used to do alot of redundant
	calls to cluster_rbuild.
Correct the ordering for vrele of .text and release of credentials.
Use the selective tlb update for 486/586/P6.
Numerous fixes to the size of objects allocated for files.  Additionally,
	fixes in the various pagers.
Fixes for proper positioning of vnode_pager_setsize in msdosfs and ext2fs.
Fixes in the swap pager for exhausted resources.  The pageout code
	will not as readily thrash.
Change the page queue flags (PG_ACTIVE, PG_INACTIVE, PG_FREE, PG_CACHE) into
	page queue indices (PQ_ACTIVE, PQ_INACTIVE, PQ_FREE, PQ_CACHE),
	thereby improving efficiency of several routines.
Eliminate even more unnecessary vm_page_protect operations.
Significantly speed up process forks.
Make vm_object_page_clean more efficient, thereby eliminating the pause
	that happens every 30seconds.
Make sequential clustered writes B_ASYNC instead of B_DELWRI even in the
	case of filesystems mounted async.
Fix a panic with busy pages when write clustering is done for non-VMIO
	buffers.
1996-01-19 04:00:31 +00:00
Garrett Wollman
0e41ee3037 Convert DDB to new-style option. 1996-01-04 21:13:23 +00:00
Garrett Wollman
50c73f3620 Convert SYSV IPC to new-style options. (I hope I got everything...)
The LKMs will need an extra file, to come later.
1996-01-04 20:29:06 +00:00
David Greenman
a2d5b14236 Increased vm_object_cache_max by about 50% to yield better utilization of
memory when lots of small files are cached.

Reviewed by:	dyson
1996-01-04 18:32:31 +00:00
Peter Wemm
a5b996a7ec recording cvs-1.6 file death 1995-12-30 19:02:48 +00:00
Julian Elischer
1dfcbb0ce3 i386/i386/conf.c is no longer needed.. remove it from files.i386
redistribute a few last routines to beter places and shoot the file

I haven't act actually 'deleted' the file yet togive people time
to
have done a config.. I.e. they are likely to have done one in a week or so
so I'll remove it then..
 it's now empty.
makes the question of a USL copyright rather moot.
1995-12-21 20:09:46 +00:00
John Dyson
d63596ce9e Fix paging from ext2fs (and other fs w/block size < PAGE_SIZE). This
should fix kern/900.
1995-12-17 23:29:56 +00:00
Bruce Evans
927c0b1131 Cleaned up prototypes in pmap headers: removed ones for nonexistent
functions; moved misplaced ones; restored most of KNFish formatting
from 4.4lite version; removed bogus __BEGIN/END_DECLS.
1995-12-17 07:39:05 +00:00
Bruce Evans
f2c6b65bab Fixed 1TB filesize changes. Some pindexes had bogus names and types
but worked because vm_pindex_t is indistinuishable from vm_offset_t.
1995-12-17 07:19:58 +00:00
Poul-Henning Kamp
f708ef1b9e Another mega commit to staticize things. 1995-12-14 09:55:16 +00:00
Poul-Henning Kamp
87b6de2b76 A Major staticize sweep. Generates a couple of warnings that I'll deal
with later.
A number of unused vars removed.
A number of unused procs removed or #ifdefed.
1995-12-14 08:32:45 +00:00
Julian Elischer
6ba9ebce28 devsw tables are now arrays of POINTERS to struct [cb]devsw
seems to work hre just fine though  I can't check every file
that changed due to limmited h/w, however I've checked enught to be petty
happy withe hte code..

WARNING... struct lkm[mumble] has changed
so it might be an idea to recompile any lkm related programs
1995-12-13 15:13:57 +00:00
John Dyson
3048c51216 There was a bug that the size for an msync'ed region was not rounded
up.  The effect of this was that msync with a size would generally sync
1 page less than it should.  This problem was brought to my attention
by Darrel Herbst <dherbst@gradin.cis.upenn.edu> and Ron Minnich
<rminnich@sarnoff.com>.
1995-12-13 12:28:39 +00:00
John Dyson
cb6962cdee Some new anti-deadlock code ended up messing up the paging stats. A modified
version of the code is now in place, and gausspage performance is back
up to where it should be.
1995-12-11 15:43:33 +00:00
John Dyson
ec07c60c6a Some DIAGNOSTIC code was enabled all of the time in error. The
diagnostic code is now conditional on #ifdef DIAGNOSTIC again.
1995-12-11 15:41:50 +00:00
John Dyson
a316d390bd Changes to support 1Tb filesizes. Pages are now named by an
(object,index) pair instead of (object,offset) pair.
1995-12-11 04:58:34 +00:00
Bruce Evans
4439655d52 Replaced nxdump by nodump (if the dump function gets called, then the
device must be configured, so ENXIO is a bogus errno).

Replaced zerosize by nopsize.  zerosize was a temporary alias.
1995-12-10 19:53:42 +00:00
Bruce Evans
9b4288a3d5 Restored used includes of <vm/vm_extern.h>. 1995-12-10 14:52:10 +00:00
Bruce Evans
5ad9d5a70f Moved the declaration of boolean_t from <vm/vm_param.h> to
<sys/types.h> (if KERNEL is defined).  This allows removing bogus
dependencies on vm stuff in several places (e.g., ddb) and stops
<vm_param.h> from depending on <vm_param.h>

Added declaration of boolean_t to <vm/vm.h> (if KERNEL is not
defined).  It never belonged in <vm/vm_param.h>.  Unfortunately,
it is required for some vm headers that are included by applications.

Deleted declarations of TRUE and FALSE from <vm/vm_param.h>.  They
are defined in <sys/param.h> if KERNEL is defined and we'll soon
find out if any applications depend on them being defined in a vm
header.
1995-12-10 02:34:29 +00:00
Poul-Henning Kamp
d2f265fab8 Julian forgot to make the *devsw structures static. 1995-12-08 23:23:00 +00:00
Julian Elischer
87f6c6625d Pass 3 of the great devsw changes
most devsw referenced functions are now static, as they are
in the same file as their devsw structure. I've also added DEVFS
support for nearly every device in the system, however
many of the devices have 'incorrect' names under DEVFS
because I couldn't quickly work out the correct naming conventions.
(but devfs won't be coming on line for a month or so anyhow so that doesn't
matter)

If you "OWN" a device which would normally have an entry in /dev
then search for the devfs_add_devsw() entries and munge to make them right..
check out similar devices to see what I might have done in them in you
can't see what's going on..
for a laugh compare conf.c conf.h defore and after... :)
I have not doen DEVFS entries for any DISKSLICE devices yet as that will be
a much more complicated job.. (pass 5 :)

pass 4 will be to make the devsw tables of type (cdevsw * )
rather than (cdevsw)
seems to work here..
complaints to the usual places.. :)
1995-12-08 11:19:42 +00:00
David Greenman
efeaf95a41 Untangled the vm.h include file spaghetti. 1995-12-07 12:48:31 +00:00
Bruce Evans
0ff89d5b03 Moved the declaration of vm_object_t from <vm/vm.h> to <sys/types.h>
(if KERNEL is defined).  This allows removing the #includes of vm
stuff in vnode_if.h, which will speed up the compilation of LINT by
about 5%.
1995-12-05 20:54:42 +00:00
Poul-Henning Kamp
946bb7a268 A major sweep over the sysctl stuff.
Move a lot of variables home to their own code (In good time before xmas :-)

Introduce the string descrition of format.

Add a couple more functions to poke into these marvels, while I try to
decide what the correct interface should look like.

Next is adding vars on the fly, and sysctl looking at them too.

Removed a tine bit of defunct and #ifdefed notused code in swapgeneric.
1995-12-04 16:48:58 +00:00
Bruce Evans
f31d402c5a Fixed the type mismatch in check for the bogus mmap function `nullop'.
The test should never succeed and should go away.  Temporarily print
a warning if it does succeed.
1995-12-03 18:59:55 +00:00
Bruce Evans
cac597e4c0 Completed function declarations and/or added prototypes.
Staticized some functions.

__purified some functions.  Some functions were bogusly declared as
returning `const'.  This hasn't done anything since gcc-2.5.  For
later versions of gcc, the equivalent is __attribute__((const)) at
the end of function declarations.
1995-12-03 12:18:39 +00:00
Bruce Evans
d841aaa740 Finished (?) cleaning up sysinit stuff. 1995-12-02 17:11:20 +00:00
Julian Elischer
7198bf4725 If you're going to mechanically replicate something in 50 files
it's best to not have a (compiles cleanly) typo in it! (sigh)
1995-11-29 14:41:20 +00:00
Julian Elischer
53ac6efbd8 OK, that's it..
That's EVERY SINGLE driver that has an entry in conf.c..
my next trick will be to define cdevsw[] and bdevsw[]
as empty arrays and remove all those DAMNED defines as well..

Each of these drivers has a SYSINIT linker set entry
that comes in very early.. and asks teh driver to add it's own
entry to the two devsw[] tables.

some slight reworking of the commits from yesterday (added the SYSINIT
stuff and some usually wrong but token DEVFS entries to all these
devices.

BTW does anyone know where the 'ata' entries in conf.c actually reside?
seems we don't actually have a 'ataopen() etc...

If you want to add a new device in conf.c
please  make sure I know
so I can keep it up to date too..

as before, this is all dependent on #if defined(JREMOD)
(and #ifdef DEVFS in parts)
1995-11-29 10:49:16 +00:00
Bruce Evans
512fef80a9 Completed function declarations and/or added prototypes. 1995-11-21 12:55:26 +00:00
Poul-Henning Kamp
3af768903d Remove unused vars & funcs, make things static, protoize a little bit. 1995-11-20 12:20:02 +00:00
Bruce Evans
ff98689d2d Fixed recent staticizations. Some protypes for static functions were
left in headers and not staticized.
1995-11-16 09:51:22 +00:00
Poul-Henning Kamp
f5a1271110 staticize. 1995-11-14 20:53:20 +00:00
Poul-Henning Kamp
a9ad941c7a Move all the VM sysctl stuff home where it belongs. 1995-11-14 09:29:34 +00:00
David Greenman
f81b859263 Fixed up a comment and removed some #if 0'd code. 1995-11-13 10:53:38 +00:00
David Greenman
bea41bcf3a Moved vm_map_lock call to inside the splhigh protection in vm_map_find().
This closes a probably rare but nonetheless real window that would result
in a process hanging or the system panicing.

Reviewed by:	dyson, davidg
Submitted by:	kato@eclogite.eps.nagoya-u.ac.jp (KATO Takenori)
1995-11-12 08:58:58 +00:00
Bruce Evans
d2d3e8751c Included <sys/sysproto.h> to get central declarations for syscall args
structs and prototypes for syscalls.

Ifdefed duplicated decentralized declarations of args structs.  It's
convenient to have this visible but they are hard to maintain.  Some
are already different from the central declarations.  4.4lite2 puts
them in comments in the function headers but I wanted to avoid the
large changes for that.
1995-11-12 06:43:28 +00:00
Bruce Evans
9ed1bde79c Fixed type of obreak(). The args struct member name conflicted with
the (better) machine generated one in <sys/sysproto.h>.
1995-11-11 06:40:35 +00:00
David Greenman
42e9339d38 Initialize lock struct entries explicitly rather than calling bzero(). 1995-11-06 08:44:15 +00:00
Bruce Evans
8b25681eb5 Replaced bogus macros for dummy devswitch entries by functions.
These functions went away:

	enosys (hasn't been used for some time)
	enxio
	enodev
	enoioctl (was used only once, actually for a vop)

if_tun.c:
Continued cleaning up...

conf.h:
Probably fixed the type of d_reset_t.  It is hard to tell the correct
type because there are no non-dummy device reset functions.

Removed last vestige of ambiguous sleep message strings.
1995-11-06 00:36:19 +00:00
John Dyson
aef922f514 Greatly simplify the msync code. Eliminate complications in vm_pageout
for msyncing.  Remove a bug that manifests itself primarily on NFS
(the dirty range on the buffers is not set on msync.)
1995-11-05 20:46:03 +00:00
David Greenman
23922ccaca Move page fixups (pmap_clear_modify, etc) that happen after paging input
completes out of vm_fault and into the pagers. This get rid of some
redundancy and improves the architecture.

Reviewed by:	John Dyson <dyson>
1995-11-02 06:42:47 +00:00
Bruce Evans
0b8253a746 Don't pass an extra trailing arg to some functions.
Added the prototypes that found this bug.
1995-10-30 17:56:30 +00:00
John Dyson
a91c5a7ecd Get rid of machine-dependent NBPG and replace with PAGE_SIZE. 1995-10-23 05:35:48 +00:00
John Dyson
d559b36913 Remove of now unused PG_COPYONWRITE. 1995-10-23 04:29:39 +00:00
John Dyson
e17bed1226 First phase of removing the PG_COPYONWRITE flag, and an architectural
cleanup of mapping files.
1995-10-23 03:49:43 +00:00
John Dyson
2c4488fce3 Finalize GETPAGES layering scheme. Move the device GETPAGES
interface into specfs code.  No need at this point to modify the
PUTPAGES stuff except in the layered-type (NULL/UNION) filesystems.
1995-10-23 02:23:29 +00:00
John Dyson
02c04a2f6c Implement mincore system call. 1995-10-21 17:42:28 +00:00
David Greenman
eed2d59b25 Fix initialization of "bsize" in vnode_pager_haspage(). It must happen
after the check for the mount point still existing or else the system
will panic if someone forcibly unmounted the filesystem.
1995-10-19 21:35:03 +00:00
John Dyson
0bc3a91127 Remove an unnecessary tsleep in the swapin code. This tsleep
can defer swapping in processes and is just not the right thing to do.
1995-10-16 05:45:49 +00:00
David Greenman
cd41fc123e Fix argument passing to the "freeer" routine. Added some prototypes. (bde)
Moved extern declaration of swap_pager_full into swap_pager.h and out of
the various files that reference it. (davidg)

Submitted by:	bde & davidg
1995-10-07 19:02:56 +00:00
Poul-Henning Kamp
a5eb0e27e4 Avoid a 64bit divide. 1995-10-06 09:42:11 +00:00
Bruce Evans
f2caf6f548 Fix pollution of application namespace by declarations of kernel
functions.  The application header <sys/user.h> includes <vm/vm.h>
which includes <vm/lock.h>...

vm.h:
Don't include <machine/cpufunc.h>.  It is already included by
<sys/systm.h> in the kernel and isn't designed to be included by
applications (the 2.1 version causes a syntax error in C++ and the
current version has initializers that are invalid in strict C++).

lock.h:
Only declare kernel functions if KERNEL is defined.
1995-10-05 01:11:48 +00:00
John Dyson
0e27a03ffb Perform more checking for proper loading of the UPAGES when a process
is swapped in.  Also, remove unnecessary map locking/unlocking during
selection of processes to be swapped out.

This code might afford proper panics as opposed to spontaneous reboots
on certain systems.  This should allow us to debug these problems better.
1995-09-24 19:51:50 +00:00
John Dyson
22ba64e870 Significantly simplify the fault clustering code. After some analysis by
David Greenman, it has been determined that the more sophisticated code
only made a very minor difference in fault performance.  Therefore, this
code eliminates some of the complication of the fault code, decreasing
the amount of CPU used to scan shadow chains.
1995-09-24 19:47:58 +00:00
David Greenman
2f82e60467 Check that the swap block is valid before including it in a cluster.
Submitted by:	John Dyson
1995-09-24 04:40:19 +00:00
David Greenman
c87ad37e61 Check the return value from vm_map_pageable() when mapping the process's
UPAGES and associated page table page. Panic on error. This is less than
optimial and will be fixed in the future, but is better than the old
behavior of panicing with a "kernel page directory invalid" in pmap_enter.
1995-09-17 01:46:03 +00:00
John Dyson
21bf390418 Fixed a typo in vm_fault_additional_pages. 1995-09-14 13:42:52 +00:00
John Dyson
6eab77f210 Fix really bogus casting of a block number to a long. Also change the
comparison from a "< 0" to "== -1" like it should be.
1995-09-12 14:42:43 +00:00
John Dyson
894048d71c Make sure that the prezero flag is cleared when needed. 1995-09-11 00:47:17 +00:00
John Dyson
b1fc01b7da Fix an error that can cause attempted reading beyond the end of file. 1995-09-11 00:46:19 +00:00
John Dyson
ccbb2f7226 Code cleanup and minor performance improvement in the faultin cluster
code.
1995-09-11 00:45:15 +00:00
David Greenman
4590fd3a2a Fixed init functions argument type - caddr_t -> void *. Fixed a couple of
compiler warnings.
1995-09-09 18:10:37 +00:00
John Dyson
ca56715fd8 Fixed a sign reversal problem -- might have cause some Sig-11s that
people have been seeing.
1995-09-06 07:08:45 +00:00
John Dyson
ced399ee65 Minor performance improvements, additional prototype for additional
exported symbol.
1995-09-06 05:37:43 +00:00
John Dyson
170db9c63a Allow the fault code to use additional clustering info from both
bmap and the swap pager.  Improved fault clustering performance.
1995-09-04 04:44:26 +00:00
John Dyson
c83ebe7781 Added VOP_GETPAGES/VOP_PUTPAGES and also the "backwards" block count
for VOP_BMAP.  Updated affected filesystems...
1995-09-04 00:21:16 +00:00
John Dyson
f70f05f238 Machine independent changes to support pre-zeroed free pages. This
significantly improves demand-zero performance.
1995-09-03 20:40:43 +00:00
John Dyson
10ad4d483c Added prototype for new routine "vm_page_set_validclean" and initial
declarations for the prezeroed pages mechanism.
1995-09-03 20:11:26 +00:00
John Dyson
4589a4b5dc New subroutine "vm_page_set_validclean" for a vfs_bio improvement. 1995-09-03 19:57:25 +00:00
Julian Elischer
2b14f991e6 Reviewed by: julian with quick glances by bruce and others
Submitted by:	terry (terry lambert)
This is  a composite of 3 patch sets submitted by terry.
they are:
New low-level init code that supports loadbal modules better
some cleanups in the namei code to help terry in 16-bit character support
some changes to the mount-root code to make it a little more
modular..

NOTE: mounting root off cdrom or NFS MIGHT be broken as I haven't been able
to test those cases..

certainly mounting root of disk still works just fine..
mfs should work but is untested. (tomorrows task)

The low level init stuff includes a total rewrite of init_main.c
to make it possible for new modules to have an init phase by simply
adding an entry to a TEXT_SET (or is it DATA_SET) list. thus a new module can
be added to the kernel without editing any other files other than the
'files' file.
1995-08-28 09:19:25 +00:00
Bruce Evans
187f023877 Change vm_object_print() to have the correct number and type of args
for a ddb command.
1995-08-26 23:19:49 +00:00
Bruce Evans
914181e7de Change vm_map_print() to have the correct number and type of args for
a ddb command.
1995-08-26 23:18:38 +00:00
Bruce Evans
bf25be48a5 Make everything except the unsupported network sources compile cleanly
with -Wnested-externs.
1995-08-16 16:14:28 +00:00
Bruce Evans
28f8db1403 Eliminate sloppy common-style declarations. There should be none left for
the LINT configuation.
1995-07-29 11:44:31 +00:00
David Greenman
b367ddb191 #if 0'd one of the DIAGNOSTIC checks in vm_page_alloc(). It was too
expensive for "normal" use.
1995-07-20 05:28:07 +00:00
David Greenman
2a4895f4bb 1) Merged swpager structure into vm_object.
2) Changed swap_pager internal interfaces to cope w/#1.
3) Eliminated object->copy as we no longer have copy objects.
4) Minor stylistic changes.
1995-07-16 13:28:37 +00:00
David Greenman
588d138f56 Added a copyright to this file. 1995-07-13 10:29:34 +00:00
David Greenman
2541af0bba Oops, forgot to add the "default" pager files...
NOTE: libkvm, w, ps, 'top', and any other utility which depends on struct
      proc or any VM system structure will have to be rebuilt!!!

Much needed overhaul of the VM system. Included in this first round of
changes:

1) Improved pager interfaces: init, alloc, dealloc, getpages, putpages,
   haspage, and sync operations are supported. The haspage interface now
   provides information about clusterability. All pager routines now take
   struct vm_object's instead of "pagers".

2) Improved data structures. In the previous paradigm, there is constant
   confusion caused by pagers being both a data structure ("allocate a
   pager") and a collection of routines. The idea of a pager structure has
   escentially been eliminated. Objects now have types, and this type is
   used to index the appropriate pager. In most cases, items in the pager
   structure were duplicated in the object data structure and thus were
   unnecessary. In the few cases that remained, a un_pager structure union
   was created in the object to contain these items.

3) Because of the cleanup of #1 & #2, a lot of unnecessary layering can now
   be removed. For instance, vm_object_enter(), vm_object_lookup(),
   vm_object_remove(), and the associated object hash list were some of the
   things that were removed.

4) simple_lock's removed. Discussion with several people reveals that the
   SMP locking primitives used in the VM system aren't likely the mechanism
   that we'll be adopting. Even if it were, the locking that was in the code
   was very inadequate and would have to be mostly re-done anyway. The
   locking in a uni-processor kernel was a no-op but went a long way toward
   making the code difficult to read and debug.

5) Places that attempted to kludge-up the fact that we don't have kernel
   thread support have been fixed to reflect the reality that we are really
   dealing with processes, not threads. The VM system didn't have complete
   thread support, so the comments and mis-named routines were just wrong.
   We now use tsleep and wakeup directly in the lock routines, for instance.

6) Where appropriate, the pagers have been improved, especially in the
   pager_alloc routines. Most of the pager_allocs have been rewritten and
   are now faster and easier to maintain.

7) The pagedaemon pageout clustering algorithm has been rewritten and
   now tries harder to output an even number of pages before and after
   the requested page. This is sort of the reverse of the ideal pagein
   algorithm and should provide better overall performance.

8) Unnecessary (incorrect) casts to caddr_t in calls to tsleep & wakeup
   have been removed. Some other unnecessary casts have also been removed.

9) Some almost useless debugging code removed.

10) Terminology of shadow objects vs. backing objects straightened out.
    The fact that the vm_object data structure escentially had this
    backwards really confused things. The use of "shadow" and "backing
    object" throughout the code is now internally consistent and correct
    in the Mach terminology.

11) Several minor bug fixes, including one in the vm daemon that caused
    0 RSS objects to not get purged as intended.

12) A "default pager" has now been created which cleans up the transition
    of objects to the "swap" type. The previous checks throughout the code
    for swp->pg_data != NULL were really ugly. This change also provides
    the rudiments for future backing of "anonymous" memory by something
    other than the swap pager (via the vnode pager, for example), and it
    allows the decision about which of these pagers to use to be made
    dynamically (although will need some additional decision code to do
    this, of course).

13) (dyson) MAP_COPY has been deprecated and the corresponding "copy
    object" code has been removed. MAP_COPY was undocumented and non-
    standard. It was furthermore broken in several ways which caused its
    behavior to degrade to MAP_PRIVATE. Binaries that use MAP_COPY will
    continue to work correctly, but via the slightly different semantics
    of MAP_PRIVATE.

14) (dyson) Sharing maps have been removed. It's marginal usefulness in a
    threads design can be worked around in other ways. Both #12 and #13
    were done to simplify the code and improve readability and maintain-
    ability. (As were most all of these changes)

TODO:

1) Rewrite most of the vnode pager to use VOP_GETPAGES/PUTPAGES. Doing
   this will reduce the vnode pager to a mere fraction of its current size.

2) Rewrite vm_fault and the swap/vnode pagers to use the clustering
   information provided by the new haspage pager interface. This will
   substantially reduce the overhead by eliminating a large number of
   VOP_BMAP() calls. The VOP_BMAP() filesystem interface should be
   improved to provide both a "behind" and "ahead" indication of
   contiguousness.

3) Implement the extended features of pager_haspage in swap_pager_haspage().
   It currently just says 0 pages ahead/behind.

4) Re-implement the swap device (swstrategy) in a more elegant way, perhaps
   via a much more general mechanism that could also be used for disk
   striping of regular filesystems.

5) Do something to improve the architecture of vm_object_collapse(). The
   fact that it makes calls into the swap pager and knows too much about
   how the swap pager operates really bothers me. It also doesn't allow
   for collapsing of non-swap pager objects ("unnamed" objects backed by
   other pagers).
1995-07-13 10:15:03 +00:00
David Greenman
24a1cce34f NOTE: libkvm, w, ps, 'top', and any other utility which depends on struct
proc or any VM system structure will have to be rebuilt!!!

Much needed overhaul of the VM system. Included in this first round of
changes:

1) Improved pager interfaces: init, alloc, dealloc, getpages, putpages,
   haspage, and sync operations are supported. The haspage interface now
   provides information about clusterability. All pager routines now take
   struct vm_object's instead of "pagers".

2) Improved data structures. In the previous paradigm, there is constant
   confusion caused by pagers being both a data structure ("allocate a
   pager") and a collection of routines. The idea of a pager structure has
   escentially been eliminated. Objects now have types, and this type is
   used to index the appropriate pager. In most cases, items in the pager
   structure were duplicated in the object data structure and thus were
   unnecessary. In the few cases that remained, a un_pager structure union
   was created in the object to contain these items.

3) Because of the cleanup of #1 & #2, a lot of unnecessary layering can now
   be removed. For instance, vm_object_enter(), vm_object_lookup(),
   vm_object_remove(), and the associated object hash list were some of the
   things that were removed.

4) simple_lock's removed. Discussion with several people reveals that the
   SMP locking primitives used in the VM system aren't likely the mechanism
   that we'll be adopting. Even if it were, the locking that was in the code
   was very inadequate and would have to be mostly re-done anyway. The
   locking in a uni-processor kernel was a no-op but went a long way toward
   making the code difficult to read and debug.

5) Places that attempted to kludge-up the fact that we don't have kernel
   thread support have been fixed to reflect the reality that we are really
   dealing with processes, not threads. The VM system didn't have complete
   thread support, so the comments and mis-named routines were just wrong.
   We now use tsleep and wakeup directly in the lock routines, for instance.

6) Where appropriate, the pagers have been improved, especially in the
   pager_alloc routines. Most of the pager_allocs have been rewritten and
   are now faster and easier to maintain.

7) The pagedaemon pageout clustering algorithm has been rewritten and
   now tries harder to output an even number of pages before and after
   the requested page. This is sort of the reverse of the ideal pagein
   algorithm and should provide better overall performance.

8) Unnecessary (incorrect) casts to caddr_t in calls to tsleep & wakeup
   have been removed. Some other unnecessary casts have also been removed.

9) Some almost useless debugging code removed.

10) Terminology of shadow objects vs. backing objects straightened out.
    The fact that the vm_object data structure escentially had this
    backwards really confused things. The use of "shadow" and "backing
    object" throughout the code is now internally consistent and correct
    in the Mach terminology.

11) Several minor bug fixes, including one in the vm daemon that caused
    0 RSS objects to not get purged as intended.

12) A "default pager" has now been created which cleans up the transition
    of objects to the "swap" type. The previous checks throughout the code
    for swp->pg_data != NULL were really ugly. This change also provides
    the rudiments for future backing of "anonymous" memory by something
    other than the swap pager (via the vnode pager, for example), and it
    allows the decision about which of these pagers to use to be made
    dynamically (although will need some additional decision code to do
    this, of course).

13) (dyson) MAP_COPY has been deprecated and the corresponding "copy
    object" code has been removed. MAP_COPY was undocumented and non-
    standard. It was furthermore broken in several ways which caused its
    behavior to degrade to MAP_PRIVATE. Binaries that use MAP_COPY will
    continue to work correctly, but via the slightly different semantics
    of MAP_PRIVATE.

14) (dyson) Sharing maps have been removed. It's marginal usefulness in a
    threads design can be worked around in other ways. Both #12 and #13
    were done to simplify the code and improve readability and maintain-
    ability. (As were most all of these changes)

TODO:

1) Rewrite most of the vnode pager to use VOP_GETPAGES/PUTPAGES. Doing
   this will reduce the vnode pager to a mere fraction of its current size.

2) Rewrite vm_fault and the swap/vnode pagers to use the clustering
   information provided by the new haspage pager interface. This will
   substantially reduce the overhead by eliminating a large number of
   VOP_BMAP() calls. The VOP_BMAP() filesystem interface should be
   improved to provide both a "behind" and "ahead" indication of
   contiguousness.

3) Implement the extended features of pager_haspage in swap_pager_haspage().
   It currently just says 0 pages ahead/behind.

4) Re-implement the swap device (swstrategy) in a more elegant way, perhaps
   via a much more general mechanism that could also be used for disk
   striping of regular filesystems.

5) Do something to improve the architecture of vm_object_collapse(). The
   fact that it makes calls into the swap pager and knows too much about
   how the swap pager operates really bothers me. It also doesn't allow
   for collapsing of non-swap pager objects ("unnamed" objects backed by
   other pagers).
1995-07-13 08:48:48 +00:00
David Greenman
6306c897b5 swapout_threads() -> swapout_procs(). 1995-07-10 08:53:22 +00:00
David Greenman
f2daac0c69 Increased global RSS limit to total RAM. 1995-07-10 08:48:58 +00:00
David Greenman
06cb725951 Moved call to VOP_GETATTR() out of vnode_pager_alloc() and into the places
that call vnode_pager_alloc() so that a failure return can be dealt with.
This fixes a panic seen on NFS clients when a file being opened is deleted
on the server before the open completes.
1995-07-09 06:58:03 +00:00
David Greenman
39d38f93e2 Fixed an object allocation race condition that was causing a "object
deallocated too many times" panic when using NFS.

Reviewed by:	John Dyson
1995-07-06 11:48:48 +00:00
David Greenman
aa2cabb958 1) Converted v_vmdata to v_object.
2) Removed unnecessary vm_object_lookup()/pager_cache(object, TRUE) pairs
   after vnode_pager_alloc() calls - the object is already guaranteed to be
   persistent.
3) Removed some gratuitous casts.
1995-06-28 12:01:13 +00:00
Rodney W. Grimes
d3628763db Merge RELENG_2_0_5 into HEAD 1995-06-11 19:33:05 +00:00
Rodney W. Grimes
9b2e535452 Remove trailing whitespace. 1995-05-30 08:16:23 +00:00
David Greenman
956e9ca5a0 Removed check for sw_dev == NODEV; this is a normal condition for swap
over NFS and was gratuitously panicing when it happens.

Reviewed by:	John Dyson
Submitted by:	Pierre Beyssac via Poul-Henning Kamp
1995-05-25 03:38:11 +00:00
David Greenman
61f5d51062 Changes to fix the following bugs:
1) Files weren't properly synced on filesystems other than UFS. In some
   cases, this lead to lost data. Most likely would be noticed on NFS.
   The fix is to make the VM page sync/object_clean general rather than
   in each filesystem.
2) Mixing regular and mmaped file I/O on NFS was very broken. It caused
   chunks of files to end up as zeroes rather than the intended contents.
   The fix was to fix several race conditions and to kludge up the
   "b_dirtyoff" and "b_dirtyend" that NFS relies upon - paying attention
   to page modifications that occurred via the mmapping.

Reviewed by:	David Greenman
Submitted by:	John Dyson
1995-05-21 21:39:31 +00:00
David Greenman
2976b7f19f NFS diskless operation was broken because swapdev_vp wasn't initialized.
These changes solve the problem in a general way by moving the
initialization out of the individual fs_mountroot's and into swaponvp().

Submitted by:	Poul-Henning Kamp
1995-05-19 03:27:08 +00:00
David Greenman
0e5b52868f Fixed a bug that managed to slip in during Poul's dynamic swap partition
changes. The check for nswap was bogus, but the code was so convoluted
that it was difficult to tell. It's better now. :-)

Reviewed by:	David Greenman (extensively), and John Dyson
Submitted by:	Poul-Henning Kamp, w/tweaks by me.
1995-05-18 05:09:54 +00:00
David Greenman
5f55e84104 Accessing pages beyond the end of a mapped file results in internal
inconsistencies in the VM system that eventually lead to a panic. These
changes fix the behavior to conform to the behavior in SunOS, which is
to deny faults to pages beyond the EOF (returning SIGBUS). Internally,
this is implemented by requiring faults to be within the object size
boundaries. These changes exposed another bug, namely that passing in
an offset to mmap when trying to map an unnamed anonymous region also
results in internal inconsistencies. In this case, the offset is forced
to zero.

Reviewed by:	John Dyson and others
1995-05-18 02:59:26 +00:00
David Greenman
a401ebbe32 Changed swap partition handling/allocation so that it doesn't
require specific partitions be mentioned in the kernel config
file ("swap on foo" is now obsolete).

From Poul-Henning:

The visible effect is this:

As default, unless
        options "NSWAPDEV=23"
is in your config, you will have four swap-devices.
You can swapon(2) any block device you feel like, it doesn't have
to be in the kernel config.

There is a performance/resource win available by getting the NSWAPDEV right
(but only if you have just one swap-device ??), but using that as default
would be too restrictive.

The invisible effect is that:

Swap-handling disappears from the $arch part of the kernel.
It gets a lot simpler (-145 lines) and cleaner.

Reviewed by:	John Dyson, David Greenman
Submitted by:	Poul-Henning Kamp, with minor changes by me.
1995-05-14 03:00:10 +00:00
Poul-Henning Kamp
8af2296bcb I'm about to jump on the swap-initialization, and having talked
with davidg about it, I hereby kill two undocumented misfeatures:
The code to skip a miniroot in the swapdev is not particular useful, and
if we need it we need it to be done properly, ie size the fs and skip all
of it not some hardcoded size, and subtract what we skip from the length
in the first place.
The SEQSWAP dies too.  It's not the way to do it, it doesn't work, and
nobody have expressed any great desire for it to work.  The way to
implement it correctly would be a second argument to swapon(2) to give
a priority/policy information.  Low priority swapdevs can be made so
by adding them at a far offset (0x80000000 kind of thing), with almost no
modification to the strategy routine (in particular a offset per swapdev).
But until the need is obvious, it will not be done.
1995-05-12 03:54:59 +00:00
David Greenman
ee3a64c992 Changed "handle" from type caddr_t to void *; "handle" is several different
types of pointers, and "char *" is a bad choice for the type.
1995-05-10 18:56:09 +00:00
John Dyson
11fda60b40 Another error in the correction for trimming swap allocation for
small objects.  (This code needs to be revisited.)
1995-05-07 06:36:59 +00:00
John Dyson
85b67b98fd Fixed a calculation that would once-in-a-while cause the swap_pager
to emit spurious page outside of object type messages.  It is not
a fatal condition anyway, so the message will be omitted for
release.  Also, the code that "clips" the allocation size, associated
with the above problem, was fixed.
1995-05-07 03:48:54 +00:00
David Greenman
f325917a65 Changed object hash list to be a list rather than a tailq. This saves
space for the hash list buckets and is a little faster. The features
of tailq aren't needed. Increased the size of the object hash table
to improve performance. In the future, this will be changed so that
the table is sized dynamically.
1995-05-02 05:57:11 +00:00
David Greenman
3fc3004e03 Fixed a "bswbuf" hang caused by the wakeup in relpbuf() waking up the
wrong thing.
1995-04-25 06:22:47 +00:00
Bruce Evans
7666fb4753 inline -> __inline.
Headers should always use `__inline' for inline functions to avoid
syntax errors when modules that don't even use the offending functions
are compiled with `gcc -ansi'.
1995-04-23 08:05:49 +00:00
John Dyson
7e15fd2757 Fixed a problem in _vm_object_page_clean that could cause an
infinite loop.
1995-04-21 02:48:40 +00:00
David Greenman
aba8f38eb9 New flag: B_PAGING. Added as part of the vn driver hack. 1995-04-19 10:32:11 +00:00
David Greenman
4c1f8ee9b8 Fixed a logic bug that caused the vmdaemon to not wake up when intended.
Submitted by:	John Dyson
1995-04-17 10:00:55 +00:00
David Greenman
7c0414d074 Removed obsolete/unused variable declarations. Killed externs and included
appropriate include files.
1995-04-16 14:12:15 +00:00
David Greenman
64abb5a522 Removed obsolete/unused variable declarations.
Removed some extern declarations and included the correct include files.
1995-04-16 13:58:42 +00:00
David Greenman
c3cb3e12f5 Moved some zero-initialized variables into .bss. Made code intended to be
called only from DDB #ifdef DDB. Removed some completely unused globals.
1995-04-16 12:56:22 +00:00
David Greenman
8c3d9c403c Removed gratuitous m->blah=0 assignments when initializing the vm_page
structs in vm_page_startup(). The vm_page structs are already completely
zeroed.
1995-04-16 11:50:45 +00:00
David Greenman
2fdccd5e8c Make "print_page_info" #ifdef DDB. 1995-04-16 09:59:16 +00:00
David Greenman
ec4f9fb0ee Fixed a few bugs in vm_object_page_clean, mostly related to not syncing
pages that are in FS buffers. This fixes the (believed to already have been
fixed) problem with msync() not doing it's job...in other words, the
stuff that Andrew has continuously been complaining about.

Submitted by:	John Dyson, w/minor changes by me.
1995-04-16 03:11:28 +00:00
David Greenman
f6b04d2bfb Changes from John Dyson and myself:
Fixed remaining known bugs in the buffer IO and VM system.

vfs_bio.c:
Fixed some race conditions and locking bugs. Improved performance
by removing some (now) unnecessary code and fixing some broken
logic.
Fixed process accounting of # of FS outputs.
Properly handle NFS interrupts (B_EINTR).

(various)
Replaced calls to clrbuf() with calls to an optimized routine
called vfs_bio_clrbuf().

(various FS sync)
Sync out modified vnode_pager backed pages.

ffs_vnops.c:
Do two passes: Sync out file data first, then indirect blocks.

vm_fault.c:
Fixed deadly embrace caused by acquiring locks in the wrong order.

vnode_pager.c:
Changed to use buffer I/O system for writing out modified pages. This
should fix the problem with the modification date previous not getting
updated. Also dramatically simplifies the code. Note that this is
going to change in the future and be implemented via VOP_PUTPAGES().

vm_object.c:
Fixed a pile of bugs related to cleaning (vnode) objects. The performance
of vm_object_page_clean() is terrible when dealing with huge objects,
but this will change when we implement a binary tree to keep the object
pages sorted.

vm_pageout.c:
Fixed broken clustering of pageouts. Fixed race conditions and other
lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00
Bruce Evans
3aa12267a5 Add and move declarations to fix all of the warnings from `gcc -Wimplicit'
(except in netccitt, netiso and netns) that I didn't notice when I fixed
"all" such warnings before.
1995-03-28 07:58:53 +00:00
David Greenman
17c4c40800 Fixed typo...using wrong variable in page_shortage calculation. 1995-03-28 05:58:35 +00:00
David Greenman
0bb3a0d2f4 Fixed "pages freed by daemon" statistic (again). 1995-03-28 05:35:29 +00:00
David Greenman
2ddba2158c Explicitly set page dirty if this is a write fault - reduces calls to
pmap_is_modified() later.
1995-03-27 02:41:00 +00:00
David Greenman
8953151fc8 Removed some obsolete flags.
Submitted by:   John Dyson
1995-03-26 23:33:14 +00:00
David Greenman
6c534ad868 Fix logic bug I just introduced with the flags to msync(). 1995-03-25 17:44:03 +00:00
David Greenman
492d73bd99 Pass syncio flag to vm_object_clean(). It remains unimplemented, however. 1995-03-25 17:36:57 +00:00
David Greenman
1e62bc6357 Disallow both MS_ASYNC and MS_INVALIDATE flags being set at the same time
in msync().
1995-03-25 17:36:00 +00:00
David Greenman
e6c6af1149 Added "flags" argument to msync, and implemented MS_ASYNC and MS_INVALIDATE.
The MS_ASYNC flag doesn't current work, and MS_INVALIDATE will only toss out
the pages in the address space (not all pages in the shadow chain).
1995-03-25 16:55:46 +00:00
David Greenman
7fd9a8b12f Implemented cnt.v_reactivated and moved vm_page_activate() routine to
before vm_page_deactivate().
1995-03-25 08:47:35 +00:00
David Greenman
260295f979 Removed (almost) meaningless "object cache lookups/hits" statistic. In
our framework, these numbers will usually be nearly the same, and not
because of any sort of high 'hit rate'.
1995-03-25 08:42:14 +00:00
David Greenman
a34297995a Removed cnt.v_nzfod: In our current scheme of things it is not possible
to accurately track this. It isn't an indicator of resource consumption
anyway.
Removed cnt.v_kernel_pages: We don't implement this and doing so accurately
would be very difficult (and ambiguous - since process pages are often
double mapped in the kernel and the process address spaces).
1995-03-25 06:09:33 +00:00
David Greenman
d7a0fc9360 Fixed warning caused by returning a value in a void function (introduced
in a recent commit by me). Relaxed checks before calling vm_object_remove;
a non-internal object always has a pager.
1995-03-23 05:19:44 +00:00
David Greenman
f5cf85d40a Removed unused fifth argument to vm_object_page_clean(). Fixed bug with
VTEXT not always getting cleared when it is supposed to. Added check to
make sure that vm_object_remove() isn't called with a NULL pager or for
a pager for an OBJ_INTERNAL object (neither of which will be on the hash
list). Clear OBJ_CANPERSIST if we decide to terminate it because of no
resident pages.
1995-03-22 12:24:11 +00:00
David Greenman
563128e49e Fixed potential sleep/wakeup race conditional with splhigh().
Submitted by:	John Dyson
1995-03-22 08:08:44 +00:00
David Greenman
c419d77e29 Added a check for wrong object size; print a warning, but deal with it
correctly. The warning will tell us that there is a bug somewhere else
in sizing the object correctly.

Submitted by:	John Dyson
1995-03-22 05:12:18 +00:00
David Greenman
8f4e17d4b2 Fixed bug in vm_mmap() where the object that is created in some cases
was the wrong size. This is the likely cause of panics reported by
Lars Fredriksen and Paul Richards related to a -1 blkno when paging
via the swap_pager.

Submitted by:	John Dyson
1995-03-22 05:08:41 +00:00
David Greenman
1b369d9858 Removed unused variable declaration missed in previous commit. 1995-03-21 23:09:53 +00:00
David Greenman
71263bf8db Removed do-nothing VOP_UPDATE() call. 1995-03-21 23:07:19 +00:00
David Greenman
bc9ad24756 Disallow non page-aligned file offsets in vm_mmap(). We don't support this
in either the high or low level parts of the VM system. Just return EINVAL
in this case, just like SunOS does.
1995-03-21 10:15:52 +00:00
David Greenman
fbcfcdf7db Fixed bug in the size == 0 case of msync() caused by a bogus return value
check..
1995-03-21 02:54:04 +00:00
David Greenman
7c1f6cedec Added a new boolean argument to vm_object_page_clean that causes it to
only toss out clean pages if TRUE.
1995-03-21 01:11:44 +00:00
David Greenman
0426122fa0 Don't gain/lose an object reference in vnode_pager_setsize(). It will
cause vnode locking problems in vm_object_terminate().
Implement proper vnode locking in vm_object_terminate().
1995-03-20 10:20:42 +00:00
David Greenman
9ba0a56334 Fixed "objde1" hang. It was caused by a "&" where an "&&" belonged in the
expression that decides if a wakeup should occur.
1995-03-20 10:14:55 +00:00
David Greenman
83edfd47d8 Removed an unnecessary call to vinvalbuf after the page clean. 1995-03-20 02:06:09 +00:00
David Greenman
0bdb7528db Do proper vnode locking when doing paging I/O. Removed the asynchronous
paging capability to facilitate this (we saw little or no measureable
improvement with this anyway).

Submitted by:	John Dyson
1995-03-19 23:46:25 +00:00
David Greenman
edf8a81561 Removed redundant newlines that were in some panic strings. 1995-03-19 14:29:26 +00:00
David Greenman
c01a9b8ca0 Incorporated 4.4-lite vnode_pager_uncache() and vnode_pager_umount()
routines (and merged local changes). The changed vnode_pager_uncache
gets rids of the bogosity that you can call the routine without
having the vnode locked. The changed vnode_pager_umount properly locks
the vnode before calling vnode_pager_uncache.
1995-03-19 12:36:10 +00:00
David Greenman
806e386083 In vm_page_alloc_contig: Removed a redundant semicolon and used 'm' instead
of &pga[i] in one place.
1995-03-18 01:04:36 +00:00
Bruce Evans
b5e8ce9f12 Add and move declarations to fix all of the warnings from `gcc -Wimplicit'
(except in netccitt, netiso and netns) and most of the warnings from
`gcc -Wnested-externs'.  Fix all the bugs found.  There were no serious
ones.
1995-03-16 18:17:34 +00:00
David Greenman
5eb7d0cd2e Special cased the handling of mb_map in the M_WAITOK case. kmem_malloc()
now returns NULL and sets a global 'mb_map_full' when the map is full.
m_clalloc() has further been taught to expect this and do the right thing.
This should fix the "mb_map full" panics that several people have reported.
1995-03-15 07:52:06 +00:00
Bruce Evans
1475d9283e Move a kernel inline function inside `#ifdef KERNEL' so that including
<vm/vm.h> doesn't cause warnings about nonexistent functions called
by the inline function.  Clean up the formatting of the function.
1995-03-12 12:42:00 +00:00
David Greenman
c4ed5a07a5 Fixed obsolete comment. 1995-03-12 08:11:34 +00:00
David Greenman
6879ca0fcd Deleted vm_object_setpager(). 1995-03-12 08:05:46 +00:00
David Greenman
61ca29b0e5 Deleted vm_object_setpager(). 1995-03-12 07:58:29 +00:00
David Greenman
4bb624613f Explicitly set object->flags = OBJ_CANPERSIST. 1995-03-12 07:56:06 +00:00
David Greenman
a14e8fd032 Fix completely bogus comment. 1995-03-11 22:25:20 +00:00
David Greenman
63635f5ae2 Clear OBJ_INTERNAL flag for device pager objects and named anonymous
objects.
1995-03-11 22:25:02 +00:00
David Greenman
0007244260 Set VAGE flag when pager is destroyed. This usually happens when an
object has fallen off the end of the cached list - this is likely the
last reference to the vnode and it should be reused before non file
vnodes that are already on the free list (VDIR mostly).
1995-03-07 19:04:10 +00:00
David Greenman
f2da180f8e Fixed object reference count problem that occurred in the MAP_PRIVATE
case after we rewrote vm_mmap(). Added some comments to make it easier
to follow the reference counts.
1995-03-07 17:27:49 +00:00
David Greenman
be6d5bfac7 Don't attempt to reverse collapse non OBJ_INTERNAL objects. 1995-03-07 17:25:46 +00:00
Jordan K. Hubbard
7b936a1e66 Remove a gratutious cast. 1995-03-04 21:14:19 +00:00
David Greenman
f919ebde54 Various changes from John and myself that do the following:
New functions create - vm_object_pip_wakeup and pagedaemon_wakeup that
are used to reduce the actual number of wakeups.
New function vm_page_protect which is used in conjuction with some new
page flags to reduce the number of calls to pmap_page_protect.
Minor changes to reduce unnecessary spl nesting.
Rewrote vm_page_alloc() to improve readability.
Various other mostly cosmetic changes.
1995-03-01 23:30:04 +00:00
David Greenman
54ab07303b Slight change to include file order to accommodate upcoming changes. 1995-03-01 21:37:44 +00:00
Bruce Evans
4f9fb771a5 Don't use __P(()) in a function definition. 1995-02-25 18:39:04 +00:00
David Greenman
c3a1e4250f Fixed severely broken printf (arguments out of order, no newline). 1995-02-25 17:02:48 +00:00
David Greenman
b106f3b255 Removed redundant HOLDRELE()'s. 1995-02-23 22:32:38 +00:00
David Greenman
187f007190 Changed return value from vnode_pager_addr to be in DEV_BSIZE units so
that 9 bits aren't lost in the conversion. Changed all callers to expect
this. This allows paging on large (>2GB) filesystems.

Submitted by:	John Dyson
1995-02-22 10:34:34 +00:00
David Greenman
6f2b142ed8 vm_page.c:
Use request==VM_ALLOC_NORMAL rather than object!=kmem_object in deciding
if the caller is "important" in vm_page_alloc(). Also established a new
low threshold for non-interrupt allocations via cnt.v_interrupt_free_min.

vm_pageout.c:
Various algorithmic cleanup. Some calculations simplified. Initialize
cnt.v_interrupt_free_min to 2 pages.

Submitted by:	John Dyson
1995-02-22 10:27:24 +00:00
David Greenman
5e716206c0 Just return in the case of a page not on any queue in vm_page_unqueue().
Return VM_PAGE_BITS_ALL even if size > PAGE_SIZE in vm_page_bits().

Submitted by:	John Dyson
1995-02-22 10:16:21 +00:00
David Greenman
26f5e3ab12 Removed object locking code (it was a left over from an abortion that
was done a month or so ago).

Submitted by:	John Dyson
1995-02-22 10:06:43 +00:00
David Greenman
a7ac758ec7 Removed bogus copy object collapse check (the idea is right, but the
spcific check was bogus).
Removed old copy of vm_object_page_clean and took out the #if 1 around
the remaining one.

Submitted by:	John Dyson
1995-02-22 10:00:16 +00:00
David Greenman
c0503609a0 Only do object paging_in_progress wakeups if someone is waiting on this
condition.

Submitted by:	John Dyson
1995-02-22 09:15:35 +00:00
David Greenman
50ce2102e1 Rewrote MAP_PRIVATE case of vm_mmap() - all of the COW portion of this
routine was highly convoluted.

Submitted by:	John Dyson
1995-02-22 08:40:54 +00:00
David Greenman
2c0bc39683 Panic if u_map allocation fails. 1995-02-21 12:06:06 +00:00
David Greenman
2df771e5c0 vm_extern.h: removed vm_allocate_with_pager.
Removed vm_user.c...it's now completely deprecated.
1995-02-21 01:32:47 +00:00
David Greenman
7fb0c17ed2 Deprecated remaining use of vm_deallocate. Deprecated vm_allocate_with_
pager(). Almost completely rewrote vm_mmap(); when John gets done with
the bottom half, it will be a complete rewrite. Deprecated most use of
vm_object_setpager(). Removed side effect of setting object persist
in vm_object_enter and moved this into the pager(s). A few other
cosmetic changes.
1995-02-21 01:22:48 +00:00
David Greenman
d9459480fb Set page alloced for map entries as valid. 1995-02-21 01:13:05 +00:00
David Greenman
ffeec4aefe Removed vm_allocate(), vm_deallocate(), and vm_protect() functions. The
only function remaining in this file is vm_allocate_with_pager(), and this
will be going RSN. The file will be removed when this happens.
1995-02-20 23:58:10 +00:00
David Greenman
0c1dacbc5c Moved ACT_MAX, ACT_ADVANCE, and ACT_DECLINE to vm_page.h. 1995-02-20 23:35:45 +00:00
David Greenman
a6e1b30301 vm_inherit function has been deprecated. 1995-02-20 18:08:18 +00:00
David Greenman
a1e5cf9de7 Stop using vm_allocate and vm_deallocate. 1995-02-20 17:36:36 +00:00
David Greenman
ff16eeefb0 VM for the kernel stack and page tables doesn't need to be explicitly
deallocated as it isn't inherited across the fork.
Use vm_map_find not vm_allocate.

Submitted by:	John Dyson
1995-02-20 17:26:21 +00:00
David Greenman
ba8da83911 Panic if object is deallocated too many times.
Slight change to reverse collapsing so that vm_object_deallocate doesn't
have to be called recursively.
Removed half of a previous fix - the renamed page during a collapse doesn't
need to be marked dirty because the pager backing store pointers are copied
- thus preserving the page's data. This assumes that pages without backing
store are always dirty (except perhaps for when they are first zeroed, but
this doesn't matter).
Switch order of two lines of code so that the correct pager is removed
from the hash list. The previous code bogusly passed a NULL pointer to
vm_object_remove(). The call to vm_object_remove() should be unnecessary
if named anonymous objects were being dealt with correctly. They are
currently marked as OBJ_INTERNAL, which really screws up things (such as
this).
1995-02-20 14:21:58 +00:00
David Greenman
d0686727c0 Don't allow act_count to exceed ACT_MAX when bumping it up.
Small optimization to vm_page_bits().

Submitted by:	John Dyson
1995-02-20 14:00:50 +00:00
David Greenman
d89ced81b2 Fully initialize pages returned via vm_page_alloc_contig() so that the
memory can be later freed.
1995-02-20 13:58:14 +00:00
David Greenman
9b4814bba2 1) Added protection against collapsing OBJ_DEAD objects.
2) bump reference counts by 2 instead of 1 so that an object deallocate
   doesn't try to recursively collapse the object.
3) mark pages renamed during the collapse as dirty so that their contents
   are preserved.

Submitted by:	John and me.
1995-02-18 06:48:33 +00:00
David Greenman
ca40da7417 Don't bother calling pmap_create() when creating the temporary map.
The whole COW section of vm_mmap() should be rewritten; the current
implementation is highly convoluted.
1995-02-15 09:22:17 +00:00
Poul-Henning Kamp
d2fc53150b YF fix. 1995-02-14 06:14:28 +00:00
Poul-Henning Kamp
7aa3a026ad YF Fix. 1995-02-14 06:11:21 +00:00
David Greenman
bf4bd9bd9b Fixed problem with msync causing a panic.
Submitted by:	John Dyson
1995-02-14 04:00:17 +00:00
David Greenman
0217125f2b Carefully choose the value for vm_object_cache_max. The previous calculation
was rather bogus in most cases; the new value works very well for both
large and small memory machines.
1995-02-12 09:19:44 +00:00
David Greenman
a7fa409d30 Killed MACHVMCOMPAT function prototypes as the functions don't exist. 1995-02-09 14:16:07 +00:00
David Greenman
94aade54ff Killed MACHVMCOMPAT code. It doesn't compile, and in its present state
would require some work to make it not a serious security problem. It's
non-standard and not very useful anyway.
1995-02-09 14:14:13 +00:00
David Greenman
1ed81ef2fa Minor algorithmic adjustments that reduce the CPU consumption of the
pagedaemon in half while not reducing its effectiveness.

Submitted by:	me & John
1995-02-09 07:41:42 +00:00
David Greenman
efc68ce10f Fixed bmap run-length brokeness.
Use bmap run-length extension when doing clustered paging.

Submitted by:	John Dyson
1995-02-03 06:46:28 +00:00
David Greenman
a1f6d91cc2 swap_pager.c:
Fixed long standing bug in freeing swap space during object collapses.
Fixed 'out of space' messages from printing out too often.
Modified to use new kmem_malloc() calling convention.
Implemented an additional stat in the swap pager struct to count the
amount of space allocated to that pager. This may be removed at some
point in the future.
Minimized unnecessary wakeups.

vm_fault.c:
Don't try to collect fault stats on 'swapped' processes - there aren't
any upages to store the stats in.
Changed read-ahead policy (again!).

vm_glue.c:
Be sure to gain a reference to the process's map before swapping.
Be sure to lose it when done.

kern_malloc.c:
Added the ability to specify if allocations are at interrupt time or
are 'safe'; this affects what types of pages can be allocated.

vm_map.c:
Fixed a variety of map lock problems; there's still a lurking bug that
will eventually bite.

vm_object.c:
Explicitly initialize the object fields rather than bzeroing the struct.
Eliminated the 'rcollapse' code and folded it's functionality into the
"real" collapse routine.
Moved an object_unlock() so that the backing_object is protected in
the qcollapse routine.
Make sure nobody fools with the backing_object when we're destroying it.
Added some diagnostic code which can be called from the debugger that
looks through all the internal objects and makes certain that they
all belong to someone.

vm_page.c:
Fixed a rather serious logic bug that would result in random system
crashes. Changed pagedaemon wakeup policy (again!).

vm_pageout.c:
Removed unnecessary page rotations on the inactive queue.
Changed the number of pages to explicitly free to just free_reserved
level.

Submitted by:	John Dyson
1995-02-02 09:09:15 +00:00
David Greenman
8f895206ff Completed the fix for attempting to page out pages via the device_pager.
Submitted by:	John Dyson
1995-01-28 02:02:25 +00:00
David Greenman
1e9122e529 Use the VM_PAGE_BITS_ALL in a place it can be used.
Comment out call to pmap_prefault() until stability problems can be
thoroghly analyzed.
1995-01-26 01:40:04 +00:00
David Greenman
a465acdaf8 Don't attempt to clean device_pager backed objects at terminate time.
There is similar bogusness in the pageout daemon that will be fixed soon.
This fixes a panic pointed out to me by Bruce Evans that occurs when
/dev/mem is used to map managed memory.
1995-01-25 20:36:29 +00:00
David Greenman
6d40c3d394 Added ability to detect sequential faults and DTRT. (swap_pager.c)
Added hook for pmap_prefault() and use symbolic constant for new third
argument to vm_page_alloc() (vm_fault.c, various)
Changed the way that upages and page tables are held. (vm_glue.c)
Fixed architectural flaw in allocating pages at interrupt time that was
introduced with the merged cache changes. (vm_page.c, various)
Adjusted some algorithms to acheive better paging performance and to
accomodate the fix for the architectural flaw mentioned above. (vm_pageout.c)
Fixed pbuf handling problem, changed policy on handling read-behind page.
(vnode_pager.c)

Submitted by:	John Dyson
1995-01-24 10:14:09 +00:00
David Greenman
edfab85b1a Moved some splx's down a few lines in vm_page_insert and vm_page_remove
to make the locking a bit more clear - this change is currently a NOP
as the calls to those routines are already at splhigh().
1995-01-15 07:31:34 +00:00
David Greenman
b9921222a2 Protect a qcollapse call with an object lock before calling. The locks
need to be moved into the qcollapse and rcollapse routines, but I don't
have time at the moment to make all the required changes...this will do
for now.
1995-01-13 13:30:24 +00:00
David Greenman
8b4dd3c473 Improve my previous change to use the same tests as are used in qcollapse. 1995-01-11 20:19:20 +00:00
David Greenman
a748978457 Fixed a panic that Garrett reported to me...the OBJ_INTERNAL flag wasn't
being cleared in some cases for vnode backed objects; we now do this in
vnode_pager_alloc proper to guarantee it. Also be more careful in the
rcollapse code about messing with busy/bmapped pages.
1995-01-11 20:00:10 +00:00
David Greenman
a776a317e9 Kill VM_PAGE_INIT macro as it is only used once and makes the code more
difficult to understand. Got rid of unused vm_page flags.
1995-01-10 09:19:52 +00:00
David Greenman
480dff540b Fixed some formatting weirdness that I overlooked in the previous commit. 1995-01-10 07:32:52 +00:00
David Greenman
0d94caffca These changes embody the support of the fully coherent merged VM buffer cache,
much higher filesystem I/O performance, and much better paging performance. It
represents the culmination of over 6 months of R&D.

The majority of the merged VM/cache work is by John Dyson.

The following highlights the most significant changes. Additionally, there are
(mostly minor) changes to the various filesystem modules (nfs, msdosfs, etc) to
support the new VM/buffer scheme.

vfs_bio.c:
Significant rewrite of most of vfs_bio to support the merged VM buffer cache
scheme.  The scheme is almost fully compatible with the old filesystem
interface.  Significant improvement in the number of opportunities for write
clustering.

vfs_cluster.c, vfs_subr.c
Upgrade and performance enhancements in vfs layer code to support merged
VM/buffer cache.  Fixup of vfs_cluster to eliminate the bogus pagemove stuff.

vm_object.c:
Yet more improvements in the collapse code.  Elimination of some windows that
can cause list corruption.

vm_pageout.c:
Fixed it, it really works better now.  Somehow in 2.0, some "enhancements"
broke the code.  This code has been reworked from the ground-up.

vm_fault.c, vm_page.c, pmap.c, vm_object.c
Support for small-block filesystems with merged VM/buffer cache scheme.

pmap.c vm_map.c
Dynamic kernel VM size, now we dont have to pre-allocate excessive numbers of
kernel PTs.

vm_glue.c
Much simpler and more effective swapping code.  No more gratuitous swapping.

proc.h
Fixed the problem that the p_lock flag was not being cleared on a fork.

swap_pager.c, vnode_pager.c
Removal of old vfs_bio cruft to support the past pseudo-coherency.  Now the
code doesn't need it anymore.

machdep.c
Changes to better support the parameter values for the merged VM/buffer cache
scheme.

machdep.c, kern_exec.c, vm_glue.c
Implemented a seperate submap for temporary exec string space and another one
to contain process upages. This eliminates all map fragmentation problems
that previously existed.

ffs_inode.c, ufs_inode.c, ufs_readwrite.c
Changes for merged VM/buffer cache.  Add "bypass" support for sneaking in on
busy buffers.

Submitted by:	John Dyson and David Greenman
1995-01-09 16:06:02 +00:00
David Greenman
010cf3b9e3 Make sure that the object being collapsed doesn't go away on us...by
gaining extra references to it.

Submitted by:	John Dyson
Obtained from:
1995-01-05 04:30:40 +00:00
Andreas Schulz
832f3afd12 Submitted by: Ben Jackson
just a missing newline in a kernel printf added.
1995-01-02 22:56:00 +00:00
Bruce Evans
1365ed99fb Clean up previous commits (format for 80 columns...). 1994-12-30 08:02:16 +00:00
David Greenman
45cbbb298d Do vm_page_rename more conservatively in rcollapse and qcollapse, and
change list walk so that it doesn't get stuck in an infinite loop.

Submitted by:	 John Dyson
1994-12-23 05:00:19 +00:00
David Greenman
7609ab12e4 Initialize b_vnbuf.le_next before returning a new buffer in getpbuf and
trypbuf. Move a couple of splbio's to be slightly less conservative.
1994-12-23 04:56:51 +00:00
David Greenman
6f7bc393bb Fixed a benign off by one error. 1994-12-22 05:18:12 +00:00
David Greenman
6185408336 Don't ever clear B_BUSY on a pbuf (or any other flag for that matter).
This appears to be the cause of some buffer confusion that leads to
a panic during heavy paging.

Submitted by:	 John Dyson
1994-12-19 00:02:56 +00:00