Commit Graph

357 Commits

Author SHA1 Message Date
mycroft 700dee447e DTRT if vm_object_lookup() returns NULL. This may happen while an object is
being created or destroyed.
1997-02-27 17:49:05 +00:00
tls fad978b853 a mutex on vnode_pager_sync avoids another race condition, only possible if sync were called again while we were sleeping. This should be the last of this nest of bugs. From mycroft. 1997-02-26 02:26:19 +00:00
thorpej cbb4319fb9 vm_object_shadow(): Add a DIAGNOSTIC check for shadowing a null object.
From Charles Hannum <mycroft@netbsd.org>
1997-02-25 23:28:09 +00:00
thorpej a49641e340 vm_map_copy_entry(): Don't set "needs_copy" on a zero-fill entry.
From Charles Hannum <mycroft@netbsd.org>
1997-02-25 23:27:08 +00:00
thorpej 7327c8cf62 From Charles Hannum:
During a collapse, abort if the object is paging; the pagedaemon can
trigger a collapse, causing deadlock.  (XXX)
1997-02-24 22:19:26 +00:00
mrg 82f19ac36c lite2 1997-02-23 09:58:53 +00:00
mrg 3b9c3fff2e restructure vm_object_overlay() to collapse objects in more cases.
change one diag panic() to a diag printf().  from mycroft.
1997-02-23 09:01:37 +00:00
mrg 8eba4aeacb don't clean fading objects. from mycroft. 1997-02-23 08:56:57 +00:00
thorpej a49beaa16a Possesive, Jason... possesive.... 1997-02-22 04:36:48 +00:00
thorpej 75be8504f4 More good stuff from Charles:
Eliminate a nasty race condition in vnode_pager_sync() by making
two passes.  First, find all of the objects to be cleaned and reference
them.  Second, traverse this list objects, cleaning them if there are
references other than us.
1997-02-22 04:11:42 +00:00
thorpej 255b9d74d8 Rewrite the object chain collapse code. Bugs fixed are too numerous
to describe here.  This should fix the problems with "hanging processes"
people have seen since the original object collapse code was committed.

From Charles Hannum <mycroft@netbsd.org>
1997-02-21 20:30:49 +00:00
thorpej 235a2d71f7 Use vm_object_paging() to test if an object has pages in transit.
From Charles Hannum <mycroft@netbsd.org>
1997-02-21 20:27:21 +00:00
thorpej dd8d0f1f00 Define a macro to test if an object has pages in transit. Clarify
some arithmetic, and nuke an unneeded prototype.

From Charles Hannum <mycroft@netbsd.org>.
1997-02-21 20:26:03 +00:00
thorpej 17b60fc6e9 In vnode_pager_sync(), we need to vm_object_deallocate() after cleaning
pages.  Noticed by Thor Simon <tls@netbsd.org>, confirmed by
Charles Hannum <mycroft@netbsd.org>
1997-02-21 20:22:39 +00:00
mrg 42ae012452 This is all from mycroft:
Create macros (with names borrowed from Mach 3) to manipulate
object->paging_in_progress, and use them.  When decreasing the
paging count, always make sure to wake up anyone waiting.
Fixes `vospgc', `vosca1', and `vosca2' hangs.
1997-02-18 13:39:31 +00:00
tls 127c6b54e7 Sigh. I initially had this exactly backwards: the object needs to be locked, the vnode does *not*. Avoids panic: locking against myself quickly reported by Charles. 1997-02-17 13:10:28 +00:00
tls a507fda378 vnode_pager_sync bugs: A) must lock the object, not just the vnode. B) Don't remove the object from paging queues, because we'd have to put it back on one. The pagedaemon will eventually take care of this for us anyway. This was causing us to *lose track of pages completely* every time vnode_pager_sync() was called... sigh. Credit goes to mycroft for untangling this mess. 1997-02-17 06:22:52 +00:00
tls 4b1213d27c sync needs to clean VM objects backed by vnode pagers 1997-02-13 02:54:06 +00:00
mrg 491e8af174 minor KNF from jason. 1997-02-05 08:26:25 +00:00
mrg eadd7792d9 add thread_sleep_msg() that takes a wait message, and change thread_sleep()
into a macro for backwards compat.  alter vm_object_sleep() to take advantage
of this.
1997-02-05 08:09:45 +00:00
mrg 85328f7b35 KNF. 1997-02-05 07:48:42 +00:00
thorpej 593fdd462d NFSCLIENT -> NFS 1997-01-31 03:06:06 +00:00
thorpej b7a5b84750 Treat the nswdev == 0 case like the nswap == 0 case; don't panic, just
report no swap space.  Suggested by Leo Weppleman <leo@netbsd.org>.
1997-01-22 07:28:21 +00:00
mrg 4f2a217c2b pull up vm_object_collapse() fix from mrg_vm_object_collapse branch.
many thanks to niklas hallqvist (niklas@appli.se) for this.
1997-01-03 18:03:14 +00:00
mrg f3d9c468e1 minor KNF changes. 1997-01-03 18:00:25 +00:00
cgd 8a3333b2a9 Fix an inconsistency that came in with Lite: setrq() was renamed to
setrunqueue(), but remrq() was never renamed.  Rename remrq() to
remrunqueue().
1996-11-06 20:19:19 +00:00
cgd ee08818e7a Fix an inconsistency that came in with Lite: setrq() was renamed to
setrunqueue(), but remrq() was never renamed.  Rename remrq() to
remrunqueue().  Also, move remrunqueue() prototype from vm/vm_extern.h
to sys/proc.h, so that it's in the same place as the setrunqueue() prototype
and other related prototypes.
1996-11-06 19:33:39 +00:00
gwr f5927ca260 Minor nit in my change regarding const and non-ansi code... 1996-10-27 21:50:33 +00:00
gwr 2ada6af51b Oops, part of that last commit was not meant to go in.
The definitions of va_alist, va_dcl belong elsewhere.
1996-10-27 20:50:00 +00:00
gwr a0d9bd5bb2 The hack-ish tricks to make the ANSI-style va_start macro work in
non-ANSI mode were not quite complete.  This makes non-ANSI work.
Perhaps it would be better to eliminate this hack altogether...
1996-10-27 20:30:52 +00:00
matthias 3ff0c0553e * The kernel-stack no longer needs to be wired-down on the pc532. 1996-10-23 07:20:09 +00:00
perry 9ab047883b Fix unused variable -Wxxx failure when MINIROOTSIZE defined. 1996-10-21 01:32:38 +00:00
christos 3dc7953eec back out previous kprintf changes 1996-10-12 21:50:05 +00:00
christos de1b2b437e printf -> kprintf, sprintf -> ksprintf 1996-10-10 17:16:17 +00:00
ws 2da166f660 Fix p_nice vs. NZERO code.
Change NZERO to 20 to always make p_nice positive.
On Christos' suggestion make p_nice explicitly u_char.
1996-10-02 18:04:56 +00:00
mrg 2fc2b98d83 make pageout sleep on vm_pages_needed rather than lbolt. this stops
pathological behaviour (freezes) when running short of free pages.  see
PR#2755 for a more detailed explanation (from jonathan).

thanks to john dyson for the first cut of this change.
1996-09-18 02:04:50 +00:00
cgd 62034405e6 allow kmem_malloc to return NULL if canwait is set and there's no map space
left, rather than panic()ing.  This allows callers to set their own policy
for how to handle the shortage.  Add a few comments.
1996-06-13 16:52:06 +00:00
pk 7acbd9511d Make v_swpin & v_swpout tick. 1996-05-29 21:20:11 +00:00
mrg 047dfa8524 count pagein faults in struct user. (from kenneth stailey pr#1397]) 1996-05-20 17:40:02 +00:00
ragge 621153ce0e The unexpected remove of active kstack does happen on vax also,
not only on pc532. I've verivied that this "fix" work.
1996-05-19 10:00:38 +00:00
christos a818b81625 Added missing prototypes from <sys/cpu.h> 1996-04-23 12:25:23 +00:00
pk 37a292dea0 Make prototyping of most functions conditional on `PMAP_EXCLUDE_DECLS'.
This allows MD pmap.h to implement access to those functions via pointers.
1996-03-31 22:15:32 +00:00
christos ff63e6c744 Change %r -> %: as for recursive printf's 1996-03-30 21:50:45 +00:00
christos 3cc52443b5 Fix vm_offset_t (unsigned long) and vm_size_t (unsigned long) printf arguments
to %lx and %ld instead of %x and %d.
1996-03-16 23:15:18 +00:00
gwr f6925cfbf3 Replace vm_object_prefer with PMAP_PREFER. 1996-02-28 22:39:13 +00:00
gwr fdb309615e Remove vm_object_prefer. VM code now calls PMAP_PREFER directly. 1996-02-28 22:35:35 +00:00
mycroft 8d110517c8 We no longer need the old hack for allocating the u-area on the i386. 1996-02-18 22:53:43 +00:00
christos c048d7ba68 fix a printf warning and add pmap_page_index declaration 1996-02-12 21:51:52 +00:00
christos d0fd3d7e62 Don't declare pmap_page_index for now. It will break the arm port.
This clearly needs to be fixed. I would like to declare it:

    __pure u_int     pmap_page_index __P((vm_offset_t));

but at least the arm port wants it signed [since it tests for -1 all over
the place]. Can we have this fixed please?
1996-02-10 00:30:39 +00:00
christos b7b8db60ac change pmap_page_index to return __pure u_int, like in the majority of
the ports
1996-02-10 00:20:09 +00:00
christos 03b1ad4cb4 One more cleanup pass to change %x to %p for pointers 1996-02-10 00:07:59 +00:00
christos 46d49e500e Account for __FORK_BRAINDAMAGE.. 1996-02-05 20:32:25 +00:00
christos bcabcefdcd VM prototype changes 1996-02-05 01:53:47 +00:00
christos bac60f4e4e Remove unused argument from vsunlock 1996-02-04 02:09:41 +00:00
mycroft fbb022ab48 Trivial change for consistency. 1996-01-31 03:11:16 +00:00
mycroft ebfe9ee154 Minor changes. 1995-12-21 04:44:27 +00:00
mycroft fee1a345eb Only expect vm_fork() to return if __FORK_BRAINDAMAGE is defined.
Eliminate unused third arg to vm_fork().
1995-12-09 04:28:16 +00:00
pk 785c98a0cc Oops. 1995-12-06 00:38:11 +00:00
pk 801712f01a Extend use of vm_object_prefer() to vm_allocate_with_pager().
Make vm_object_prefer() call MD aligner for "pageless" objects too,
so we can have more control over the virtual address to be used.

Implementation could be simpler if we by-pass the object to mapped, but
we'd loose the ability to adapt alignment to objects that were previously
mmap'ed with MAP_FIXED on.
1995-12-05 22:54:36 +00:00
mycroft 2dd293d3c3 Add hooks for COMPAT_FREEBSD, from Noriyuki Soda. 1995-10-10 01:26:36 +00:00
mycroft 245f292fed Prefix names of system call implementation functions with `sys_'. 1995-10-07 06:25:19 +00:00
thorpej fbbf585f72 Remove spurious (and incorrect) prototype for swapon(), pointed out
by Chris Demetriou.
1995-09-27 20:30:17 +00:00
thorpej 7160dfc805 Make system calls conform to a standard prototype and bring those
prototypes into scope.
1995-09-19 21:57:36 +00:00
gwr 5520ccdbf9 Preserve miniroot when (swdevt[i].sw_vp == rootvp) 1995-09-18 21:21:00 +00:00
mycroft 67daa4706d Add PHOLD() and PRELE() macros, used to hold a process in core and release it. 1995-08-13 09:04:43 +00:00
cgd 4ec7eec85b don't forget the 'dev' argument to physio(). pointed out by der Mouse
and others.
1995-07-19 13:04:02 +00:00
pk 7ac7bdf204 Unlock internal objects too in vm_object_terminate(). 1995-07-13 12:35:29 +00:00
cgd 21d1e8d60f fix bug in load average calculation that prevented processes which were
uninterruptibly sleeping for a short period from being counted.
1995-07-08 03:12:22 +00:00
mycroft 7263209ce6 Make each disk and tape driver define its own read and write functions.
Deprecate rawread() and rawwrite() completely.  Remove d_strategy from cdevsw to
force the abstraction barrier.
1995-07-04 07:15:28 +00:00
cgd cfbb9c7cfe no longer needs machine/cpu.h, and use new name for the 'broken swap' define 1995-06-28 02:58:51 +00:00
leo f205966027 The atari specific code can be removed now it is using MACHINE_NONCONTIG. 1995-06-26 14:40:11 +00:00
cgd 7bd5b4bfa8 oops; vm_pageout.c should include machine/cpu.h, _not_ vm_pager.c 1995-05-07 19:27:27 +00:00
cgd cf08de5cf0 needs machine/cpu.h 1995-05-07 16:22:40 +00:00
cgd 0b2ce5d881 delete some cruft, don't swap if BROKEN_SWAP defined, call cpu_swapout rather than #ifdef 1995-05-05 03:35:39 +00:00
mycroft 5f41ce9281 Correct checks for NULLs in pagertab. 1995-04-15 05:11:35 +00:00
mycroft 68c8c80929 kernel_pmap --> pmap_kernel() 1995-04-10 16:48:27 +00:00
briggs 6efcd1b6d4 KERNEL -> _KERNEL 1995-03-29 21:57:43 +00:00
cgd 9141ec70a1 change needed for atari; recycle some code for the i386 that was in Lite
that we don't use, but is used by the atari now.  (it should change
to MACHINE_NONCONTIG, too...)
1995-03-27 09:09:05 +00:00
jtc f76f1f89ad KERNEL -> _KERNEL 1995-03-26 20:35:13 +00:00
mycroft 57e6889435 Make memory map syscalls consistently use size_t. 1995-03-09 17:28:28 +00:00
cgd 94d77171f9 need compat_43_getpagesize on COMPAT_OSF1 1995-03-08 01:46:53 +00:00
mycroft 3d869b0608 Fix two off-by-one errors. 1995-01-09 17:36:48 +00:00
cgd 70f70a3122 make grow's arguments type-correct. 1995-01-09 02:48:03 +00:00
cgd 4b5b562e89 turn spaces to tabs 1995-01-09 02:47:38 +00:00
cgd d158991669 don't wrap lines 1995-01-09 02:47:23 +00:00
cgd 1af9a586a5 fix vm_page_startup's return type, kill an odd 'register' 1995-01-09 02:46:41 +00:00
cgd 86095ef7e6 no need to have two lines of arguments here. 1995-01-09 02:45:47 +00:00
cgd 688a174a72 move munmapfd into the correct section, make grow take a proper argument. 1995-01-09 02:45:29 +00:00
cgd 2d8771217f can move FALSE up a line. 1995-01-09 01:33:17 +00:00
cgd b1f6bfa080 kill extraneous label 1995-01-09 01:30:46 +00:00
pk 3e4bf211c6 Use vm_object_prefer() to find a good range of virtual addresses to
map a memory object to.
1994-12-10 11:48:12 +00:00
pk d114baba9a Introduce vm_object_prefer(), which interrogates the PMAP layer in search
for an optimal virtual address for a given object.
1994-12-10 11:46:25 +00:00
mycroft 331b95bd8a Garbage collect unneeded variable. 1994-12-01 15:39:17 +00:00
gwr 1a7920257d Make sure vm_mmap creates mappings with page-aligned file offsets.
Handle non-aligned file offsets in the mmap/munmap system calls.
1994-12-01 00:23:11 +00:00
gwr 0a0c02cd38 Allow vm_page_array to have holes in it so systems that use
MACHINE_NONCONTIG can use a fast, simple pmap_page_index().
1994-12-01 00:19:57 +00:00
cgd 0a930e5ce6 redo that last change, correctly. (suggested by mycroft.) 1994-10-30 19:11:09 +00:00
cgd aa151f86b3 change some strategic casts to (*long) from (*int). 1994-10-29 07:35:04 +00:00
cgd 6b86130410 update for new syscall args description mechanism 1994-10-20 04:22:35 +00:00
cgd 4eb752cd51 make the "wait on swbuf" message SDB_IO -- it's I/O related, and it's
not at all anomalous (and bloody annoying!) on 4M machines.
1994-10-18 06:42:28 +00:00
mycroft ab9caa0e9a Fix the munmapfd() prototype, too. 1994-09-16 02:01:49 +00:00
mycroft d5fdf729f2 All of the users of munmapfd() expect to pass it
two args.  Make it work correctly.
1994-09-16 01:57:57 +00:00
mycroft 6dd2f9eed2 vm_object_page_clean() expects to be able to use vm_page_deactivate() on a
locked page.  Make that work.  This also obviates the need for vm_fault() to
bogusly activate a page before deactivating it.  Finally, make sure the
semantics of vm_object_deactive_pages() don't change.
1994-09-07 20:25:07 +00:00
mycroft 2f710a54ee Convert process, file, and namei lists and hash tables to use queue.h. 1994-08-30 03:04:28 +00:00
deraadt 72b3d408b9 replace "ctob(UPAGES)" and "UPAGES * NBPG" with "USPACE" 1994-08-23 22:07:42 +00:00
mycroft fe18d806d2 Don't panic if primary swap device is not configured. 1994-07-21 07:12:13 +00:00
cgd fccfa11af5 New RCS ID's, take two. they're more aesthecially pleasant, and use 'NetBSD' 1994-06-29 06:39:25 +00:00
cgd 9608c5b657 clean up slightly; change RCS ID's to be minimally intrusive 1994-06-27 04:55:46 +00:00
cgd e4c9e4b2cd these were supposed to be written in C; they should follow its syntax 1994-06-27 04:03:15 +00:00
pk bccc2c3c1b straighten out diskless swap code somewhat. 1994-06-22 14:01:45 +00:00
cgd bf43549e44 fix a thinko that kept msync from working if no size specified. 1994-06-20 00:17:28 +00:00
mycroft 0d8affb278 Turn P_NOSWAP and P_PHYSIO into a hold count, as suggested by a comment. 1994-06-15 19:59:21 +00:00
pk 11ebd2e9c8 #ifdef FIFO's 1994-06-13 16:31:54 +00:00
pk 5d4bd29314 Check for all NFS related vnode ops. 1994-06-10 20:01:20 +00:00
mycroft 264b874c14 Update to 4.4-Lite fs code, with local changes. 1994-06-08 11:41:58 +00:00
mycroft 9eb7a8a711 Fix minor NFS glitch introduced with new VM code. 1994-05-30 13:10:06 +00:00
pk 7672b8b95b New style core dumps. 1994-05-25 10:56:56 +00:00
cgd 0094e3719e change warnings 1994-05-24 01:07:50 +00:00
mycroft 5a9d09c5d5 Clean up deleted files. 1994-05-23 03:12:14 +00:00
cgd 3495827959 new VM code, from 4.4-Lite 1994-05-23 03:11:20 +00:00
cgd 11b6c06344 no need to include <sys/vmmeter.h> 1994-05-21 04:00:13 +00:00
mycroft 72edd4b918 For MAP_ANON, force the file offset to 0. I can't think of any any bad side
effects of this, and it fixes an interesting bug.
1994-05-19 18:03:04 +00:00
cgd 207339d53d define maxdmap and maxsmap 1994-05-19 08:08:46 +00:00
cgd c6abdc2054 setrq -> setrunqueue, sched -> scheduler 1994-05-13 00:50:41 +00:00
cgd 6fea3fab9a oops; spaces at ends of lines, etc. 1994-05-11 01:01:03 +00:00
cgd 904bdb2ea9 sysctl 1994-05-07 00:40:07 +00:00
cgd ca14e4af10 stubs 1994-05-07 00:39:58 +00:00
cgd d0a066a438 functions soon to be added 1994-05-06 22:45:29 +00:00
cgd 162c8601c0 latent sysctl support 1994-05-06 22:44:22 +00:00
mycroft 31393b8acd Rearrange some code, and fix a condition where the pageout scanner would start
grabbing pages from the active queue.
1994-05-05 20:35:11 +00:00
cgd 0f2ecb72f6 lots of changes: prototype migration, move lots of variables, definitions,
and structure elements around.  kill some unnecessary type and macro
definitions.  standardize clock handling.  More changes than you'd want.
1994-05-05 05:40:40 +00:00
cgd 4bbabd8e7a Rename a lot of process flags. 1994-05-04 03:41:12 +00:00
cgd f8dd4f370d expand the rlimit struct, kill last vestiges of off_t bogosity. 1994-05-04 01:38:25 +00:00
mycroft 06a4a02877 Add some DIAGNOSTIC code to detect trying to free a page that is being faulted
in.
1994-04-29 08:21:49 +00:00
cgd f50f031448 kill syscall name aliases. no user-visible changes 1994-04-29 04:41:02 +00:00
cgd e5ad8074f2 minor code structure nit that i pointed out to hibler a while ago... 1994-04-29 03:56:26 +00:00
cgd 10b63d74bc SHUT UP! 1994-04-28 03:51:58 +00:00
cgd 95fcf16b66 add sequential swap device support, supplied by hibler many moons ago.
(Note that it needs config support, too, which isn't there yet, but
might be soon...)  Also, general cleanup.  use NODEV to terminate swdevt,
so that sparc port (and future) is happy.
1994-04-25 23:53:51 +00:00
cgd 3dda0064a5 Convert mount, vnode, and buf structs to use <sys/queue.h>. Also,
some knf and structure frobbing to do along with it.
1994-04-21 07:47:31 +00:00
cgd ee3e1168b7 undo a mistaken deletion. at least i found it... 1994-04-20 21:46:15 +00:00
cgd 551e92c16a start to phase out temp. off_t syscalls 1994-04-16 23:19:06 +00:00
cgd ad6c540f90 cancel some of my stupidity, thanks to charles. 1994-04-15 22:49:15 +00:00
cgd ae41edb3e8 typo, pointed out by charles. 1994-04-15 18:02:02 +00:00
cgd ece66170de somehow, these didn't get committed!!! 1994-04-15 08:14:27 +00:00
mycroft c51913cd19 Clean up deleted files. 1994-04-15 07:05:05 +00:00
cgd e937bc9bfd convert vm system to use new queues. I'll never write code w/queues again. 1994-04-15 07:04:03 +00:00
cgd ed740c7de8 frob arguments a little bit 1994-04-02 08:39:20 +00:00
cgd b7e76677c6 expand uid_t/gid_t/off_t 1994-03-27 09:09:57 +00:00
chopps b8f74df1a5 default type is MAP_FILE if none given. 1994-03-23 01:54:02 +00:00
cgd 70dabb7d43 don't be so agressive w/renaming, clean for consistency, types in one place. 1994-03-17 02:51:57 +00:00
mycroft a8f3db1d79 Use b_actf, not av_forw. 1994-02-06 10:13:02 +00:00