1474 Commits

Author SHA1 Message Date
alfred
bf8e8a6e8f Remove M_TRYWAIT/M_WAITOK/M_WAIT. Callers should use 0.
Merge M_NOWAIT/M_DONTWAIT into a single flag M_NOWAIT.
2003-01-21 08:56:16 +00:00
dillon
091797e7e5 Fix swapping to a file, it was broken when SPECSTRATEGY was introduced. 2003-01-20 20:00:32 +00:00
dillon
e7be7a0432 Close the remaining user address mapping races for physical
I/O, CAM, and AIO.  Still TODO: streamline useracc() checks.

Reviewed by:	alc, tegge
MFC after:	7 days
2003-01-20 17:46:48 +00:00
alc
43a8628b32 - Hold the page queues lock around vm_page_hold().
- Assert that the page queues lock rather than Giant is held in
   vm_page_hold().
2003-01-20 09:24:03 +00:00
jeff
b584d46930 - M_WAITOK is 0 and not a real flag. Test for this properly.
Submitted by:	tmm
Pointy hat to:	jeff
2003-01-20 01:32:56 +00:00
obrien
e3fd0e351e Rev 1.16 renamed VM_METER to VM_TOTAL. This is breaking 3rd-party apps.
So add a VM_METER compat define.

Submitted by:	Andy Fawcett <andy@athame.co.uk>
2003-01-18 21:14:02 +00:00
dillon
aba727244c Merge all the various copies of vm_fault_quick() into a single
portable copy.
2003-01-16 00:02:21 +00:00
alc
c7ca47fcc7 - Update vm_pageout_deficit using atomic operations. It's a simple
counter outside the scope of existing locks.
 - Eliminate a redundant clearing of vm_pageout_deficit.
2003-01-14 06:57:03 +00:00
alc
f65aba1418 Make vm_pageout_page_free() static. 2003-01-14 02:28:39 +00:00
dillon
ce710d36cc It is possible for an active aio to prevent shared memory from being
dereferenced when a process exits due to the vmspace ref-count being
bumped.  Change shmexit() and shmexit_myhook() to take a vmspace instead
of a process and call it in vmspace_dofree().  This way if it is missed
in exit1()'s early-resource-free it will still be caught when the zombie is
reaped.

Also fix a potential race in shmexit_myhook() by NULLing out
vmspace->vm_shm prior to calling shm_delete_mapping() and free().

MFC after:	7 days
2003-01-13 23:04:32 +00:00
phk
12e99581a7 We can get past here on a normal vnode as well, so use VOP_STRATEGY if so. 2003-01-13 21:32:16 +00:00
dillon
ccd5574cc6 Bow to the whining masses and change a union back into void *. Retain
removal of unnecessary casts and throw in some minor cleanups to see if
anyone complains, just for the hell of it.
2003-01-13 00:33:17 +00:00
alc
fe540b81bb Make vm_page_alloc() return PG_ZERO only if VM_ALLOC_ZERO is specified.
The objective being to eliminate some cases of page queues locking.
(See, for example, vm/vm_fault.c revision 1.160.)

Reviewed by:	tegge

(Also, pointed out by tegge that I changed vm_fault.c before changing
vm_page.c.  Oops.)
2003-01-12 23:32:46 +00:00
alc
e45ec3c803 vm_fault_copy_entry() needn't clear PG_ZERO because it didn't pass
VM_ALLOC_ZERO to vm_page_alloc().
2003-01-12 07:33:16 +00:00
dillon
ddf9ef103e Change struct file f_data to un_data, a union of the correct struct
pointer types, and remove a huge number of casts from code using it.

Change struct xfile xf_data to xun_data (ABI is still compatible).

If we need to add a #define for f_data and xf_data we can, but I don't
think it will be necessary.  There are no operational changes in this
commit.
2003-01-12 01:37:13 +00:00
alc
d833c390ba In vm_page_alloc(), fuse two if statements that are conditioned on the same
expression.
2003-01-11 20:07:17 +00:00
dillon
187e654d9a Make 'sysctl vm.vmtotal' work properly using updated patch from Hiten.
(the patch in the PR was stale).

PR:             kern/5689
Submitted by:   Hiten Pandya <hiten@unixdaemons.com>
2003-01-11 07:29:47 +00:00
alc
03f7564d9e In vm_page_alloc(), honor VM_ALLOC_ZERO for system and interrupt class
requests when the number of free pages is below the reserved threshold.
Previously, VM_ALLOC_ZERO was only honored when the number of free pages
was above the reserved threshold.  Honoring it in all cases generally
makes sense, does no harm, and simplifies the code.
2003-01-08 19:58:42 +00:00
phk
0552c4bec8 Convert VOP_STRATEGY to VOP_SPECSTRATEGY in the generic getpages and
the pager input for small filesystems.
2003-01-05 20:32:03 +00:00
alc
e9e799ad84 Use atomic add and subtract to update the global wired page count,
cnt.v_wire_count.
2003-01-05 01:31:45 +00:00
phk
131885aa2f Temporarily introduce a new VOP_SPECSTRATEGY operation while I try
to sort out disk-io from file-io in the vm/buffer/filesystem space.

The intent is to sort VOP_STRATEGY calls into those which operate
on "real" vnodes and those which operate on VCHR vnodes.  For
the latter kind, the call will be changed to VOP_SPECSTRATEGY,
possibly conditionally for those places where dual-use happens.

Add a default VOP_SPECSTRATEGY method which will call the normal
VOP_STRATEGY.  First time it is called it will print debugging
information.  This will only happen if a normal vnode is passed
to VOP_SPECSTRATEGY by mistake.

Add a real VOP_SPECSTRATEGY in specfs, which does what VOP_STRATEGY
does on a VCHR vnode today.

Add a new VOP_STRATEGY method in specfs to catch instances where
the conversion to VOP_SPECSTRATEGY has not yet happened.  Handle
the request just like we always did, but first time called print
debugging information.

Apart up to two instances of console messages per boot, this amounts
to a glorified no-op commit.

If you get any of the messages on your console I would very much
like a copy of them mailed to phk@freebsd.org
2003-01-04 22:10:36 +00:00
alc
6412b92ef7 Allow kmem_malloc() without Giant if M_NOWAIT is specified. 2003-01-04 19:26:35 +00:00
alc
d34fb3af16 Use vm_object_lock() and vm_object_unlock() in vm_object_deallocate().
(This procedure needs further work, but this change is sufficient for
locking the kmem_object.)
2003-01-04 19:23:19 +00:00
alc
d5b9d0271f Refine the assertions in vm_page_alloc(). 2003-01-04 19:07:13 +00:00
alc
bb9d453ab0 Refine the assertion in vm_object_clear_flag() to allow operation on the
kmem_object without Giant.  In that case, assert that the kmem_object's
mutex is held.
2003-01-03 19:19:08 +00:00
phk
997c254192 Revert use of dmmax_mask, I had overlooked a '~'.
Spotted by:	bde
2003-01-03 19:16:48 +00:00
phk
b884c995db Make struct swblock kernel only, to make vm/swap_pager.h userland includable.
Move struct swdevt from sys/conf.h to the more appropriate vm/swap_pager.h.
Adjust #include use in libkvm and pstat(8) to match.
2003-01-03 16:23:12 +00:00
phk
88bd74d184 Avoid extern decls in .c files by putting them in the vm/swap_pager.h
include file where they belong.
Share the dmmax_mask variable.
2003-01-03 14:30:46 +00:00
phk
2f7baffed6 Use correct _VM_SWAP_PAGER_H_ to check for multiple inclusion. 2003-01-03 14:22:52 +00:00
phk
83581fdc02 Retire sys/dmap.h by including the two lines of it which matters
directly in vm/vm_swap.c.
2003-01-03 09:55:05 +00:00
alc
2db78e226f Lock the vm object when performing vm_object_clear_flag(). 2003-01-03 09:15:43 +00:00
phk
daf6948653 Convert calls to BUF_STRATEGY to VOP_STRATEGY calls. This is a no-op since
all BUF_STRATEGY did in the first place was call VOP_STRATEGY.
2003-01-03 06:32:15 +00:00
alc
ed9a027a4a Add vm map and vm object locking to vmtotal(). 2003-01-03 05:52:02 +00:00
alc
73597ffcc2 Lock the vm object when performing vm_object_clear_flag(). 2003-01-02 09:09:27 +00:00
alc
0912d72a88 Update the assertions in vm_page_insert() and vm_page_lookup() to reflect
locking of the kmem_object.
2003-01-01 19:45:36 +00:00
schweikh
d3367c5f5d Correct typos, mostly s/ a / an / where appropriate. Some whitespace cleanup,
especially in troff files.
2003-01-01 18:49:04 +00:00
alc
c28c69ebf9 Add a needed #include.
Reported by:	ia64 tinderbox
2003-01-01 00:13:01 +00:00
alc
1842adc340 Implement a variant locking scheme for vm maps: Access to system maps
is now synchronized by a mutex, whereas access to user maps is still
synchronized by a lockmgr()-based lock.  Why?  No single type of lock,
including sx locks, meets the requirements of both types of vm map.
Sometimes we sleep while holding the lock on a user map.  Thus, a
a mutex isn't appropriate.  On the other hand, both lockmgr()-based
and sx locks release Giant when a thread/process blocks during
contention for a lock.  This could lead to a race condition in a legacy
driver (that relies on Giant for synchronization) if it attempts to
kmem_malloc() and fails to immediately obtain the lock.  Fortunately,
we never sleep while holding a system map lock.
2002-12-31 19:38:04 +00:00
alc
71afac1c55 - Mark the kernel_map as a system map immediately after its creation.
- Correct a cast.
2002-12-30 05:55:41 +00:00
alc
bc746e81d2 - Increment the vm_map's timestamp if _vm_map_trylock() succeeds.
- Introduce map_sleep_mtx and use it to replace Giant in
   vm_map_unlock_and_wait() and vm_map_wakeup().  (Original
   version by: tegge.)
2002-12-30 00:41:33 +00:00
alc
18e9372579 - Remove vm_object_init2(). It is unused.
- Add a mtx_destroy() to vm_object_collapse().  (This allows a bzero()
   to migrate from _vm_object_allocate() to vm_object_zinit(), where it
   will be performed less often.)
2002-12-29 21:01:14 +00:00
alc
3f894298eb Reduce the number of times that we acquire and release the page queues
lock by making vm_page_rename()'s caller, rather than vm_page_rename(),
responsible for acquiring it.
2002-12-29 07:17:06 +00:00
alc
e053499669 Assert that the page queues lock rather than Giant is held in
vm_page_flag_clear().
2002-12-28 22:49:37 +00:00
dillon
0fe137bfcf vm_pager_put_pages() takes VM_PAGER_* flags, not OBJPC_* flags. It just
so happens that OBJPC_SYNC has the same value as VM_PAGER_PUT_SYNC so no
harm done.  But fix it :-)

No operational changes.

MFC after:	1 day
2002-12-28 21:15:39 +00:00
dillon
1f367998c6 Allow the VM object flushing code to cluster. When the filesystem syncer
comes along and flushes a file which has been mmap()'d SHARED/RW, with
dirty pages, it was flushing the underlying VM object asynchronously,
resulting in thousands of 8K writes.  With this change the VM Object flushing
code will cluster dirty pages in 64K blocks.

Note that until the low memory deadlock issue is reviewed, it is not safe
to allow the pageout daemon to use this feature.  Forced pageouts still
use fs block size'd ops for the moment.

MFC after:	3 days
2002-12-28 21:03:42 +00:00
alc
98df99a71b Two changes to kmem_malloc():
- Use VM_ALLOC_WIRED.
 - Perform vm_page_wakeup() after pmap_enter(), like we do everywhere else.
2002-12-28 19:03:54 +00:00
alc
9a94897a13 - Change vm_object_page_collect_flush() to assert rather than
acquire the page queues lock.
 - Acquire the page queues lock in vm_object_page_clean().
2002-12-27 20:16:13 +00:00
alc
b9ffe85ed2 Increase the scope of the page queues lock in phys_pager_getpages(). 2002-12-27 06:09:56 +00:00
alc
43045f3d3b - Hold the page queues lock around calls to vm_page_flag_clear(). 2002-12-24 19:02:03 +00:00
alc
f47bf7602e - Hold the page queues lock around vm_page_wakeup(). 2002-12-24 04:24:58 +00:00