Commit Graph

387 Commits

Author SHA1 Message Date
Alan Cox
59677d3c0e Prevent the leakage of wired pages in the following circumstances:
First, a file is mmap(2)ed and then mlock(2)ed.  Later, it is truncated.
Under "normal" circumstances, i.e., when the file is not mlock(2)ed, the
pages beyond the EOF are unmapped and freed.  However, when the file is
mlock(2)ed, the pages beyond the EOF are unmapped but not freed because
they have a non-zero wire count.  This can be a mistake.  Specifically,
it is a mistake if the sole reason why the pages are wired is because of
wired, managed mappings.  Previously, unmapping the pages destroys these
wired, managed mappings, but does not reduce the pages' wire count.
Consequently, when the file is unmapped, the pages are not unwired
because the wired mapping has been destroyed.  Moreover, when the vm
object is finally destroyed, the pages are leaked because they are still
wired.  The fix is to reduce the pages' wired count by the number of
wired, managed mappings destroyed.  To do this, I introduce a new pmap
function pmap_page_wired_mappings() that returns the number of managed
mappings to the given physical page that are wired, and I use this
function in vm_object_page_remove().

Reviewed by: tegge
MFC after: 6 weeks
2007-11-17 22:52:29 +00:00
Alan Cox
2573269111 The previous revision, updating vm_object_page_remove() for the new page
cache, did not account for the case where the vm object has nothing but
cached pages.

Reported by:	kris, tegge
Reviewed by:	tegge
MFC after:	3 days
2007-10-18 23:02:18 +00:00
Alan Cox
c944491426 Correct an error of omission in the reimplementation of the page
cache: vm_object_page_remove() should convert any cached pages that
fall with the specified range to free pages.  Otherwise, there could
be a problem if a file is first truncated and then regrown.
Specifically, some old data from prior to the truncation might reappear.

Generalize vm_page_cache_free() to support the conversion of either a
subset or the entirety of an object's cached pages.

Reported by: tegge
Reviewed by: tegge
Approved by: re (kensmith)
2007-09-27 04:21:59 +00:00
Alan Cox
f3a2ed4bd9 Correct an error in the previous revision, specifically,
vm_object_madvise() should request that the reactivated, cached page
not be busied.

Reported by: Rink Springer
Approved by: re (kensmith)
2007-09-25 21:01:10 +00:00
Alan Cox
7bfda801a8 Change the management of cached pages (PQ_CACHE) in two fundamental
ways:

(1) Cached pages are no longer kept in the object's resident page
splay tree and memq.  Instead, they are kept in a separate per-object
splay tree of cached pages.  However, access to this new per-object
splay tree is synchronized by the _free_ page queues lock, not to be
confused with the heavily contended page queues lock.  Consequently, a
cached page can be reclaimed by vm_page_alloc(9) without acquiring the
object's lock or the page queues lock.

This solves a problem independently reported by tegge@ and Isilon.
Specifically, they observed the page daemon consuming a great deal of
CPU time because of pages bouncing back and forth between the cache
queue (PQ_CACHE) and the inactive queue (PQ_INACTIVE).  The source of
this problem turned out to be a deadlock avoidance strategy employed
when selecting a cached page to reclaim in vm_page_select_cache().
However, the root cause was really that reclaiming a cached page
required the acquisition of an object lock while the page queues lock
was already held.  Thus, this change addresses the problem at its
root, by eliminating the need to acquire the object's lock.

Moreover, keeping cached pages in the object's primary splay tree and
memq was, in effect, optimizing for the uncommon case.  Cached pages
are reclaimed far, far more often than they are reactivated.  Instead,
this change makes reclamation cheaper, especially in terms of
synchronization overhead, and reactivation more expensive, because
reactivated pages will have to be reentered into the object's primary
splay tree and memq.

(2) Cached pages are now stored alongside free pages in the physical
memory allocator's buddy queues, increasing the likelihood that large
allocations of contiguous physical memory (i.e., superpages) will
succeed.

Finally, as a result of this change long-standing restrictions on when
and where a cached page can be reclaimed and returned by
vm_page_alloc(9) are eliminated.  Specifically, calls to
vm_page_alloc(9) specifying VM_ALLOC_INTERRUPT can now reclaim and
return a formerly cached page.  Consequently, a call to malloc(9)
specifying M_NOWAIT is less likely to fail.

Discussed with: many over the course of the summer, including jeff@,
   Justin Husted @ Isilon, peter@, tegge@
Tested by: an earlier version by kris@
Approved by: re (kensmith)
2007-09-25 06:25:06 +00:00
Alan Cox
2446e4f02c Enable the new physical memory allocator.
This allocator uses a binary buddy system with a twist.  First and
foremost, this allocator is required to support the implementation of
superpages.  As a side effect, it enables a more robust implementation
of contigmalloc(9).  Moreover, this reimplementation of
contigmalloc(9) eliminates the acquisition of Giant by
contigmalloc(..., M_NOWAIT, ...).

The twist is that this allocator tries to reduce the number of TLB
misses incurred by accesses through a direct map to small, UMA-managed
objects and page table pages.  Roughly speaking, the physical pages
that are allocated for such purposes are clustered together in the
physical address space.  The performance benefits vary.  In the most
extreme case, a uniprocessor kernel running on an Opteron, I measured
an 18% reduction in system time during a buildworld.

This allocator does not implement page coloring.  The reason is that
superpages have much the same effect.  The contiguous physical memory
allocation necessary for a superpage is inherently colored.

Finally, the one caveat is that this allocator does not effectively
support prezeroed pages.  I hope this is temporary.  On i386, this is
a slight pessimization.  However, on amd64, the beneficial effects of
the direct-map optimization outweigh the ill effects.  I speculate
that this is true in general of machines with a direct map.

Approved by:	re
2007-06-16 04:57:06 +00:00
Attilio Rao
393a081d42 Optimize vmmeter locking.
In particular:
- Add an explicative table for locking of struct vmmeter members
- Apply new rules for some of those members
- Remove some unuseful comments

Heavily reviewed by: alc, bde, jeff
Approved by: jeff (mentor)
2007-06-10 21:59:14 +00:00
Attilio Rao
b4b7081961 Do proper "locking" for missing vmmeters part.
Now, we assume no more sched_lock protection for some of them and use the
distribuited loads method for vmmeter (distribuited through CPUs).

Reviewed by: alc, bde
Approved by: jeff (mentor)
2007-06-04 21:45:18 +00:00
Attilio Rao
2feb50bf7d Revert VMCNT_* operations introduction.
Probabilly, a general approach is not the better solution here, so we should
solve the sched_lock protection problems separately.

Requested by: alc
Approved by: jeff (mentor)
2007-05-31 22:52:15 +00:00
Jeff Roberson
222d01951f - define and use VMCNT_{GET,SET,ADD,SUB,PTR} macros for manipulating
vmcnts.  This can be used to abstract away pcpu details but also changes
   to use atomics for all counters now.  This means sched lock is no longer
   responsible for protecting counts in the switch routines.

Contributed by:		Attilio Rao <attilio@FreeBSD.org>
2007-05-18 07:10:50 +00:00
Alan Cox
19c244d064 Prevent a race between vm_object_collapse() and vm_object_split() from
causing a crash.

Suppose that we have two objects, obj and backing_obj, where
backing_obj is obj's backing object.  Further, suppose that
backing_obj has a reference count of two.  One being the reference
held by obj and the other by a map entry.  Now, suppose that the map
entry is deallocated and its reference removed by
vm_object_deallocate().  vm_object_deallocate() recognizes that the
only remaining reference is from a shadow object, obj, and calls
vm_object_collapse() on obj.  vm_object_collapse() executes

                if (backing_object->ref_count == 1) {
                        /*
                         * If there is exactly one reference to the backing
                         * object, we can collapse it into the parent.
                         */
                        vm_object_backing_scan(object, OBSC_COLLAPSE_WAIT);

vm_object_backing_scan(OBSC_COLLAPSE_WAIT) executes

        if (op & OBSC_COLLAPSE_WAIT) {
                vm_object_set_flag(backing_object, OBJ_DEAD);
        }

Finally, suppose that either vm_object_backing_scan() or
vm_object_collapse() sleeps releasing its locks.  At this instant,
another thread executes vm_object_split().  It crashes in
vm_object_reference_locked() on the assertion that the object is not
dead.  If, however, assertions are not enabled, it crashes much later,
after the object has been recycled, in vm_object_deallocate() because
the shadow count and shadow list are inconsistent.

Reviewed by: tegge
Reported by: jhb
MFC after: 1 week
2007-03-27 08:55:17 +00:00
Alan Cox
c5474b8f18 Change the order of lock reacquisition in vm_object_split() in order to
simplify the code slightly.  Add a comment concerning lock ordering.
2007-03-22 07:02:43 +00:00
John Baldwin
8db5fc58ff Use pause() in vm_object_deallocate() to yield the CPU to the lock holder
rather than a tsleep() on &proc0.  The only wakeup on &proc0 is intended
to awaken the swapper, not random threads blocked in
vm_object_deallocate().
2007-02-27 19:40:26 +00:00
Alan Cox
9f5c801b94 Change the way that unmanaged pages are created. Specifically,
immediately flag any page that is allocated to a OBJT_PHYS object as
unmanaged in vm_page_alloc() rather than waiting for a later call to
vm_page_unmanage().  This allows for the elimination of some uses of
the page queues lock.

Change the type of the kernel and kmem objects from OBJT_DEFAULT to
OBJT_PHYS.  This allows us to take advantage of the above change to
simplify the allocation of unmanaged pages in kmem_alloc() and
kmem_malloc().

Remove vm_page_unmanage().  It is no longer used.
2007-02-25 06:14:58 +00:00
Alan Cox
0cd31a0d75 Change the page's CLEANCHK flag from being a page queue mutex synchronized
flag to a vm object mutex synchronized flag.
2007-02-22 06:15:52 +00:00
Xin LI
f67af5c918 Use FOREACH_PROC_IN_SYSTEM instead of using its unrolled form. 2007-01-17 15:05:52 +00:00
Alan Cox
73000556e8 Optimize vm_object_split(). Specifically, make the number of iterations
equal to the number of physical pages that are renamed to the new object
rather than the new object's virtual size.
2006-12-17 20:14:43 +00:00
Alan Cox
95442adf05 Simplify the computation of the new object's size in vm_object_split(). 2006-12-16 08:17:07 +00:00
Alan Cox
2a53696fb8 The page queues lock is no longer required by vm_page_busy() or
vm_page_wakeup().  Reduce or eliminate its use accordingly.
2006-10-22 21:18:48 +00:00
Alan Cox
9af80719db Replace PG_BUSY with VPO_BUSY. In other words, changes to the page's
busy flag, i.e., VPO_BUSY, are now synchronized by the per-vm object
lock instead of the global page queues lock.
2006-10-22 04:28:14 +00:00
Alan Cox
b276ae6f6a Add _vm_stats and _vm_stats_misc to the sysctl declarations in sysctl.h and
eliminate their declarations from various source files.
2006-08-21 06:27:28 +00:00
Alan Cox
b146f9e5d2 Reimplement the page's NOSYNC flag as an object-synchronized instead of a
page queues-synchronized flag.  Reduce the scope of the page queues lock in
vm_fault() accordingly.

Move vm_fault()'s call to vm_object_set_writeable_dirty() outside of the
scope of the page queues lock.  Reviewed by: tegge
Additionally, eliminate an unnecessary dereference in computing the
argument that is passed to vm_object_set_writeable_dirty().
2006-08-13 00:11:09 +00:00
Alan Cox
5786be7cc7 Introduce a field to struct vm_page for storing flags that are
synchronized by the lock on the object containing the page.

Transition PG_WANTED and PG_SWAPINPROG to use the new field,
eliminating the need for holding the page queues lock when setting
or clearing these flags.  Rename PG_WANTED and PG_SWAPINPROG to
VPO_WANTED and VPO_SWAPINPROG, respectively.

Eliminate the assertion that the page queues lock is held in
vm_page_io_finish().

Eliminate the acquisition and release of the page queues lock
around calls to vm_page_io_finish() in kern_sendfile() and
vfs_unbusy_pages().
2006-08-09 17:43:27 +00:00
Alan Cox
91449ce98c When sleeping on a busy page, use the lock from the containing object
rather than the global page queues lock.
2006-08-03 23:56:11 +00:00
Alan Cox
78985e424a Complete the transition from pmap_page_protect() to pmap_remove_write().
Originally, I had adopted sparc64's name, pmap_clear_write(), for the
function that is now pmap_remove_write().  However, this function is more
like pmap_remove_all() than like pmap_clear_modify() or
pmap_clear_reference(), hence, the name change.

The higher-level rationale behind this change is described in
src/sys/amd64/amd64/pmap.c revision 1.567.  The short version is that I'm
trying to clean up and fix our support for execute access.

Reviewed by: marcel@ (ia64)
2006-08-01 19:06:06 +00:00
Alan Cox
604c2bbc34 Export the number of object bypasses and collapses through sysctl. 2006-07-22 22:31:57 +00:00
Alan Cox
af51d7bf57 Eliminate OBJ_WRITEABLE. It hasn't been used in a long time. 2006-07-21 06:40:29 +00:00
Alan Cox
2e9f4a698d Ensure that vm_object_deallocate() doesn't dereference a stale object
pointer: When vm_object_deallocate() sleeps because of a non-zero
paging in progress count on either object or object's shadow,
vm_object_deallocate() must ensure that object is still the shadow's
backing object when it reawakens.  In fact, object may have been
deallocated while vm_object_deallocate() slept.  If so, reacquiring
the lock on object can lead to a deadlock.

Submitted by: ups@
MFC after: 3 weeks
2006-07-17 06:45:03 +00:00
Tor Egge
3b582b4e72 Eliminate a deadlock when creating snapshots. Blocking vn_start_write() must
be called without any vnode locks held.  Remove calls to vn_start_write() and
vn_finished_write() in vnode_pager_putpages() and add these calls before the
vnode lock is obtained to most of the callers that don't already have them.
2006-03-02 22:13:28 +00:00
John Baldwin
ca95b5146a Lock the vm_object while checking its type to see if it is a vnode-backed
object that requires Giant in vm_object_deallocate().  This is somewhat
hairy in that if we can't obtain Giant directly, we have to drop the
object lock, then lock Giant, then relock the object lock and verify that
we still need Giant.  If we don't (because the object changed to OBJT_DEAD
for example), then we drop Giant before continuing.

Reviewed by:	alc
Tested by:	kris
2006-02-21 22:09:54 +00:00
Jeff Roberson
c05e22d44b - Install a temporary bandaid in vm_object_reference() that will stop
mtx_assert()s from triggering until I find a real long-term solution.
2006-02-01 09:47:02 +00:00
Alan Cox
997e1c252b Use the new macros abstracting the page coloring/queues implementation.
(There are no functional changes.)
2006-01-27 07:28:51 +00:00
Jeff Roberson
df59a0fee7 - Avoid calling vm_object_backing_scan() when collapsing an object when
the resident page count matches the object size.  We know it fully backs
   its parent in this case.

Reviewed by:	acl, tegge
Sponsored by:	Isilon Systems, Inc.
2006-01-25 08:42:58 +00:00
Alan Cox
02dd83311a Make vm_object_vndeallocate() static. The external calls to it were
eliminated in ufs/ffs/ffs_vnops.c's revision 1.125.
2006-01-22 23:56:20 +00:00
Alexander Leidinger
ef39c05baa MI changes:
- provide an interface (macros) to the page coloring part of the VM system,
   this allows to try different coloring algorithms without the need to
   touch every file [1]
 - make the page queue tuning values readable: sysctl vm.stats.pagequeue
 - autotuning of the page coloring values based upon the cache size instead
   of options in the kernel config (disabling of the page coloring as a
   kernel option is still possible)

MD changes:
 - detection of the cache size: only IA32 and AMD64 (untested) contains
   cache size detection code, every other arch just comes with a dummy
   function (this results in the use of default values like it was the
   case without the autotuning of the page coloring)
 - print some more info on Intel CPU's (like we do on AMD and Transmeta
   CPU's)

Note to AMD owners (IA32 and AMD64): please run "sysctl vm.stats.pagequeue"
and report if the cache* values are zero (= bug in the cache detection code)
or not.

Based upon work by:	Chad David <davidc@acns.ab.ca> [1]
Reviewed by:		alc, arch (in 2004)
Discussed with:		alc, Chad David, arch (in 2004)
2005-12-31 14:39:20 +00:00
Alan Cox
8215781ba2 Eliminate unneeded preallocation at initialization.
Reviewed by: tegge
2005-12-03 19:37:29 +00:00
Alan Cox
f6d8983846 Use of the ZERO_COPY_SOCKETS options can result in an unusual state that
vm_object_backing_scan() was not written to handle.  Specifically, a wired
page within a backing object that is shadowed by a page within the shadow
object.  Handle this state by removing the wired page from the backing
object.  The wired page will be freed by socow_iodone().

Stop masking errors: If a page is being freed by vm_object_backing_scan(),
assert that it is no longer mapped rather than quietly destroying any
mappings.

Tested by: Harald Schmalzbauer
2005-10-22 18:46:38 +00:00
Tor Egge
8dbca793a9 Don't allow pagedaemon to skip pages while scanning PQ_ACTIVE or PQ_INACTIVE
due to the vm object being locked.

When a process writes large amounts of data to a file, the vm object associated
with that file can contain most of the physical pages on the machine.  If the
process is preempted while holding the lock on the vm object, pagedaemon would
be able to move very few pages from PQ_INACTIVE to PQ_CACHE or from PQ_ACTIVE
to PQ_INACTIVE, resulting in unlimited cleaning of dirty pages belonging to
other vm objects.

Temporarily unlock the page queues lock while locking vm objects to avoid lock
order violation.  Detect and handle relevant page queue changes.

This change depends on both the lock portion of struct vm_object and normal
struct vm_page being type stable.

Reviewed by:	alc
2005-08-10 00:17:36 +00:00
Jeff Roberson
b8a0b997fd - We need to inhert the OBJ_NEEDGIANT flag from the original object in
vm_object_split().

Spotted by:	alc
2005-05-04 20:54:16 +00:00
Jeff Roberson
ed4fe4f4f5 - Add a new object flag "OBJ_NEEDSGIANT". We set this flag if the
underlying vnode requires Giant.
 - In vm_fault only acquire Giant if the underlying object has NEEDSGIANT
   set.
 - In vm_object_shadow inherit the NEEDSGIANT flag from the backing object.
2005-05-03 11:11:26 +00:00
Alan Cox
c6ec6a7cae Eliminate (now) unnecessary acquisition and release of the global page
queues lock in vm_object_backing_scan().  Updates to the page's PG_BUSY
flag and busy field are synchronized by the containing object's lock.

Testing the page's hold_count and wire_count in vm_object_backing_scan()'s
OBSC_COLLAPSE_NOWAIT case is unnecessary.  There is no reason why the held
or wired pages cannot be migrated to the shadow object.

Reviewed by: tegge
2005-03-30 05:40:02 +00:00
Jeff Roberson
ee39666a76 - Don't lock the vnode interlock in vm_object_set_writeable_dirty() if
we've already set the object flags.

Reviewed by:	alc
2005-03-17 12:03:42 +00:00
Alan Cox
8e99783b25 Update the text of an assertion to reflect changes made in revision 1.148.
Submitted by: tegge

Eliminate an unnecessary, temporary increment of the backing object's
reference count in vm_object_qcollapse().  Reviewed by: tegge
2005-01-30 21:29:47 +00:00
Jeff Roberson
ae51ff1127 - Remove GIANT_REQUIRED where giant is no longer required.
- Use VFS_LOCK_GIANT() rather than directly acquiring giant in places
   where giant is only held because vfs requires it.

Sponsored By:   Isilon Systems, Inc.
2005-01-24 10:48:29 +00:00
Alan Cox
d936694f09 Consider three objects, O, BO, and BBO, where BO is O's backing object
and BBO is BO's backing object.  Now, suppose that O and BO are being
collapsed.  Furthermore, suppose that BO has been marked dead
(OBJ_DEAD) by vm_object_backing_scan() and that either
vm_object_backing_scan() has been forced to sleep due to encountering
a busy page or vm_object_collapse() has been forced to sleep due to
memory allocation in the swap pager.  If vm_object_deallocate() is
then called on BBO and BO is BBO's only shadow object,
vm_object_deallocate() will collapse BO and BBO.  In doing so, it adds
a necessary temporary reference to BO.  If this collapse also sleeps
and the prior collapse resumes first, the temporary reference will
cause vm_object_collapse to panic with the message "backing_object %p
was somehow re-referenced during collapse!"

Resolve this race by changing vm_object_deallocate() such that it
doesn't collapse BO and BBO if BO is marked dead.  Once O and BO are
collapsed, vm_object_collapse() will attempt to collapse O and BBO.
So, vm_object_deallocate() on BBO need do nothing.

Reported by: Peter Holm on 20050107
URL: http://www.holm.cc/stress/log/cons102.html

In collaboration with: tegge@
Candidate for RELENG_4 and RELENG_5
MFC after: 2 weeks
2005-01-15 21:12:47 +00:00
Poul-Henning Kamp
7c0745eeae Eliminate unused and unnecessary "cred" argument from vinvalbuf() 2005-01-14 07:33:51 +00:00
Poul-Henning Kamp
8df6bac4c7 Remove the unused credential argument from VOP_FSYNC() and VFS_SYNC().
I'm not sure why a credential was added to these in the first place, it is
not used anywhere and it doesn't make much sense:

	The credentials for syncing a file (ability to write to the
	file) should be checked at the system call level.

	Credentials for syncing one or more filesystems ("none")
	should be checked at the system call level as well.

	If the filesystem implementation needs a particular credential
	to carry out the syncing it would logically have to the
	cached mount credential, or a credential cached along with
	any delayed write data.

Discussed with:	rwatson
2005-01-11 07:36:22 +00:00
Alan Cox
5ba514bc89 Move the acquisition and release of the page queues lock outside of a loop
in vm_object_split() to avoid repeated acquisition and release.
2005-01-08 23:41:11 +00:00
Warner Losh
60727d8b86 /* -> /*- for license, minor formatting changes 2005-01-07 02:29:27 +00:00
Alan Cox
98fe9a0ddf Eliminate another unnecessary call to vm_page_busy(). (See revision 1.333
for a detailed explanation.)
2004-12-17 18:54:51 +00:00