aboutsummaryrefslogtreecommitdiff
path: root/sys/vm
Commit message (Expand)AuthorAgeFilesLines
* Replace the linear search in vm_map_findspace() with an O(log n)Alan Cox2004-08-132-98/+213
* The vm map lock is needed in vm_fault() after the page has been found,Tor Egge2004-08-123-58/+141
* Re-delete the comment from r1.352.Brian Feldman2004-08-121-3/+0
* Back out all behavioral chnages.Brian Feldman2004-08-102-36/+17
* Revamp VM map wiring.Brian Feldman2004-08-093-14/+50
* Make two changes to vm_fault().Alan Cox2004-08-091-16/+7
* Remove a stale comment from vm_map_lookup() that pertains to share maps.Alan Cox2004-08-091-3/+0
* Make two changes to vm_fault().Alan Cox2004-08-091-11/+6
* Tag all geom classes in the tree with a version number.Poul-Henning Kamp2004-08-081-3/+5
* Remove dead code. A vm_map's first_free is never NULL (even if the map isAlan Cox2004-08-071-7/+2
* Generate KTR trace records for uma_zalloc_arg() and uma_zfree_arg().Robert Watson2004-08-061-0/+6
* Turn on the new contigmalloc(9) by default. There should not actuallyBrian Feldman2004-08-051-1/+1
* Remove a product specific workaround for wrong modes when mmap(2)'ingPoul-Henning Kamp2004-08-051-15/+2
* - Push down the acquisition and release of Giant into pmap_enter_quick()Alan Cox2004-08-041-2/+0
* In dev_pager_updatefake, m->valid is typically 0 on entry. ItDoug Rabson2004-08-041-1/+2
* Eliminate the acquisition and release of Giant around the call toAlan Cox2004-08-021-2/+0
* * Add a "how" argument to uma_zone constructors and initialization functionsBrian Feldman2004-08-026-86/+167
* - Push down the acquisition and release of Giant into pmap_protect() onAlan Cox2004-07-301-2/+0
* Giant is no longer required by vm_waitproc() and vmspace_exitfree().Alan Cox2004-07-301-1/+0
* Fix a memory leak in the device pager which is exposed by the NVIDIADoug Rabson2004-07-301-13/+41
* Fix handling of msync(2) for character special files.Doug Rabson2004-07-301-1/+3
* Get rid of another lockmgr(9) consumer by using sx locks for the userMaxime Henrion2004-07-302-27/+19
* Advance the state of pmap locking on alpha, amd64, and i386.Alan Cox2004-07-291-6/+3
* Rework the way slab header storage space is calculated in UMA.Bosko Milekic2004-07-292-52/+176
* Correct a very old error in both vm_object_madvise() (originating inAlan Cox2004-07-281-2/+2
* - Use atomic ops for updating the vmspace's refcnt and exitingcnt.Alan Cox2004-07-272-8/+13
* For years, kmem_alloc_pageable() has been misused. Now that the last ofAlan Cox2004-07-252-25/+0
* Remove spl calls.Alan Cox2004-07-251-11/+0
* Make the code and comments for vm_object_coalesce() consistent.Alan Cox2004-07-253-9/+6
* Simplify vmspace initialization. The bcopy() of fields from the oldAlan Cox2004-07-242-14/+8
* - Change uma_zone_set_obj() to call kmem_alloc_nofault() instead ofAlan Cox2004-07-224-13/+10
* Fix a race in vm_page_sleep_if_busy(). Due to vm_object lockingBrian Feldman2004-07-211-4/+12
* Semi-gratuitous change. Move two refcount operations to their own linesPeter Wemm2004-07-211-2/+4
* Move the initialization and teardown of pmaps to the vmspace zone'sPeter Wemm2004-07-211-3/+2
* Remove extraneous locks on the VM free page queue mutex; it is notBrian Feldman2004-07-191-2/+0
* - Eliminate the pte object from the pmap. Instead, page table pages areAlan Cox2004-07-191-2/+0
* Since breakage of malloc(9)/uma_zalloc(9) is totally non-optional inBrian Feldman2004-07-191-0/+6
* Reimplement contigmalloc(9) with an algorithm which stands a greatly-Brian Feldman2004-07-192-36/+273
* Remove the GIANT_REQUIRED preceding pmap_remove() inAlan Cox2004-07-181-1/+0
* Push down the acquisition and release of the page queues lock intoAlan Cox2004-07-152-10/+0
* Remove an unused and unimplemented sysctl. (For the record, it was markedAlan Cox2004-07-121-10/+1
* Increase the scope of the page queues lock in vm_page_alloc() to coverAlan Cox2004-07-101-1/+1
* Micro-optimize vmspace for 64-bit architectures: Colocate vm_refcnt andAlan Cox2004-07-061-1/+1
* Properly brucify a string by outdenting it.Bruce M Simpson2004-07-061-2/+2
* Introduce debug.nosleepwithlocks sysctl, 0 by default. If set to 1Bosko Milekic2004-07-041-11/+9
* Reextend the M_WAITOK-disabling-hack to all three of the mbuf-relatedBrian Feldman2004-07-041-2/+4
* Limit mbuma damage. Suddenly ALL allocations with M_WAITOK are subjectBrian Feldman2004-07-031-4/+8
* Implement preemption of kernel threads natively in the scheduler ratherJohn Baldwin2004-07-021-0/+2
* - Change mi_switch() and sched_switch() to accept an optional thread toJohn Baldwin2004-07-021-1/+1
* - Don't use a variable to point to the user area that we only use once.John Baldwin2004-07-021-15/+10