summaryrefslogtreecommitdiffstats
path: root/sys/vm/vm_fault.c
Commit message (Expand)AuthorAgeFilesLines
* Don't call vm_fault_prefault() on zero-fill faults. It's a waste of time.alc2014-02-091-1/+4
* Make prefaulting more aggressive on hard faults. Previously, we would onlyalc2014-02-021-24/+32
* Revert back to use int for the page counts. In vn_io_fault(), the i/okib2013-11-201-4/+4
* Avoid overflow for the page counts.kib2013-11-121-1/+1
* Remove zero-copy sockets code. It only worked for anonymous memory,kib2013-09-161-19/+1
* On all the architectures, avoid to preallocate the physical memoryattilio2013-08-091-3/+5
* The soft and hard busy mechanism rely on the vm object lock to work.attilio2013-08-091-26/+20
* Revert r253939:attilio2013-08-051-8/+5
* The page hold mechanism is fast but it has couple of fallouts:attilio2013-08-041-5/+8
* The vm_fault() should not be allowed to proceed on the map entry whichkib2013-07-111-0/+13
* Acquire read lock on the src object for vm_fault_copy_entry().attilio2013-05-221-4/+4
* Relax the object locking in vm_fault_prefault(). A read lock suffices.alc2013-05-171-5/+5
* Hide the details for the assertion for VM_OBJECT_LOCK operations.attilio2013-02-211-2/+2
* Rename VM_OBJECT_LOCK(), VM_OBJECT_UNLOCK() and VM_OBJECT_TRYLOCK() toattilio2013-02-201-39/+39
* Switch vm_object lock to be a rwlock.attilio2013-02-201-3/+3
* - Add system wide page faults requiring I/O counter.zont2013-01-281-2/+3
* In the past four years, we've added two new vm object types. Each time,alc2012-12-091-2/+2
* Replace the single, global page queues lock with per-queue locks on thealc2012-11-131-1/+1
* Commit the actual text provided by Alan, instead of the wrong updatekib2012-10-241-5/+7
* Dirty the newly copied anonymous pages after the wired region iskib2012-10-241-3/+6
* Remove the support for using non-mpsafe filesystem modules.kib2012-10-221-21/+0
* Calculate the count of per-process cow faults. Export the count tokib2012-05-231-0/+1
* Give vm_fault()'s sequential access optimization a makeover.alc2012-05-101-68/+98
* Add new ktrace records for the start and end of VM faults. This givesjhb2012-04-051-2/+19
* Handle spurious page faults that may occur in no-fault sections of thealc2012-03-221-1/+7
* Use the trick of performing the atomic operation on the contained alignedkib2011-09-281-10/+2
* Split the vm_page flags PG_WRITEABLE and PG_REFERENCED into atomickib2011-09-061-3/+1
* Add a facility to disable processing page faults. When activated,kib2011-07-091-0/+16
* Revert to using the page queues lock in vm_page_clear_dirty_mask() onalc2011-06-231-2/+1
* Precisely document the synchronization rules for the page's dirty field.alc2011-06-191-0/+10
* Handle the corner case in vm_fault_quick_hold_pages().kib2011-03-251-0/+2
* For some time now, the kernel and kmem objects have been ordinaryalc2011-01-151-4/+1
* Correct a typo in vm_fault_quick_hold_pages().alc2010-12-281-1/+1
* Retire vm_fault_quick(). It's no longer used.alc2010-12-251-18/+0
* Introduce and use a new VM interface for temporarily pinning pages. Thisalc2010-12-251-0/+75
* Introduce vm_fault_hold() and use it to (1) eliminate a long-standing racealc2010-12-201-5/+18
* Replace pointer to "struct uidinfo" with pointer to "struct ucred"trasz2010-12-021-5/+5
* Use vm_page_prev() instead of vm_page_lookup() in the implementation ofalc2010-07-021-10/+12
* When waiting for the busy page, do not unlock the object unless unlockkib2010-05-201-3/+6
* Push down the acquisition of the page queues lock into vm_pageq_remove().alc2010-05-091-2/+0
* Minimize the scope of the page queues lock in vm_fault().alc2010-05-081-1/+2
* Push down the page queues into vm_page_cache(), vm_page_try_to_cache(), andalc2010-05-081-12/+3
* Push down the page queues lock into vm_page_activate().alc2010-05-071-6/+1
* Push down the page queues lock into vm_page_deactivate(). Eliminate analc2010-05-071-2/+0
* Eliminate page queues locking around most calls to vm_page_free().alc2010-05-061-11/+1
* Acquire the page lock around all remaining calls to vm_page_free() onalc2010-05-051-2/+0
* Push down the acquisition of the page queues lock into vm_page_unwire().alc2010-05-051-9/+5
* Add page locking to the vm_page_cow* functions.alc2010-05-041-6/+0
* Simplify vm_fault(). The introduction of the new page lock renders a bit ofalc2010-05-021-13/+5
* It makes no sense for vm_page_sleep_if_busy()'s helper, vm_page_sleep(),alc2010-05-021-0/+6
OpenPOWER on IntegriCloud