summaryrefslogtreecommitdiffstats
path: root/sys/vm
Commit message (Expand)AuthorAgeFilesLines
* Export various helper variables describing the layout and size ofjhb2015-11-121-0/+3
* Ensure that deactivated pages that are not expected to be reused aremarkj2015-11-083-9/+16
* Reduce the amount of calls to VOP_BMAP() made from the local vnodekib2015-10-241-81/+39
* Fix capitalizationjah2015-10-231-1/+1
* Remove unclear comment about address truncation in busdma. Add (hopefully mu...jah2015-10-231-0/+5
* Only marker is guaranteed to be present on the queue after the relockkib2015-10-181-5/+17
* Revert r289302, invalid pages can be queued, e.g. by vfs_vmio_unwire().kib2015-10-151-5/+5
* Invalid pages should not appear on the inactive queue. Change thekib2015-10-141-5/+5
* Parallelize the buffer cache and rewrite getnewbuf(). This results in ajeff2015-10-141-1/+5
* Exploit r288122 to avoid pointlessly enqueueing a page that is about to bealc2015-10-091-1/+1
* Exploit r288122 to address a cosmetic issue. Pages belonging to eitheralc2015-10-061-1/+1
* Mark swap_pager_putpages static at its definition. It was alreadyimp2015-10-051-3/+1
* Reduce the scope of a variable to the only file where it is used.alc2015-10-033-2/+2
* As a step towards the elimination of PG_CACHED pages, rework the handlingmarkj2015-09-304-15/+28
* The conversion of kmem_alloc_attr() from operating on a vm map to a vmemalc2015-09-261-28/+13
* Exploit r288122 to address a cosmetic issue. Since the pages allocatedalc2015-09-261-1/+1
* Change vm_page_unwire() such that it (1) accepts PQ_NONE as the specifiedalc2015-09-222-17/+21
* Correct a non-fatal error in vm_pageout_worker(). vm_pageout_worker()alc2015-09-201-6/+11
* Eliminate (many) unnecessary calls to pmap_remove_all(). Pages from objectsalc2015-09-171-1/+2
* Remove the v_cache_min and v_cache_max sysctls. They are unused and havemarkj2015-09-113-9/+2
* Remove a check which caused spurious SIGSEGV on usermode access to thekib2015-09-091-10/+4
* The swap pager is compatible with direct dispatch. It does its ownimp2015-09-081-11/+42
* To simplify upcoming changes to the inactive queue scan, change the codealc2015-09-081-18/+8
* Eliminate pointless requeueing of pages from terminated objects. Thesealc2015-09-051-23/+25
* Don't trash memory from UMA_ZONE_NOFREE zones.mjg2015-09-021-2/+2
* Handle held pages earlier in the inactive queue scan.alc2015-09-011-31/+33
* Remove weighted page handling from vm_page_advise().markj2015-08-281-54/+25
* In vm_pageout_scan(), simplify the logic for determining if a page can bealc2015-08-271-15/+15
* Testing whether a page is dirty does not require the page lock. Moreover,alc2015-08-251-6/+10
* Make the UMA harvesting go away completely if not wanted. Default to "not wan...markm2015-08-221-4/+4
* Eliminate pointless assignments to rtvals[] in swap_pager_putpages().alc2015-08-211-12/+12
* Prevent ticks rollover from preventing vm_lowmem eventrstone2015-08-201-3/+4
* Add the kernel support for minidumps on arm64.andrew2015-08-201-7/+7
* As another piece of PG_CACHE page elimination, remove an LRU-defeating callalc2015-08-161-5/+0
* Make kstack_pages a tunable on arm, x86, and powepc. On i386, thekib2015-08-101-5/+5
* Avoid sign extension of value passed to kva_alloc from uma_zone_reserve_kvazbb2015-08-101-2/+2
* Introduce a sysctl for reporting the number of fully populated reservations.alc2015-08-061-0/+32
* Properly sort the function declarations added in r286296jah2015-08-051-2/+2
* Add two new pmap functions:jah2015-08-041-0/+2
* Refinements to r281079's sequential access optimization: Prefetched pages,alc2015-08-031-2/+12
* Do not pretend that vm_fault(9) supports unwiring the address. Renamekib2015-07-303-26/+21
* - Make 'struct buf *buf' private to vfs_bio.c. Having a global variablejeff2015-07-291-0/+2
* Revert r173708's modifications to vm_object_page_remove().kib2015-07-252-24/+5
* Refactor unmapped buffer address handling.jeff2015-07-233-29/+18
* Add an initial NUMA affinity/policy configuration for threads and processes.adrian2015-07-114-11/+596
* The intention of r254304 was to scan the active queue continuously.alc2015-07-082-15/+20
* Add a local variable initialization needed in the OBJT_DEFAULT case.markj2015-07-051-0/+1
* vm: don't lock proc around accesses to vm_{t,d}addr and RLIMIT_DATA in sys_mmapmjg2015-07-021-4/+2
* Account for the main process stack being one page below the highestkib2015-07-021-1/+2
* Huge cleanup of random(4) code.markm2015-06-301-32/+4
OpenPOWER on IntegriCloud