summaryrefslogtreecommitdiff
path: root/sys/vm/vm_object.c
AgeCommit message (Collapse)Author
1998-04-25typoNiklas Hallqvist
1998-03-01Merge of MACHINE_NEW_CONTIG (aka MNN) code from Chuck Cranor,Niklas Hallqvist
<chuck@openbsd.org>. This code is as of yet disabled on all platforms, actually not yet supported on more than mvme68k, although other platforms are expected soon, as code is already available. This code makes handling of multiple physical memory regions consistent over all platforms, as well as keeping the performance of maintaining a single continuous memory chunk. It is also a requirement for the upcoming UVM replacement VM system. What I did in this merge: just declared the pmap_map function in a MD include file per port that needs it. It's not an exported pmap interface, says Chuck. It ended up in differnt include files on differnet ports, as I tried to follow the current policy on a per-arch basis.
1997-11-06Updates for VFS Lite 2 + soft update.Constantine Sapuntzakis
1997-10-06back out vfs lite2 till after 2.2Theo de Raadt
1997-10-06VFS Lite2 ChangesConstantine Sapuntzakis
1997-07-25tabifyMichael Shalayeff
some /lx/p/ printf changes
1997-05-28move vm_cache_max into param.cMichael Shalayeff
make it maxusers adaptive note, that NTEXT approximation is probably not the best idea
1997-04-17Removal of race conditions. Inspired by Charles Hannum'sNiklas Hallqvist
<mycroft@netbsd.org> reorganization of the vm_collapse logic, although not used verbatim. We no longer collapse objects from the pagedaemon as that is not necessary anymore with the more aggressive collapses that gets done. This also increases performance of loaded systems. Much KNF too.
1997-03-26Just some initial KNF in preparation for some other changes coming soonNiklas Hallqvist
1996-12-24Make termination of objects wait for possible collapses before alteringNiklas Hallqvist
the object's shadow linkage. Also note that the object is going to die so that a possible collapse can finish early.
1996-11-06Be more restrictive with concurrent uses of the objects involved whileNiklas Hallqvist
paging in.
1996-08-19A front pager can be lost during pagein, check for this caseNiklas Hallqvist
1996-08-18Restructured vm_object_collapse_aux so sharing object's pageins can go onNiklas Hallqvist
simultaneously without interfering. Also maintain PG_BUSY at pagein time. Removed some statics so gdb will be happier.
1996-08-14IMPORTANT FIX: arbitrary page could be removed during collapse.Niklas Hallqvist
Sanitize "fake" page handling once more.
1996-08-13Clean up vm_object_collapse_aux's dirty page game. Pun intended :-)Niklas Hallqvist
1996-08-12Fixed CRITICAL bug hitting in former swap leak scenariosNiklas Hallqvist
when collapsing two objects a formerly COWed page might lose it's backing store causing stuff like lost ld.so fixups of shared lib images. Also handle fake pages in all cases. In short: UPDATE!
1996-08-02Fix long-standing swap-leak. Add OpenBSD tags. Optimize thread_wakeup.Niklas Hallqvist
1996-07-23make printf/addlog return 0, for compat to userlandTheo de Raadt
1996-04-19NetBSD 960317 mergeNiklas Hallqvist
1996-03-03From NetBSD: merge with 960217Niklas Hallqvist
1995-12-14from netbsd:Theo de Raadt
Extend use of vm_object_prefer() to vm_allocate_with_pager(). Make vm_object_prefer() call MD aligner for "pageless" objects too, so we can have more control over the virtual address to be used. Implementation could be simpler if we by-pass the object to mapped, but we'd loose the ability to adapt alignment to objects that were previously mmap'ed with MAP_FIXED on. Only expect vm_fork() to return if __FORK_BRAINDAMAGE is defined. Eliminate unused third arg to vm_fork().
1995-10-18initial import of NetBSD treeTheo de Raadt