| Commit message (Expand) | Author | Age | Files | Lines |
... | |
* | On all the architectures, avoid to preallocate the physical memory | attilio | 2013-08-09 | 10 | -116/+408 |
* | The soft and hard busy mechanism rely on the vm object lock to work. | attilio | 2013-08-09 | 10 | -233/+421 |
* | Split the pagequeues per NUMA domains, and split pageademon process | kib | 2013-08-07 | 6 | -177/+414 |
* | Replace kernel virtual address space allocation with vmem. This provides | jeff | 2013-08-07 | 12 | -367/+278 |
* | Fill in the description fields for M_FICT_PAGES. | markj | 2013-08-07 | 1 | -1/+1 |
* | Revert r253939: | attilio | 2013-08-05 | 4 | -18/+16 |
* | The page hold mechanism is fast but it has couple of fallouts: | attilio | 2013-08-04 | 4 | -16/+18 |
* | Unbreak sysctl ABI changes introduced in r253662 | zont | 2013-07-29 | 1 | -2/+4 |
* | Improve page LRU quality and simplify the logic. | jeff | 2013-07-26 | 1 | -71/+57 |
* | Remove define and documentation for vm_pageout_algorithm missed in r253587 | zont | 2013-07-26 | 1 | -4/+2 |
* | Clear entire map structure including locks so that the | kientzle | 2013-07-25 | 1 | -2/+1 |
* | rename scheduler->swapper and SI_SUB_RUN_SCHEDULER->SI_SUB_LAST | avg | 2013-07-24 | 2 | -18/+4 |
* | Since r251709 a slab no longer use 8-bit indicies to manage items, | glebius | 2013-07-24 | 1 | -8/+0 |
* | - Remove the long obsolete 'vm_pageout_algorithm' experiment. | jeff | 2013-07-24 | 1 | -9/+2 |
* | - Correct a stale comment. We don't have vclean() anymore. The work is | jeff | 2013-07-23 | 1 | -5/+0 |
* | Revert r249590 and in case if mp_ncpus isn't initialized use MAXCPU. This | glebius | 2013-07-23 | 1 | -2/+3 |
* | Fix previous commit when option RACCT is not used. | jlh | 2013-07-22 | 1 | -0/+2 |
* | Fix a panic in the racct code when munlock(2) is called with incorrect values. | jlh | 2013-07-22 | 1 | -1/+4 |
* | Be more aggressive in using superpages in all mappings of objects: | jhb | 2013-07-19 | 3 | -8/+20 |
* | When swap pager allocates metadata in the pagedaemon context, allow it | kib | 2013-07-11 | 1 | -1/+2 |
* | The vm_fault() should not be allowed to proceed on the map entry which | kib | 2013-07-11 | 1 | -0/+13 |
* | The mlockall() or VM_MAP_WIRE_HOLESOK does not interact properly with | kib | 2013-07-11 | 2 | -11/+52 |
* | Never remove user-wired pages from an object when doing | kib | 2013-07-11 | 2 | -9/+11 |
* | In the vm_page_set_invalid() function, do not assert that the page is | kib | 2013-07-11 | 1 | -2/+0 |
* | Fix typo in comment. | kib | 2013-07-09 | 1 | -1/+1 |
* | vm_phys_fictitious_reg_range() was losing the 'memattr' because it would be | neel | 2013-07-03 | 2 | -1/+1 |
* | Remove a spurious keg lock acquisition. | davide | 2013-06-28 | 1 | -1/+1 |
* | - Add a general purpose resource allocator, vmem, from NetBSD. It was | jeff | 2013-06-28 | 7 | -40/+29 |
* | - Resolve bucket recursion issues by passing a cookie with zone flags | jeff | 2013-06-26 | 3 | -43/+86 |
* | Typo in comment. | glebius | 2013-06-24 | 1 | -1/+1 |
* | - Add a per-zone lock for zones without kegs. | jeff | 2013-06-20 | 4 | -105/+104 |
* | - Persist the caller's flags in the bucket allocation flags so we don't | jeff | 2013-06-19 | 1 | -1/+1 |
* | Fix a bug that allowed a tracing process (e.g. gdb) to write | des | 2013-06-18 | 1 | -0/+6 |
* | Refine UMA bucket allocation to reduce space consumption and improve | jeff | 2013-06-18 | 2 | -309/+264 |
* | - Add a new UMA API: uma_zcache_create(). This makes a zone without any | jeff | 2013-06-17 | 3 | -216/+293 |
* | - Convert the slab free item list from a linked array of indices to a | jeff | 2013-06-13 | 3 | -295/+149 |
* | Revise the interface between vm_object_madvise() and vm_page_dontneed() so | alc | 2013-06-10 | 3 | -31/+30 |
* | Make sys_mlock() function just a wrapper around vm_mlock() function | glebius | 2013-06-08 | 2 | -5/+11 |
* | Complete r251452: | attilio | 2013-06-06 | 2 | -4/+7 |
* | In vm_object_split(), busy and consequently unbusy the pages only when | attilio | 2013-06-04 | 1 | -3/+4 |
* | Update a comment. | alc | 2013-06-04 | 1 | -2/+2 |
* | Relax the object locking in vm_pageout_map_deactivate_pages() and | alc | 2013-06-04 | 1 | -11/+11 |
* | Remove irrelevant comments. | kib | 2013-06-03 | 1 | -7/+0 |
* | Require that the page lock is held, instead of the object lock, when | alc | 2013-06-03 | 2 | -7/+16 |
* | Now that access to the page's "act_count" field is synchronized by the page | alc | 2013-06-01 | 1 | -1/+0 |
* | Simplify the definition of vm_page_lock_assert(). There is no compelling | alc | 2013-05-31 | 1 | -7/+6 |
* | After the object lock was dropped, the object' reference count could | kib | 2013-05-30 | 1 | -5/+5 |
* | Remove the capitalization in the assertion message. Print the address | kib | 2013-05-30 | 1 | -1/+1 |
* | o Change the locking scheme for swp_bcount. | attilio | 2013-05-28 | 1 | -5/+7 |
* | Acquire read lock on the src object for vm_fault_copy_entry(). | attilio | 2013-05-22 | 1 | -4/+4 |