| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Simplify the implementation of the failure case in kmem_alloc_attr(). | alc | 2011-11-04 | 1 | -8/+7 |
| * | Eliminate vm_phys_bootstrap_alloc(). It was a failed attempt at | alc | 2011-10-30 | 1 | -1/+2 |
| * | Use "u_long" instead of "unsigned long". | alc | 2011-10-28 | 1 | -3/+2 |
| * | contigmalloc(9) and contigfree(9) are now implemented in terms of other | alc | 2011-10-27 | 1 | -28/+0 |
| * | Fix an error in kmem_alloc_attr(). Unless "tries" is updated, | alc | 2011-08-07 | 1 | -0/+1 |
| * | Remove unneeded includes of <sys/linker_set.h>. Other headers that use | jhb | 2011-01-11 | 1 | -1/+0 |
| * | There is no point in vm_contig_launder{,_page}() flushing held pages, | alc | 2010-12-29 | 1 | -5/+4 |
| * | vm_pageout_flush() might cache the pages that finished write to the | kib | 2010-11-18 | 1 | -1/+1 |
| * | With the demise of page coloring, the page queue macros no longer serve any | alc | 2010-07-02 | 1 | -1/+1 |
| * | Make vm_contig_grow_cache() extern, and use it when vm_phys_alloc_contig() | jchandra | 2010-06-04 | 1 | -9/+12 |
| * | Do not leak vm page lock in vm_contig_launder(), vm_pageout_page_lock() | kib | 2010-06-03 | 1 | -1/+3 |
| * | Continue cleaning the queue instead of moving to the next queue or | kib | 2010-05-10 | 1 | -4/+2 |
| * | Add a helper function vm_pageout_page_lock(), similar to tegge' | kib | 2010-05-06 | 1 | -2/+4 |
| * | The pages allocated by kmem_alloc_attr() and kmem_malloc() are unmanaged. | alc | 2010-05-03 | 1 | -4/+0 |
| * | This change addresses the race condition that was introduced by the previous | alc | 2010-05-02 | 1 | -14/+16 |
| * | - acquire the page lock in vm_contig_launder_page before checking page fields | kmacy | 2010-04-30 | 1 | -6/+15 |
| * | On Alan's advice, rather than do a wholesale conversion on a single | kmacy | 2010-04-30 | 1 | -0/+2 |
| * | Introduce the function kmem_alloc_attr(), which allocates kernel virtual | alc | 2010-04-09 | 1 | -18/+97 |
| * | Add support to the virtual memory system for configuring machine- | alc | 2009-07-12 | 1 | -6/+8 |
| * | This change is the next step in implementing the cache control functionality | alc | 2009-06-26 | 1 | -3/+5 |
| * | Refactor contigmalloc() into two functions: a simple front-end that deals | alc | 2009-06-17 | 1 | -8/+19 |
| * | Pass the size of the mapping to contigmapping() as a "vm_size_t" rather | alc | 2009-06-17 | 1 | -16/+13 |
| * | Make the maintenance of a page's valid bits by contigmalloc() more like | alc | 2009-06-17 | 1 | -0/+1 |
| * | Simplify contigfree(). | alc | 2009-06-05 | 1 | -3/+1 |
| * | - back out direct map hack | kmacy | 2009-05-19 | 1 | -6/+1 |
| * | apply band-aid to x86_64 systems with more physical memory than kmem by alloc... | kmacy | 2009-05-16 | 1 | -1/+6 |
| * | VOP_LOCK1() (and so VOP_LOCK()) and VOP_UNLOCK() are only used in | attilio | 2008-01-13 | 1 | -1/+1 |
| * | vn_lock() is currently only used with the 'curthread' passed as argument. | attilio | 2008-01-10 | 1 | -1/+1 |
| * | Make contigmalloc(9)'s page laundering more robust. Specifically, use | alc | 2007-11-25 | 1 | -3/+6 |
| * | Tidy up: Add comments. Eliminate the pointless | alc | 2007-11-25 | 1 | -19/+23 |
| * | Change the management of cached pages (PQ_CACHE) in two fundamental | alc | 2007-09-25 | 1 | -10/+1 |
| * | Enable the new physical memory allocator. | alc | 2007-06-16 | 1 | -374/+42 |
| * | Conditionally acquire Giant in vm_contig_launder_page(). | alc | 2007-06-11 | 1 | -0/+4 |
| * | Revert VMCNT_* operations introduction. | attilio | 2007-05-31 | 1 | -2/+2 |
| * | - define and use VMCNT_{GET,SET,ADD,SUB,PTR} macros for manipulating | jeff | 2007-05-18 | 1 | -2/+2 |
| * | Correct contigmalloc2()'s implementation of M_ZERO. Specifically, | alc | 2007-04-19 | 1 | -1/+1 |
| * | Change the free page queue lock from a spin mutex to a default (blocking) | alc | 2007-02-05 | 1 | -7/+7 |
| * | Ensure that the page's oflags field is initialized by contigmalloc(). | alc | 2006-11-08 | 1 | -0/+1 |
| * | Replace PG_BUSY with VPO_BUSY. In other words, changes to the page's | alc | 2006-10-22 | 1 | -2/+2 |
| * | sun4v requires TSBs (translation storage buffers) to be contiguous and be | kmacy | 2006-10-12 | 1 | -4/+6 |
| * | Make vm_page_release_contig() static. | alc | 2006-09-03 | 1 | -1/+1 |
| * | Prevent a call to contigmalloc() that asks for more physical memory than | alc | 2006-08-26 | 1 | -1/+1 |
| * | Ignore dirty pages owned by "dead" objects. | tegge | 2006-03-08 | 1 | -0/+4 |
| * | Eliminate a deadlock when creating snapshots. Blocking vn_start_write() must | tegge | 2006-03-02 | 1 | -0/+3 |
| * | Hold extra reference to vm object while cleaning pages. | tegge | 2006-03-02 | 1 | -0/+2 |
| * | The change a few years ago of having contigmalloc start its scan at the top | scottl | 2006-01-29 | 1 | -2/+19 |
| * | Plug a leak in the newer contigmalloc() implementation. Specifically, if | alc | 2006-01-26 | 1 | -10/+5 |
| * | The previous revision incorrectly changed a switch statement into an if | alc | 2006-01-25 | 1 | -3/+3 |
| * | MI changes: | netchild | 2005-12-31 | 1 | -6/+5 |
| * | Check for marker pages when scanning active and inactive page queues. | tegge | 2005-08-12 | 1 | -0/+5 |
