Lines Matching defs:from
35 * University Acknowledgment- Portions of this document are derived from
161 * else from moving pages from one list to the other while the
183 * from happening. This drain condition will occur when pcf_count decrement
211 int pc_free_pageout; /* free's from pageout */
215 int pc_get_cache; /* get's from cache list */
216 int pc_get_free; /* get's from free list */
412 * time, use a delta from availrmem_initial.
807 * cachelist it must be pulled now. To pull the page from
812 * page from whatever free list (the cache list in this
1023 * page from disappearing or appearing immediately after
1024 * the return from this routine. Subsequently, we don't
1043 * with these pages locked SHARED. If necessary reclaim pages from
1292 * a hint since nothing prevents the page from disappearing
1293 * or appearing immediately after the return from this routine.
1335 * sets freemem to the total of all NCPU buckets. It is called from
1431 * Throttle for page_create(): try to prevent freemem from dropping
1522 * page_create_wait() is called to either coalesce pages from the
1526 * Sadly, this is called from platform/vm/vm_machdep.c
1726 * Called from page_create_va() when both the cache and free lists
1775 * Requests for free pages from critical threads
1789 * pages from the freelists.
1800 * We failed to get a `correctly colored' page from both the
1808 * stop deletions from the lists. This will help because
1812 * move from looking at the cache list to the free list. This
1824 * are created by converting pages from the global page pool. This
1947 * from page_create() to page_create_va(), and created this wrapper.
2081 * fill it in with all the constituent pages from the large page. But
2164 * If cage is on, dampen draw from cage when available
2312 * If cage is on, dampen draw from cage when available
2382 * Try to get a page from the freelist (ie,
2396 * from and use it to specify where to allocate
2413 * Since this page came from the
2876 * Reclaim the given page from the free list.
2903 * If `freemem' is 0, we cannot reclaim this page from the
3266 * lock, again preventing another page from being created with
3370 * Transfer the lock count from the old page (if any).
3485 * Remove page ``pp'' from the hash and vp chains and remove vp association.
3516 * Now remove it from its associated vnode.
3530 * Remove page ``pp'' from the hash and vp chains and remove vp association.
3631 * Remove this page from a linked list of pages
3769 * Remove this page from the linked list of pages
3957 * The caller subtracted from availrmem for read only mapping.
4477 * to prevent it from being freed and re-used by
4598 * and need not be hashed out from anywhere.
4625 /* The following comment preserved from page_flip(). */
5295 * PAGESIZE level from the file system side and end up in page_free() or
5326 * p_szc starting from the last constituent page of the large page. By working
5327 * from the tail of a large page in pfn decreasing order allows one looking at
5371 lgrp_t *from;
5443 from = lgrp_pfn_to_lgrp(pfn);
5460 lgrp_stat_add(from->lgrp_id, LGRP_PMM_FAIL_PGS, pages);
5470 lgrp_stat_add(from->lgrp_id, LGRP_PMM_FAIL_PGS,
5512 lgrp_stat_add(from->lgrp_id, LGRP_PMM_FAIL_PGS,
5536 lgrp_stat_add(from->lgrp_id, LGRP_PMM_PGS, nlocked);
5552 lgrp_t *from;
5576 * migrate bit by not having to clear it from each
5598 from = lgrp_pfn_to_lgrp(pfn);
5611 lgrp_stat_add(from->lgrp_id,
5623 if (i == 0 && to == from) {
5663 * Keep stats for number of pages migrated from and to
5666 lgrp_stat_add(from->lgrp_id, LGRP_PM_SRC_PGS, page_cnt);
5745 * If the hint misses, we start from the beginning of the sorted memseg list
5764 * cpu_vm_data field in order to prevent us from being switched to
5766 * This will keep us on cpu and prevent it from being removed while
5858 * n structs away from the current one in the global page
5873 * cpu_vm_data field in order to prevent us from being switched to
5875 * This will keep us on cpu and prevent it from being removed while
6110 * The following functions is called from free_vp_pages()
6159 * the item from the hash if it has expired.
6165 * you can add or remove an entry from the hash.
6212 * from the active list to the walked list. Thus if there is an async request
6336 /* Remove from old list */
6544 * Since this page came from the
6764 * Remove the entry from the page_capture hash, but don't free it yet
6806 * We need to remove the entry from the page capture hash and turn off
7012 * Note that in the failure cases from page_capture_take_action, the
7120 * prevent a latent error trap from occurring on the next boot.
7431 * Go and try to satisfy the page request from any number of buckets.
7447 /* try to collect pages from several pcf bins */