Commit 72a9defa authored by Andrew Morton's avatar Andrew Morton Committed by Linus Torvalds

[PATCH] vmscan.c: struct scan_control

From: Nick Piggin <nickpiggin@yahoo.com.au>

Replace lots of parameters to functions in mm/vmscan.c with a structure
struct scan_control.
Signed-off-by: default avatarNick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: default avatarAndrew Morton <akpm@osdl.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@osdl.org>
parent 690c9c04
...@@ -311,17 +311,35 @@ static pageout_t pageout(struct page *page, struct address_space *mapping) ...@@ -311,17 +311,35 @@ static pageout_t pageout(struct page *page, struct address_space *mapping)
return PAGE_CLEAN; return PAGE_CLEAN;
} }
struct scan_control {
/* Ask refill_inactive_zone, or shrink_cache to scan this many pages */
unsigned long nr_to_scan;
/* Incremented by the number of inactive pages that were scanned */
unsigned long nr_scanned;
/* Incremented by the number of pages reclaimed */
unsigned long nr_reclaimed;
/* Ask shrink_caches, or shrink_zone to scan at this priority */
unsigned int priority;
/* This context's GFP mask */
unsigned int gfp_mask;
struct page_state ps;
int may_writepage;
};
/* /*
* shrink_list returns the number of reclaimed pages * shrink_list adds the number of reclaimed pages to sc->nr_reclaimed
*/ */
static int static int shrink_list(struct list_head *page_list, struct scan_control *sc)
shrink_list(struct list_head *page_list, unsigned int gfp_mask,
int *nr_scanned, int do_writepage)
{ {
LIST_HEAD(ret_pages); LIST_HEAD(ret_pages);
struct pagevec freed_pvec; struct pagevec freed_pvec;
int pgactivate = 0; int pgactivate = 0;
int ret = 0; int reclaimed = 0;
cond_resched(); cond_resched();
...@@ -338,15 +356,16 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask, ...@@ -338,15 +356,16 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask,
if (TestSetPageLocked(page)) if (TestSetPageLocked(page))
goto keep; goto keep;
/* Double the slab pressure for mapped and swapcache pages */
if (page_mapped(page) || PageSwapCache(page))
(*nr_scanned)++;
BUG_ON(PageActive(page)); BUG_ON(PageActive(page));
if (PageWriteback(page)) if (PageWriteback(page))
goto keep_locked; goto keep_locked;
sc->nr_scanned++;
/* Double the slab pressure for mapped and swapcache pages */
if (page_mapped(page) || PageSwapCache(page))
sc->nr_scanned++;
page_map_lock(page); page_map_lock(page);
referenced = page_referenced(page); referenced = page_referenced(page);
if (referenced && page_mapping_inuse(page)) { if (referenced && page_mapping_inuse(page)) {
...@@ -371,8 +390,8 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask, ...@@ -371,8 +390,8 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask,
#endif /* CONFIG_SWAP */ #endif /* CONFIG_SWAP */
mapping = page_mapping(page); mapping = page_mapping(page);
may_enter_fs = (gfp_mask & __GFP_FS) || may_enter_fs = (sc->gfp_mask & __GFP_FS) ||
(PageSwapCache(page) && (gfp_mask & __GFP_IO)); (PageSwapCache(page) && (sc->gfp_mask & __GFP_IO));
/* /*
* The page is mapped into the page tables of one or more * The page is mapped into the page tables of one or more
...@@ -397,7 +416,7 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask, ...@@ -397,7 +416,7 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask,
goto keep_locked; goto keep_locked;
if (!may_enter_fs) if (!may_enter_fs)
goto keep_locked; goto keep_locked;
if (laptop_mode && !do_writepage) if (laptop_mode && !sc->may_writepage)
goto keep_locked; goto keep_locked;
/* Page is dirty, try to write it out here */ /* Page is dirty, try to write it out here */
...@@ -445,7 +464,7 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask, ...@@ -445,7 +464,7 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask,
* Otherwise, leave the page on the LRU so it is swappable. * Otherwise, leave the page on the LRU so it is swappable.
*/ */
if (PagePrivate(page)) { if (PagePrivate(page)) {
if (!try_to_release_page(page, gfp_mask)) if (!try_to_release_page(page, sc->gfp_mask))
goto activate_locked; goto activate_locked;
if (!mapping && page_count(page) == 1) if (!mapping && page_count(page) == 1)
goto free_it; goto free_it;
...@@ -483,7 +502,7 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask, ...@@ -483,7 +502,7 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask,
free_it: free_it:
unlock_page(page); unlock_page(page);
ret++; reclaimed++;
if (!pagevec_add(&freed_pvec, page)) if (!pagevec_add(&freed_pvec, page))
__pagevec_release_nonlru(&freed_pvec); __pagevec_release_nonlru(&freed_pvec);
continue; continue;
...@@ -501,7 +520,8 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask, ...@@ -501,7 +520,8 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask,
if (pagevec_count(&freed_pvec)) if (pagevec_count(&freed_pvec))
__pagevec_release_nonlru(&freed_pvec); __pagevec_release_nonlru(&freed_pvec);
mod_page_state(pgactivate, pgactivate); mod_page_state(pgactivate, pgactivate);
return ret; sc->nr_reclaimed += reclaimed;
return reclaimed;
} }
/* /*
...@@ -509,19 +529,16 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask, ...@@ -509,19 +529,16 @@ shrink_list(struct list_head *page_list, unsigned int gfp_mask,
* a batch of pages and working on them outside the lock. Any pages which were * a batch of pages and working on them outside the lock. Any pages which were
* not freed will be added back to the LRU. * not freed will be added back to the LRU.
* *
* shrink_cache() is passed the number of pages to scan and returns the number * shrink_cache() adds the number of pages reclaimed to sc->nr_reclaimed
* of pages which were reclaimed.
* *
* For pagecache intensive workloads, the first loop here is the hottest spot * For pagecache intensive workloads, the first loop here is the hottest spot
* in the kernel (apart from the copy_*_user functions). * in the kernel (apart from the copy_*_user functions).
*/ */
static int static void shrink_cache(struct zone *zone, struct scan_control *sc)
shrink_cache(struct zone *zone, unsigned int gfp_mask,
int max_scan, int *total_scanned, int do_writepage)
{ {
LIST_HEAD(page_list); LIST_HEAD(page_list);
struct pagevec pvec; struct pagevec pvec;
int ret = 0; int max_scan = sc->nr_to_scan;
pagevec_init(&pvec, 1); pagevec_init(&pvec, 1);
...@@ -567,17 +584,11 @@ shrink_cache(struct zone *zone, unsigned int gfp_mask, ...@@ -567,17 +584,11 @@ shrink_cache(struct zone *zone, unsigned int gfp_mask,
mod_page_state_zone(zone, pgscan_kswapd, nr_scan); mod_page_state_zone(zone, pgscan_kswapd, nr_scan);
else else
mod_page_state_zone(zone, pgscan_direct, nr_scan); mod_page_state_zone(zone, pgscan_direct, nr_scan);
nr_freed = shrink_list(&page_list, gfp_mask, nr_freed = shrink_list(&page_list, sc);
total_scanned, do_writepage);
*total_scanned += nr_taken;
if (current_is_kswapd()) if (current_is_kswapd())
mod_page_state(kswapd_steal, nr_freed); mod_page_state(kswapd_steal, nr_freed);
mod_page_state_zone(zone, pgsteal, nr_freed); mod_page_state_zone(zone, pgsteal, nr_freed);
ret += nr_freed;
if (nr_freed <= 0 && list_empty(&page_list))
goto done;
spin_lock_irq(&zone->lru_lock); spin_lock_irq(&zone->lru_lock);
/* /*
* Put back any unfreeable pages. * Put back any unfreeable pages.
...@@ -601,7 +612,6 @@ shrink_cache(struct zone *zone, unsigned int gfp_mask, ...@@ -601,7 +612,6 @@ shrink_cache(struct zone *zone, unsigned int gfp_mask,
spin_unlock_irq(&zone->lru_lock); spin_unlock_irq(&zone->lru_lock);
done: done:
pagevec_release(&pvec); pagevec_release(&pvec);
return ret;
} }
/* /*
...@@ -622,12 +632,12 @@ shrink_cache(struct zone *zone, unsigned int gfp_mask, ...@@ -622,12 +632,12 @@ shrink_cache(struct zone *zone, unsigned int gfp_mask,
* But we had to alter page->flags anyway. * But we had to alter page->flags anyway.
*/ */
static void static void
refill_inactive_zone(struct zone *zone, const int nr_pages_in, refill_inactive_zone(struct zone *zone, struct scan_control *sc)
struct page_state *ps)
{ {
int pgmoved; int pgmoved;
int pgdeactivate = 0; int pgdeactivate = 0;
int nr_pages = nr_pages_in; int pgscanned = 0;
int nr_pages = sc->nr_to_scan;
LIST_HEAD(l_hold); /* The pages which were snipped off */ LIST_HEAD(l_hold); /* The pages which were snipped off */
LIST_HEAD(l_inactive); /* Pages to go onto the inactive_list */ LIST_HEAD(l_inactive); /* Pages to go onto the inactive_list */
LIST_HEAD(l_active); /* Pages to go onto the active_list */ LIST_HEAD(l_active); /* Pages to go onto the active_list */
...@@ -641,7 +651,7 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in, ...@@ -641,7 +651,7 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in,
lru_add_drain(); lru_add_drain();
pgmoved = 0; pgmoved = 0;
spin_lock_irq(&zone->lru_lock); spin_lock_irq(&zone->lru_lock);
while (nr_pages && !list_empty(&zone->active_list)) { while (pgscanned < nr_pages && !list_empty(&zone->active_list)) {
page = lru_to_page(&zone->active_list); page = lru_to_page(&zone->active_list);
prefetchw_prev_lru_page(page, &zone->active_list, flags); prefetchw_prev_lru_page(page, &zone->active_list, flags);
if (!TestClearPageLRU(page)) if (!TestClearPageLRU(page))
...@@ -661,7 +671,7 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in, ...@@ -661,7 +671,7 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in,
list_add(&page->lru, &l_hold); list_add(&page->lru, &l_hold);
pgmoved++; pgmoved++;
} }
nr_pages--; pgscanned++;
} }
zone->nr_active -= pgmoved; zone->nr_active -= pgmoved;
spin_unlock_irq(&zone->lru_lock); spin_unlock_irq(&zone->lru_lock);
...@@ -677,7 +687,7 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in, ...@@ -677,7 +687,7 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in,
* mapped memory instead of just pagecache. Work out how much memory * mapped memory instead of just pagecache. Work out how much memory
* is mapped. * is mapped.
*/ */
mapped_ratio = (ps->nr_mapped * 100) / total_memory; mapped_ratio = (sc->ps.nr_mapped * 100) / total_memory;
/* /*
* Now decide how much we really want to unmap some pages. The mapped * Now decide how much we really want to unmap some pages. The mapped
...@@ -776,7 +786,7 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in, ...@@ -776,7 +786,7 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in,
spin_unlock_irq(&zone->lru_lock); spin_unlock_irq(&zone->lru_lock);
pagevec_release(&pvec); pagevec_release(&pvec);
mod_page_state_zone(zone, pgrefill, nr_pages_in - nr_pages); mod_page_state_zone(zone, pgrefill, pgscanned);
mod_page_state(pgdeactivate, pgdeactivate); mod_page_state(pgdeactivate, pgdeactivate);
} }
...@@ -784,13 +794,14 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in, ...@@ -784,13 +794,14 @@ refill_inactive_zone(struct zone *zone, const int nr_pages_in,
* Scan `nr_pages' from this zone. Returns the number of reclaimed pages. * Scan `nr_pages' from this zone. Returns the number of reclaimed pages.
* This is a basic per-zone page freer. Used by both kswapd and direct reclaim. * This is a basic per-zone page freer. Used by both kswapd and direct reclaim.
*/ */
static int static void
shrink_zone(struct zone *zone, int max_scan, unsigned int gfp_mask, shrink_zone(struct zone *zone, struct scan_control *sc)
int *total_scanned, struct page_state *ps, int do_writepage)
{ {
unsigned long scan_active; unsigned long scan_active, scan_inactive;
int count; int count;
scan_inactive = (zone->nr_active + zone->nr_inactive) >> sc->priority;
/* /*
* Try to keep the active list 2/3 of the size of the cache. And * Try to keep the active list 2/3 of the size of the cache. And
* make sure that refill_inactive is given a decent number of pages. * make sure that refill_inactive is given a decent number of pages.
...@@ -803,13 +814,13 @@ shrink_zone(struct zone *zone, int max_scan, unsigned int gfp_mask, ...@@ -803,13 +814,13 @@ shrink_zone(struct zone *zone, int max_scan, unsigned int gfp_mask,
*/ */
if (zone->nr_active >= 4*(zone->nr_inactive*2 + 1)) { if (zone->nr_active >= 4*(zone->nr_inactive*2 + 1)) {
/* Don't scan more than 4 times the inactive list scan size */ /* Don't scan more than 4 times the inactive list scan size */
scan_active = 4*max_scan; scan_active = 4*scan_inactive;
} else { } else {
unsigned long long tmp; unsigned long long tmp;
/* Cast to long long so the multiply doesn't overflow */ /* Cast to long long so the multiply doesn't overflow */
tmp = (unsigned long long)max_scan * zone->nr_active; tmp = (unsigned long long)scan_inactive * zone->nr_active;
do_div(tmp, zone->nr_inactive*2 + 1); do_div(tmp, zone->nr_inactive*2 + 1);
scan_active = (unsigned long)tmp; scan_active = (unsigned long)tmp;
} }
...@@ -818,17 +829,17 @@ shrink_zone(struct zone *zone, int max_scan, unsigned int gfp_mask, ...@@ -818,17 +829,17 @@ shrink_zone(struct zone *zone, int max_scan, unsigned int gfp_mask,
count = atomic_read(&zone->nr_scan_active); count = atomic_read(&zone->nr_scan_active);
if (count >= SWAP_CLUSTER_MAX) { if (count >= SWAP_CLUSTER_MAX) {
atomic_set(&zone->nr_scan_active, 0); atomic_set(&zone->nr_scan_active, 0);
refill_inactive_zone(zone, count, ps); sc->nr_to_scan = count;
refill_inactive_zone(zone, sc);
} }
atomic_add(max_scan, &zone->nr_scan_inactive); atomic_add(scan_inactive, &zone->nr_scan_inactive);
count = atomic_read(&zone->nr_scan_inactive); count = atomic_read(&zone->nr_scan_inactive);
if (count >= SWAP_CLUSTER_MAX) { if (count >= SWAP_CLUSTER_MAX) {
atomic_set(&zone->nr_scan_inactive, 0); atomic_set(&zone->nr_scan_inactive, 0);
return shrink_cache(zone, gfp_mask, count, sc->nr_to_scan = count;
total_scanned, do_writepage); return shrink_cache(zone, sc);
} }
return 0;
} }
/* /*
...@@ -847,29 +858,23 @@ shrink_zone(struct zone *zone, int max_scan, unsigned int gfp_mask, ...@@ -847,29 +858,23 @@ shrink_zone(struct zone *zone, int max_scan, unsigned int gfp_mask,
* If a zone is deemed to be full of pinned pages then just give it a light * If a zone is deemed to be full of pinned pages then just give it a light
* scan then give up on it. * scan then give up on it.
*/ */
static int static void
shrink_caches(struct zone **zones, int priority, int *total_scanned, shrink_caches(struct zone **zones, struct scan_control *sc)
int gfp_mask, struct page_state *ps, int do_writepage)
{ {
int ret = 0;
int i; int i;
for (i = 0; zones[i] != NULL; i++) { for (i = 0; zones[i] != NULL; i++) {
struct zone *zone = zones[i]; struct zone *zone = zones[i];
int max_scan;
zone->temp_priority = priority; zone->temp_priority = sc->priority;
if (zone->prev_priority > priority) if (zone->prev_priority > sc->priority)
zone->prev_priority = priority; zone->prev_priority = sc->priority;
if (zone->all_unreclaimable && priority != DEF_PRIORITY) if (zone->all_unreclaimable && sc->priority != DEF_PRIORITY)
continue; /* Let kswapd poll it */ continue; /* Let kswapd poll it */
max_scan = (zone->nr_active + zone->nr_inactive) >> priority; shrink_zone(zone, sc);
ret += shrink_zone(zone, max_scan, gfp_mask,
total_scanned, ps, do_writepage);
} }
return ret;
} }
/* /*
...@@ -890,11 +895,13 @@ int try_to_free_pages(struct zone **zones, ...@@ -890,11 +895,13 @@ int try_to_free_pages(struct zone **zones,
{ {
int priority; int priority;
int ret = 0; int ret = 0;
int nr_reclaimed = 0; int total_scanned = 0, total_reclaimed = 0;
struct reclaim_state *reclaim_state = current->reclaim_state; struct reclaim_state *reclaim_state = current->reclaim_state;
struct scan_control sc;
int i; int i;
unsigned long total_scanned = 0;
int do_writepage = 0; sc.gfp_mask = gfp_mask;
sc.may_writepage = 0;
inc_page_state(allocstall); inc_page_state(allocstall);
...@@ -902,21 +909,23 @@ int try_to_free_pages(struct zone **zones, ...@@ -902,21 +909,23 @@ int try_to_free_pages(struct zone **zones,
zones[i]->temp_priority = DEF_PRIORITY; zones[i]->temp_priority = DEF_PRIORITY;
for (priority = DEF_PRIORITY; priority >= 0; priority--) { for (priority = DEF_PRIORITY; priority >= 0; priority--) {
int scanned = 0; get_page_state(&sc.ps);
struct page_state ps; sc.nr_scanned = 0;
sc.nr_reclaimed = 0;
get_page_state(&ps); sc.priority = priority;
nr_reclaimed += shrink_caches(zones, priority, &scanned, shrink_caches(zones, &sc);
gfp_mask, &ps, do_writepage); shrink_slab(sc.nr_scanned, gfp_mask);
shrink_slab(scanned, gfp_mask);
if (reclaim_state) { if (reclaim_state) {
nr_reclaimed += reclaim_state->reclaimed_slab; sc.nr_reclaimed += reclaim_state->reclaimed_slab;
reclaim_state->reclaimed_slab = 0; reclaim_state->reclaimed_slab = 0;
} }
if (nr_reclaimed >= SWAP_CLUSTER_MAX) { if (sc.nr_reclaimed >= SWAP_CLUSTER_MAX) {
ret = 1; ret = 1;
goto out; goto out;
} }
total_scanned += sc.nr_scanned;
total_reclaimed += sc.nr_reclaimed;
/* /*
* Try to write back as many pages as we just scanned. This * Try to write back as many pages as we just scanned. This
* tends to cause slow streaming writers to write data to the * tends to cause slow streaming writers to write data to the
...@@ -924,14 +933,13 @@ int try_to_free_pages(struct zone **zones, ...@@ -924,14 +933,13 @@ int try_to_free_pages(struct zone **zones,
* that's undesirable in laptop mode, where we *want* lumpy * that's undesirable in laptop mode, where we *want* lumpy
* writeout. So in laptop mode, write out the whole world. * writeout. So in laptop mode, write out the whole world.
*/ */
total_scanned += scanned;
if (total_scanned > SWAP_CLUSTER_MAX + SWAP_CLUSTER_MAX/2) { if (total_scanned > SWAP_CLUSTER_MAX + SWAP_CLUSTER_MAX/2) {
wakeup_bdflush(laptop_mode ? 0 : total_scanned); wakeup_bdflush(laptop_mode ? 0 : total_scanned);
do_writepage = 1; sc.may_writepage = 1;
} }
/* Take a nap, wait for some writeback to complete */ /* Take a nap, wait for some writeback to complete */
if (scanned && priority < DEF_PRIORITY - 2) if (sc.nr_scanned && priority < DEF_PRIORITY - 2)
blk_congestion_wait(WRITE, HZ/10); blk_congestion_wait(WRITE, HZ/10);
} }
if ((gfp_mask & __GFP_FS) && !(gfp_mask & __GFP_NORETRY)) if ((gfp_mask & __GFP_FS) && !(gfp_mask & __GFP_NORETRY))
...@@ -967,15 +975,18 @@ int try_to_free_pages(struct zone **zones, ...@@ -967,15 +975,18 @@ int try_to_free_pages(struct zone **zones,
* the page allocator fallback scheme to ensure that aging of pages is balanced * the page allocator fallback scheme to ensure that aging of pages is balanced
* across the zones. * across the zones.
*/ */
static int balance_pgdat(pg_data_t *pgdat, int nr_pages, struct page_state *ps) static int balance_pgdat(pg_data_t *pgdat, int nr_pages)
{ {
int to_free = nr_pages; int to_free = nr_pages;
int priority; int priority;
int i; int i;
int total_scanned = 0, total_reclaimed = 0;
struct reclaim_state *reclaim_state = current->reclaim_state; struct reclaim_state *reclaim_state = current->reclaim_state;
unsigned long total_scanned = 0; struct scan_control sc;
unsigned long total_reclaimed = 0;
int do_writepage = 0; sc.gfp_mask = GFP_KERNEL;
sc.may_writepage = 0;
get_page_state(&sc.ps);
inc_page_state(pageoutrun); inc_page_state(pageoutrun);
...@@ -1023,9 +1034,6 @@ static int balance_pgdat(pg_data_t *pgdat, int nr_pages, struct page_state *ps) ...@@ -1023,9 +1034,6 @@ static int balance_pgdat(pg_data_t *pgdat, int nr_pages, struct page_state *ps)
*/ */
for (i = 0; i <= end_zone; i++) { for (i = 0; i <= end_zone; i++) {
struct zone *zone = pgdat->node_zones + i; struct zone *zone = pgdat->node_zones + i;
int max_scan;
int reclaimed;
int scanned = 0;
if (zone->all_unreclaimable && priority != DEF_PRIORITY) if (zone->all_unreclaimable && priority != DEF_PRIORITY)
continue; continue;
...@@ -1037,16 +1045,14 @@ static int balance_pgdat(pg_data_t *pgdat, int nr_pages, struct page_state *ps) ...@@ -1037,16 +1045,14 @@ static int balance_pgdat(pg_data_t *pgdat, int nr_pages, struct page_state *ps)
zone->temp_priority = priority; zone->temp_priority = priority;
if (zone->prev_priority > priority) if (zone->prev_priority > priority)
zone->prev_priority = priority; zone->prev_priority = priority;
max_scan = (zone->nr_active + zone->nr_inactive) sc.nr_scanned = 0;
>> priority; sc.nr_reclaimed = 0;
reclaimed = shrink_zone(zone, max_scan, GFP_KERNEL, sc.priority = priority;
&scanned, ps, do_writepage); shrink_zone(zone, &sc);
total_scanned += scanned;
reclaim_state->reclaimed_slab = 0; reclaim_state->reclaimed_slab = 0;
shrink_slab(scanned, GFP_KERNEL); shrink_slab(sc.nr_scanned, GFP_KERNEL);
reclaimed += reclaim_state->reclaimed_slab; sc.nr_reclaimed += reclaim_state->reclaimed_slab;
total_reclaimed += reclaimed; total_reclaimed += sc.nr_reclaimed;
to_free -= reclaimed;
if (zone->all_unreclaimable) if (zone->all_unreclaimable)
continue; continue;
if (zone->pages_scanned > zone->present_pages * 2) if (zone->pages_scanned > zone->present_pages * 2)
...@@ -1058,9 +1064,9 @@ static int balance_pgdat(pg_data_t *pgdat, int nr_pages, struct page_state *ps) ...@@ -1058,9 +1064,9 @@ static int balance_pgdat(pg_data_t *pgdat, int nr_pages, struct page_state *ps)
*/ */
if (total_scanned > SWAP_CLUSTER_MAX * 2 && if (total_scanned > SWAP_CLUSTER_MAX * 2 &&
total_scanned > total_reclaimed+total_reclaimed/2) total_scanned > total_reclaimed+total_reclaimed/2)
do_writepage = 1; sc.may_writepage = 1;
} }
if (nr_pages && to_free > 0) if (nr_pages && to_free > total_reclaimed)
continue; /* swsusp: need to do more work */ continue; /* swsusp: need to do more work */
if (all_zones_ok) if (all_zones_ok)
break; /* kswapd: all done */ break; /* kswapd: all done */
...@@ -1124,15 +1130,13 @@ int kswapd(void *p) ...@@ -1124,15 +1130,13 @@ int kswapd(void *p)
tsk->flags |= PF_MEMALLOC|PF_KSWAPD; tsk->flags |= PF_MEMALLOC|PF_KSWAPD;
for ( ; ; ) { for ( ; ; ) {
struct page_state ps;
if (current->flags & PF_FREEZE) if (current->flags & PF_FREEZE)
refrigerator(PF_FREEZE); refrigerator(PF_FREEZE);
prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE); prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
schedule(); schedule();
finish_wait(&pgdat->kswapd_wait, &wait); finish_wait(&pgdat->kswapd_wait, &wait);
get_page_state(&ps);
balance_pgdat(pgdat, 0, &ps); balance_pgdat(pgdat, 0);
} }
} }
...@@ -1165,10 +1169,7 @@ int shrink_all_memory(int nr_pages) ...@@ -1165,10 +1169,7 @@ int shrink_all_memory(int nr_pages)
current->reclaim_state = &reclaim_state; current->reclaim_state = &reclaim_state;
for_each_pgdat(pgdat) { for_each_pgdat(pgdat) {
int freed; int freed;
struct page_state ps; freed = balance_pgdat(pgdat, nr_to_free);
get_page_state(&ps);
freed = balance_pgdat(pgdat, nr_to_free, &ps);
ret += freed; ret += freed;
nr_to_free -= freed; nr_to_free -= freed;
if (nr_to_free <= 0) if (nr_to_free <= 0)
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment