mm/workingset: remove unused @mapping argument in workingset_eviction()
Subsystems (13)
PATCH v2 1/4 mm/workingset: remove unused @mapping argument in workingset_eviction()
workingset_eviction() doesn't use and never did use the @mapping argument. Remove it.
---
Changes since v1:
- s/@mapping/@page->mapping in comment
- Acks
include/linux/swap.h | 2 +- mm/vmscan.c | 2 +- mm/workingset.c | 5 ++--- 3 files changed, 4 insertions(+), 5 deletions(-)
include/linux/swap.h +1 -1
@@ -307,7 +307,7 @@ struct vma_swap_readahead { };/* linux/mm/workingset.c */ -void *workingset_eviction (struct address_space *mapping,struct page *page); +void *workingset_eviction (struct page *page);void workingset_refault (struct page *page,void *shadow);void workingset_activation (struct page *page);
mm/vmscan.c +1 -1
@@ -952,7 +952,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page,*/ if (reclaimed &&page_is_file_cache (page) && !mapping_exiting (mapping) && !dax_mapping (mapping)) - shadow =workingset_eviction (mapping, page); + shadow =workingset_eviction (page);__delete_from_page_cache (page, shadow);xa_unlock_irqrestore (&mapping->i_pages, flags);
mm/workingset.c +2 -3
@@ -215,13 +215,12 @@ static void unpack_shadow(void *shadow, int *memcgidp, pg_data_t **pgdat,/** * workingset_eviction - note the eviction of a page from memory -* @mapping: address space the page was backing * @page: the page being evicted * -* Returns a shadow entry to be stored in @mapping->i_pages in place +* Returns a shadow entry to be stored in @page->mapping->i_pages in place * of the evicted @page so that a later refault can be detected. */ -void *workingset_eviction (struct address_space *mapping,struct page *page) +void *workingset_eviction (struct page *page) {struct pglist_data *pgdat =page_pgdat (page);struct mem_cgroup *memcg =page_memcg (page);
Signature · 1 line
2.19.2
PATCH v2 2/4 mm: remove zone_lru_lock() function access ->lru_lock directly
We have common pattern to access lru_lock from a page pointer:
zone_lru_lock(page_zone(page))
Which is silly, because it unfolds to this:
&NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)]->zone_pgdat->lru_lock while we can simply do &NODE_DATA(page_to_nid(page))->lru_lock
Remove zone_lru_lock() function, since it's only complicate things. Use 'page_pgdat(page)->lru_lock' pattern instead.
---
Changes since v1:
- Added ack.
Documentation/cgroup-v1/memcg_test.txt | 4 ++-- Documentation/cgroup-v1/memory.txt | 4 ++-- include/linux/mm_types.h | 2 +- include/linux/mmzone.h | 4 ---- mm/compaction.c | 15 ++++++++------- mm/filemap.c | 4 ++-- mm/huge_memory.c | 6 +++--- mm/memcontrol.c | 14 +++++++------- mm/mlock.c | 14 +++++++------- mm/page_idle.c | 8 ++++---- mm/rmap.c | 2 +- mm/swap.c | 16 ++++++++-------- mm/vmscan.c | 16 ++++++++-------- 13 files changed, 53 insertions(+), 56 deletions(-)
Documentation/cgroup-v1/memcg_test.txt +2 -2
@@ -107,9 +107,9 @@ Under below explanation, we assume CONFIG_MEM_RES_CTRL_SWAP=y. 8. LRU Each memcg has its own private LRU. Now, its handling is under global - VM's control (means that it's handled under global zone_lru_lock). + VM's control (means that it's handled under global pgdat->lru_lock). Almost all routines around memcg's LRU is called by global LRU's - list management functions under zone_lru_lock(). + list management functions under pgdat->lru_lock. A special function is mem_cgroup_isolate_pages(). This scans memcg's private LRU and call __isolate_lru_page() to extract a page
Documentation/cgroup-v1/memory.txt +2 -2
@@ -267,11 +267,11 @@ When oom event notifier is registered, event will be delivered. Other lock order is following: PG_locked. mm->page_table_lock - zone_lru_lock + pgdat->lru_lock lock_page_cgroup. In many cases, just lock_page_cgroup() is called. per-zone-per-cgroup LRU (cgroup's private LRU) is just guarded by - zone_lru_lock, it has no lock of its own. + pgdat->lru_lock, it has no lock of its own. 2.7 Kernel Memory Extension (CONFIG_MEMCG_KMEM)
include/linux/mm_types.h +1 -1
@@ -79,7 +79,7 @@ struct page {struct {/* Page cache and anonymous pages */ /** * @lru: Pageout list, eg. active_list protected by -* zone_lru_lock. Sometimes used as a generic list +* pgdat->lru_lock. Sometimes used as a generic list * by the page owner. */ struct list_head lru;
include/linux/mmzone.h +0 -4
@@ -788,10 +788,6 @@ typedef struct pglist_data {#define node_start_pfn (nid) (NODE_DATA(nid)->node_start_pfn)#define node_end_pfn (nid) pgdat_end_pfn(NODE_DATA(nid)) -static inline spinlock_t *zone_lru_lock (struct zone *zone) -{ -return &zone->zone_pgdat->lru_lock; -}static inline struct lruvec *node_lruvec (struct pglist_data *pgdat) {
mm/compaction.c +8 -7
@@ -775,6 +775,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn, unsigned long end_pfn, isolate_mode_t isolate_mode) {struct zone *zone = cc->zone; +pg_data_t *pgdat = zone->zone_pgdat;unsigned long nr_scanned =0 , nr_isolated =0 ;struct lruvec *lruvec;unsigned long flags =0 ;
@@ -839,8 +840,8 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,* if contended. */ if (!(low_pfn %SWAP_CLUSTER_MAX ) - &&compact_unlock_should_abort (zone_lru_lock (zone), flags, - &locked, cc)) + &&compact_unlock_should_abort (&pgdat->lru_lock, + flags, &locked, cc))break ;if (!pfn_valid_within (low_pfn))
@@ -910,7 +911,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,if (unlikely (__PageMovable (page)) && !PageIsolated (page)) {if (locked) { -spin_unlock_irqrestore (zone_lru_lock (zone), +spin_unlock_irqrestore (&pgdat->lru_lock, flags); locked = false; }
@@ -940,7 +941,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,/* If we already hold the lock, we can skip some rechecking */ if (!locked) { - locked =compact_lock_irqsave (zone_lru_lock (zone), + locked =compact_lock_irqsave (&pgdat->lru_lock, &flags, cc);/* Try get exclusive access under lock */
@@ -965,7 +966,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn, } } - lruvec =mem_cgroup_page_lruvec (page, zone->zone_pgdat); + lruvec =mem_cgroup_page_lruvec (page, pgdat);/* Try isolate the page */ if (__isolate_lru_page (page, isolate_mode) !=0 )
@@ -1007,7 +1008,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,*/ if (nr_isolated) {if (locked) { -spin_unlock_irqrestore (zone_lru_lock (zone), flags); +spin_unlock_irqrestore (&pgdat->lru_lock, flags); locked = false; }putback_movable_pages (&cc->migratepages);
@@ -1034,7 +1035,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn, isolate_abort:if (locked) -spin_unlock_irqrestore (zone_lru_lock (zone), flags); +spin_unlock_irqrestore (&pgdat->lru_lock, flags);/* * Updated the cached scanner pfn once the pageblock has been scanned
mm/filemap.c +2 -2
@@ -98,8 +98,8 @@* ->swap_lock (try_to_unmap_one) * ->private_lock (try_to_unmap_one) * ->i_pages lock (try_to_unmap_one) -* ->zone_lru_lock(zone) (follow_page->mark_page_accessed) -* ->zone_lru_lock(zone) (check_pte_range->isolate_lru_page) +* ->pgdat->lru_lock (follow_page->mark_page_accessed) +* ->pgdat->lru_lock (check_pte_range->isolate_lru_page) * ->private_lock (page_remove_rmap->set_page_dirty) * ->i_pages lock (page_remove_rmap->set_page_dirty) * bdi.wb->list_lock (page_remove_rmap->set_page_dirty)
mm/huge_memory.c +3 -3
@@ -2475,7 +2475,7 @@ static void __split_huge_page(struct page *page, struct list_head *list,xa_unlock (&head->mapping->i_pages); } -spin_unlock_irqrestore (zone_lru_lock (page_zone (head)), flags); +spin_unlock_irqrestore (&page_pgdat (head)->lru_lock, flags);remap_page (head);
@@ -2686,7 +2686,7 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)lru_add_drain ();/* prevent PageLRU to go away from under us, and freeze lru stats */ -spin_lock_irqsave (zone_lru_lock (page_zone (head)), flags); +spin_lock_irqsave (&pgdata->lru_lock, flags);if (mapping) {XA_STATE (xas, &mapping->i_pages,page_index (head));
@@ -2731,7 +2731,7 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)spin_unlock (&pgdata->split_queue_lock); fail:if (mapping)xa_unlock (&mapping->i_pages); -spin_unlock_irqrestore (zone_lru_lock (page_zone (head)), flags); +spin_unlock_irqrestore (&pgdata->lru_lock, flags);remap_page (head); ret = -EBUSY ; }
mm/memcontrol.c +7 -7
@@ -2377,13 +2377,13 @@ static void cancel_charge(struct mem_cgroup *memcg, unsigned int nr_pages)static void lock_page_lru (struct page *page,int *isolated) { -struct zone *zone =page_zone (page); +pg_data_t *pgdat =page_pgdat (page); -spin_lock_irq (zone_lru_lock (zone)); +spin_lock_irq (&pgdat->lru_lock);if (PageLRU (page)) {struct lruvec *lruvec; - lruvec =mem_cgroup_page_lruvec (page, zone->zone_pgdat); + lruvec =mem_cgroup_page_lruvec (page, pgdat);ClearPageLRU (page);del_page_from_lru_list (page, lruvec,page_lru (page)); *isolated =1 ;
@@ -2393,17 +2393,17 @@ static void lock_page_lru(struct page *page, int *isolated)static void unlock_page_lru (struct page *page,int isolated) { -struct zone *zone =page_zone (page); +pg_data_t *pgdat =page_pgdat (page);if (isolated) {struct lruvec *lruvec; - lruvec =mem_cgroup_page_lruvec (page, zone->zone_pgdat); + lruvec =mem_cgroup_page_lruvec (page, pgdat);VM_BUG_ON_PAGE (PageLRU (page), page);SetPageLRU (page);add_page_to_lru_list (page, lruvec,page_lru (page)); } -spin_unlock_irq (zone_lru_lock (zone)); +spin_unlock_irq (&pgdat->lru_lock); } static void commit_charge(struct page *page, struct mem_cgroup *memcg,
@@ -2689,7 +2689,7 @@ void __memcg_kmem_uncharge(struct page *page, int order)/* * Because tail pages are not marked as "used", set it. We're under -* zone_lru_lock and migration entries setup in all page mappings. +* pgdat->lru_lock and migration entries setup in all page mappings. */ void mem_cgroup_split_huge_fixup (struct page *head) {
mm/mlock.c +7 -7
@@ -182,7 +182,7 @@ static void __munlock_isolation_failed(struct page *page)unsigned int munlock_vma_page (struct page *page) {int nr_pages; -struct zone *zone =page_zone (page); +pg_data_t *pgdat =page_pgdat (page);/* For try_to_munlock() and to serialize with page migration */ BUG_ON (!PageLocked (page));
@@ -194,7 +194,7 @@ unsigned int munlock_vma_page(struct page *page)* might otherwise copy PageMlocked to part of the tail pages before * we clear it in the head page. It also stabilizes hpage_nr_pages(). */ -spin_lock_irq (zone_lru_lock (zone)); +spin_lock_irq (&pgdat->lru_lock);if (!TestClearPageMlocked (page)) {/* Potentially, PTE-mapped THP: do not skip the rest PTEs */
@@ -203,17 +203,17 @@ unsigned int munlock_vma_page(struct page *page) } nr_pages =hpage_nr_pages (page); -__mod_zone_page_state (zone,NR_MLOCK , -nr_pages); +__mod_zone_page_state (page_zone (page),NR_MLOCK , -nr_pages);if (__munlock_isolate_lru_page (page, true)) { -spin_unlock_irq (zone_lru_lock (zone)); +spin_unlock_irq (&pgdat->lru_lock);__munlock_isolated_page (page); goto out; }__munlock_isolation_failed (page); unlock_out: -spin_unlock_irq (zone_lru_lock (zone)); +spin_unlock_irq (&pgdat->lru_lock); out:return nr_pages -1 ;
@@ -298,7 +298,7 @@ static void __munlock_pagevec(struct pagevec *pvec, struct zone *zone)pagevec_init (&pvec_putback);/* Phase 1: page isolation */ -spin_lock_irq (zone_lru_lock (zone)); +spin_lock_irq (&zone->zone_pgdat->lru_lock);for (i =0 ; i < nr; i++) {struct page *page = pvec->pages[i];
@@ -325,7 +325,7 @@ static void __munlock_pagevec(struct pagevec *pvec, struct zone *zone) pvec->pages[i] =NULL ; }__mod_zone_page_state (zone,NR_MLOCK , delta_munlocked); -spin_unlock_irq (zone_lru_lock (zone)); +spin_unlock_irq (&zone->zone_pgdat->lru_lock);/* Now we can release pins of pages that we are not munlocking */ pagevec_release (&pvec_putback);
mm/page_idle.c +4 -4
@@ -31,7 +31,7 @@static struct page *page_idle_get_page (unsigned long pfn) {struct page *page; -struct zone *zone; +pg_data_t *pgdat;if (!pfn_valid (pfn))return NULL ;
@@ -41,13 +41,13 @@ static struct page *page_idle_get_page(unsigned long pfn) !get_page_unless_zero(page))return NULL ; - zone =page_zone (page); -spin_lock_irq (zone_lru_lock (zone)); + pgdat =page_pgdat (page); +spin_lock_irq (&pgdat->lru_lock);if (unlikely (!PageLRU (page))) {put_page (page); page =NULL ; } -spin_unlock_irq (zone_lru_lock (zone)); +spin_unlock_irq (&pgdat->lru_lock);return page; }
mm/rmap.c +1 -1
@@ -27,7 +27,7 @@* mapping->i_mmap_rwsem * anon_vma->rwsem * mm->page_table_lock or pte_lock -* zone_lru_lock (in mark_page_accessed, isolate_lru_page) +* pgdat->lru_lock (in mark_page_accessed, isolate_lru_page) * swap_lock (in swap_duplicate, swap_info_get) * mmlist_lock (in mmput, drain_mmlist and others) * mapping->private_lock (in __set_page_dirty_buffers)
mm/swap.c +8 -8
@@ -58,16 +58,16 @@ static DEFINE_PER_CPU(struct pagevec, activate_page_pvecs);static void __page_cache_release (struct page *page) {if (PageLRU (page)) { -struct zone *zone =page_zone (page); +pg_data_t *pgdat =page_pgdat (page);struct lruvec *lruvec;unsigned long flags; -spin_lock_irqsave (zone_lru_lock (zone), flags); - lruvec =mem_cgroup_page_lruvec (page, zone->zone_pgdat); +spin_lock_irqsave (&pgdat->lru_lock, flags); + lruvec =mem_cgroup_page_lruvec (page, pgdat);VM_BUG_ON_PAGE (!PageLRU (page), page);__ClearPageLRU (page);del_page_from_lru_list (page, lruvec,page_off_lru (page)); -spin_unlock_irqrestore (zone_lru_lock (zone), flags); +spin_unlock_irqrestore (&pgdat->lru_lock, flags); }__ClearPageWaiters (page);mem_cgroup_uncharge (page);
@@ -322,12 +322,12 @@ static inline void activate_page_drain(int cpu)void activate_page (struct page *page) { -struct zone *zone =page_zone (page); +pg_data_t *pgdat =page_pgdat (page); page =compound_head (page); -spin_lock_irq (zone_lru_lock (zone)); -__activate_page (page,mem_cgroup_page_lruvec (page, zone->zone_pgdat),NULL ); -spin_unlock_irq (zone_lru_lock (zone)); +spin_lock_irq (&pgdat->lru_lock); +__activate_page (page,mem_cgroup_page_lruvec (page, pgdat),NULL ); +spin_unlock_irq (&pgdat->lru_lock); }#endif
mm/vmscan.c +8 -8
@@ -1614,8 +1614,8 @@ static __always_inline void update_lru_sizes(struct lruvec *lruvec, } -/* -* zone_lru_lock is heavily contended. Some of the functions that +/** +* pgdat->lru_lock is heavily contended. Some of the functions that * shrink the lists perform better by taking out a batch of pages * and working on them outside the LRU lock. *
@@ -1750,11 +1750,11 @@ int isolate_lru_page(struct page *page)WARN_RATELIMIT (PageTail (page),"trying to isolate tail page" );if (PageLRU (page)) { -struct zone *zone =page_zone (page); +pg_data_t *pgdat =page_pgdat (page);struct lruvec *lruvec; -spin_lock_irq (zone_lru_lock (zone)); - lruvec =mem_cgroup_page_lruvec (page, zone->zone_pgdat); +spin_lock_irq (&pgdat->lru_lock); + lruvec =mem_cgroup_page_lruvec (page, pgdat);if (PageLRU (page)) {int lru =page_lru (page);get_page (page);
@@ -1762,7 +1762,7 @@ int isolate_lru_page(struct page *page)del_page_from_lru_list (page, lruvec, lru); ret =0 ; } -spin_unlock_irq (zone_lru_lock (zone)); +spin_unlock_irq (&pgdat->lru_lock); }return ret; }
@@ -1990,9 +1990,9 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,* processes, from rmap. * * If the pages are mostly unmapped, the processing is fast and it is -* appropriate to hold zone_lru_lock across the whole operation. But if +* appropriate to hold pgdat->lru_lock across the whole operation. But if * the pages are mapped, the processing is slow (page_referenced()) so we -* should drop zone_lru_lock around each page. It's impossible to balance +* should drop pgdat->lru_lock around each page. It's impossible to balance * this, so instead we remove the pages from the LRU while processing them. * It is safe to rely on PG_active against the non-LRU pages in here because * nobody will play with that bit on a non-LRU page.
Signature · 1 line
2.19.2
Re: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
Quoted 11 lines · Andrey Ryabinin
We have common pattern to access
lru_lockfrom a page pointer:
zone_lru_lock(page_zone(page))Which is silly, because it unfolds to this:
&NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)]->zone_pgdat->lru_lock while we can simply do &NODE_DATA(page_to_nid(page))->lru_lockRemove
zone_lru_lock()function, since it's only complicate things. Use 'page_pgdat(page)->lru_lock' pattern instead.
Quoted 6 lines
Signed-off-by Andrey Ryabinin <[email protected]>Acked-by Vlastimil Babka <[email protected]>Cc Johannes Weiner <[email protected]>Cc Michal Hocko <[email protected]>Cc Rik van Riel <[email protected]>Cc Mel Gorman <[email protected]>
Signature · 2 lines
Mel Gorman SUSE Labs
Re: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
In reply to Andrey RyabininQuoted 9 lines · Andrey Ryabinin
mm/vmscan.c +0
-0
mm/vmscan.c static __always_inline void update_lru_sizes(struct lruvec *lruvec, · line 1617
Quoted 6 lines
mm/vmscan.c +1
-2@@ -1614,8 +1614,8 @@ static __always_inline void update_lru_sizes(struct lruvec *lruvec, } -/* -* zone_lru_lock is heavily contended. Some of the functions that +/**
Nit: Remove the extra asterisk here; the line should then revert to being unchanged from the original.
Andrew Morton
maintainerRe: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
Quoted 8 lines · On Thu, 28 Feb 2019 05:53:37 -0700 William Kucharski…
Quoted 8 lines · Andrey Ryabinin
mm/vmscan.c +0
-0@@ -1614,8 +1614,8 @@ static __always_inline void update_lru_sizes(struct lruvec *lruvec,
Quoted 8 lines
mm/vmscan.c static __always_inline void update_lru_sizes(struct lruvec *lruvec, · line 1617
Quoted 5 lines
mm/vmscan.c +1
-2} -/* -* zone_lru_lock is heavily contended. Some of the functions that +/** Nit: Remove the extra asterisk here; the line should then revert to being unchanged from the original.
I don't think so. This kernedoc comment was missing its leading /**. The patch fixes that.
Re: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
Quoted 6 lines · Andrew Morton
I don't think so. This kernedoc comment was missing its leading /**. The patch fixes that.
That makes sense; it had looked like just an extraneous asterisk.
Re: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
In reply to Andrey RyabininQuoted 8 lines · Andrey Ryabinin
We have common pattern to access
lru_lockfrom a page pointer:
zone_lru_lock(page_zone(page))Which is silly, because it unfolds to this:
&NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)]->zone_pgdat->lru_lock while we can simply do &NODE_DATA(page_to_nid(page))->lru_lock
Hi Andrey,
Nice. I like it so much that I immediately want to tweak it. :)
Quoted 2 lines
Remove
zone_lru_lock()function, since it's only complicate things. Use 'page_pgdat(page)->lru_lock' pattern instead.
Here, I think the zone_lru_lock() is actually a nice way to add a touch of clarity at the call sites. How about, see below:
[snip]
include/linux/mmzone.h typedef struct pglist_data { · line 791
Quoted 13 lines
include/linux/mmzone.h +0
-4@@ -788,10 +788,6 @@ typedef struct pglist_data {#define node_start_pfn (nid) (NODE_DATA(nid)->node_start_pfn)#define node_end_pfn (nid) pgdat_end_pfn(NODE_DATA(nid)) -static inline spinlock_t *zone_lru_lock (struct zone *zone) -{ -return &zone->zone_pgdat->lru_lock; -}
Instead of removing that function, let's change it, and add another (since you have two cases: either a page* or a pgdat* is available), and move it to where it can compile, like this:
include/linux/mm.h +10 -0
@@ -1167,6 +1167,16 @@ static inline pg_data_t *page_pgdat(const struct page *page)return NODE_DATA (page_to_nid (page)); } +static inline spinlock_t *zone_lru_lock (pg_data_t *pgdat) +{ +return &pgdat->lru_lock; +} + +static inline spinlock_t *zone_lru_lock_from_page (struct page *page) +{ +return zone_lru_lock (page_pgdat (page)); +} +#ifdef SECTION_IN_PAGE_FLAGS static inline void set_page_section (struct page *page,unsigned long section) {
include/linux/mmzone.h +0 -5
@@ -728,11 +728,6 @@ typedef struct pglist_data {#define node_start_pfn (nid) (NODE_DATA(nid)->node_start_pfn)#define node_end_pfn (nid) pgdat_end_pfn(NODE_DATA(nid)) -static inline spinlock_t *zone_lru_lock (struct zone *zone) -{ -return &zone->zone_pgdat->lru_lock; -} -static inline struct lruvec *node_lruvec (struct pglist_data *pgdat) {return &pgdat->lruvec;
Like it?
thanks,
Signature · 2 lines
John Hubbard NVIDIA
Re: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
Quoted 10 lines · John Hubbard
Instead of removing that function, let's change it, and add another (since you have two cases: either a page* or a pgdat* is available), and move it to where it can compile, like this:
include/linux/mm.h +0
-0
include/linux/mm.h static inline pg_data_t *page_pgdat(const struct page *page) · line 1170
Quoted 4 lines
include/linux/mm.h +1
-0@@ -1167,6 +1167,16 @@ static inline pg_data_t *page_pgdat(const struct page *page)return NODE_DATA (page_to_nid (page)); } +static inline spinlock_t *zone_lru_lock (pg_data_t *pgdat)
In that case it should now be named node_lru_lock(). zone_lru_lock() was a wrapper introduced to make the conversion of per-zone to per-node lru_lock smoother.
include/linux/mm.h static inline pg_data_t *page_pgdat(const struct page *page) · line 1175
Quoted 5 lines
include/linux/mm.h +5
-0+{ +return &pgdat->lru_lock; +} + +static inline spinlock_t *zone_lru_lock_from_page (struct page *page)
Ditto. Or maybe even page_node_lru_lock()?
Quoted 30 lines
include/linux/mm.h +4
-0+{ +return zone_lru_lock (page_pgdat (page)); +} +#ifdef SECTION_IN_PAGE_FLAGS static inline void set_page_section (struct page *page,unsigned long section) {include/linux/mmzone.h +0
-5@@ -728,11 +728,6 @@ typedef struct pglist_data {#define node_start_pfn (nid) (NODE_DATA(nid)->node_start_pfn)#define node_end_pfn (nid) pgdat_end_pfn(NODE_DATA(nid)) -static inline spinlock_t *zone_lru_lock (struct zone *zone) -{ -return &zone->zone_pgdat->lru_lock; -} -static inline struct lruvec *node_lruvec (struct pglist_data *pgdat) {return &pgdat->lruvec;Like it?
thanks,
Re: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
Quoted 14 lines · Vlastimil Babka
Quoted 13 lines · John Hubbard
Instead of removing that function, let's change it, and add another (since you have two cases: either a page* or a pgdat* is available), and move it to where it can compile, like this:
include/linux/mm.h +0
-0@@ -1167,6 +1167,16 @@ static inline pg_data_t *page_pgdat(const struct page *page)return NODE_DATA (page_to_nid (page)); }
Quoted 5 lines
include/linux/mm.h static inline pg_data_t *page_pgdat(const struct page *page) · line 1170
Quoted 1 line
include/linux/mm.h +1
-0+static inline spinlock_t *zone_lru_lock (pg_data_t *pgdat)In that case it should now be named
node_lru_lock().zone_lru_lock()was a wrapper introduced to make the conversion of per-zone to per-nodelru_locksmoother.
Sounds good to me.
thanks,
Signature · 2 lines
John Hubbard NVIDIA
Re: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
In reply to John HubbardQuoted 47 lines · John Hubbard
Quoted 8 lines · Andrey Ryabinin
We have common pattern to access
lru_lockfrom a page pointer:
zone_lru_lock(page_zone(page))Which is silly, because it unfolds to this:
&NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)]->zone_pgdat->lru_lock while we can simply do &NODE_DATA(page_to_nid(page))->lru_lockHi Andrey,
Nice. I like it so much that I immediately want to tweak it. :)
Quoted 2 lines
Remove
zone_lru_lock()function, since it's only complicate things. Use 'page_pgdat(page)->lru_lock' pattern instead.Here, I think the
zone_lru_lock()is actually a nice way to add a touch of clarity at the call sites. How about, see below:[snip]
include/linux/mmzone.h typedef struct pglist_data { · line 791
Quoted 13 lines
include/linux/mmzone.h +0
-4@@ -788,10 +788,6 @@ typedef struct pglist_data {#define node_start_pfn (nid) (NODE_DATA(nid)->node_start_pfn)#define node_end_pfn (nid) pgdat_end_pfn(NODE_DATA(nid)) -static inline spinlock_t *zone_lru_lock (struct zone *zone) -{ -return &zone->zone_pgdat->lru_lock; -}Instead of removing that function, let's change it, and add another (since you have two cases: either a page* or a pgdat* is available), and move it to where it can compile, like this:
include/linux/mm.h +0
-0
include/linux/mm.h static inline pg_data_t *page_pgdat(const struct page *page) · line 1174
Quoted 8 lines
include/linux/mm.h +5
-0@@ -1167,6 +1167,16 @@ static inline pg_data_t *page_pgdat(const struct page *page)return NODE_DATA (page_to_nid (page)); } +static inline spinlock_t *zone_lru_lock (pg_data_t *pgdat) +{ +return &pgdat->lru_lock; +} +
I don't think wrapper for a simple plain access to the struct member is reasonable. Besides, there are plenty of "spin_lock(&pgdat->lru_lock)" even without this patch, so for consistency reasons &pgdat->lru_lock seems like a better choice to me.
Also "&pgdat->lru_lock" is just shorter than:
"node_lru_lock(pgdat)"
include/linux/mm.h static inline pg_data_t *page_pgdat(const struct page *page) · line 1179
Quoted 5 lines
include/linux/mm.h +5
-0+static inline spinlock_t *zone_lru_lock_from_page (struct page *page) +{ +return zone_lru_lock (page_pgdat (page)); +} +
I don't think such function would find any use. Usually lru_lock is taken to perform some manipulations with page *and* pgdat, thus it's better to remember page_pgdat(page) in local variable.
Re: [PATCH v2 2/4] mm: remove zone_lru_lock() function access ->lru_lock directly
Quoted 69 lines · Andrey Ryabinin
Quoted 47 lines · John Hubbard
Quoted 8 lines · Andrey Ryabinin
We have common pattern to access
lru_lockfrom a page pointer:
zone_lru_lock(page_zone(page))Which is silly, because it unfolds to this:
&NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)]->zone_pgdat->lru_lock while we can simply do &NODE_DATA(page_to_nid(page))->lru_lockHi Andrey,
Nice. I like it so much that I immediately want to tweak it. :)
Quoted 2 lines
Remove
zone_lru_lock()function, since it's only complicate things. Use 'page_pgdat(page)->lru_lock' pattern instead.Here, I think the
zone_lru_lock()is actually a nice way to add a touch of clarity at the call sites. How about, see below:[snip]
include/linux/mmzone.h typedef struct pglist_data { · line 791
Quoted 13 lines
include/linux/mmzone.h +0
-4@@ -788,10 +788,6 @@ typedef struct pglist_data {#define node_start_pfn (nid) (NODE_DATA(nid)->node_start_pfn)#define node_end_pfn (nid) pgdat_end_pfn(NODE_DATA(nid)) -static inline spinlock_t *zone_lru_lock (struct zone *zone) -{ -return &zone->zone_pgdat->lru_lock; -}Instead of removing that function, let's change it, and add another (since you have two cases: either a page* or a pgdat* is available), and move it to where it can compile, like this:
include/linux/mm.h +0
-0include/linux/mm.h static inline pg_data_t *page_pgdat(const struct page *page) · line 1174
Quoted 8 lines
include/linux/mm.h +5
-0@@ -1167,6 +1167,16 @@ static inline pg_data_t *page_pgdat(const struct page *page)return NODE_DATA (page_to_nid (page)); } +static inline spinlock_t *zone_lru_lock (pg_data_t *pgdat) +{ +return &pgdat->lru_lock; +} +I don't think wrapper for a simple plain access to the struct member is reasonable. Besides, there are plenty of "
spin_lock(&pgdat->lru_lock)" even without this patch, so for consistency reasons&pgdat->lru_lockseems like a better choice to me.Also "
&pgdat->lru_lock" is just shorter than:"node_lru_lock(pgdat)"
Quoted 10 lines
include/linux/mm.h static inline pg_data_t *page_pgdat(const struct page *page) · line 1179
Quoted 5 lines
include/linux/mm.h +5
-0+static inline spinlock_t *zone_lru_lock_from_page (struct page *page) +{ +return zone_lru_lock (page_pgdat (page)); +} +I don't think such function would find any use. Usually
lru_lockis taken to perform some manipulations with page *and* pgdat, thus it's better to rememberpage_pgdat(page)in local variable.
That's a good argument.
thanks,
Signature · 2 lines
John Hubbard NVIDIA
PATCH v2 3/4 mm/compaction: pass pgdat to too_many_isolated() instead of zone
In reply to Andrey Ryabinintoo_many_isolated() in mm/compaction.c looks only at node state, so it makes more sense to change argument to pgdat instead of zone.
---
Changes since v1:
- Added acks
mm/compaction.c | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-)
mm/compaction.c +9 -10
@@ -738,16 +738,16 @@ isolate_freepages_range(struct compact_control *cc, }/* Similar to reclaim, but different enough that they don't share logic */ -static bool too_many_isolated (struct zone *zone) +static bool too_many_isolated (pg_data_t *pgdat) {unsigned long active, inactive, isolated; - inactive =node_page_state (zone->zone_pgdat,NR_INACTIVE_FILE ) + -node_page_state (zone->zone_pgdat,NR_INACTIVE_ANON ); - active =node_page_state (zone->zone_pgdat,NR_ACTIVE_FILE ) + -node_page_state (zone->zone_pgdat,NR_ACTIVE_ANON ); - isolated =node_page_state (zone->zone_pgdat,NR_ISOLATED_FILE ) + -node_page_state (zone->zone_pgdat,NR_ISOLATED_ANON ); + inactive =node_page_state (pgdat,NR_INACTIVE_FILE ) + +node_page_state (pgdat,NR_INACTIVE_ANON ); + active =node_page_state (pgdat,NR_ACTIVE_FILE ) + +node_page_state (pgdat,NR_ACTIVE_ANON ); + isolated =node_page_state (pgdat,NR_ISOLATED_FILE ) + +node_page_state (pgdat,NR_ISOLATED_ANON );return isolated > (inactive + active) /2 ; }
@@ -774,8 +774,7 @@ static unsigned long isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn, unsigned long end_pfn, isolate_mode_t isolate_mode) { - struct zone *zone = cc->zone; -pg_data_t *pgdat = zone->zone_pgdat; + pg_data_t *pgdat = cc->zone->zone_pgdat;unsigned long nr_scanned =0 , nr_isolated =0 ;struct lruvec *lruvec;unsigned long flags =0 ;
@@ -791,7 +790,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,* list by either parallel reclaimers or compaction. If there are, * delay for some time until fewer pages are isolated */ -while (unlikely (too_many_isolated (zone))) { +while (unlikely (too_many_isolated (pgdat))) {/* async migration should just abort */ if (cc->mode ==MIGRATE_ASYNC )return 0 ;
Signature · 1 line
2.19.2
Re: [PATCH v2 3/4] mm/compaction: pass pgdat to too_many_isolated() instead of zone
Quoted 9 lines · Andrey Ryabinin
too_many_isolated()inmm/compaction.clooks only at node state, so it makes more sense to change argument to pgdat instead of zone.Signed-off-by Andrey Ryabinin <[email protected]>Acked-by Vlastimil Babka <[email protected]>Acked-by Rik van Riel <[email protected]>Cc Johannes Weiner <[email protected]>Cc Michal Hocko <[email protected]>Cc Mel Gorman <[email protected]>
Signature · 2 lines
Mel Gorman SUSE Labs
PATCH v2 4/4 mm/vmscan: remove unused lru_pages argument
In reply to Andrey RyabininSince commit 9092c71bb724 ("mm: use sc->priority for slab shrink targets") the argument 'unsigned long *lru_pages' passed around with no purpose. Remove it.
---
Changes since v1:
- Changelog update
- Added acks
mm/vmscan.c | 17 +++++------------ 1 file changed, 5 insertions(+), 12 deletions(-)
mm/vmscan.c +5 -12
@@ -2257,8 +2257,7 @@ enum scan_balance {* nr[2] = file inactive pages to scan; nr[3] = file active pages to scan */ static void get_scan_count (struct lruvec *lruvec,struct mem_cgroup *memcg, -struct scan_control *sc,unsigned long *nr, -unsigned long *lru_pages) +struct scan_control *sc,unsigned long *nr) {int swappiness =mem_cgroup_swappiness (memcg);struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
@@ -2409,7 +2408,6 @@ static void get_scan_count(struct lruvec *lruvec, struct mem_cgroup *memcg, fraction[1 ] = fp; denominator = ap + fp +1 ; out: - *lru_pages =0 ;for_each_evictable_lru (lru) {int file =is_file_lru (lru);unsigned long lruvec_size;
@@ -2525,7 +2523,6 @@ static void get_scan_count(struct lruvec *lruvec, struct mem_cgroup *memcg,BUG (); } - *lru_pages += lruvec_size; nr[lru] = scan; } }
@@ -2534,7 +2531,7 @@ static void get_scan_count(struct lruvec *lruvec, struct mem_cgroup *memcg,* This is a basic per-node page freer. Used by both kswapd and direct reclaim. */ static void shrink_node_memcg (struct pglist_data *pgdat,struct mem_cgroup *memcg, -struct scan_control *sc,unsigned long *lru_pages) +struct scan_control *sc) {struct lruvec *lruvec =mem_cgroup_lruvec (pgdat, memcg);unsigned long nr[NR_LRU_LISTS ];
@@ -2546,7 +2543,7 @@ static void shrink_node_memcg(struct pglist_data *pgdat, struct mem_cgroup *memcstruct blk_plug plug;bool scan_adjusted; -get_scan_count (lruvec, memcg, sc, nr, lru_pages); +get_scan_count (lruvec, memcg, sc, nr);/* Record the original scan target for proportional adjustments later */ memcpy (targets, nr,sizeof (nr));
@@ -2751,7 +2748,6 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc) .pgdat = pgdat, .priority = sc->priority, }; -unsigned long node_lru_pages =0 ;struct mem_cgroup *memcg;memset (&sc->nr,0 ,sizeof (sc->nr));
@@ -2761,7 +2757,6 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc) memcg =mem_cgroup_iter (root,NULL , &reclaim); do { - unsigned long lru_pages; unsigned long reclaimed; unsigned long scanned;
@@ -2798,8 +2793,7 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc) reclaimed = sc->nr_reclaimed; scanned = sc->nr_scanned; -shrink_node_memcg (pgdat, memcg, sc, &lru_pages); - node_lru_pages += lru_pages; +shrink_node_memcg (pgdat, memcg, sc);if (sc->may_shrinkslab) { shrink_slab(sc->gfp_mask, pgdat->node_id,
@@ -3332,7 +3326,6 @@ unsigned long mem_cgroup_shrink_node(struct mem_cgroup *memcg, .may_swap = !noswap, .may_shrinkslab = 1, }; -unsigned long lru_pages; sc.gfp_mask = (gfp_mask &GFP_RECLAIM_MASK ) | (GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK );
@@ -3349,7 +3342,7 @@ unsigned long mem_cgroup_shrink_node(struct mem_cgroup *memcg,* will pick up pages from other mem cgroup's as well. We hack * the priority and make it zero. */ -shrink_node_memcg (pgdat, memcg, &sc, &lru_pages); +shrink_node_memcg (pgdat, memcg, &sc);trace_mm_vmscan_memcg_softlimit_reclaim_end (sc.nr_reclaimed);
Signature · 1 line
2.19.2
Re: [PATCH v2 4/4] mm/vmscan: remove unused lru_pages argument
Quoted 10 lines · Andrey Ryabinin
Since commit
9092c71bb724("mm: usesc->priorityfor slab shrink targets") the argument 'unsigned long*lru_pages' passed around with no purpose. Remove it.Signed-off-by Andrey Ryabinin <[email protected]>Acked-by Johannes Weiner <[email protected]>Acked-by Vlastimil Babka <[email protected]>Cc Michal Hocko <[email protected]>Cc Rik van Riel <[email protected]>Cc Mel Gorman <[email protected]>
Signature · 2 lines
Mel Gorman SUSE Labs
Quoted 6 lines · Andrey Ryabinin
workingset_eviction()doesn't use and never did use the @mapping argument. Remove it.Signed-off-by Andrey Ryabinin <[email protected]>Acked-by Johannes Weiner <[email protected]>Acked-by Rik van Riel <[email protected]>
Quoted 61 lines
Cc Michal Hocko <[email protected]>Cc Vlastimil Babka <[email protected]>Cc Mel Gorman <[email protected]>---
Changes since v1:
- s/@mapping/@page->mapping in comment
- Acksinclude/linux/swap.h | 2 +- mm/vmscan.c | 2 +- mm/workingset.c | 5 ++--- 3 files changed, 4 insertions(+), 5 deletions(-)include/linux/swap.h +1
-1@@ -307,7 +307,7 @@ struct vma_swap_readahead { };/* linux/mm/workingset.c */ -void *workingset_eviction (struct address_space *mapping,struct page *page); +void *workingset_eviction (struct page *page);void workingset_refault (struct page *page,void *shadow);void workingset_activation (struct page *page);mm/vmscan.c +1
-1@@ -952,7 +952,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page,*/ if (reclaimed &&page_is_file_cache (page) && !mapping_exiting (mapping) && !dax_mapping (mapping)) - shadow =workingset_eviction (mapping, page); + shadow =workingset_eviction (page);__delete_from_page_cache (page, shadow);xa_unlock_irqrestore (&mapping->i_pages, flags);mm/workingset.c +2
-3@@ -215,13 +215,12 @@ static void unpack_shadow(void *shadow, int *memcgidp, pg_data_t **pgdat,/** * workingset_eviction - note the eviction of a page from memory -* @mapping: address space the page was backing * @page: the page being evicted * -* Returns a shadow entry to be stored in @mapping->i_pages in place +* Returns a shadow entry to be stored in @page->mapping->i_pages in place * of the evicted @page so that a later refault can be detected. */ -void *workingset_eviction (struct address_space *mapping,struct page *page) +void *workingset_eviction (struct page *page) {struct pglist_data *pgdat =page_pgdat (page);struct mem_cgroup *memcg =page_memcg (page);
Quoted 9 lines · Andrey Ryabinin
workingset_eviction()doesn't use and never did use the @mapping argument. Remove it.Signed-off-by Andrey Ryabinin <[email protected]>Acked-by Johannes Weiner <[email protected]>Acked-by Rik van Riel <[email protected]>Cc Michal Hocko <[email protected]>Cc Vlastimil Babka <[email protected]>Cc Mel Gorman <[email protected]>
Signature · 2 lines
Mel Gorman SUSE Labs