Vm Pageout Scalability Improvements (V8) - continued
Subsystems (19)
cover Vm Pageout Scalability Improvements (V8) - continued
The patches to follow are a continuation of the V8 "VM pageout scalability improvements" series that Rik van Riel posted to LKML on 23May08. These patches apply atop Rik's series with the following overlap:
Patches 13 through 16 replace the corresponding patches in Rik's posting.
Patch 13, the noreclaim lru infrastructure, now includes Kosaki Motohiro's memcontrol enhancements to track nonreclaimable pages.
Patches 14 and 15 are largely unchanged, except for refresh. Includes some minor statistics formatting cleanup.
Patch 16 includes a fix for an potential [unobserved] race condition during SHM_UNLOCK.
---
Additional patches in this series:
Patches 17 through 20 keep mlocked pages off the normal [in]active LRU
lists using the noreclaim lru infrastructure. These patches represent
a fairly significant rework of an RFC patch originally posted by Nick Piggin.
Patches 21 and 22 are optional, but recommended, enhancements to the overall noreclaim series.
Patches 23 and 24 are optional enhancements useful during debug and testing.
Patch 25 is a rather verbose document describing the noreclaim lru infrastructure and the use thereof to keep ramfs, SHM_LOCKED and mlocked pages off the normal LRU lists.
---
The entire stack, including Rik's split lru patches, are holding up very well under stress loads. E.g., ran for over 90+ hours over the weekend on both x86_64 [32GB, 8core] and ia64 [32GB, 16cpu] platforms without error over last weekend.
I think these are ready for a spin in -mm atop Rik's patches.
Lee
PATCH 13/25 Noreclaim LRU Infrastructure
From: Lee Schermerhorn <[email protected]>
Infrastructure to manage pages excluded from reclaim--i.e., hidden from vmscan. Based on a patch by Larry Woodman of Red Hat. Reworked to maintain "nonreclaimable" pages on a separate per-zone LRU list, to "hide" them from vmscan.
Kosaki Motohiro added the support for the memory controller noreclaim lru list.
Pages on the noreclaim list have both PG_noreclaim and PG_lru set. Thus, PG_noreclaim is analogous to and mutually exclusive with PG_active--it specifies which LRU list the page is on.
The noreclaim infrastructure is enabled by a new mm Kconfig option [CONFIG_]NORECLAIM_LRU.
A new function 'page_reclaimable(page, vma)' in vmscan.c tests whether or not a page is reclaimable. Subsequent patches will add the various !reclaimable tests. We'll want to keep these tests light-weight for use in shrink_active_list() and, possibly, the fault path.
To avoid races between tasks putting pages [back] onto an LRU list and tasks that might be moving the page from nonreclaimable to reclaimable state, one should test reclaimability under page lock and place nonreclaimable pages directly on the noreclaim list before dropping the lock. Otherwise, we risk "stranding" reclaimable pages on the noreclaim list. It's OK to use the pagevec caches for reclaimable pages. The new function 'putback_lru_page()'--inverse to 'isolate_lru_page()'--handles this transition, including potential page truncation while the page is unlocked.
include/linux/memcontrol.h | 2 include/linux/mm_inline.h | 13 ++- include/linux/mmzone.h | 24 ++++++ include/linux/page-flags.h | 13 +++ include/linux/pagevec.h | 1 include/linux/swap.h | 12 +++ mm/Kconfig | 10 ++ mm/internal.h | 26 +++++++ mm/memcontrol.c | 73 ++++++++++++-------- mm/mempolicy.c | 2 mm/migrate.c | 68 ++++++++++++------ mm/page_alloc.c | 9 ++ mm/swap.c | 52 +++++++++++--- mm/vmscan.c | 164 +++++++++++++++++++++++++++++++++++++++------ 14 files changed, 382 insertions(+), 87 deletions(-)
mm/Kconfig +10 -0
@@ -205,3 +205,13 @@ config NR_QUICK config VIRT_TO_BUS def_bool y depends on !ARCH_NO_VIRT_TO_BUS + +config NORECLAIM_LRU + bool "Add LRU list to track non-reclaimable pages (EXPERIMENTAL, 64BIT only)" + depends on EXPERIMENTAL && 64BIT + help + Supports tracking of non-reclaimable pages off the [in]active lists + to avoid excessive reclaim overhead on large memory systems. Pages + may be non-reclaimable because: they are locked into memory, they + are anonymous pages for which no swap space exists, or they are anon + pages that are expensive to unmap [long anon_vma "related vma" list.]
include/linux/page-flags.h +13 -0
@@ -94,6 +94,9 @@ enum pageflags { PG_reclaim,/* To be reclaimed asap */ PG_buddy,/* Page is free, on buddy lists */ PG_swapbacked,/* Page is backed by RAM/swap */ +#ifdef CONFIG_NORECLAIM_LRU + PG_noreclaim,/* Page is "non-reclaimable" */ +#endif #ifdef CONFIG_IA64_UNCACHED_ALLOCATOR PG_uncached, /* Page has been mapped as uncached */ #endif
@@ -167,6 +170,7 @@ PAGEFLAG(Referenced, referenced) TESTCLE PAGEFLAG(Dirty, dirty) TESTSCFLAG(Dirty, dirty) __CLEARPAGEFLAG(Dirty, dirty) PAGEFLAG(LRU, lru) __CLEARPAGEFLAG(LRU, lru) PAGEFLAG(Active, active) __CLEARPAGEFLAG(Active, active) + TESTCLEARFLAG(Active, active) __PAGEFLAG(Slab, slab) PAGEFLAG(Checked, owner_priv_1) /* Used by some filesystems */ PAGEFLAG(Pinned, owner_priv_1) TESTSCFLAG(Pinned, owner_priv_1) /* Xen */
@@ -203,6 +207,15 @@ PAGEFLAG(SwapCache, swapcache) PAGEFLAG_FALSE(SwapCache)#endif +#ifdef CONFIG_NORECLAIM_LRU +PAGEFLAG(Noreclaim, noreclaim) __CLEARPAGEFLAG(Noreclaim, noreclaim) + TESTCLEARFLAG(Noreclaim, noreclaim) +#else +PAGEFLAG_FALSE(Noreclaim) TESTCLEARFLAG_FALSE(Noreclaim) + SETPAGEFLAG_NOOP(Noreclaim) CLEARPAGEFLAG_NOOP(Noreclaim) + __CLEARPAGEFLAG_NOOP(Noreclaim) +#endif + #ifdef CONFIG_IA64_UNCACHED_ALLOCATOR PAGEFLAG(Uncached, uncached)#else
include/linux/mmzone.h +23 -1
@@ -85,6 +85,11 @@ enum zone_stat_item {NR_ACTIVE_ANON ,/* " " " " " */ NR_INACTIVE_FILE ,/* " " " " " */ NR_ACTIVE_FILE ,/* " " " " " */ +#ifdef CONFIG_NORECLAIM_LRU +NR_NORECLAIM ,/* " " " " " */ +#else +NR_NORECLAIM =NR_ACTIVE_FILE ,/* avoid compiler errors in dead code */ +#endifNR_ANON_PAGES ,/* Mapped anonymous pages */ NR_FILE_MAPPED ,/* pagecache pages mapped into pagetables. only modified from process context */
@@ -124,10 +129,18 @@ enum lru_list {LRU_ACTIVE_ANON =LRU_BASE +LRU_ACTIVE ,LRU_INACTIVE_FILE =LRU_BASE +LRU_FILE ,LRU_ACTIVE_FILE =LRU_BASE +LRU_FILE +LRU_ACTIVE , - NR_LRU_LISTS }; +#ifdef CONFIG_NORECLAIM_LRU +LRU_NORECLAIM , +#else +LRU_NORECLAIM =LRU_ACTIVE_FILE ,/* avoid compiler errors in dead code */ +#endif + NR_LRU_LISTS +};#define for_each_lru (l) for (l = 0; l < NR_LRU_LISTS; l++) +#define for_each_reclaimable_lru (l) for (l = 0; l <= LRU_ACTIVE_FILE; l++) +static inline int is_file_lru (enum lru_list l) {return (l ==LRU_INACTIVE_FILE || l ==LRU_ACTIVE_FILE );
@@ -138,6 +151,15 @@ static inline int is_active_lru(enum lrureturn (l ==LRU_ACTIVE_ANON || l ==LRU_ACTIVE_FILE ); } +static inline int is_noreclaim_lru (enum lru_list l) +{ +#ifdef CONFIG_NORECLAIM_LRU +return l ==LRU_NORECLAIM ; +#else +return 0 ; +#endif +} +enum lru_list page_lru (struct page *page);struct per_cpu_pages {
mm/page_alloc.c +9 -0
@@ -256,6 +256,9 @@ static void bad_page(struct page *page) 1 << PG_private | 1 << PG_locked | 1 << PG_active | +#ifdef CONFIG_NORECLAIM_LRU + 1 << PG_noreclaim | +#endif 1 << PG_dirty | 1 << PG_reclaim | 1 << PG_slab |
@@ -491,6 +494,9 @@ static inline int free_pages_check(struc 1 << PG_swapcache | 1 << PG_writeback | 1 << PG_reserved | +#ifdef CONFIG_NORECLAIM_LRU + 1 << PG_noreclaim | +#endif 1 << PG_buddy ))))bad_page (page);if (PageDirty (page))
@@ -642,6 +648,9 @@ static int prep_new_page(struct page *pa 1 << PG_private | 1 << PG_locked | 1 << PG_active | +#ifdef CONFIG_NORECLAIM_LRU + 1 << PG_noreclaim | +#endif 1 << PG_dirty | 1 << PG_slab | 1 << PG_swapcache |
include/linux/mm_inline.h +9 -4
@@ -89,11 +89,16 @@ del_page_from_lru(struct zone *zone, strenum lru_list l =LRU_INACTIVE_ANON ;list_del (&page->lru); -if (PageActive (page)) { -__ClearPageActive (page); - l +=LRU_ACTIVE ; +if (PageNoreclaim (page)) { +__ClearPageNoreclaim (page); + l =LRU_NORECLAIM ; + }else { +if (PageActive (page)) { +__ClearPageActive (page); + l +=LRU_ACTIVE ; + } + l +=page_file_cache (page); } - l +=page_file_cache (page);__dec_zone_state (zone,NR_INACTIVE_ANON + l); }
include/linux/swap.h +12 -0
@@ -180,6 +180,8 @@ extern int lru_add_drain_all(void);extern void rotate_reclaimable_page (struct page *page);extern void swap_setup (void ); +extern void add_page_to_noreclaim_list (struct page *page); +/** * lru_cache_add: add a page to the page lists * @page: the page to add
@@ -228,6 +230,16 @@ static inline int zone_reclaim(struct zo }#endif +#ifdef CONFIG_NORECLAIM_LRU +extern int page_reclaimable (struct page *page,struct vm_area_struct *vma); +#else +static inline int page_reclaimable (struct page *page, +struct vm_area_struct *vma) +{ +return 1 ; +} +#endif +extern int kswapd_run (int nid);#ifdef CONFIG_MMU
include/linux/pagevec.h +0 -1
@@ -101,7 +101,6 @@ static inline void __pagevec_lru_add_act____pagevec_lru_add (pvec,LRU_ACTIVE_FILE ); } -static inline void pagevec_lru_add_file (struct pagevec *pvec) {if (pagevec_count (pvec))
mm/swap.c +43 -9
118 lines not on this page · raw message
mm/migrate.c +45 -23
151 lines not on this page · raw message
mm/vmscan.c +145 -19
262 lines not on this page · raw message
mm/mempolicy.c +1 -1
@@ -2199,7 +2199,7 @@ static void gather_stats(struct page *paif (PageSwapCache (page)) md->swapcache++; -if (PageActive (page)) +if (PageActive (page) ||PageNoreclaim (page)) md->active++;if (PageWriteback (page))
mm/internal.h +26 -0
@@ -34,8 +34,15 @@ static inline void __put_page(struct pagatomic_dec (&page->_count); } +/* +* in mm/vmscan.c: +*/ extern int isolate_lru_page (struct page *page); +extern int putback_lru_page (struct page *page); +/* +* in mm/page_alloc.c +*/ extern void __free_pages_bootmem (struct page *page,unsigned int order);/*
@@ -49,6 +56,25 @@ static inline unsigned long page_order(sreturn page_private (page); } +#ifdef CONFIG_NORECLAIM_LRU +/* +* noreclaim_migrate_page() called only from migrate_page_copy() to +* migrate noreclaim flag to new page. +* Note that the old page has been isolated from the LRU lists at this +* point so we don't need to worry about LRU statistics. +*/ +static inline void noreclaim_migrate_page (struct page *new,struct page *old) +{ +if (TestClearPageNoreclaim (old)) +SetPageNoreclaim (new); +} +#else +static inline void noreclaim_migrate_page (struct page *new,struct page *old) +{ +} +#endif + +/* * FLATMEM and DISCONTIGMEM configurations use alloc_bootmem_node, * so all functions starting at paging_init should be marked __init
mm/memcontrol.c +45 -28
121 lines not on this page · raw message
include/linux/memcontrol.h +1 -1
@@ -35,7 +35,7 @@ extern int mem_cgroup_charge(struct pageextern int mem_cgroup_cache_charge (struct page *page,struct mm_struct *mm,gfp_t gfp_mask);extern void mem_cgroup_uncharge_page (struct page *page); -extern void mem_cgroup_move_lists (struct page *page,bool active); +extern void mem_cgroup_move_lists (struct page *page,enum lru_list lru); extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan, struct list_head *dst, unsigned long *scanned, int order,
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 14/25 Noreclaim LRU Page Statistics
In reply to Lee SchermerhornFrom: Lee Schermerhorn <[email protected]>
Report non-reclaimable pages per zone and system wide.
Kosaki Motohiro added support for memory controller noreclaim statistics.
drivers/base/node.c | 6 ++++++ fs/proc/proc_misc.c | 6 ++++++ mm/memcontrol.c | 6 ++++++ mm/page_alloc.c | 16 +++++++++++++++- mm/vmstat.c | 3 +++ 5 files changed, 36 insertions(+), 1 deletion(-)
mm/page_alloc.c +15 -1
@@ -1918,12 +1918,20 @@ void show_free_areas(void) } printk("Active_anon:%lu active_file:%lu inactive_anon%lu\n" - " inactive_file:%lu dirty:%lu writeback:%lu unstable:%lu\n" + " inactive_file:%lu" +//TODO: check/adjust line lengths +#ifdef CONFIG_NORECLAIM_LRU +" noreclaim:%lu" +#endif +" dirty:%lu writeback:%lu unstable:%lu\n" " free:%lu slab:%lu mapped:%lu pagetables:%lu bounce:%lu\n" ,global_page_state (NR_ACTIVE_ANON ),global_page_state (NR_ACTIVE_FILE ),global_page_state (NR_INACTIVE_ANON ),global_page_state (NR_INACTIVE_FILE ), +#ifdef CONFIG_NORECLAIM_LRU +global_page_state (NR_NORECLAIM ), +#endifglobal_page_state (NR_FILE_DIRTY ),global_page_state (NR_WRITEBACK ), global_page_state(NR_UNSTABLE_NFS),
@@ -1950,6 +1958,9 @@ void show_free_areas(void) " inactive_anon:%lukB" " active_file:%lukB" " inactive_file:%lukB" +#ifdef CONFIG_NORECLAIM_LRU + " noreclaim:%lukB" +#endif" present:%lukB" " pages_scanned:%lu" " all_unreclaimable? %s"
@@ -1963,6 +1974,9 @@ void show_free_areas(void)K (zone_page_state (zone,NR_INACTIVE_ANON )),K (zone_page_state (zone,NR_ACTIVE_FILE )),K (zone_page_state (zone,NR_INACTIVE_FILE )), +#ifdef CONFIG_NORECLAIM_LRU +K (zone_page_state (zone,NR_NORECLAIM )), +#endifK (zone->present_pages), zone->pages_scanned, (zone_is_all_unreclaimable (zone) ?"yes" :"no" )
mm/vmstat.c +3 -0
@@ -699,6 +699,9 @@ static const char * const vmstat_text[]"nr_active_anon" ,"nr_inactive_file" ,"nr_active_file" , +#ifdef CONFIG_NORECLAIM_LRU +"nr_noreclaim" , +#endif"nr_anon_pages" ,"nr_mapped" , "nr_file_pages",
drivers/base/node.c +6 -0
@@ -67,6 +67,9 @@ static ssize_t node_read_meminfo(struct "Node %d Inactive(anon): %8lu kB\n" "Node %d Active(file): %8lu kB\n" "Node %d Inactive(file): %8lu kB\n" +#ifdef CONFIG_NORECLAIM_LRU + "Node %d Noreclaim: %8lu kB\n" +#endif#ifdef CONFIG_HIGHMEM "Node %d HighTotal: %8lu kB\n" "Node %d HighFree: %8lu kB\n"
@@ -96,6 +99,9 @@ static ssize_t node_read_meminfo(struct nid,node_page_state (nid,NR_INACTIVE_ANON ), nid,node_page_state (nid,NR_ACTIVE_FILE ), nid,node_page_state (nid,NR_INACTIVE_FILE ), +#ifdef CONFIG_NORECLAIM_LRU + nid,node_page_state (nid,NR_NORECLAIM ), +#endif #ifdef CONFIG_HIGHMEM nid,K (i.totalhigh), nid, K(i.freehigh),
fs/proc/proc_misc.c +6 -0
@@ -174,6 +174,9 @@ static int meminfo_read_proc(char *page, "Inactive(anon): %8lu kB\n" "Active(file): %8lu kB\n" "Inactive(file): %8lu kB\n" +#ifdef CONFIG_NORECLAIM_LRU + "Noreclaim: %8lu kB\n" +#endif#ifdef CONFIG_HIGHMEM "HighTotal: %8lu kB\n" "HighFree: %8lu kB\n"
@@ -209,6 +212,9 @@ static int meminfo_read_proc(char *page,K (inactive_anon),K (active_file),K (inactive_file), +#ifdef CONFIG_NORECLAIM_LRU +K (global_page_state (NR_NORECLAIM )), +#endif #ifdef CONFIG_HIGHMEMK (i.totalhigh), K(i.freehigh),
mm/memcontrol.c +6 -0
@@ -905,6 +905,7 @@ static int mem_control_stat_show(struct {unsigned long active_anon, inactive_anon;unsigned long active_file, inactive_file; +unsigned long noreclaim; inactive_anon =mem_cgroup_get_all_zonestat (mem_cont,LRU_INACTIVE_ANON );
@@ -914,10 +915,15 @@ static int mem_control_stat_show(struct LRU_INACTIVE_FILE); active_file =mem_cgroup_get_all_zonestat (mem_cont,LRU_ACTIVE_FILE ); + noreclaim =mem_cgroup_get_all_zonestat (mem_cont, +LRU_NORECLAIM ); + cb->fill (cb,"active_anon" , (active_anon) *PAGE_SIZE ); cb->fill (cb,"inactive_anon" , (inactive_anon) *PAGE_SIZE ); cb->fill (cb,"active_file" , (active_file) *PAGE_SIZE ); cb->fill (cb,"inactive_file" , (inactive_file) *PAGE_SIZE ); + cb->fill (cb,"noreclaim" , noreclaim *PAGE_SIZE ); + }return 0 ; }
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 15/25 Ramfs and Ram Disk pages are non-reclaimable
In reply to Lee SchermerhornFrom: Lee Schermerhorn <[email protected]>
Christoph Lameter pointed out that ram disk pages also clutter the LRU lists. When vmscan finds them dirty and tries to clean them, the ram disk writeback function just redirties the page so that it goes back onto the active list. Round and round she goes...
Define new address_space flag [shares address_space flags member with mapping's gfp mask] to indicate that the address space contains all non-reclaimable pages. This will provide for efficient testing of ramdisk pages in page_reclaimable().
Also provide wrapper functions to set/test the noreclaim state to minimize #ifdefs in ramdisk driver and any other users of this facility.
Set the noreclaim state on address_space structures for new ramdisk inodes. Test the noreclaim state in page_reclaimable() to cull non-reclaimable pages.
Similarly, ramfs pages are non-reclaimable. Set the 'noreclaim' address_space flag for new ramfs inodes.
These changes depend on [CONFIG_]NORECLAIM_LRU.
drivers/block/brd.c | 13 +++++++++++++ fs/ramfs/inode.c | 1 + include/linux/pagemap.h | 22 ++++++++++++++++++++++ mm/vmscan.c | 5 +++++ 4 files changed, 41 insertions(+)
include/linux/pagemap.h +22 -0
@@ -30,6 +30,28 @@ static inline void mapping_set_error(str } } +#ifdef CONFIG_NORECLAIM_LRU +#define AS_NORECLAIM (__GFP_BITS_SHIFT + 2)/* e.g., ramdisk, SHM_LOCK */ + +static inline void mapping_set_noreclaim (struct address_space *mapping) +{ +set_bit (AS_NORECLAIM , &mapping->flags); +} + +static inline int mapping_non_reclaimable (struct address_space *mapping) +{ +if (mapping && (mapping->flags &AS_NORECLAIM )) +return 1 ; +return 0 ; +} +#else +static inline void mapping_set_noreclaim (struct address_space *mapping) { } +static inline int mapping_non_reclaimable (struct address_space *mapping) +{ +return 0 ; +} +#endif +static inline gfp_t mapping_gfp_mask (struct address_space * mapping) {return (__force gfp_t )mapping->flags & __GFP_BITS_MASK;
mm/vmscan.c +5 -0
@@ -2308,6 +2308,8 @@ int zone_reclaim(struct zone *zone, gfp_* lists vs noreclaim list. * * Reasons page might not be reclaimable: +* (1) page's mapping marked non-reclaimable +* * TODO - later patches */ int page_reclaimable (struct page *page,struct vm_area_struct *vma)
@@ -2315,6 +2317,9 @@ int page_reclaimable(struct page *page,VM_BUG_ON (PageNoreclaim (page)); +if (mapping_non_reclaimable (page_mapping (page))) +return 0 ; +/* TODO: test page [!]reclaimable conditions */ return 1 ;
fs/ramfs/inode.c +1 -0
@@ -61,6 +61,7 @@ struct inode *ramfs_get_inode(struct sup inode->i_mapping->a_ops = &ramfs_aops; inode->i_mapping->backing_dev_info = &ramfs_backing_dev_info;mapping_set_gfp_mask (inode->i_mapping,GFP_HIGHUSER ); +mapping_set_noreclaim (inode->i_mapping); inode->i_atime = inode->i_mtime = inode->i_ctime =CURRENT_TIME ;switch (mode &S_IFMT ) {default :
drivers/block/brd.c +13 -0
@@ -374,8 +374,21 @@ static int brd_ioctl(struct inode *inodereturn error; } +/* +* brd_open(): +* Just mark the mapping as containing non-reclaimable pages +*/ +static int brd_open (struct inode *inode,struct file *filp) +{ +struct address_space *mapping = inode->i_mapping; + +mapping_set_noreclaim (mapping); +return 0 ; +} +static struct block_device_operations brd_fops = { .owner =THIS_MODULE , + .open = brd_open, .ioctl = brd_ioctl, #ifdef CONFIG_BLK_DEV_XIP .direct_access = brd_direct_access,
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 16/25 SHM_LOCKED pages are non-reclaimable
In reply to Lee SchermerhornAgainst: 2.6.26-rc2-mm1
While working with Nick Piggin's mlock patches, I noticed that shmem segments locked via shmctl(SHM_LOCKED) were not being handled. SHM_LOCKed pages work like ramdisk pages--the writeback function just redirties the page so that it can't be reclaimed. Deal with these using the same approach as for ram disk pages.
Use the AS_NORECLAIM flag to mark address_space of SHM_LOCKed shared memory regions as non-reclaimable. Then these pages will be culled off the normal LRU lists during vmscan.
Add new wrapper function to clear the mapping's noreclaim state when/if shared memory segment is munlocked.
Add 'scan_mapping_noreclaim_page()' to mm/vmscan.c to scan all pages in the shmem segment's mapping [struct address_space] for reclaimability now that they're no longer locked. If so, move them to the appropriate zone lru list. Note that scan_mapping_noreclaim_page() must be able to sleep on page_lock(), so we can't call it holding the shmem info spinlock nor the shmid spinlock. So, we pass the mapping [address_space] back to shmctl() on SHM_UNLOCK for rescuing any nonreclaimable pages after dropping the spinlocks. Once we drop the shmid lock, the backing shmem file can be deleted if the calling task doesn't have the shm area attached. To handle this, we take an extra reference on the file before dropping the shmid lock and drop the reference after scanning the mapping's noreclaim pages.
Changes depend on [CONFIG_]NORECLAIM_LRU.
Signed-off-by: Lee Schermerhorn <[email protected]> Signed-off-by: Rik van Riel <[email protected]> Signed-off-by: Kosaki Motohiro <[email protected]>
include/linux/mm.h | 9 ++-- include/linux/pagemap.h | 12 ++++-- include/linux/swap.h | 4 ++ ipc/shm.c | 20 +++++++++- mm/shmem.c | 10 +++-- mm/vmscan.c | 93 ++++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 136 insertions(+), 12 deletions(-)
mm/shmem.c +7 -3
@@ -1458,23 +1458,27 @@ static struct mempolicy *shmem_get_polic }#endif -int shmem_lock (struct file *file,int lock,struct user_struct *user) +struct address_space *shmem_lock (struct file *file,int lock, +struct user_struct *user) {struct inode *inode = file->f_path.dentry->d_inode;struct shmem_inode_info *info =SHMEM_I (inode); -int retval = -ENOMEM ; +struct address_space *retval =ERR_PTR (-ENOMEM );spin_lock (&info->lock);if (lock && !(info->flags &VM_LOCKED )) {if (!user_shm_lock (inode->i_size, user)) goto out_nomem; info->flags |=VM_LOCKED ; +mapping_set_noreclaim (file->f_mapping); + retval =NULL ; }if (!lock && (info->flags &VM_LOCKED ) && user) {user_shm_unlock (inode->i_size, user); info->flags &= ~VM_LOCKED ; +mapping_clear_noreclaim (file->f_mapping); + retval = file->f_mapping; } - retval =0 ; out_nomem:spin_unlock (&info->lock);return retval;
include/linux/pagemap.h +9 -3
@@ -38,14 +38,20 @@ static inline void mapping_set_noreclaim set_bit(AS_NORECLAIM, &mapping->flags); } +static inline void mapping_clear_noreclaim (struct address_space *mapping) +{ +clear_bit (AS_NORECLAIM , &mapping->flags); +} +static inline int mapping_non_reclaimable (struct address_space *mapping) { -if (mapping && (mapping->flags &AS_NORECLAIM )) -return 1 ; -return 0 ; +if (likely (mapping)) +return test_bit (AS_NORECLAIM , &mapping->flags); +return !!mapping; }#else static inline void mapping_set_noreclaim (struct address_space *mapping) { } +static inline void mapping_clear_noreclaim (struct address_space *mapping) { }static inline int mapping_non_reclaimable (struct address_space *mapping) {return 0 ;
mm/vmscan.c +93 -0
97 lines not on this page · raw message
include/linux/swap.h +4 -0
@@ -232,12 +232,16 @@ static inline int zone_reclaim(struct zo#ifdef CONFIG_NORECLAIM_LRU extern int page_reclaimable (struct page *page,struct vm_area_struct *vma); +extern void scan_mapping_noreclaim_pages (struct address_space *);#else static inline int page_reclaimable (struct page *page,struct vm_area_struct *vma) {return 1 ; } +static inline void scan_mapping_noreclaim_pages (struct address_space *mapping) +{ +}#endif extern int kswapd_run (int nid);
include/linux/mm.h +5 -4
@@ -694,12 +694,13 @@ static inline int page_mapped(struct pagextern void show_free_areas (void );#ifdef CONFIG_SHMEM -int shmem_lock (struct file *file,int lock,struct user_struct *user); +extern struct address_space *shmem_lock (struct file *file,int lock, +struct user_struct *user);#else -static inline int shmem_lock (struct file *file,int lock, -struct user_struct *user) +static inline struct address_space *shmem_lock (struct file *file,int lock, +struct user_struct *user) { -return 0 ; +return NULL ; }#endif struct file *shmem_file_setup (char *name,loff_t size,unsigned long flags);
ipc/shm.c +18 -2
@@ -736,6 +736,11 @@ asmlinkage long sys_shmctl(int shmid, incase SHM_LOCK :case SHM_UNLOCK : { +struct address_space *mapping =NULL ; +struct file *uninitialized_var (shm_file ); + +lru_add_drain_all ();/* drain pagevecs to lru lists */ + shp =shm_lock_check (ns, shmid);if (IS_ERR (shp)) { err =PTR_ERR (shp);
@@ -763,18 +768,29 @@ asmlinkage long sys_shmctl(int shmid, inif (cmd==SHM_LOCK ) {struct user_struct * user = current->user;if (!is_file_hugepages (shp->shm_file)) { - err =shmem_lock (shp->shm_file,1 , user); + mapping =shmem_lock (shp->shm_file,1 , user); +if (IS_ERR (mapping)) + err =PTR_ERR (mapping); + mapping =NULL ;if (!err && !(shp->shm_perm.mode &SHM_LOCKED )){ shp->shm_perm.mode |=SHM_LOCKED ; shp->mlock_user = user; } } }else if (!is_file_hugepages (shp->shm_file)) { -shmem_lock (shp->shm_file,0 , shp->mlock_user); + mapping =shmem_lock (shp->shm_file,0 , shp->mlock_user); shp->shm_perm.mode &= ~SHM_LOCKED ; shp->mlock_user =NULL ; +if (mapping) { + shm_file = shp->shm_file; +get_file (shm_file);/* hold across unlock */ + } }shm_unlock (shp); +if (mapping) { +scan_mapping_noreclaim_pages (mapping); +fput (shm_file); + } goto out; }case IPC_RMID :
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 17/25 Mlocked Pages are non-reclaimable
In reply to Lee SchermerhornOriginally
From: Nick Piggin <[email protected]>
Against: 2.6.26-rc2-mm1
V8:
+ more refinement of rmap interaction, including attempt to handle mlocked pages in non-linear mappings.
+ cleanup of lockdep reported errors.
+ enhancement of munlock page table walker to detect and handle pages under migration [migration ptes].
V6:
+ Kosaki-san and Rik van Riel: added check for "page mapped in vma" to try_to_unlock() processing in try_to_unmap_anon().
+ Kosaki-san added munlock page table walker to avoid use of get_user_pages() for munlock. get_user_pages() proved to be unreliable for some types of vmas.
+ added filtering of "special" vmas. Some [_IO||_PFN] we skip altogether. Others, we just "make_pages_present" to simulate old behavior--i.e., populate page tables. Clear/don't set VM_LOCKED in non-mlockable vmas so that we don't try to unlock at exit/unmap time.
+ rework PG_mlock page flag definitions for new page flags macros.
+ Clear PageMlocked when COWing a page into a VM_LOCKED vma so we don't leave an mlocked page in another non-mlocked vma. If the other vma[s] had the page mlocked, we'll re-mlock it if/when we try to reclaim it. This is less expensive than walking the rmap in the COW/fault path.
+ in vmscan:shrink_page_list(), avoid adding anon page to the swap cache if it's in a VM_LOCKED vma, even tho' PG_mlocked might not be set. Call try_to_unlock() to determine this. As a result, we'll never try to unmap an mlocked anon page.
+ in support of the above change, updated try_to_unlock() to use same logic as try_to_unmap() when it encounters a VM_LOCKED vma--call mlock_vma_page() directly. Added stub try_to_unlock() for vmscan when NORECLAIM_MLOCK not configured.
V4 -> V5:
+ fixed problem with placement of #ifdef CONFIG_NORECLAIM_MLOCK in prep_new_page() [Thanks, minchan Kim!].
V3 -> V4:
+ Added #ifdef CONFIG_NORECLAIM_MLOCK, #endif around use of PG_mlocked in free_page_check(), et al. Not defined for 32-bit builds.
V2 -> V3:
+ rebase to 23-mm1 atop RvR's split lru series
+ fix page flags macros for *PageMlocked() when not configured.
+ ensure lru_add_drain_all() runs on all cpus when NORECLAIM_MLOCK configured. Was just for NUMA.
V1 -> V2:
+ moved this patch [and related patches] up to right after ramdisk/ramfs and SHM_LOCKed patches.
+ add [back] missing put_page() in putback_lru_page(). This solved page leakage as seen by stats in previous version.
+ fix up munlock_vma_page() to isolate page from lru before calling try_to_unlock(). Think I detected a race here.
+ use TestClearPageMlock() on old page in migrate.c's migrate_page_copy() to clean up old page.
+ live dangerously: remove TestSetPageLocked() in is_mlocked_vma()--should only be called on new pages in the fault path--iff we chose to cull there [later patch].
+ Add PG_mlocked to free_pages_check() etc to detect mlock state mismanagement.
NOTE: temporarily [???] commented out--tripping over it under load. Why?
Rework of Nick Piggins's "mm: move mlocked pages off the LRU" patch
-- part 1 of 2.
This patch:
1) defines the [CONFIG_]NORECLAIM_MLOCK sub-option and the stub version of the mlock/noreclaim APIs when it's not configured. Depends on [CONFIG_]NORECLAIM_LRU.
2) add yet another page flag--PG_mlocked--to indicate that the page is locked for efficient testing in vmscan and, optionally, fault path. This allows early culling of nonreclaimable pages, preventing them from getting to page_referenced()/try_to_unmap(). Also allows separate accounting of mlock'd pages, as Nick's original patch did.
Note: Nick's original mlock patch used a PG_mlocked flag. I had removed this in favor of the PG_noreclaim flag + an mlock_count [new page struct member]. I restored the PG_mlocked flag to eliminate the new count field.
3) add the mlock/noreclaim infrastructure to mm/mlock.c, with internal APIs in mm/internal.h. This is a rework of Nick's original patch to these files, taking into account that mlocked pages are now kept on noreclaim LRU list.
4) update vmscan.c:page_reclaimable() to check PageMlocked() and, if vma passed in, the vm_flags. Note that the vma will only be passed in for new pages in the fault path; and then only if the "cull nonreclaimable pages in fault path" patch is included.
5) add try_to_unlock() to rmap.c to walk a page's rmap and ClearPageMlocked() if no other vmas have it mlocked. Reuses as much of try_to_unmap() as possible. This effectively replaces the use of one of the lru list links as an mlock count. If this mechanism let's pages in mlocked vmas leak through w/o PG_mlocked set [I don't know that it does], we should catch them later in try_to_unmap(). One hopes this will be rare, as it will be relatively expensive.
6) Kosaki: added munlock page table walk to avoid using get_user_pages() for unlock. get_user_pages() is unreliable for some vma protections.
Lee: modified to wait for in-flight migration to complete to close munlock/migration race that could strand pages.
Original mm/internal.h, mm/rmap.c and mm/mlock.c changes:
include/linux/mm.h | 5 include/linux/page-flags.h | 16 + include/linux/rmap.h | 14 + mm/Kconfig | 14 + mm/internal.h | 70 ++++++++ mm/memory.c | 19 ++ mm/migrate.c | 2 mm/mlock.c | 386 ++++++++++++++++++++++++++++++++++++++++++--- mm/mmap.c | 1 mm/page_alloc.c | 15 + mm/rmap.c | 252 +++++++++++++++++++++++++---- mm/swap.c | 2 mm/vmscan.c | 40 +++- 13 files changed, 767 insertions(+), 69 deletions(-)
mm/Kconfig +14 -0
@@ -215,3 +215,17 @@ config NORECLAIM_LRU may be non-reclaimable because: they are locked into memory, they are anonymous pages for which no swap space exists, or they are anon pages that are expensive to unmap [long anon_vma "related vma" list.] + +config NORECLAIM_MLOCK + bool "Exclude mlock'ed pages from reclaim" + depends on NORECLAIM_LRU + help + Treats mlock'ed pages as no-reclaimable. Removing these pages from + the LRU [in]active lists avoids the overhead of attempting to reclaim + them. Pages marked non-reclaimable for this reason will become + reclaimable again when the last mlock is removed. + when no swap space exists. Removing these pages from the LRU lists + avoids the overhead of attempting to reclaim them. Pages marked + non-reclaimable for this reason will become reclaimable again when/if + sufficient swap space is added to the system. +
mm/internal.h +70 -0
82 lines not on this page · raw message
mm/mlock.c +368 -18
427 lines not on this page · raw message
mm/vmscan.c +29 -11
@@ -537,11 +537,8 @@ static unsigned long shrink_page_list(st sc->nr_scanned++; -if (unlikely (!page_reclaimable (page,NULL ))) { -if (putback_lru_page (page)) -unlock_page (page); -continue ; - } +if (unlikely (!page_reclaimable (page,NULL ))) + goto cull_mlocked;if (!sc->may_swap &&page_mapped (page)) goto keep_locked;
@@ -578,9 +575,19 @@ static unsigned long shrink_page_list(st* Anonymous process memory has backing store? * Try to allocate it some swap space here. */ -if (PageAnon (page) && !PageSwapCache (page)) +if (PageAnon (page) && !PageSwapCache (page)) { +switch (try_to_unlock (page)) { +case SWAP_FAIL :/* shouldn't happen */ +case SWAP_AGAIN : + goto keep_locked; +case SWAP_MLOCK : + goto cull_mlocked; +case SWAP_SUCCESS : + ;/* fall thru'; add to swap cache */ + }if (!add_to_swap (page,GFP_ATOMIC )) goto activate_locked; + }#endif /* CONFIG_SWAP */ mapping =page_mapping (page);
@@ -595,6 +602,8 @@ static unsigned long shrink_page_list(st goto activate_locked;case SWAP_AGAIN : goto keep_locked; +case SWAP_MLOCK : + goto cull_mlocked;case SWAP_SUCCESS : ;/* try to free the page below */ }
@@ -667,6 +676,11 @@ free_it:__pagevec_release_nonlru (&freed_pvec);continue ; +cull_mlocked: +if (putback_lru_page (page)) +unlock_page (page); +continue ; + activate_locked:/* Not a candidate for swapping, so reclaim swap space. */ if (PageSwapCache (page) &&vm_swap_full ())
@@ -678,7 +692,7 @@ keep_locked:unlock_page (page); keep:list_add (&page->lru, &ret_pages); -VM_BUG_ON (PageLRU (page)); +VM_BUG_ON (PageLRU (page) ||PageNoreclaim (page)); }list_splice (&ret_pages, page_list);if (pagevec_count (&freed_pvec))
@@ -2305,12 +2319,13 @@ int zone_reclaim(struct zone *zone, gfp_* @vma: the VMA in which the page is or will be mapped, may be NULL * * Test whether page is reclaimable--i.e., should be placed on active/inactive -* lists vs noreclaim list. +* lists vs noreclaim list. The vma argument is !NULL when called from the +* fault path to determine how to instantate a new page. * * Reasons page might not be reclaimable: * (1) page's mapping marked non-reclaimable +* (2) page is part of an mlocked VMA * -* TODO - later patches */ int page_reclaimable (struct page *page,struct vm_area_struct *vma) {
@@ -2320,13 +2335,16 @@ int page_reclaimable(struct page *page,if (mapping_non_reclaimable (page_mapping (page)))return 0 ; -/* TODO: test page [!]reclaimable conditions */ +#ifdef CONFIG_NORECLAIM_MLOCK +if (PageMlocked (page) || (vma &&is_mlocked_vma (vma, page))) +return 0 ; +#endif return 1 ; }/** -* check_move_noreclaim_page - check page for reclaimability and move to appropriate zone lru list +* check_move_noreclaim_page - check page for reclaimability and move to appropriate lru list * @page: page to check reclaimability and move to appropriate lru list * @zone: zone page is in *
include/linux/page-flags.h +16 -0
@@ -96,6 +96,9 @@ enum pageflags { PG_swapbacked, /* Page is backed by RAM/swap */#ifdef CONFIG_NORECLAIM_LRU PG_noreclaim, /* Page is "non-reclaimable" */ +#ifdef CONFIG_NORECLAIM_MLOCK + PG_mlocked, /* Page is vma mlocked */ +#endif #endif #ifdef CONFIG_IA64_UNCACHED_ALLOCATOR PG_uncached, /* Page has been mapped as uncached */
@@ -210,12 +213,25 @@ PAGEFLAG_FALSE(SwapCache)#ifdef CONFIG_NORECLAIM_LRU PAGEFLAG(Noreclaim, noreclaim) __CLEARPAGEFLAG(Noreclaim, noreclaim) TESTCLEARFLAG(Noreclaim, noreclaim) + +#ifdef CONFIG_NORECLAIM_MLOCK +#define MLOCK_PAGES 1 +PAGEFLAG(Mlocked, mlocked) __CLEARPAGEFLAG(Mlocked, mlocked) + TESTSCFLAG(Mlocked, mlocked) +#endif +#else PAGEFLAG_FALSE(Noreclaim) TESTCLEARFLAG_FALSE(Noreclaim) SETPAGEFLAG_NOOP(Noreclaim) CLEARPAGEFLAG_NOOP(Noreclaim) __CLEARPAGEFLAG_NOOP(Noreclaim) #endif +#if !defined(CONFIG_NORECLAIM_MLOCK) +#define MLOCK_PAGES 0 +PAGEFLAG_FALSE(Mlocked) + SETPAGEFLAG_NOOP(Mlocked) TESTCLEARFLAG_FALSE(Mlocked) +#endif +#ifdef CONFIG_IA64_UNCACHED_ALLOCATOR PAGEFLAG (Uncached, uncached) #else
include/linux/rmap.h +14 -0
@@ -97,6 +97,19 @@ unsigned long page_address_in_vma(struct*/ int page_mkclean (struct page *); +#ifdef CONFIG_NORECLAIM_MLOCK +/* +* called in munlock()/munmap() path to check for other vmas holding +* the page mlocked. +*/ +int try_to_unlock (struct page *); +#else +static inline int try_to_unlock (struct page *page) +{ +return 0 ;/* a.k.a. SWAP_SUCCESS */ +} +#endif +#else /* !CONFIG_MMU */ #define anon_vma_init () do {} while (0)
@@ -120,5 +133,6 @@ static inline int page_mkclean(struct pa#define SWAP_SUCCESS 0#define SWAP_AGAIN 1#define SWAP_FAIL 2 +#define SWAP_MLOCK 3#endif /* _LINUX_RMAP_H */
mm/rmap.c +216 -36
406 lines not on this page · raw message
mm/migrate.c +2 -0
@@ -359,6 +359,8 @@ static void migrate_page_copy(struct pag__set_page_dirty_nobuffers (newpage); } +mlock_migrate_page (newpage, page); +#ifdef CONFIG_SWAP ClearPageSwapCache (page);#endif
mm/page_alloc.c +14 -1
@@ -258,6 +258,9 @@ static void bad_page(struct page *page) 1 << PG_active |#ifdef CONFIG_NORECLAIM_LRU 1 << PG_noreclaim | +#ifdef CONFIG_NORECLAIM_MLOCK + 1 << PG_mlocked | +#endif #endif 1 << PG_dirty | 1 << PG_reclaim |
@@ -497,6 +500,9 @@ static inline int free_pages_check(struc#ifdef CONFIG_NORECLAIM_LRU 1 << PG_noreclaim |#endif +#ifdef CONFIG_NORECLAIM_MLOCK + 1 << PG_mlocked | +#endif 1 << PG_buddy ))))bad_page (page);if (PageDirty (page))
@@ -650,6 +656,9 @@ static int prep_new_page(struct page *pa 1 << PG_active |#ifdef CONFIG_NORECLAIM_LRU 1 << PG_noreclaim | +#ifdef CONFIG_NORECLAIM_MLOCK + 1 << PG_mlocked | +#endif #endif 1 << PG_dirty | 1 << PG_slab |
@@ -669,7 +678,11 @@ static int prep_new_page(struct page *pa page->flags &= ~(1 << PG_uptodate |1 << PG_error |1 << PG_reclaim |1 << PG_referenced |1 << PG_arch_1 | -1 << PG_owner_priv_1 |1 << PG_mappedtodisk); +1 << PG_owner_priv_1 |1 << PG_mappedtodisk +#ifdef CONFIG_NORECLAIM_MLOCK + |1 << PG_mlocked +#endif + );set_page_private (page,0 );set_page_refcounted (page);
mm/swap.c +1 -1
@@ -307,7 +307,7 @@ void lru_add_drain(void)put_cpu (); } -#ifdef CONFIG_NUMA +#if defined(CONFIG_NUMA ) || defined(CONFIG_NORECLAIM_MLOCK )static void lru_add_drain_per_cpu (struct work_struct *dummy) {lru_add_drain ();
mm/memory.c +18 -1
@@ -61,6 +61,8 @@#include <linux/swapops.h> #include <linux/elf.h> +#include "internal.h" +#ifndef CONFIG_NEED_MULTIPLE_NODES /* use the per-pgdat data instead for discontigmem - mbligh */ unsigned long max_mapnr;
@@ -1734,6 +1736,15 @@ gotten: new_page =alloc_page_vma (GFP_HIGHUSER_MOVABLE , vma, address);if (!new_page) goto oom; +/* +* Don't let another task, with possibly unlocked vma, +* keep the mlocked page. +*/ +if (vma->vm_flags &VM_LOCKED ) { +lock_page (old_page);/* for LRU manipulation */ +clear_page_mlock (old_page); +unlock_page (old_page); + }cow_user_page (new_page, old_page, address, vma);__SetPageUptodate (new_page);
@@ -2176,7 +2187,7 @@ static int do_swap_page(struct mm_structpage_add_anon_rmap (page, vma, address);swap_free (entry); -if (vm_swap_full ()) +if (vm_swap_full () || (vma->vm_flags &VM_LOCKED ) ||PageMlocked (page))remove_exclusive_swap_page (page);unlock_page (page);
@@ -2316,6 +2327,12 @@ static int __do_fault(struct mm_struct * ret =VM_FAULT_OOM ; goto out; } +/* +* Don't let another task, with possibly unlocked vma, +* keep the mlocked page. +*/ +if (vma->vm_flags &VM_LOCKED ) +clear_page_mlock (vmf.page);copy_user_highpage (page, vmf.page, address, vma);__SetPageUptodate (page); }else {
mm/mmap.c +0 -1
@@ -652,7 +652,6 @@ again: remove_next = 1 + (end > next->* If the vma has a ->close operation then the driver probably needs to release * per-vma resources, so we don't attempt to merge those. */ -#define VM_SPECIAL (VM_IO | VM_DONTEXPAND | VM_RESERVED | VM_PFNMAP)static inline int is_mergeable_vma (struct vm_area_struct *vma,struct file *file,unsigned long vm_flags)
include/linux/mm.h +5 -0
@@ -126,6 +126,11 @@ extern unsigned int kobjsize(const void#define VM_RandomReadHint (v) ((v)->vm_flags & VM_RAND_READ)/* +* special vmas that are non-mergable, non-mlock()able +*/ +#define VM_SPECIAL (VM_IO | VM_DONTEXPAND | VM_RESERVED | VM_PFNMAP) + +/* * mapping from the currently active vm_flags protection bits (the * low four bits) to a page protection mask.. */
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 18/25 Downgrade mmap sem while populating mlocked regions
In reply to Lee SchermerhornAgainst: 2.6.26-rc2-mm1
V2 -> V3:
+ rebase to 23-mm1 atop RvR's split lru series [no change]
+ fix function return types [void -> int] to fix build when not configured.
New in V2.
We need to hold the mmap_sem for write to initiatate mlock()/munlock() because we may need to merge/split vmas. However, this can lead to very long lock hold times attempting to fault in a large memory region
to mlock it into memory. This can hold off other faults against the
mm [multithreaded tasks] and other scans of the mm, such as via /proc. To alleviate this, downgrade the mmap_sem to read mode during the population of the region for locking. This is especially the case if we need to reclaim memory to lock down the region. We [probably?] don't need to do this for unlocking as all of the pages should be resident--they're already mlocked.
Now, the caller's of the mlock functions [mlock_fixup() and mlock_vma_pages_range()] expect the mmap_sem to be returned in write mode. Changing all callers appears to be way too much effort at this point. So, restore write mode before returning. Note that this opens a window where the mmap list could change in a multithreaded process. So, at least for mlock_fixup(), where we could be called in a loop over multiple vmas, we check that a vma still exists at the start address and that vma still covers the page range [start,end). If not, we return an error, -EAGAIN, and let the caller deal with it.
Return -EAGAIN from mlock_vma_pages_range() function and mlock_fixup() if the vma at 'start' disappears or changes so that the page range [start,end) is no longer contained in the vma. Again, let the caller deal with it. Looks like only sys_remap_file_pages() [via mmap_region()] should actually care.
With this patch, I no longer see processes like ps(1) blocked for seconds or minutes at a time waiting for a large [multiple gigabyte] region to be locked down. However, I occassionally see delays while unlocking or unmapping a large mlocked region. Should we also downgrade the mmap_sem for the unlock path?
mm/mlock.c | 43 +++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 41 insertions(+), 2 deletions(-)
mm/mlock.c +41 -2
@@ -309,6 +309,7 @@ static void __munlock_vma_pages_range(stint mlock_vma_pages_range (struct vm_area_struct *vma,unsigned long start,unsigned long end) { +struct mm_struct *mm = vma->vm_mm;int nr_pages = (end - start) /PAGE_SIZE ;BUG_ON (!(vma->vm_flags &VM_LOCKED ));
@@ -323,7 +324,17 @@ int mlock_vma_pages_range(struct vm_area vma == get_gate_vma(current)) goto make_present; -return __mlock_vma_pages_range (vma, start, end); +downgrade_write (&mm->mmap_sem); + nr_pages =__mlock_vma_pages_range (vma, start, end); + +up_read (&mm->mmap_sem); +/* vma can change or disappear */ +down_write (&mm->mmap_sem); + vma =find_vma (mm, start); +/* non-NULL vma must contain @start, but need to check @end */ +if (!vma || end > vma->vm_end) +return -EAGAIN ; +return nr_pages; make_present:/*
@@ -418,13 +429,41 @@ success: vma->vm_flags = newflags;if (lock) { +/* +* mmap_sem is currently held for write. Downgrade the write +* lock to a read lock so that other faults, mmap scans, ... +* while we fault in all pages. +*/ +downgrade_write (&mm->mmap_sem); + ret =__mlock_vma_pages_range (vma, start, end);if (ret >0 ) { mm->locked_vm -= ret; ret =0 ; } - }else +/* +* Need to reacquire mmap sem in write mode, as our callers +* expect this. We have no support for atomically upgrading +* a sem to write, so we need to check for ranges while sem +* is unlocked. +*/ +up_read (&mm->mmap_sem); +/* vma can change or disappear */ +down_write (&mm->mmap_sem); + *prev =find_vma (mm, start); +/* non-NULL *prev must contain @start, but need to check @end */ +if (!(*prev) || end > (*prev)->vm_end) + ret = -EAGAIN ; + }else { +/* +* TODO: for unlocking, pages will already be resident, so +* we don't need to wait for allocations/reclaim/pagein, ... +* However, unlocking a very large region can still take a +* while. Should we downgrade the semaphore for both lock +* AND unlock ? +*/ __munlock_vma_pages_range (vma, start, end); + } out: *prev = vma;
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 19/25 Handle mlocked pages during map, remap, unmap
In reply to Lee SchermerhornOriginally
From: Nick Piggin <[email protected]>
Against: 2.6.26-rc2-mm1
V6:
+ munlock page in range of VM_LOCKED vma being covered by remap_file_pages(), as this is an implied unmap of the range.
+ in support of special vma filtering, don't account for non-mlockable vmas as locked_vm.
V2 -> V3:
+ rebase to 23-mm1 atop RvR's split lru series [no changes]
V1 -> V2:
+ modified mmap.c:mmap_region() to return error if mlock_vma_pages_range() does. This can only occur if the vma gets removed/changed while
we're switching mmap_sem lock modes. Most callers don't care, but
sys_remap_file_pages() appears to.
Rework of Nick Piggins's "mm: move mlocked pages off the LRU" patch -- part 2 0f 2.
Remove mlocked pages from the LRU using "NoReclaim infrastructure" during mmap(), munmap(), mremap() and truncate(). Try to move back to normal LRU lists on munmap() when last mlocked mapping removed. Removed PageMlocked() status when page truncated from file.
Originally Signed-off-by: Nick Piggin <[email protected]>
mm/fremap.c | 26 +++++++++++++++++--- mm/internal.h | 13 ++++++++-- mm/mlock.c | 10 ++++--- mm/mmap.c | 75 ++++++++++++++++++++++++++++++++++++++++++++-------------- mm/mremap.c | 8 +++--- mm/truncate.c | 4 +++ 6 files changed, 106 insertions(+), 30 deletions(-)
mm/mmap.c +58 -17
@@ -32,6 +32,8 @@#include <asm/tlb.h> #include <asm/mmu_context.h> +#include "internal.h" +#ifndef arch_mmap_check#define arch_mmap_check (addr, len, flags) (0)#endif
@@ -961,6 +963,7 @@ unsigned long do_mmap_pgoff(struct filereturn -EPERM ; vm_flags |=VM_LOCKED ; } +/* mlock MCL_FUTURE? */ if (vm_flags &VM_LOCKED ) {unsigned long locked, lock_limit;
@@ -1121,10 +1124,12 @@ munmap_back:* The VM_SHARED test is necessary because shmem_zero_setup * will create the file object for a shared anonymous map below. */ -if (!file && !(vm_flags &VM_SHARED ) && -vma_merge (mm, prev, addr, addr + len, vm_flags, -NULL ,NULL , pgoff,NULL )) - goto out; +if (!file && !(vm_flags &VM_SHARED )) { + vma =vma_merge (mm, prev, addr, addr + len, vm_flags, +NULL ,NULL , pgoff,NULL ); +if (vma) + goto out; + }/* * Determine the object being mapped and call the appropriate
@@ -1206,10 +1211,14 @@ out: mm->total_vm += len >>PAGE_SHIFT ;vm_stat_account (mm, vm_flags, file, len >>PAGE_SHIFT );if (vm_flags &VM_LOCKED ) { - mm->locked_vm += len >>PAGE_SHIFT ; -make_pages_present (addr, addr + len); - } -if ((flags &MAP_POPULATE ) && !(flags &MAP_NONBLOCK )) +/* +* makes pages present; downgrades, drops, reacquires mmap_sem +*/ +int nr_pages =mlock_vma_pages_range (vma, addr, addr + len); +if (nr_pages <0 ) +return nr_pages;/* vma gone! */ + mm->locked_vm += (len >>PAGE_SHIFT ) - nr_pages; + }else if ((flags &MAP_POPULATE ) && !(flags &MAP_NONBLOCK ))make_pages_present (addr, addr + len);return addr;
@@ -1682,8 +1691,11 @@ find_extend_vma(struct mm_struct *mm, unreturn vma;if (!prev ||expand_stack (prev, addr))return NULL ; -if (prev->vm_flags &VM_LOCKED ) -make_pages_present (addr, prev->vm_end); +if (prev->vm_flags &VM_LOCKED ) { +int nr_pages =mlock_vma_pages_range (prev, addr, prev->vm_end); +if (nr_pages <0 ) +return NULL ;/* vma gone! */ + } return prev; }#else
@@ -1709,8 +1721,11 @@ find_extend_vma(struct mm_struct * mm, u start = vma->vm_start;if (expand_stack (vma, addr))return NULL ; -if (vma->vm_flags &VM_LOCKED ) -make_pages_present (addr, start); +if (vma->vm_flags &VM_LOCKED ) { +int nr_pages =mlock_vma_pages_range (vma, addr, start); +if (nr_pages <0 ) +return NULL ;/* vma gone! */ + } return vma; }#endif
@@ -1895,6 +1910,18 @@ int do_munmap(struct mm_struct *mm, unsi vma = prev? prev->vm_next: mm->mmap;/* +* unlock any mlock()ed ranges before detaching vmas +*/ +if (mm->locked_vm) { +struct vm_area_struct *tmp = vma; +while (tmp && tmp->vm_start < end) { +if (tmp->vm_flags &VM_LOCKED ) +munlock_vma_pages_all (tmp); + tmp = tmp->vm_next; + } + } + +/* * Remove the vma's, and unmap the actual pages */ detach_vmas_to_be_unmapped (mm, vma, prev, end);
@@ -2006,8 +2033,9 @@ unsigned long do_brk(unsigned long addr,return -ENOMEM ;/* Can we just expand an old private anonymous mapping? */ -if (vma_merge (mm, prev, addr, addr + len, flags, -NULL ,NULL , pgoff,NULL )) + vma =vma_merge (mm, prev, addr, addr + len, flags, +NULL ,NULL , pgoff,NULL ); +if (vma) goto out;/*
@@ -2029,8 +2057,9 @@ unsigned long do_brk(unsigned long addr, out: mm->total_vm += len >>PAGE_SHIFT ;if (flags &VM_LOCKED ) { - mm->locked_vm += len >>PAGE_SHIFT ; -make_pages_present (addr, addr + len); +int nr_pages =mlock_vma_pages_range (vma, addr, addr + len); +if (nr_pages >=0 ) + mm->locked_vm += (len >>PAGE_SHIFT ) - nr_pages; }return addr; }
@@ -2041,13 +2070,25 @@ EXPORT_SYMBOL(do_brk);void exit_mmap (struct mm_struct *mm) {struct mmu_gather *tlb; -struct vm_area_struct *vma = mm->mmap; +struct vm_area_struct *vma;unsigned long nr_accounted =0 ;unsigned long end;/* mm's last user has gone, and its about to be pulled down */ arch_exit_mmap (mm); +if (mm->locked_vm) { + vma = mm->mmap; +while (vma) { +if (vma->vm_flags &VM_LOCKED ) +munlock_vma_pages_all (vma); + vma = vma->vm_next; + } + } + + vma = mm->mmap; + +lru_add_drain ();flush_cache_mm (mm); tlb =tlb_gather_mmu (mm,1 );
mm/mremap.c +5 -3
@@ -23,6 +23,8 @@#include <asm/cacheflush.h> #include <asm/tlbflush.h> +#include "internal.h" +static pmd_t *get_old_pmd (struct mm_struct *mm,unsigned long addr) {pgd_t *pgd;
@@ -232,8 +234,8 @@ static unsigned long move_vma(struct vm_if (vm_flags &VM_LOCKED ) { mm->locked_vm += new_len >>PAGE_SHIFT ;if (new_len > old_len) -make_pages_present (new_addr + old_len, - new_addr + new_len); +mlock_vma_pages_range (new_vma, new_addr + old_len, + new_addr + new_len); }return new_addr;
@@ -373,7 +375,7 @@ unsigned long do_mremap(unsigned long advm_stat_account (mm, vma->vm_flags, vma->vm_file, pages);if (vma->vm_flags &VM_LOCKED ) { mm->locked_vm += pages; -make_pages_present (addr + old_len, +mlock_vma_pages_range (vma, addr + old_len, addr + new_len); } ret = addr;
mm/truncate.c +4 -0
@@ -18,6 +18,7 @@#include <linux/task_io_accounting_ops.h> #include <linux/buffer_head.h> /* grr. try_to_release_page, do_invalidatepage */ +#include "internal.h" /**
@@ -104,6 +105,7 @@ truncate_complete_page(struct address_spcancel_dirty_page (page,PAGE_CACHE_SIZE );remove_from_page_cache (page); +clear_page_mlock (page);ClearPageUptodate (page);ClearPageMappedToDisk (page);page_cache_release (page);/* pagecache ref */
@@ -128,6 +130,7 @@ invalidate_complete_page(struct address_if (PagePrivate (page) && !try_to_release_page (page,0 ))return 0 ; +clear_page_mlock (page); ret =remove_mapping (mapping, page);return ret;
@@ -353,6 +356,7 @@ invalidate_complete_page2(struct addressif (PageDirty (page)) goto failed; +clear_page_mlock (page);BUG_ON (PagePrivate (page));__remove_from_page_cache (page);write_unlock_irq (&mapping->tree_lock);
mm/mlock.c +6 -4
@@ -270,7 +270,8 @@ static void __munlock_vma_pages_range(ststruct munlock_page_walk mpw;VM_BUG_ON (start & ~PAGE_MASK || end & ~PAGE_MASK ); -VM_BUG_ON (!rwsem_is_locked (&vma->vm_mm->mmap_sem)); +VM_BUG_ON ((!rwsem_is_locked (&vma->vm_mm->mmap_sem)) && + (atomic_read (&mm->mm_users) !=0 ));VM_BUG_ON (start < vma->vm_start);VM_BUG_ON (end > vma->vm_end);
@@ -354,12 +355,13 @@ no_mlock:/* -* munlock all pages in vma. For munmap() and exit(). +* munlock all pages in the vma range. For mremap(), munmap() and exit(). */ -void munlock_vma_pages_all (struct vm_area_struct *vma) +void munlock_vma_pages_range (struct vm_area_struct *vma, +unsigned long start,unsigned long end) { vma->vm_flags &= ~VM_LOCKED ; -__munlock_vma_pages_range (vma, vma->vm_start, vma->vm_end); +__munlock_vma_pages_range (vma, start, end); }/*
mm/fremap.c +22 -4
@@ -20,6 +20,8 @@#include <asm/cacheflush.h> #include <asm/tlbflush.h> +#include "internal.h" +static void zap_pte (struct mm_struct *mm,struct vm_area_struct *vma,unsigned long addr,pte_t *ptep) {
@@ -214,13 +216,29 @@ asmlinkage long sys_remap_file_pages(unsspin_unlock (&mapping->i_mmap_lock); } +if (vma->vm_flags &VM_LOCKED ) { +/* +* drop PG_Mlocked flag for over-mapped range +*/ +unsigned int saved_flags = vma->vm_flags; +munlock_vma_pages_range (vma, start, start + size); + vma->vm_flags = saved_flags; + } + err =populate_range (mm, vma, start, size, pgoff);if (!err && !(flags &MAP_NONBLOCK )) { -if (unlikely (has_write_lock)) { -downgrade_write (&mm->mmap_sem); - has_write_lock =0 ; +if (vma->vm_flags &VM_LOCKED ) { +/* +* might be mapping previously unmapped range of file +*/ +mlock_vma_pages_range (vma, start, start + size); + }else { +if (unlikely (has_write_lock)) { +downgrade_write (&mm->mmap_sem); + has_write_lock =0 ; + } +make_pages_present (start, start+size); } -make_pages_present (start, start+size); }/*
mm/internal.h +11 -2
@@ -63,9 +63,18 @@ extern int mlock_vma_pages_range(struct unsigned long start, unsigned long end);/* -* munlock all pages in vma. For munmap() and exit(). +* munlock all pages in vma range. For mremap(). */ -extern void munlock_vma_pages_all (struct vm_area_struct *vma); +extern void munlock_vma_pages_range (struct vm_area_struct *vma, +unsigned long start,unsigned long end); + +/* +* munlock all pages in vma. For munmap and exit(). +*/ +static inline void munlock_vma_pages_all (struct vm_area_struct *vma) +{ +munlock_vma_pages_range (vma, vma->vm_start, vma->vm_end); +}#ifdef CONFIG_NORECLAIM_LRU /*
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 20/25 Mlocked Pages statistics
In reply to Lee SchermerhornAgainst: 2.6.26-rc2-mm1
V2 -> V3:
+ rebase to 23-mm1 atop RvR's split lru series
+ fix definitions of NR_MLOCK to fix build errors when not configured.
V1 -> V2:
+ new in V2 -- pulled in & reworked from Nick's previous series
Add NR_MLOCK zone page state, which provides a (conservative) count of mlocked pages (actually, the number of mlocked pages moved off the LRU).
Reworked by lts to fit in with the modified mlock page support in the Reclaim Scalability series.
drivers/base/node.c | 24 +++++++++++++++--------- fs/proc/proc_misc.c | 6 ++++++ include/linux/mmzone.h | 5 +++++ mm/internal.h | 14 +++++++++++--- mm/mlock.c | 22 ++++++++++++++++++---- mm/vmstat.c | 3 +++ 6 files changed, 58 insertions(+), 16 deletions(-)
drivers/base/node.c +15 -9
@@ -69,6 +69,9 @@ static ssize_t node_read_meminfo(struct "Node %d Inactive(file): %8lu kB\n" #ifdef CONFIG_NORECLAIM_LRU "Node %d Noreclaim: %8lu kB\n" +#ifdef CONFIG_NORECLAIM_MLOCK + "Node %d Mlocked: %8lu kB\n" +#endif #endif#ifdef CONFIG_HIGHMEM "Node %d HighTotal: %8lu kB\n"
@@ -91,16 +94,19 @@ static ssize_t node_read_meminfo(struct nid,K (i.totalram), nid,K (i.freeram), nid,K (i.totalram - i.freeram), - nid,node_page_state (nid,NR_ACTIVE_ANON ) + -node_page_state (nid,NR_ACTIVE_FILE ), - nid,node_page_state (nid,NR_INACTIVE_ANON ) + -node_page_state (nid,NR_INACTIVE_FILE ), - nid,node_page_state (nid,NR_ACTIVE_ANON ), - nid,node_page_state (nid,NR_INACTIVE_ANON ), - nid,node_page_state (nid,NR_ACTIVE_FILE ), - nid,node_page_state (nid,NR_INACTIVE_FILE ), + nid,K (node_page_state (nid,NR_ACTIVE_ANON ) + +node_page_state (nid,NR_ACTIVE_FILE )), + nid,K (node_page_state (nid,NR_INACTIVE_ANON ) + +node_page_state (nid,NR_INACTIVE_FILE )), + nid,K (node_page_state (nid,NR_ACTIVE_ANON )), + nid,K (node_page_state (nid,NR_INACTIVE_ANON )), + nid,K (node_page_state (nid,NR_ACTIVE_FILE )), + nid,K (node_page_state (nid,NR_INACTIVE_FILE )), #ifdef CONFIG_NORECLAIM_LRU - nid,node_page_state (nid,NR_NORECLAIM ), + nid,K (node_page_state (nid,NR_NORECLAIM )), +#ifdef CONFIG_NORECLAIM_MLOCK + nid,K (node_page_state (nid,NR_MLOCK )), +#endif #endif #ifdef CONFIG_HIGHMEM nid, K(i.totalhigh),
fs/proc/proc_misc.c +6 -0
@@ -176,6 +176,9 @@ static int meminfo_read_proc(char *page, "Inactive(file): %8lu kB\n" #ifdef CONFIG_NORECLAIM_LRU "Noreclaim: %8lu kB\n" +#ifdef CONFIG_NORECLAIM_MLOCK + "Mlocked: %8lu kB\n" +#endif #endif#ifdef CONFIG_HIGHMEM "HighTotal: %8lu kB\n"
@@ -214,6 +217,9 @@ static int meminfo_read_proc(char *page, K(inactive_file),#ifdef CONFIG_NORECLAIM_LRU K(global_page_state(NR_NORECLAIM)), +#ifdef CONFIG_NORECLAIM_MLOCK + K(global_page_state(NR_MLOCK)), +#endif #endif #ifdef CONFIG_HIGHMEM K(i.totalhigh),
include/linux/mmzone.h +5 -0
@@ -90,6 +90,11 @@ enum zone_stat_item {#else NR_NORECLAIM = NR_ACTIVE_FILE, /* avoid compiler errors in dead code */#endif +#ifdef CONFIG_NORECLAIM_MLOCK +NR_MLOCK ,/* mlock()ed pages found and moved off LRU */ +#else +NR_MLOCK =NR_ACTIVE_FILE ,/* avoid compier errors... */ +#endifNR_ANON_PAGES ,/* Mapped anonymous pages */ NR_FILE_MAPPED, /* pagecache pages mapped into pagetables.only modified from process context */
mm/mlock.c +18 -4
@@ -56,6 +56,7 @@ void __clear_page_mlock(struct page *pag {VM_BUG_ON (!PageLocked (page));/* for LRU islolate/putback */ +dec_zone_page_state (page,NR_MLOCK );if (!isolate_lru_page (page)) {putback_lru_page (page); }else {
@@ -76,8 +77,11 @@ void mlock_vma_page(struct page *page) {BUG_ON (!PageLocked (page)); -if (!TestSetPageMlocked (page) && !isolate_lru_page (page)) +if (!TestSetPageMlocked (page)) { +inc_zone_page_state (page,NR_MLOCK ); +if (!isolate_lru_page (page))putback_lru_page (page); + } }/*
@@ -102,9 +106,19 @@ static void munlock_vma_page(struct page {BUG_ON (!PageLocked (page)); -if (TestClearPageMlocked (page) && !isolate_lru_page (page)) { -try_to_unlock (page); -putback_lru_page (page); +if (TestClearPageMlocked (page)) { +dec_zone_page_state (page,NR_MLOCK ); +if (!isolate_lru_page (page)) { +try_to_unlock (page);/* maybe relock the page */ +putback_lru_page (page); + } +/* +* Else we lost the race. let try_to_unmap() deal with it. +* At least we get the page state and mlock stats right. +* However, page is still on the noreclaim list. We'll fix +* that up when the page is eventually freed or we scan the +* noreclaim list. +*/ } }
mm/vmstat.c +3 -0
@@ -702,6 +702,9 @@ static const char * const vmstat_text[]#ifdef CONFIG_NORECLAIM_LRU "nr_noreclaim",#endif +#ifdef CONFIG_NORECLAIM_MLOCK +"nr_mlock" , +#endif"nr_anon_pages" ,"nr_mapped" , "nr_file_pages",
mm/internal.h +11 -3
@@ -107,7 +107,8 @@ static inline int is_mlocked_vma(structif (likely ((vma->vm_flags & (VM_LOCKED |VM_SPECIAL )) !=VM_LOCKED ))return 0 ; -SetPageMlocked (page); +if (!TestSetPageMlocked (page)) +inc_zone_page_state (page,NR_MLOCK );return 1 ; }
@@ -134,12 +135,19 @@ static inline void clear_page_mlock(stru/* * mlock_migrate_page - called only from migrate_page_copy() to -* migrate the Mlocked page flag +* migrate the Mlocked page flag; update statistics. */ static inline void mlock_migrate_page (struct page *newpage,struct page *page) { -if (TestClearPageMlocked (page)) +if (TestClearPageMlocked (page)) { +unsigned long flags; + +local_irq_save (flags); +__dec_zone_page_state (page,NR_MLOCK );SetPageMlocked (newpage); +__inc_zone_page_state (newpage,NR_MLOCK ); +local_irq_restore (flags); + } }
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 21/25 Cull non-reclaimable pages in fault path
In reply to Lee SchermerhornAgainst: 2.6.26-rc2-mm1
V2 -> V3:
+ rebase to 23-mm1 atop RvR's split lru series.
V1 -> V2:
+ no changes
"Optional" part of "noreclaim infrastructure"
In the fault paths that install new anonymous pages, check whether the page is reclaimable or not using lru_cache_add_active_or_noreclaim(). If the page is reclaimable, just add it to the active lru list [via the pagevec cache], else add it to the noreclaim list.
This "proactive" culling in the fault path mimics the handling of mlocked pages in Nick Piggin's series to keep mlocked pages off the lru lists.
Notes:
1) This patch is optional--e.g., if one is concerned about the additional test in the fault path. We can defer the moving of nonreclaimable pages until when vmscan [shrink_*_list()] encounters them. Vmscan will only need to handle such pages once.
2) The 'vma' argument to page_reclaimable() is require to notice that we're faulting a page into an mlock()ed vma w/o having to scan the
page's rmap in the fault path. Culling mlock()ed anon pages is
currently the only reason for this patch.
3) We can't cull swap pages in read_swap_cache_async() because the vma argument doesn't necessarily correspond to the swap cache offset passed in by swapin_readahead(). This could [did!] result in mlocking pages in non-VM_LOCKED vmas if [when] we tried to cull in this path.
4) Move set_pte_at() to after where we add page to lru to keep it hidden from other tasks that might walk the page table. We already do it in this order in do_anonymous() page. And, these are COW'd anon pages. Is this safe?
include/linux/swap.h | 2 ++ mm/memory.c | 20 ++++++++++++-------- mm/swap.c | 21 +++++++++++++++++++++ 3 files changed, 35 insertions(+), 8 deletions(-)
mm/memory.c +12 -8
@@ -1774,12 +1774,15 @@ gotten:* thread doing COW. */ ptep_clear_flush (vma, address, page_table); -set_pte_at (mm, address, page_table, entry); -update_mmu_cache (vma, address, entry); +SetPageSwapBacked (new_page); -lru_cache_add_active_anon (new_page); +lru_cache_add_active_or_noreclaim (new_page, vma);page_add_new_anon_rmap (new_page, vma, address); +//TODO: is this safe? do_anonymous_page() does it this way. +set_pte_at (mm, address, page_table, entry); +update_mmu_cache (vma, address, entry); +/* Free the old page.. */ new_page = old_page; ret |=VM_FAULT_WRITE ;
@@ -2246,7 +2249,7 @@ static int do_anonymous_page(struct mm_s goto release;inc_mm_counter (mm, anon_rss);SetPageSwapBacked (page); -lru_cache_add_active_anon (page); +lru_cache_add_active_or_noreclaim (page, vma);page_add_new_anon_rmap (page, vma, address);set_pte_at (mm, address, page_table, entry);
@@ -2390,12 +2393,11 @@ static int __do_fault(struct mm_struct * entry =mk_pte (page, vma->vm_page_prot);if (flags &FAULT_FLAG_WRITE ) entry =maybe_mkwrite (pte_mkdirty (entry), vma); -set_pte_at (mm, address, page_table, entry);if (anon) { -inc_mm_counter (mm, anon_rss); +inc_mm_counter (mm, anon_rss);SetPageSwapBacked (page); -lru_cache_add_active_anon (page); -page_add_new_anon_rmap (page, vma, address); +lru_cache_add_active_or_noreclaim (page, vma); +page_add_new_anon_rmap (page, vma, address); }else {inc_mm_counter (mm, file_rss);page_add_file_rmap (page);
@@ -2404,6 +2406,8 @@ static int __do_fault(struct mm_struct *get_page (dirty_page); } } +//TODO: is this safe? do_anonymous_page() does it this way. +set_pte_at (mm, address, page_table, entry);/* no need to invalidate: a not-present page won't be cached */ update_mmu_cache (vma, address, entry);
include/linux/swap.h +2 -0
@@ -173,6 +173,8 @@ extern unsigned int nr_free_pagecache_pa/* linux/mm/swap.c */ extern void __lru_cache_add (struct page *,enum lru_list lru);extern void lru_cache_add_lru (struct page *,enum lru_list lru); +extern void lru_cache_add_active_or_noreclaim (struct page *, +struct vm_area_struct *);extern void activate_page (struct page *);extern void mark_page_accessed (struct page *);extern void lru_add_drain (void );
mm/swap.c +21 -0
@@ -31,6 +31,8 @@#include <linux/backing-dev.h> #include <linux/memcontrol.h> +#include "internal.h" +/* How many pages do we try to swap or page in/out together? */ int page_cluster;
@@ -273,6 +275,25 @@ void add_page_to_noreclaim_list(struct pspin_unlock_irq (&zone->lru_lock); } +/** +* lru_cache_add_active_or_noreclaim +* @page: the page to be added to LRU +* @vma: vma in which page is mapped for determining reclaimability +* +* place @page on active or noreclaim LRU list, depending on +* page_reclaimable(). Note that if the page is not reclaimable, +* it goes directly back onto it's zone's noreclaim list. It does +* NOT use a per cpu pagevec. +*/ +void lru_cache_add_active_or_noreclaim (struct page *page, +struct vm_area_struct *vma) +{ +if (page_reclaimable (page, vma)) +lru_cache_add_lru (page,LRU_ACTIVE +page_file_cache (page)); +else +add_page_to_noreclaim_list (page); +} +/* * Drain pages out of the cpu's pagevecs. * Either "cpu" is the current CPU, and preemption has already been
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 22/25 Noreclaim and Mlocked pages vm events
In reply to Lee SchermerhornAgainst: 2.6.26-rc2-mm1
Add some event counters to vmstats for testing noreclaim/mlock. Some of these might be interesting enough to keep around.
include/linux/vmstat.h | 11 +++++++++++ mm/internal.h | 4 +++- mm/mlock.c | 33 +++++++++++++++++++++++++-------- mm/vmscan.c | 16 +++++++++++++++- mm/vmstat.c | 12 ++++++++++++ 5 files changed, 66 insertions(+), 10 deletions(-)
include/linux/vmstat.h +11 -0
@@ -41,6 +41,17 @@ enum vm_event_item { PGPGIN, PGPGOUT, PS#ifdef CONFIG_HUGETLB_PAGE HTLB_BUDDY_PGALLOC, HTLB_BUDDY_PGALLOC_FAIL,#endif +#ifdef CONFIG_NORECLAIM_LRU +NORECL_PGCULLED ,/* culled to noreclaim list */ +NORECL_PGSCANNED ,/* scanned for reclaimability */ +NORECL_PGRESCUED ,/* rescued from noreclaim list */ +#ifdef CONFIG_NORECLAIM_MLOCK +NORECL_PGMLOCKED , +NORECL_PGMUNLOCKED , +NORECL_PGCLEARED , +NORECL_PGSTRANDED ,/* unable to isolate on unlock */ +#endif +#endifNR_VM_EVENT_ITEMS };
mm/vmscan.c +15 -1
@@ -453,12 +453,13 @@ int putback_lru_page(struct page *page) {int lru;int ret =1 ; +int was_nonreclaimable;VM_BUG_ON (!PageLocked (page));VM_BUG_ON (PageLRU (page)); lru = !!TestClearPageActive (page); -ClearPageNoreclaim (page);/* for page_reclaimable() */ + was_nonreclaimable =TestClearPageNoreclaim (page);if (unlikely (!page->mapping)) {/*
@@ -478,6 +479,10 @@ int putback_lru_page(struct page *page) lru +=page_file_cache (page);lru_cache_add_lru (page, lru);mem_cgroup_move_lists (page, lru); +#ifdef CONFIG_NORECLAIM_LRU +if (was_nonreclaimable) +count_vm_event (NORECL_PGRESCUED ); +#endif }else {/* * Put non-reclaimable pages directly on zone's noreclaim
@@ -485,6 +490,10 @@ int putback_lru_page(struct page *page)*/ add_page_to_noreclaim_list (page);mem_cgroup_move_lists (page,LRU_NORECLAIM ); +#ifdef CONFIG_NORECLAIM_LRU +if (!was_nonreclaimable) +count_vm_event (NORECL_PGCULLED ); +#endif }put_page (page);/* drop ref from isolate */
@@ -2363,6 +2372,7 @@ static void check_move_noreclaim_page(st__dec_zone_state (zone,NR_NORECLAIM );list_move (&page->lru, &zone->list[l]);__inc_zone_state (zone,NR_INACTIVE_ANON + l); +__count_vm_event (NORECL_PGRESCUED ); }else {/* * rotate noreclaim list
@@ -2394,6 +2404,7 @@ void scan_mapping_noreclaim_pages(structwhile (next < end &&pagevec_lookup (&pvec, mapping, next,PAGEVEC_SIZE )) {int i; +int pg_scanned =0 ; zone =NULL ;
@@ -2402,6 +2413,7 @@ void scan_mapping_noreclaim_pages(structpgoff_t page_index = page->index;struct zone *pagezone =page_zone (page); + pg_scanned++;if (page_index > next) next = page_index; next++;
@@ -2432,6 +2444,8 @@ void scan_mapping_noreclaim_pages(structif (zone)spin_unlock_irq (&zone->lru_lock);pagevec_release (&pvec); + +count_vm_events (NORECL_PGSCANNED , pg_scanned); } }
mm/vmstat.c +12 -0
@@ -759,6 +759,18 @@ static const char * const vmstat_text[] "htlb_buddy_alloc_success", "htlb_buddy_alloc_fail", #endif + +#ifdef CONFIG_NORECLAIM_LRU + "noreclaim_pgs_culled", + "noreclaim_pgs_scanned", + "noreclaim_pgs_rescued", +#ifdef CONFIG_NORECLAIM_MLOCK + "noreclaim_pgs_mlocked", + "noreclaim_pgs_munlocked", + "noreclaim_pgs_cleared", + "noreclaim_pgs_stranded", +#endif +#endif #endif };
mm/mlock.c +25 -8
@@ -18,6 +18,7 @@#include <linux/rmap.h> #include <linux/mmzone.h> #include <linux/hugetlb.h> +#include <linux/vmstat.h> #include "internal.h"
@@ -57,6 +58,7 @@ void __clear_page_mlock(struct page *pagVM_BUG_ON (!PageLocked (page));/* for LRU islolate/putback */ dec_zone_page_state (page,NR_MLOCK ); +count_vm_event (NORECL_PGCLEARED );if (!isolate_lru_page (page)) {putback_lru_page (page); }else {
@@ -66,6 +68,8 @@ void __clear_page_mlock(struct page *paglru_add_drain_all ();if (!isolate_lru_page (page))putback_lru_page (page); +else if (PageNoreclaim (page)) +count_vm_event (NORECL_PGSTRANDED ); } }
@@ -79,6 +83,7 @@ void mlock_vma_page(struct page *page)if (!TestSetPageMlocked (page)) {inc_zone_page_state (page,NR_MLOCK ); +count_vm_event (NORECL_PGMLOCKED );if (!isolate_lru_page (page))putback_lru_page (page); }
@@ -109,16 +114,28 @@ static void munlock_vma_page(struct pageif (TestClearPageMlocked (page)) {dec_zone_page_state (page,NR_MLOCK );if (!isolate_lru_page (page)) { -try_to_unlock (page);/* maybe relock the page */ +int ret =try_to_unlock (page); +/* +* did try_to_unlock() succeed or punt? +*/ +if (ret ==SWAP_SUCCESS || ret ==SWAP_AGAIN ) +count_vm_event (NORECL_PGMUNLOCKED ); +putback_lru_page (page); + }else { +/* +* We lost the race. let try_to_unmap() deal +* with it. At least we get the page state and +* mlock stats right. However, page is still on +* the noreclaim list. We'll fix that up when +* the page is eventually freed or we scan the +* noreclaim list. +*/ +if (PageNoreclaim (page)) +count_vm_event (NORECL_PGSTRANDED ); +else +count_vm_event (NORECL_PGMUNLOCKED ); } -/* -* Else we lost the race. let try_to_unmap() deal with it. -* At least we get the page state and mlock stats right. -* However, page is still on the noreclaim list. We'll fix -* that up when the page is eventually freed or we scan the -* noreclaim list. -*/ } }
mm/internal.h +3 -1
@@ -107,8 +107,10 @@ static inline int is_mlocked_vma(structif (likely ((vma->vm_flags & (VM_LOCKED |VM_SPECIAL )) !=VM_LOCKED ))return 0 ; -if (!TestSetPageMlocked (page)) +if (!TestSetPageMlocked (page)) {inc_zone_page_state (page,NR_MLOCK ); +count_vm_event (NORECL_PGMLOCKED ); + }return 1 ; }
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 23/25 Noreclaim LRU scan sysctl
In reply to Lee SchermerhornAgainst: 2.6.26-rc2-mm1
V6:
+ moved to end of series as optional debug patch
V2 -> V3:
+ rebase to 23-mm1 atop RvR's split LRU series
New in V2
This patch adds a function to scan individual or all zones' noreclaim lists and move any pages that have become reclaimable onto the respective zone's inactive list, where shrink_inactive_list() will deal with them.
Adds sysctl to scan all nodes, and per node attributes to individual nodes' zones.
Kosaki:
If reclaimable page found in noreclaim lru when write /proc/sys/vm/scan_noreclaim_pages, print filename and file offset of these pages.
TODO: DEBUGGING ONLY: NOT FOR UPSTREAM MERGE
drivers/base/node.c | 5 + include/linux/rmap.h | 3 include/linux/swap.h | 15 ++++ kernel/sysctl.c | 10 +++ mm/rmap.c | 4 - mm/vmscan.c | 161 +++++++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 196 insertions(+), 2 deletions(-)
include/linux/swap.h +15 -0
@@ -7,6 +7,7 @@#include <linux/list.h> #include <linux/memcontrol.h> #include <linux/sched.h> +#include <linux/node.h> #include <asm/atomic.h> #include <asm/page.h>
@@ -235,15 +236,29 @@ static inline int zone_reclaim(struct zo#ifdef CONFIG_NORECLAIM_LRU extern int page_reclaimable (struct page *page,struct vm_area_struct *vma);extern void scan_mapping_noreclaim_pages (struct address_space *); + +extern unsigned long scan_noreclaim_pages; +extern int scan_noreclaim_handler (struct ctl_table *,int ,struct file *, +void __user *,size_t *,loff_t *); +extern int scan_noreclaim_register_node (struct node *node); +extern void scan_noreclaim_unregister_node (struct node *node);#else static inline int page_reclaimable (struct page *page,struct vm_area_struct *vma) {return 1 ; } +static inline void scan_mapping_noreclaim_pages (struct address_space *mapping) { } + +static inline int scan_noreclaim_register_node (struct node *node) +{ +return 0 ; +} + +static inline void scan_noreclaim_unregister_node (struct node *node) { }#endif extern int kswapd_run (int nid);
mm/vmscan.c +161 -0
183 lines not on this page · raw message
kernel/sysctl.c +10 -0
@@ -1151,6 +1151,16 @@ static struct ctl_table vm_table[] = { .extra2 = &one, },#endif +#ifdef CONFIG_NORECLAIM_LRU + { + .ctl_name = CTL_UNNUMBERED, + .procname = "scan_noreclaim_pages", + .data = &scan_noreclaim_pages, + .maxlen = sizeof(scan_noreclaim_pages), + .mode = 0644, + .proc_handler = &scan_noreclaim_handler, + }, +#endif /* * NOTE: do not add new entries to this table unless you have read * Documentation/sysctl/ctl_unnumbered.txt
drivers/base/node.c +5 -0
@@ -13,6 +13,7 @@#include <linux/nodemask.h> #include <linux/cpu.h> #include <linux/device.h> +#include <linux/swap.h> static struct sysdev_class node_class = { .name ="node" ,
@@ -190,6 +191,8 @@ int register_node(struct node *node, intsysdev_create_file (&node->sysdev, &attr_meminfo);sysdev_create_file (&node->sysdev, &attr_numastat);sysdev_create_file (&node->sysdev, &attr_distance); + +scan_noreclaim_register_node (node); }return error; }
@@ -209,6 +212,8 @@ void unregister_node(struct node *node)sysdev_remove_file (&node->sysdev, &attr_numastat);sysdev_remove_file (&node->sysdev, &attr_distance); +scan_noreclaim_unregister_node (node); +sysdev_unregister (&node->sysdev); }
include/linux/rmap.h +3 -0
@@ -55,6 +55,9 @@ void anon_vma_unlink(struct vm_area_struvoid anon_vma_link (struct vm_area_struct *);void __anon_vma_link (struct vm_area_struct *); +extern struct anon_vma *page_lock_anon_vma (struct page *page); +extern void page_unlock_anon_vma (struct anon_vma *anon_vma); +/* * rmap interfaces called when adding or removing pte of page */
mm/rmap.c +2 -2
@@ -168,7 +168,7 @@ void __init anon_vma_init(void)* Getting a lock on a stable anon_vma from a page off the LRU is * tricky: page_lock_anon_vma rely on RCU to guard against the races. */ -static struct anon_vma *page_lock_anon_vma (struct page *page) +struct anon_vma *page_lock_anon_vma (struct page *page) {struct anon_vma *anon_vma;unsigned long anon_mapping;
@@ -188,7 +188,7 @@ out:return NULL ; } -static void page_unlock_anon_vma (struct anon_vma *anon_vma) +void page_unlock_anon_vma (struct anon_vma *anon_vma) {spin_unlock (&anon_vma->lock);rcu_read_unlock ();
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 24/25 Mlocked Pages: count attempts to free mlocked page
In reply to Lee SchermerhornFrom: Lee Schermerhorn <[email protected]>
Against: 2.6.26-rc2-mm1
Allow free of mlock()ed pages. This shouldn't happen, but during developement, it occasionally did.
This patch allows us to survive that condition, while keeping the statistics and events correct for debug.
include/linux/vmstat.h | 1 + mm/internal.h | 17 +++++++++++++++++ mm/page_alloc.c | 1 + mm/vmstat.c | 1 + 4 files changed, 20 insertions(+)
mm/internal.h +17 -0
@@ -152,6 +152,22 @@ static inline void mlock_migrate_page(st } } +/* +* free_page_mlock() -- clean up attempts to free and mlocked() page. +* Page should not be on lru, so no need to fix that up. +* free_pages_check() will verify... +*/ +static inline void free_page_mlock (struct page *page) +{ +if (unlikely (TestClearPageMlocked (page))) { +unsigned long flags; + +local_irq_save (flags); +__dec_zone_page_state (page,NR_MLOCK ); +__count_vm_event (NORECL_MLOCKFREED ); +local_irq_restore (flags); + } +}#else /* CONFIG_NORECLAIM_MLOCK */ static inline int is_mlocked_vma (struct vm_area_struct *v,struct page *p)
@@ -161,6 +177,7 @@ static inline int is_mlocked_vma(structstatic inline void clear_page_mlock (struct page *page) { }static inline void mlock_vma_page (struct page *page) { }static inline void mlock_migrate_page (struct page *new,struct page *old) { } +static inline void free_page_mlock (struct page *page) { }#endif /* CONFIG_NORECLAIM_MLOCK */
mm/page_alloc.c +1 -0
@@ -484,6 +484,7 @@ static inline void __free_one_page(strucstatic inline int free_pages_check (struct page *page) { +free_page_mlock (page); if (unlikely(page_mapcount(page) | (page->mapping != NULL) | (page_get_page_cgroup(page) != NULL) |
include/linux/vmstat.h +1 -0
@@ -50,6 +50,7 @@ enum vm_event_item { PGPGIN, PGPGOUT, PS NORECL_PGMUNLOCKED, NORECL_PGCLEARED, NORECL_PGSTRANDED, /* unable to isolate on unlock */ + NORECL_MLOCKFREED, #endif #endif NR_VM_EVENT_ITEMS
mm/vmstat.c +1 -0
@@ -769,6 +769,7 @@ static const char * const vmstat_text[] "noreclaim_pgs_munlocked", "noreclaim_pgs_cleared", "noreclaim_pgs_stranded", + "noreclaim_pgs_mlockfreed", #endif #endif #endif
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
PATCH 25/25 Noreclaim LRU and Mlocked Pages Documentation
In reply to Lee SchermerhornFrom: Lee Schermerhorn <[email protected]>
Documentation for noreclaim lru list and its usage.
Documentation/vm/noreclaim-lru.txt | 609 +++++++++++++++++++++++++++++++++++++ 1 file changed, 609 insertions(+)
Documentation/vm/noreclaim-lru.txt new file +609 -0
@@ -0,0 +1,609 @@ + +This document describes the Linux memory management "Noreclaim LRU" +infrastructure and the use of this infrastructure to manage several types +of "non-reclaimable" pages. The document attempts to provide the overall +rationale behind this mechanism and the rationale for some of the design +decisions that drove the implementation. The latter design rationale is +discussed in the context of an implementation description. Admittedly, one +can obtain the implementation details--the "what does it do?"--by reading the +code. One hopes that the descriptions below add value by provide the answer +to "why does it do that?". + +Noreclaim LRU Infrastructure: + +The Noreclaim LRU adds an additional LRU list to track non-reclaimable pages +and to hide these pages from vmscan. This mechanism is based on a patch by +Larry Woodman of Red Hat to address several scalability problems with page +reclaim in Linux. The problems have been observed at customer sites on large +memory x86_64 systems. For example, a non-numal x86_64 platform with 128GB +of main memory will have over 32 million 4k pages in a single zone. When a +large fraction of these pages are not reclaimable for any reason [see below], +vmscan will spend a lot of time scanning the LRU lists looking for the small +fraction of pages that are reclaimable. This can result in a situation where +all cpus are spending 100% of their time in vmscan for hours or days on end, +with the system completely unresponsive. + +The Noreclaim LRU infrastructure addresses the following classes of +non-reclaimable pages: + ++ page owned by ram disks or ramfs ++ page mapped into SHM_LOCKed shared memory regions ++ page mapped into VM_LOCKED [mlock()ed] vmas + +The infrastructure might be able to handle other conditions that make pages +nonreclaimable, either by definition or by circumstance, in the future. + + +The Noreclaim LRU List + +The Noreclaim LRU infrastructure consists of an additional, per-zone, LRU list +called the "noreclaim" list and an associated page flag, PG_noreclaim, to +indicate that the page is being managed on the noreclaim list. The PG_noreclaim +flag is analogous to, and mutually exclusive with, the PG_active flag in that +it indicates on which LRU list a page resides when PG_lru is set. The +noreclaim LRU list is source configurable based on the NORECLAIM_LRU Kconfig +option. + +Why maintain nonreclaimable pages on an additional LRU list? The Linux memory +management subsystem has well established protocols for managing pages on the +LRU. Vmscan is based on LRU lists. LRU list exist per zone, and we want to +maintain pages relative to their "home zone". All of these make the use of +an additional list, parallel to the LRU active and inactive lists, a natural +mechanism to employ. Note, however, that the noreclaim list does not +differentiate between file backed and swap backed [anon] pages. This +differentiation is only important while the pages are, in fact, reclaimable. + +The noreclaim LRU list benefits from the "arrayification" of the per-zone +LRU lists and statistics originally proposed and posted by Christoph Lameter. + +Note that the noreclaim list does not use the lru pagevec mechanism. Rather, +nonreclaimable pages are placed directly on the page's zone's noreclaim +list under the zone lru_lock. The reason for this is to prevent stranding +of pages on the noreclaim list when one task has the page isolated from the +lru and other tasks are changing the "reclaimability" state of the page. + + +Noreclaim LRU and Memory Controller Interaction + +The memory controller data structure automatically gets a per zone noreclaim +lru list as a result of the "arrayification" of the per-zone LRU lists. The +memory controller tracks the movement of pages to and from the noreclaim list. +When a memory control group comes under memory pressure, the controller will +not attempt to reclaim pages on the noreclaim list. This has a couple of +effects. Because the pages are "hidden" from reclaim on the noreclaim list, +the reclaim process can be more efficient, dealing only with pages that have +a chance of being reclaimed. On the other hand, if too many of the pages +charged to the control group are non-reclaimable, the reclaimable portion of the +working set of the tasks in the control group may not fit into the available +memory. This can cause the control group to thrash or to oom-kill tasks. + + +Noreclaim LRU: Detecting Non-reclaimable Pages + +The function page_reclaimable(page, vma) in vmscan.c determines whether a +page is reclaimable or not. For ramfs and ram disk [brd] pages and pages in +SHM_LOCKed regions, page_reclaimable() tests a new address space flag, +AS_NORECLAIM, in the page's address space using a wrapper function. +Wrapper functions are used to set, clear and test the flag to reduce the +requirement for #ifdef's throughout the source code. AS_NORECLAIM is set on +ramfs inode/mapping when it is created and on ram disk inode/mappings at open +time. This flag remains for the life of the inode. + +For shared memory regions, AS_NORECLAIM is set when an application successfully +SHM_LOCKs the region and is removed when the region is SHM_UNLOCKed. Note that +shmctl(SHM_LOCK, ...) does not populate the page tables for the region as does, +for example, mlock(). So, we make no special effort to push any pages in the +SHM_LOCKed region to the noreclaim list. Vmscan will do this when/if it +encounters the pages during reclaim. On SHM_UNLOCK, shmctl() scans the pages +in the region and "rescues" them from the noreclaim list if no other condition +keeps them non-reclaimable. If a SHM_LOCKed region is destroyed, the pages +are also "rescued" from the noreclaim list in the process of freeing them. + +page_reclaimable() detects mlock()ed pages by testing an additional page flag, +PG_mlocked via the PageMlocked() wrapper. If the page is NOT mlocked, and a +non-NULL vma is supplied, page_reclaimable() will check whether the vma is +VM_LOCKED via is_mlocked_vma(). is_mlocked_vma() will SetPageMlocked() and +update the appropriate statistics if the vma is VM_LOCKED. This method allows +efficient "culling" of pages in the fault path that are being faulted in to +VM_LOCKED vmas. + + +Non-reclaimable Pages and Vmscan [shrink_*_list()] + +If non-reclaimable pages are culled in the fault path, or moved to the +noreclaim list at mlock() or mmap() time, vmscan will never encounter the pages +until they have become reclaimable again, for example, via munlock() and have +been "rescued" from the noreclaim list. However, there may be situations where +we decide, for the sake of expediency, to leave a non-reclaimable page on one of +the regular active/inactive LRU lists for vmscan to deal with. Vmscan checks +for such pages in all of the shrink_{active|inactive|page}_list() functions and +will "cull" such pages that it encounters--that is, it diverts those pages to +the noreclaim list for the zone being scanned. + +There may be situations where a page is mapped into a VM_LOCKED vma, but the +page is not marked as PageMlocked. Such pages will make it all the way to +shrink_page_list() where they will be detected when vmscan walks the reverse +map in try_to_unmap(). If try_to_unmap() returns SWAP_MLOCK, shrink_page_list() +will cull the page at that point. + +Note that for anonymous pages, shrink_page_list() attempts to add the page to +the swap cache before it tries to unmap the page. To avoid this unnecessary +consumption of swap space, shrink_page_list() calls try_to_unlock() to check +whether any VM_LOCKED vmas map the page without attempting to unmap the page. +If try_to_unlock() returns SWAP_MLOCK, shrink_page_list() will cull the page +without consuming swap space. try_to_unlock() will be described below. + + +Mlocked Page: Prior Work + +The "Noreclaim Mlocked Pages" infrastructure is based on work originally posted +by Nick Piggin in an RFC patch entitled "mm: mlocked pages off LRU". Nick's +posted his patch as an alternative to a patch posted by Christoph Lameter to +achieve the same objective--hiding mlocked pages from vmscan. In Nick's patch, +he used one of the struct page lru list link fields as a count of VM_LOCKED +vmas that map the page. This use of the link field for a count prevent the +management of the pages on an LRU list. When Nick's patch was integrated with +the Noreclaim LRU work, the count was replaced by walking the reverse map to +determine whether any VM_LOCKED vmas mapped the page. More on this below. +The primary reason for wanting to keep mlocked pages on an LRU list is that +mlocked pages are migratable, and the LRU list is used to arbitrate tasks +attempting to migrate the same page. Whichever task succeeds in "isolating" +the page from the LRU performs the migration. + + +Mlocked Pages: Basic Management + +Mlocked pages--pages mapped into a VM_LOCKED vma--represent one class of +nonreclaimable pages. When such a page has been "noticed" by the memory +management subsystem, the page is marked with the PG_mlocked [PageMlocked()] +flag. A PageMlocked() page will be placed on the noreclaim LRU list when +it is added to the LRU. Pages can be "noticed" by memory management in +several places: + +1) in the mlock()/mlockall() system call handlers. +2) in the mmap() system call handler when mmap()ing a region with the + MAP_LOCKED flag, or mmap()ing a region in a task that has called + mlockall() with the MCL_FUTURE flag. Both of these conditions result + in the VM_LOCKED flag being set for the vma. +3) in the fault path, if mlocked pages are "culled" in the fault path, + and when a VM_LOCKED stack segment is expanded. +4) as mentioned above, in vmscan:shrink_page_list() with attempting to + reclaim a page in a VM_LOCKED vma--via try_to_unmap() or try_to_unlock(). + +Mlocked pages become unlocked and rescued from the noreclaim list when: + +1) mapped in a range unlocked via the munlock()/munlockall() system calls. +2) munmapped() out of the last VM_LOCKED vma that maps the page, including + unmapping at task exit. +3) when the page is truncated from the last VM_LOCKED vma of an mmap()ed file. +4) before a page is COWed in a VM_LOCKED vma. + + +Mlocked Pages: mlock()/mlockall() System Call Handling + +Both [do_]mlock() and [do_]mlockall() system call handlers call mlock_fixup() +for each vma in the range specified by the call. In the case of mlockall(), +this is the entire active address space of the task. Note that mlock_fixup() +is used for both mlock()ing and munlock()ing a range of memory. A call to +mlock() an already VM_LOCKED vma, or to munlock() a vma that is not VM_LOCKED +is treated as a no-op--mlock_fixup() simply returns. + +If the vma passes some filtering described in "Mlocked Pages: Filtering Vmas" +below, mlock_fixup() will attempt to merge the vma with its neighbors or split +off a subset of the vma if the range does not cover the entire vma. Once the +vma has been merged or split or neither, mlock_fixup() will call +__mlock_vma_pages_range() to fault in the pages via get_user_pages() and +to mark the pages as mlocked via mlock_vma_page(). + +Note that the vma being mlocked might be mapped with PROT_NONE. In this case, +get_user_pages() will be unable to fault in the pages. That's OK. If pages +do end up getting faulted into this VM_LOCKED vma, we'll handle them in the +fault path or in vmscan. + +Also note that a page returned by get_user_pages() could be truncated or +migrated out from under us, while we're trying to mlock it. To detect +this, __mlock_vma_pages_range() tests the page_mapping after acquiring +the page lock. If the page is still associated with its mapping, we'll +go ahead and call mlock_vma_page(). If the mapping is gone, we just +unlock the page and move on. Worse case, this results in page mapped +in a VM_LOCKED vma remaining on a normal LRU list without being +PageMlocked(). Again, vmscan will detect and cull such pages. + +mlock_vma_page(), called with the page locked [N.B., not "mlocked"] will +TestSetPageMlocked() for each page returned by get_user_pages(). We use +TestSetPageMlocked() because the page might already be mlocked by another +task/vma and we don't want to do extra work. We especially do not want to +count an mlocked page more than once in the statistics. If the page was +already mlocked, mlock_vma_page() is done. + +If the page was NOT already mlocked, mlock_vma_page() attempts to isolate the +page from the LRU, as it is likely on the appropriate active or inactive list +at that time. If the isolate_lru_page() succeeds, mlock_vma_page() will +putback the page--putback_lru_page()--which will notice that the page is now +mlocked and divert the page to the zone's noreclaim LRU list. If +mlock_vma_page() is unable to isolate the page from the LRU, vmscan will handle +it later if/when it attempts to reclaim the page. + + +Mlocked Pages: Filtering Vmas + +mlock_fixup() filters several classes of "special" vmas: + +1) vmas with VM_IO|VM_PFNMAP set are skipped entirely. The pages behind + these mappings are inherently pinned, so we don't need to mark them as + mlocked. In any case, most of the pages have no struct page in which to + so mark the page. Because of this, get_user_pages() will fail for these + vmas, so there is no sense in attempting to visit them. + +2) vmas mapping hugetlbfs page are already effectively pinned into memory. + We don't need nor want to mlock() these pages. However, to preserve the + prior behavior of mlock()--before the noreclaim/mlock changes--mlock_fixup() + will call make_pages_present() in the hugetlbfs vma range to allocate the + huge pages and populate the ptes. + +3) vmas with VM_DONTEXPAND|VM_RESERVED are generally user space mappings of + kernel pages, such as the vdso page, relay channel pages, etc. These pages + are inherently non-reclaimable and are not managed on the LRU lists. + mlock_fixup() treats these vmas the same as hugetlbfs vmas. It calls + make_pages_present() to populate the ptes. + +Note that for all of these special vmas, mlock_fixup() does not set the +VM_LOCKED flag. Therefore, we won't have to deal with them later during +munlock() or munmap()--for example, at task exit. Neither does mlock_fixup() +account these vmas against the task's "locked_vm". + +Mlocked Pages: Downgrading the Mmap Semaphore. + +mlock_fixup() must be called with the mmap semaphore held for write, because +it may have to merge or split vmas. However, mlocking a large region of +memory can take a long time--especially if vmscan must reclaim pages to +satisfy the regions requirements. Faulting in a large region with the mmap +semaphore held for write can hold off other faults on the address space, in +the case of a multi-threaded task. It can also hold off scans of the task's +address space via /proc. While testing under heavy load, it was observed that +the ps(1) command could be held off for many minutes while a large segment was +mlock()ed down. + +To address this issue, and to make the system more responsive during mlock()ing +of large segments, mlock_fixup() downgrades the mmap semaphore to read mode +during the call to __mlock_vma_pages_range(). This works fine. However, the +callers of mlock_fixup() expect the semaphore to be returned in write mode. +So, mlock_fixup() "upgrades" the semphore to write mode. Linux does not +support an atomic upgrade_sem() call, so mlock_fixup() must drop the semaphore +and reacquire it in write mode. In a multi-threaded task, it is possible for +the task memory map to change while the semaphore is dropped. Therefore, +mlock_fixup() looks up the vma at the range start address after reacquiring +the semaphore in write mode and verifies that it still covers the original +range. If not, mlock_fixup() returns an error [-EAGAIN]. All callers of +mlock_fixup() have been changed to deal with this new error condition. + +Note: when munlocking a region, all of the pages should already be resident-- +unless we have racing threads mlocking() and munlocking() regions. So, +unlocking should not have to wait for page allocations nor faults of any kind. +Therefore mlock_fixup() does not downgrade the semaphore for munlock(). + + +Mlocked Pages: munlock()/munlockall() System Call Handling + +The munlock() and munlockall() system calls are handled by the same functions-- +do_mlock[all]()--as the mlock() and mlockall() system calls with the unlock +vs lock operation indicated by an argument. So, these system calls are also +handled by mlock_fixup(). Again, if called for an already munlock()ed vma, +mlock_fixup() simply returns. Because of the vma filtering discussed above, +VM_LOCKED will not be set in any "special" vmas. So, these vmas will be +ignored for munlock. + +If the vma is VM_LOCKED, mlock_fixup() again attempts to merge or split off +the specified range. The range is then munlocked via the function +__munlock_vma_pages_range(). Because the vma access protections could have +been changed to PROT_NONE after faulting in and mlocking some pages, +get_user_pages() is unreliable for visiting these pages for munlocking. We +don't want to leave pages mlocked(), so __munlock_vma_pages_range() uses a +custom page table walker to find all pages mapped into the specified range. +Note that this again assumes that all pages in the mlocked() range are resident +and mapped by the task's page table. + +As with __mlock_vma_pages_range(), unlocking can race with truncation and +migration. It is very important that munlock of a page succeeds, lest we +leak pages by stranding them in the mlocked state on the noreclaim list. +The munlock page walk pte handler resolves the race with page migration +by checking the pte for a special swap pte indicating that the page is +being migrated. If this is the case, the pte handler will wait for the +migration entry to be replaced and then refetch the pte for the new page. +Once the pte handler has locked the page, it checks the page_mapping to +ensure that it still exists. If not, the handler unlocks the page and +retries the entire process after refetching the pte. + +The munlock page walk pte handler unlocks individual pages by calling +munlock_vma_page(). munlock_vma_page() unconditionally clears the PG_mlocked +flag using TestClearPageMlocked(). As with mlock_vma_page(), munlock_vma_page() +use the Test*PageMlocked() function to handle the case where the page might +have already been unlocked by another task. If the page was mlocked, +munlock_vma_page() updates that zone statistics for the number of mlocked +pages. Note, however, that at this point we haven't checked whether the page +is mapped by other VM_LOCKED vmas. + +We can't call try_to_unlock(), the function that walks the reverse map to check +for other VM_LOCKED vmas, without first isolating the page from the LRU. +try_to_unlock() is a variant of try_to_unmap() and thus requires that the page +not be on an lru list. [More on these below.] However, the call to +isolate_lru_page() could fail, in which case we couldn't try_to_unlock(). +So, we go ahead and clear PG_mlocked up front, as this might be the only chance +we have. If we can successfully isolate the page, we go ahead and +try_to_unlock(), which will restore the PG_mlocked flag and update the zone +page statistics if it finds another vma holding the page mlocked. If we fail +to isolate the page, we'll have left a potentially mlocked page on the LRU. +This is fine, because we'll catch it later when/if vmscan tries to reclaim the +page. This should be relatively rare. + +Mlocked Pages: Migrating Them... + +A page that is being migrated has been isolated from the lru lists and is +held locked across unmapping of the page, updating the page's mapping +[address_space] entry and copying the contents and state, until the +page table entry has been replaced with an entry that refers to the new +page. Linux supports migration of mlocked pages and other non-reclaimable +pages. This involves simply moving the PageMlocked and PageNoreclaim states +from the old page to the new page. + +Note that page migration can race with mlocking or munlocking of the same +page. This has been discussed from the mlock/munlock perspective in the +respective sections above. Both processes [migration, m[un]locking], hold +the page locked. This provides the first level of synchronization. Page +migration zeros out the page_mapping of the old page before unlocking it, +so m[un]lock can skip these pages. However, as discussed above, munlock +must wait for a migrating page to be replaced with the new page to prevent +the new page from remaining mlocked outside of any VM_LOCKED vma. + +To ensure that we don't strand pages on the noreclaim list because of a +race between munlock and migration, we must also prevent the munlock pte +handler from acquiring the old or new page lock from the time that the +migration subsystem acquires the old page lock, until either migration +succeeds and the new page is added to the lru or migration fails and +the old page is putback to the lru. The achieve this coordination, +the migration subsystem places the new page on success, or the old +page on failure, back on the lru lists before dropping the respective +page's lock. It uses the putback_lru_page() function to accomplish this, +which rechecks the page's overall reclaimability and adjusts the page +flags accordingly. To free the old page on success or the new page on +failure, the migration subsystem just drops what it knows to be the last +page reference via put_page(). + + +Mlocked Pages: mmap(MAP_LOCKED) System Call Handling + +In addition the the mlock()/mlockall() system calls, an application can request +that a region of memory be mlocked using the MAP_LOCKED flag with the mmap() +call. Furthermore, any mmap() call or brk() call that expands the heap by a +task that has previously called mlockall() with the MCL_FUTURE flag will result +in the newly mapped memory being mlocked. Before the noreclaim/mlock changes, +the kernel simply called make_pages_present() to allocate pages and populate +the page table. + +To mlock a range of memory under the noreclaim/mlock infrastructure, the +mmap() handler and task address space expansion functions call +mlock_vma_pages_range() specifying the vma and the address range to mlock. +mlock_vma_pages_range() filters vmas like mlock_fixup(), as described above in +"Mlocked Pages: Filtering Vmas". It will clear the VM_LOCKED flag, which will +have already been set by the caller, in filtered vmas. Thus these vma's need +not be visited for munlock when the region is unmapped. + +For "normal" vmas, mlock_vma_pages_range() calls __mlock_vma_pages_range() to +fault/allocate the pages and mlock them. Again, like mlock_fixup(), +mlock_vma_pages_range() downgrades the mmap semaphore to read mode before +attempting to fault/allocate and mlock the pages; and "upgrades" the semaphore +back to write mode before returning. + +The callers of mlock_vma_pages_range() will have already added the memory +range to be mlocked to the task's "locked_vm". To account for filtered vmas, +mlock_vma_pages_range() returns the number of pages NOT mlocked. All of the +callers then subtract a non-negative return value from the task's locked_vm. +A negative return value represent an error--for example, from get_user_pages() +attempting to fault in a vma with PROT_NONE access. In this case, we leave +the memory range accounted as locked_vm, as the protections could be changed +later and pages allocated into that region. + + +Mlocked Pages: munmap()/exit()/exec() System Call Handling + +When unmapping an mlocked region of memory, whether by an explicit call to +munmap() or via an internal unmap from exit() or exec() processing, we must +munlock the pages if we're removing the last VM_LOCKED vma that maps the pages. +Before the noreclaim/mlock changes, mlocking did not mark the pages in any way, +so unmapping them required no processing. + +To munlock a range of memory under the noreclaim/mlock infrastructure, the +munmap() hander and task address space tear down function call +munlock_vma_pages_all(). The name reflects the observation that one always +specifies the entire vma range when munlock()ing during unmap of a region. +Because of the vma filtering when mlocking() regions, only "normal" vmas that +actually contain mlocked pages will be passed to munlock_vma_pages_all(). + +munlock_vma_pages_all() clears the VM_LOCKED vma flag and, like mlock_fixup() +for the munlock case, calls __munlock_vma_pages_range() to walk the page table +for the vma's memory range and munlock_vma_page() each resident page mapped by +the vma. This effectively munlocks the page, only if this is the last +VM_LOCKED vma that maps the page. + + +Mlocked Page: try_to_unmap() + +[Note: the code changes represented by this section are really quite small +compared to the text to describe what happening and why, and to discuss the +implications.] + +Pages can, of course, be mapped into multiple vmas. Some of these vmas may +have VM_LOCKED flag set. It is possible for a page mapped into one or more +VM_LOCKED vmas not to have the PG_mlocked flag set and therefore reside on one +of the active or inactive LRU lists. This could happen if, for example, a +task in the process of munlock()ing the page could not isolate the page from +the LRU. As a result, vmscan/shrink_page_list() might encounter such a page +as described in "Non-reclaimable Pages and Vmscan [shrink_*_list()]". To +handle this situation, try_to_unmap() has been enhanced to check for VM_LOCKED +vmas while it is walking a page's reverse map. + +try_to_unmap() is always called, by either vmscan for reclaim or for page +migration, with the argument page locked and isolated from the LRU. BUG_ON() +assertions enforce this requirement. Separate functions handle anonymous and +mapped file pages, as these types of pages have different reverse map +mechanisms. + + try_to_unmap_anon() + +To unmap anonymous pages, each vma in the list anchored in the anon_vma must be +visited--at least until a VM_LOCKED vma is encountered. If the page is being +unmapped for migration, VM_LOCKED vmas do not stop the process because mlocked +pages are migratable. However, for reclaim, if the page is mapped into a +VM_LOCKED vma, the scan stops. try_to_unmap() attempts to acquire the mmap +semphore of the mm_struct to which the vma belongs in read mode. If this is +successful, try_to_unmap() will mlock the page via mlock_vma_page()--we +wouldn't have gotten to try_to_unmap() if the page were already mlocked--and +will return SWAP_MLOCK, indicating that the page is nonreclaimable. If the +mmap semaphore cannot be acquired, we are not sure whether the page is really +nonreclaimable or not. In this case, try_to_unmap() will return SWAP_AGAIN. + + try_to_unmap_file() -- linear mappings + +Unmapping of a mapped file page works the same, except that the scan visits +all vmas that maps the page's index/page offset in the page's mapping's +reverse map priority search tree. It must also visit each vma in the page's +mapping's non-linear list, if the list is non-empty. As for anonymous pages, +on encountering a VM_LOCKED vma for a mapped file page, try_to_unmap() will +attempt to acquire the associated mm_struct's mmap semaphore to mlock the page, +returning SWAP_MLOCK if this is successful, and SWAP_AGAIN, if not. + + try_to_unmap_file() -- non-linear mappings + +If a page's mapping contains a non-empty non-linear mapping vma list, then +try_to_un{map|lock}() must also visit each vma in that list to determine +whether the page is mapped in a VM_LOCKED vma. Again, the scan must visit +all vmas in the non-linear list to ensure that the pages is not/should not be +mlocked. If a VM_LOCKED vma is found in the list, the scan could terminate. +However, there is no easy way to determine whether the page is actually mapped +in a given vma--either for unmapping or testing whether the VM_LOCKED vma +actually pins the page. + +So, try_to_unmap_file() handles non-linear mappings by scanning a certain +number of pages--a "cluster"--in each non-linear vma associated with the page's +mapping, for each file mapped page that vmscan tries to unmap. If this happens +to unmap the page we're trying to unmap, try_to_unmap() will notice this on +return--(page_mapcount(page) == 0)--and return SWAP_SUCCESS. Otherwise, it +will return SWAP_AGAIN, causing vmscan to recirculate this page. We take +advantage of the cluster scan in try_to_unmap_cluster() as follows: + +For each non-linear vma, try_to_unmap_cluster() attempts to acquire the mmap +semaphore of the associated mm_struct for read without blocking. If this +attempt is successful and the vma is VM_LOCKED, try_to_unmap_cluster() will +retain the mmap semaphore for the scan; otherwise it drops it here. Then, +for each page in the cluster, if we're holding the mmap semaphore for a locked +vma, try_to_unmap_cluster() calls mlock_vma_page() to mlock the page. This +call is a no-op if the page is already locked, but will mlock any pages in +the non-linear mapping that happen to be unlocked. If one of the pages so +mlocked is the page passed in to try_to_unmap(), try_to_unmap_cluster() will +return SWAP_MLOCK, rather than the default SWAP_AGAIN. This will allow vmscan +to cull the page, rather than recirculating it on the inactive list. Again, +if try_to_unmap_cluster() cannot acquire the vma's mmap sem, it returns +SWAP_AGAIN, indicating that the page is mapped by a VM_LOCKED vma, but +couldn't be mlocked. + + +Mlocked pages: try_to_unlock() Reverse Map Scan + +TODO/FIXME: a better name might be page_mlocked()--analogous to the +page_referenced() reverse map walker--especially if we continue to call this +from shrink_page_list(). See related TODO/FIXME below. + +When munlock_vma_page()--see "Mlocked Pages: munlock()/munlockall() System +Call Handling" above--tries to munlock a page, or when shrink_page_list() +encounters an anonymous page that is not yet in the swap cache, they need to +determine whether or not the page is mapped by any VM_LOCKED vma, without +actually attempting to unmap all ptes from the page. For this purpose, the +noreclaim/mlock infrastructure introduced a variant of try_to_unmap() called +try_to_unlock(). + +try_to_unlock() calls the same functions as try_to_unmap() for anonymous and +mapped file pages with an additional argument specifing unlock versus unmap +processing. Again, these functions walk the respective reverse maps looking +for VM_LOCKED vmas. When such a vma is found for anonymous pages and file +pages mapped in linear VMAs, as in the try_to_unmap() case, the functions +attempt to acquire the associated mmap semphore, mlock the page via +mlock_vma_page() and return SWAP_MLOCK. This effectively undoes the +pre-clearing of the page's PG_mlocked done by munlock_vma_page() and informs +shrink_page_list() that the anonymous page should be culled rather than added +to the swap cache in preparation for a try_to_unmap() that will almost +certainly fail. + +If try_to_unmap() is unable to acquire a VM_LOCKED vma's associated mmap +semaphore, it will return SWAP_AGAIN. This will allow shrink_page_list() +to recycle the page on the inactive list and hope that it has better luck +with the page next time. + +For file pages mapped into non-linear vmas, the try_to_unlock() logic works +slightly differently. On encountering a VM_LOCKED non-linear vma that might +map the page, try_to_unlock() returns SWAP_AGAIN without actually mlocking +the page. munlock_vma_page() will just leave the page unlocked and let +vmscan deal with it--the usual fallback position. + +Note that try_to_unlock()'s reverse map walk must visit every vma in a pages' +reverse map to determine that a page is NOT mapped into any VM_LOCKED vma. +However, the scan can terminate when it encounters a VM_LOCKED vma and can +successfully acquire the vma's mmap semphore for read and mlock the page. +Although try_to_unlock() can be called many [very many!] times when +munlock()ing a large region or tearing down a large address space that has been +mlocked via mlockall(), overall this is a fairly rare event. In addition, +although shrink_page_list() calls try_to_unlock() for every anonymous page that +it handles that is not yet in the swap cache, on average anonymous pages will +have very short reverse map lists. + +Mlocked Page: Page Reclaim in shrink_*_list() + +shrink_active_list() culls any obviously nonreclaimable pages--i.e., +!page_reclaimable(page, NULL)--diverting these to the noreclaim lru +list. However, shrink_active_list() only sees nonreclaimable pages that +made it onto the active/inactive lru lists. Note that these pages do not +have PageNoreclaim set--otherwise, they would be on the noreclaim list and +shrink_active_list would never see them. + +Some examples of these nonreclaimable pages on the LRU lists are: + +1) ramfs and ram disk pages that have been placed on the lru lists when + first allocated. + +2) SHM_LOCKed shared memory pages. shmctl(SHM_LOCK) does not attempt to + allocate or fault in the pages in the shared memory region. This happens + when an application accesses the page the first time after SHM_LOCKing + the segment. + +3) Mlocked pages that could not be isolated from the lru and moved to the + noreclaim list in mlock_vma_page(). + +3) Pages mapped into multiple VM_LOCKED vmas, but try_to_unlock() couldn't + acquire the vma's mmap semaphore to test the flags and set PageMlocked. + munlock_vma_page() was forced to let the page back on to the normal + LRU list for vmscan to handle. + +shrink_inactive_list() also culls any nonreclaimable pages that it finds +on the inactive lists, again diverting them to the appropriate zone's noreclaim +lru list. shrink_inactive_list() should only see SHM_LOCKed pages that became +SHM_LOCKed after shrink_active_list() had moved them to the inactive list, or +pages mapped into VM_LOCKED vmas that munlock_vma_page() couldn't isolate from +the lru to recheck via try_to_unlock(). shrink_inactive_list() won't notice +the latter, but will pass on to shrink_page_list(). + +shrink_page_list() again culls obviously nonreclaimable pages that it could +encounter for similar reason to shrink_inactive_list(). As already discussed, +shrink_page_list() proactively looks for anonymous pages that should have +PG_mlocked set but don't--these would not be detected by page_reclaimable()--to +avoid adding them to the swap cache unnecessarily. File pages mapped into +VM_LOCKED vmas but without PG_mlocked set will make it all the way to +try_to_unmap(). shrink_page_list() will divert them to the noreclaim list when +try_to_unmap() returns SWAP_MLOCK, as discussed above. + +TODO/FIXME: If we can enhance the swap cache to reliably remove entries +with page_count(page) > 2, as long as all ptes are mapped to the page and +not the swap entry, we can probably remove the call to try_to_unlock() in +shrink_page_list() and just remove the page from the swap cache when +try_to_unmap() returns SWAP_MLOCK. Currently, remove_exclusive_swap_page() +doesn't seem to allow that. + +
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Andrew Morton
maintainerQuoted 7 lines · On Thu, 29 May 2008 15:50:30 -0400 Lee Schermerhorn <[email protected]>…
The patches to follow are a continuation of the V8 "VM pageout scalability improvements" series that Rik van Riel posted to LKML on 23May08. These patches apply atop Rik's series with the following overlap:
Patches 13 through 16 replace the corresponding patches in Rik's posting.
Quoted 9 lines
Patch 13, the noreclaim lru infrastructure, now includes Kosaki Motohiro's memcontrol enhancements to track nonreclaimable pages.
Patches 14 and 15 are largely unchanged, except for refresh. Includes some minor statistics formatting cleanup.
Patch 16 includes a fix for an potential [unobserved] race condition during
SHM_UNLOCK.
<head spins a bit>
Quoted 16 lines
Additional patches in this series:
Patches 17 through 20 keep mlocked pages off the normal [in]active LRU
lists using the noreclaim lru infrastructure. These patches represent
a fairly significant rework of an RFC patch originally posted by Nick Piggin.Patches 21 and 22 are optional, but recommended, enhancements to the overall noreclaim series.
Patches 23 and 24 are optional enhancements useful during debug and testing.
Patch 25 is a rather verbose document describing the noreclaim lru infrastructure and the use thereof to keep ramfs,
SHM_LOCKEDand mlocked pages off the normal LRU lists.
Quoted 8 lines
---
The entire stack, including Rik's split lru patches, are holding up very well under stress loads. E.g., ran for over 90+ hours over the weekend on both
x86_64[32GB, 8core] and ia64 [32GB, 16cpu] platforms without error over last weekend.I think these are ready for a spin in -mm atop Rik's patches.
I was >this< close to getting onto Rik's patches (honest) but a few other people have been kicking the tyres and seem to have caused some punctures so I'm expecting V9?
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 3 lines · On Thu, 29 May 2008 13:16:24 -0700 Andrew Morton <[email protected]>…
I was >this< close to getting onto Rik's patches (honest) but a few other people have been kicking the tyres and seem to have caused some punctures so I'm expecting V9?
If I send you a V9 up to patch 12, you can apply Lee's patches straight over my V9 :)
*fidgets with quilt mail*
Signature · 7 lines
All rights reversed. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 10 lines · Rik van Riel
Quoted 3 lines · On Thu, 29 May 2008 13:16:24 -0700 Andrew Morton <[email protected]>…
I was >this< close to getting onto Rik's patches (honest) but a few other people have been kicking the tyres and seem to have caused some punctures so I'm expecting V9?
If I send you a V9 up to patch 12, you can apply Lee's patches straight over my V9 :)
I failed to patch Lee's patches over your V9.
barrios@barrios-desktop:~/linux-2.6$ patch -p1 < /tmp/msg0_13.txt patching file mm/Kconfig patching file include/linux/page-flags.h patching file include/linux/mmzone.h patching file mm/page_alloc.c patching file include/linux/mm_inline.h patching file include/linux/swap.h patching file include/linux/pagevec.h patching file mm/swap.c patching file mm/migrate.c patching file mm/vmscan.c Hunk #10 FAILED at 1162. Hunk #11 succeeded at 1210 (offset 3 lines). Hunk #12 succeeded at 1242 (offset 3 lines). Hunk #13 succeeded at 1380 (offset 3 lines). Hunk #14 succeeded at 1411 (offset 3 lines). Hunk #15 succeeded at 1962 (offset 3 lines). Hunk #16 succeeded at 2300 (offset 3 lines). 1 out of 16 hunks FAILED -- saving rejects to file mm/vmscan.c.rej patching file mm/mempolicy.c patching file mm/internal.h patching file mm/memcontrol.c patching file include/linux/memcontrol.h
Signature · 8 lines
Kinds regards, MinChan Kim -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Rik" == Rik van Riel <[email protected]> writes:
Rik> On Thu, 29 May 2008 13:16:24 -0700
Quoted 3 lines · Rik> Andrew Morton <[email protected]> wrote:
Quoted 3 lines
I was >this< close to getting onto Rik's patches (honest) but a few other people have been kicking the tyres and seem to have caused some punctures so I'm expecting V9?
Rik> If I send you a V9 up to patch 12, you can apply Lee's patches Rik> straight over my V9 :)
I haven't seen any performance numbers talking about how well this stuff works on single or dual CPU machines with smaller amounts of memory, or whether it's worth using on these machines at all?
The big machines with lots of memory and lots of CPUs are certainly becoming more prevalent, but for my home machine with 4Gb RAM and dual core, what's the advantage?
Let's not slow down the common case for the sake of the bigger guys if possible.
John
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 10 lines · On Fri, 30 May 2008 09:52:48 -0400 "John Stoffel" <[email protected]> wrote:
I haven't seen any performance numbers talking about how well this stuff works on single or dual CPU machines with smaller amounts of memory, or whether it's worth using on these machines at all?
The big machines with lots of memory and lots of CPUs are certainly becoming more prevalent, but for my home machine with 4Gb RAM and dual core, what's the advantage?
Let's not slow down the common case for the sake of the bigger guys if possible.
I wouldn't call your home system with 4GB RAM "small".
After all, the VM that Linux currently has was developed mostly on machines with less than 1GB of RAM and later encrusted in bandaids to make sure the large systems did not fail too badly.
As for small system performance, I believe that my patch series should cause no performance regressions on those systems and has a framework that allows us to improve performance on those systems too.
If you manage to break performance with my patch set somehow, please let me know so I can fix it. Something like the VM is very subtle and any change is pretty much guaranteed to break something, so I am very interested in feedback.
Signature · 7 lines
All rights reversed. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Rik> On Fri, 30 May 2008 09:52:48 -0400
Quoted 10 lines · Rik> "John Stoffel" <[email protected]> wrote:
Quoted 10 lines
I haven't seen any performance numbers talking about how well this stuff works on single or dual CPU machines with smaller amounts of memory, or whether it's worth using on these machines at all?
The big machines with lots of memory and lots of CPUs are certainly becoming more prevalent, but for my home machine with 4Gb RAM and dual core, what's the advantage?
Let's not slow down the common case for the sake of the bigger guys if possible.
Rik> I wouldn't call your home system with 4GB RAM "small".
*grin* me either in some ways. But my other main linux box, which acts as an NFS server has 2Gb of RAM, but a pair of PIII Xeons at 550mhz. This is the box I'd be worried about in some ways, since it handles a bunch of stuff like backups, mysql, apache, NFS server, etc.
Rik> After all, the VM that Linux currently has was developed mostly Rik> on machines with less than 1GB of RAM and later encrusted in Rik> bandaids to make sure the large systems did not fail too badly.
Sure, I understand.
Rik> As for small system performance, I believe that my patch series Rik> should cause no performance regressions on those systems and has Rik> a framework that allows us to improve performance on those Rik> systems too.
Great! It would be nice to just be able to track this nicely.
Rik> If you manage to break performance with my patch set somehow, Rik> please let me know so I can fix it. Something like the VM is Rik> very subtle and any change is pretty much guaranteed to break Rik> something, so I am very interested in feedback.
What are you using to test/benchmark your changes as you develop this patchset? What would you suggest as a test load to help check performance?
John
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 8 lines · On Fri, 30 May 2008 10:36:05 -0400 "John Stoffel" <[email protected]> wrote:
Rik> If you manage to break performance with my patch set somehow, Rik> please let me know so I can fix it. Something like the VM is Rik> very subtle and any change is pretty much guaranteed to break Rik> something, so I am very interested in feedback.
What are you using to test/benchmark your changes as you develop this patchset? What would you suggest as a test load to help check performance?
Your normal workload.
I am doing some IO throughput, swap throughput and database tests, however those are probably not representative of what YOU throw at the VM.
There are no VM benchmarks that cover everything, so what is needed most at this point is real world exposure. I cannot promise that the code is perfect; all I can promise is that I will try to fix any performance issue that people find.
Signature · 7 lines
All rights reversed. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 4 lines
The entire stack, including Rik's split lru patches, are holding up very well under stress loads. E.g., ran for over 90+ hours over the weekend on both
x86_64[32GB, 8core] and ia64 [32GB, 16cpu] platforms without error over last weekend.
Note:
On fujitsu server(IA64 8CPU 8GB), this patch series works well 48+ hours too :)
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>