From 818cf5909701806285d977f7a9365c5cadb062a7 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 23 Apr 2009 09:58:22 +0300 Subject: slub: enforce MAX_ORDER slub_max_order may not be equal to or greater than MAX_ORDER. Additionally, if a single object cannot be placed in a slab of slub_max_order, it still must allocate slabs below MAX_ORDER. Acked-by: Christoph Lameter Signed-off-by: David Rientjes Signed-off-by: Pekka Enberg --- mm/slub.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/slub.c b/mm/slub.c index 7ab54ecbd3f3..0e1247ed2a06 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -1909,7 +1909,7 @@ static inline int calculate_order(int size) * Doh this slab cannot be placed using slub_max_order. */ order = slab_order(size, 1, MAX_ORDER, 1); - if (order <= MAX_ORDER) + if (order < MAX_ORDER) return order; return -ENOSYS; } @@ -2522,6 +2522,7 @@ __setup("slub_min_order=", setup_slub_min_order); static int __init setup_slub_max_order(char *str) { get_option(&str, &slub_max_order); + slub_max_order = min(slub_max_order, MAX_ORDER - 1); return 1; } -- cgit v1.2.3 From 1eb5ac6466d4be7b15b38ce3ab709600f1bc891f Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Tue, 5 May 2009 19:13:44 +1000 Subject: mm: SLUB fix reclaim_state SLUB does not correctly account reclaim_state.reclaimed_slab, so it will break memory reclaim. Account it like SLAB does. Cc: stable@kernel.org Cc: linux-mm@kvack.org Cc: Matt Mackall Acked-by: Christoph Lameter Signed-off-by: Nick Piggin Signed-off-by: Pekka Enberg --- mm/slub.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'mm') diff --git a/mm/slub.c b/mm/slub.c index 7ab54ecbd3f3..aa34913a41ea 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -9,6 +9,7 @@ */ #include +#include /* struct reclaim_state */ #include #include #include @@ -1170,6 +1171,8 @@ static void __free_slab(struct kmem_cache *s, struct page *page) __ClearPageSlab(page); reset_page_mapcount(page); + if (current->reclaim_state) + current->reclaim_state->reclaimed_slab += pages; __free_pages(page, order); } -- cgit v1.2.3 From 1f0532eb617d28f65c93593a1491f662f14f7eac Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Tue, 5 May 2009 19:13:45 +1000 Subject: mm: SLOB fix reclaim_state SLOB does not correctly account reclaim_state.reclaimed_slab, so it will break memory reclaim. Account it like SLAB does. Cc: stable@kernel.org Cc: linux-mm@kvack.org Acked-by: Matt Mackall Acked-by: Christoph Lameter Signed-off-by: Nick Piggin Signed-off-by: Pekka Enberg --- mm/slob.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/slob.c b/mm/slob.c index a2d4ab32198d..f92e66d558bd 100644 --- a/mm/slob.c +++ b/mm/slob.c @@ -60,6 +60,7 @@ #include #include #include +#include /* struct reclaim_state */ #include #include #include @@ -255,6 +256,8 @@ static void *slob_new_pages(gfp_t gfp, int order, int node) static void slob_free_pages(void *b, int order) { + if (current->reclaim_state) + current->reclaim_state->reclaimed_slab += 1 << order; free_pages((unsigned long)b, order); } @@ -407,7 +410,7 @@ static void slob_free(void *block, int size) spin_unlock_irqrestore(&slob_lock, flags); clear_slob_page(sp); free_slob_page(sp); - free_page((unsigned long)b); + slob_free_pages(b, 0); return; } -- cgit v1.2.3 From 0f181328287db30671e9997329cff71395d4af8b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 13 May 2009 08:29:12 -0700 Subject: Revert "Ignore madvise(MADV_WILLNEED) for hugetlbfs-backed regions" This reverts commit a425a638c858fd10370b573bde81df3ba500e271. Now that the previous commit removed the "readpage" actor for hugetlb files, read-ahead will no longer mess up the mapping, and there's no longer any reason to treat hugetlbfs mappings specially. Tested-and-acked-by: Mel Gorman Signed-off-by: Linus Torvalds --- mm/madvise.c | 8 -------- 1 file changed, 8 deletions(-) (limited to 'mm') diff --git a/mm/madvise.c b/mm/madvise.c index 36d6ea2b6340..b9ce574827c8 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -112,14 +112,6 @@ static long madvise_willneed(struct vm_area_struct * vma, if (!file) return -EBADF; - /* - * Page cache readahead assumes page cache pages are order-0 which - * is not the case for hugetlbfs. Do not give a bad return value - * but ignore the advice. - */ - if (vma->vm_flags & VM_HUGETLB) - return 0; - if (file->f_mapping->a_ops->get_xip_mem) { /* no bad return value, but ignore advice */ return 0; -- cgit v1.2.3 From cd17cbfda004fe5f406c01b318c6378d9895896f Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 15 May 2009 11:32:24 +0200 Subject: Revert "mm: add /proc controls for pdflush threads" This reverts commit fafd688e4c0c34da0f3de909881117d374e4c7af. Work is progressing to switch away from pdflush as the process backing for flushing out dirty data. So it seems pointless to add more knobs to control pdflush threads. The original author of the patch did not have any specific use cases for adding the knobs, so we can easily revert this before 2.6.30 to avoid having to maintain this API forever. Signed-off-by: Jens Axboe --- Documentation/sysctl/vm.txt | 28 ---------------------------- include/linux/writeback.h | 2 -- kernel/sysctl.c | 23 ----------------------- mm/pdflush.c | 31 ++++++++++++------------------- 4 files changed, 12 insertions(+), 72 deletions(-) (limited to 'mm') diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt index b716d33912d8..c302ddf629a0 100644 --- a/Documentation/sysctl/vm.txt +++ b/Documentation/sysctl/vm.txt @@ -39,8 +39,6 @@ Currently, these files are in /proc/sys/vm: - nr_hugepages - nr_overcommit_hugepages - nr_pdflush_threads -- nr_pdflush_threads_min -- nr_pdflush_threads_max - nr_trim_pages (only if CONFIG_MMU=n) - numa_zonelist_order - oom_dump_tasks @@ -469,32 +467,6 @@ The default value is 0. ============================================================== -nr_pdflush_threads_min - -This value controls the minimum number of pdflush threads. - -At boot time, the kernel will create and maintain 'nr_pdflush_threads_min' -threads for the kernel's lifetime. - -The default value is 2. The minimum value you can specify is 1, and -the maximum value is the current setting of 'nr_pdflush_threads_max'. - -See 'nr_pdflush_threads_max' below for more information. - -============================================================== - -nr_pdflush_threads_max - -This value controls the maximum number of pdflush threads that can be -created. The pdflush algorithm will create a new pdflush thread (up to -this maximum) if no pdflush threads have been available for >= 1 second. - -The default value is 8. The minimum value you can specify is the -current value of 'nr_pdflush_threads_min' and the -maximum is 1000. - -============================================================== - overcommit_memory: This value contains a flag that enables memory overcommitment. diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 9c1ed1fb6ddb..93445477f86a 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -168,8 +168,6 @@ void writeback_set_ratelimit(void); /* pdflush.c */ extern int nr_pdflush_threads; /* Global so it can be exported to sysctl read-only. */ -extern int nr_pdflush_threads_max; /* Global so it can be exported to sysctl */ -extern int nr_pdflush_threads_min; /* Global so it can be exported to sysctl */ #endif /* WRITEBACK_H */ diff --git a/kernel/sysctl.c b/kernel/sysctl.c index ea78fa101ad6..b2970d56fb76 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -101,7 +101,6 @@ static int __maybe_unused one = 1; static int __maybe_unused two = 2; static unsigned long one_ul = 1; static int one_hundred = 100; -static int one_thousand = 1000; /* this is needed for the proc_doulongvec_minmax of vm_dirty_bytes */ static unsigned long dirty_bytes_min = 2 * PAGE_SIZE; @@ -1033,28 +1032,6 @@ static struct ctl_table vm_table[] = { .mode = 0444 /* read-only*/, .proc_handler = &proc_dointvec, }, - { - .ctl_name = CTL_UNNUMBERED, - .procname = "nr_pdflush_threads_min", - .data = &nr_pdflush_threads_min, - .maxlen = sizeof nr_pdflush_threads_min, - .mode = 0644 /* read-write */, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, - .extra1 = &one, - .extra2 = &nr_pdflush_threads_max, - }, - { - .ctl_name = CTL_UNNUMBERED, - .procname = "nr_pdflush_threads_max", - .data = &nr_pdflush_threads_max, - .maxlen = sizeof nr_pdflush_threads_max, - .mode = 0644 /* read-write */, - .proc_handler = &proc_dointvec_minmax, - .strategy = &sysctl_intvec, - .extra1 = &nr_pdflush_threads_min, - .extra2 = &one_thousand, - }, { .ctl_name = VM_SWAPPINESS, .procname = "swappiness", diff --git a/mm/pdflush.c b/mm/pdflush.c index f2caf96993f8..235ac440c44e 100644 --- a/mm/pdflush.c +++ b/mm/pdflush.c @@ -57,14 +57,6 @@ static DEFINE_SPINLOCK(pdflush_lock); */ int nr_pdflush_threads = 0; -/* - * The max/min number of pdflush threads. R/W by sysctl at - * /proc/sys/vm/nr_pdflush_threads_max/min - */ -int nr_pdflush_threads_max __read_mostly = MAX_PDFLUSH_THREADS; -int nr_pdflush_threads_min __read_mostly = MIN_PDFLUSH_THREADS; - - /* * The time at which the pdflush thread pool last went empty */ @@ -76,7 +68,7 @@ static unsigned long last_empty_jifs; * Thread pool management algorithm: * * - The minimum and maximum number of pdflush instances are bound - * by nr_pdflush_threads_min and nr_pdflush_threads_max. + * by MIN_PDFLUSH_THREADS and MAX_PDFLUSH_THREADS. * * - If there have been no idle pdflush instances for 1 second, create * a new one. @@ -142,13 +134,14 @@ static int __pdflush(struct pdflush_work *my_work) * To throttle creation, we reset last_empty_jifs. */ if (time_after(jiffies, last_empty_jifs + 1 * HZ)) { - if (list_empty(&pdflush_list) && - nr_pdflush_threads < nr_pdflush_threads_max) { - last_empty_jifs = jiffies; - nr_pdflush_threads++; - spin_unlock_irq(&pdflush_lock); - start_one_pdflush_thread(); - spin_lock_irq(&pdflush_lock); + if (list_empty(&pdflush_list)) { + if (nr_pdflush_threads < MAX_PDFLUSH_THREADS) { + last_empty_jifs = jiffies; + nr_pdflush_threads++; + spin_unlock_irq(&pdflush_lock); + start_one_pdflush_thread(); + spin_lock_irq(&pdflush_lock); + } } } @@ -160,7 +153,7 @@ static int __pdflush(struct pdflush_work *my_work) */ if (list_empty(&pdflush_list)) continue; - if (nr_pdflush_threads <= nr_pdflush_threads_min) + if (nr_pdflush_threads <= MIN_PDFLUSH_THREADS) continue; pdf = list_entry(pdflush_list.prev, struct pdflush_work, list); if (time_after(jiffies, pdf->when_i_went_to_sleep + 1 * HZ)) { @@ -266,9 +259,9 @@ static int __init pdflush_init(void) * Pre-set nr_pdflush_threads... If we fail to create, * the count will be decremented. */ - nr_pdflush_threads = nr_pdflush_threads_min; + nr_pdflush_threads = MIN_PDFLUSH_THREADS; - for (i = 0; i < nr_pdflush_threads_min; i++) + for (i = 0; i < MIN_PDFLUSH_THREADS; i++) start_one_pdflush_thread(); return 0; } -- cgit v1.2.3 From 22ef37eed673587ac984965dc88ba94c68873291 Mon Sep 17 00:00:00 2001 From: Toshiyuki Okajima Date: Sat, 16 May 2009 22:56:28 -0700 Subject: page-writeback: fix the calculation of the oldest_jif in wb_kupdate() wb_kupdate() function has a bug on linux-2.6.30-rc5. This bug causes generic_sync_sb_inodes() to start to write inodes back much earlier than our expectations because it miscalculates oldest_jif in wb_kupdate(). This bug was introduced in 704503d836042d4a4c7685b7036e7de0418fbc0f ('mm: fix proc_dointvec_userhz_jiffies "breakage"'). Signed-off-by: Toshiyuki Okajima Cc: Alexey Dobriyan Cc: Peter Zijlstra Cc: Nick Piggin Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page-writeback.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'mm') diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 30351f0063ac..bb553c3e955d 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -94,12 +94,12 @@ unsigned long vm_dirty_bytes; /* * The interval between `kupdate'-style writebacks */ -unsigned int dirty_writeback_interval = 5 * 100; /* sentiseconds */ +unsigned int dirty_writeback_interval = 5 * 100; /* centiseconds */ /* * The longest time for which data is allowed to remain dirty */ -unsigned int dirty_expire_interval = 30 * 100; /* sentiseconds */ +unsigned int dirty_expire_interval = 30 * 100; /* centiseconds */ /* * Flag that makes the machine dump writes/reads and block dirtyings. @@ -770,7 +770,7 @@ static void wb_kupdate(unsigned long arg) sync_supers(); - oldest_jif = jiffies - msecs_to_jiffies(dirty_expire_interval); + oldest_jif = jiffies - msecs_to_jiffies(dirty_expire_interval * 10); start_jif = jiffies; next_jif = start_jif + msecs_to_jiffies(dirty_writeback_interval * 10); nr_to_write = global_page_state(NR_FILE_DIRTY) + -- cgit v1.2.3 From eb33575cf67d3f35fa2510210ef92631266e2465 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 13 May 2009 17:34:48 +0100 Subject: [ARM] Double check memmap is actually valid with a memmap has unexpected holes V2 pfn_valid() is meant to be able to tell if a given PFN has valid memmap associated with it or not. In FLATMEM, it is expected that holes always have valid memmap as long as there is valid PFNs either side of the hole. In SPARSEMEM, it is assumed that a valid section has a memmap for the entire section. However, ARM and maybe other embedded architectures in the future free memmap backing holes to save memory on the assumption the memmap is never used. The page_zone linkages are then broken even though pfn_valid() returns true. A walker of the full memmap must then do this additional check to ensure the memmap they are looking at is sane by making sure the zone and PFN linkages are still valid. This is expensive, but walkers of the full memmap are extremely rare. This was caught before for FLATMEM and hacked around but it hits again for SPARSEMEM because the page_zone linkages can look ok where the PFN linkages are totally screwed. This looks like a hatchet job but the reality is that any clean solution would end up consumning all the memory saved by punching these unexpected holes in the memmap. For example, we tried marking the memmap within the section invalid but the section size exceeds the size of the hole in most cases so pfn_valid() starts returning false where valid memmap exists. Shrinking the size of the section would increase memory consumption offsetting the gains. This patch identifies when an architecture is punching unexpected holes in the memmap that the memory model cannot automatically detect and sets ARCH_HAS_HOLES_MEMORYMODEL. At the moment, this is restricted to EP93xx which is the model sub-architecture this has been reported on but may expand later. When set, walkers of the full memmap must call memmap_valid_within() for each PFN and passing in what it expects the page and zone to be for that PFN. If it finds the linkages to be broken, it assumes the memmap is invalid for that PFN. Signed-off-by: Mel Gorman Signed-off-by: Russell King --- arch/arm/Kconfig | 6 +++--- include/linux/mmzone.h | 26 ++++++++++++++++++++++++++ mm/mmzone.c | 15 +++++++++++++++ mm/vmstat.c | 19 ++++--------------- 4 files changed, 48 insertions(+), 18 deletions(-) (limited to 'mm') diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index e60ec54df334..9d02cdb15b23 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -273,6 +273,7 @@ config ARCH_EP93XX select HAVE_CLK select COMMON_CLKDEV select ARCH_REQUIRE_GPIOLIB + select ARCH_HAS_HOLES_MEMORYMODEL help This enables support for the Cirrus EP93xx series of CPUs. @@ -976,10 +977,9 @@ config OABI_COMPAT UNPREDICTABLE (in fact it can be predicted that it won't work at all). If in doubt say Y. -config ARCH_FLATMEM_HAS_HOLES +config ARCH_HAS_HOLES_MEMORYMODEL bool - default y - depends on FLATMEM + default n # Discontigmem is deprecated config ARCH_DISCONTIGMEM_ENABLE diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 186ec6ab334d..a47c879e1304 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -1097,6 +1097,32 @@ unsigned long __init node_memmap_size_bytes(int, unsigned long, unsigned long); #define pfn_valid_within(pfn) (1) #endif +#ifdef CONFIG_ARCH_HAS_HOLES_MEMORYMODEL +/* + * pfn_valid() is meant to be able to tell if a given PFN has valid memmap + * associated with it or not. In FLATMEM, it is expected that holes always + * have valid memmap as long as there is valid PFNs either side of the hole. + * In SPARSEMEM, it is assumed that a valid section has a memmap for the + * entire section. + * + * However, an ARM, and maybe other embedded architectures in the future + * free memmap backing holes to save memory on the assumption the memmap is + * never used. The page_zone linkages are then broken even though pfn_valid() + * returns true. A walker of the full memmap must then do this additional + * check to ensure the memmap they are looking at is sane by making sure + * the zone and PFN linkages are still valid. This is expensive, but walkers + * of the full memmap are extremely rare. + */ +int memmap_valid_within(unsigned long pfn, + struct page *page, struct zone *zone); +#else +static inline int memmap_valid_within(unsigned long pfn, + struct page *page, struct zone *zone) +{ + return 1; +} +#endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */ + #endif /* !__GENERATING_BOUNDS.H */ #endif /* !__ASSEMBLY__ */ #endif /* _LINUX_MMZONE_H */ diff --git a/mm/mmzone.c b/mm/mmzone.c index 16ce8b955dcf..f5b7d1760213 100644 --- a/mm/mmzone.c +++ b/mm/mmzone.c @@ -6,6 +6,7 @@ #include +#include #include #include @@ -72,3 +73,17 @@ struct zoneref *next_zones_zonelist(struct zoneref *z, *zone = zonelist_zone(z); return z; } + +#ifdef CONFIG_ARCH_HAS_HOLES_MEMORYMODEL +int memmap_valid_within(unsigned long pfn, + struct page *page, struct zone *zone) +{ + if (page_to_pfn(page) != pfn) + return 0; + + if (page_zone(page) != zone) + return 0; + + return 1; +} +#endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */ diff --git a/mm/vmstat.c b/mm/vmstat.c index 66f6130976cb..74d66dba0cbe 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -509,22 +509,11 @@ static void pagetypeinfo_showblockcount_print(struct seq_file *m, continue; page = pfn_to_page(pfn); -#ifdef CONFIG_ARCH_FLATMEM_HAS_HOLES - /* - * Ordinarily, memory holes in flatmem still have a valid - * memmap for the PFN range. However, an architecture for - * embedded systems (e.g. ARM) can free up the memmap backing - * holes to save memory on the assumption the memmap is - * never used. The page_zone linkages are then broken even - * though pfn_valid() returns true. Skip the page if the - * linkages are broken. Even if this test passed, the impact - * is that the counters for the movable type are off but - * fragmentation monitoring is likely meaningless on small - * systems. - */ - if (page_zone(page) != zone) + + /* Watch for unexpected holes punched in the memmap */ + if (!memmap_valid_within(pfn, page, zone)) continue; -#endif + mtype = get_pageblock_migratetype(page); if (mtype < MIGRATE_TYPES) -- cgit v1.2.3