diff options
-rw-r--r-- | MAINTAINERS | 2 | ||||
-rw-r--r-- | drivers/leds/leds-pca9532.c | 2 | ||||
-rw-r--r-- | drivers/mfd/asic3.c | 1 | ||||
-rw-r--r-- | drivers/mfd/htc-pasic3.c | 1 | ||||
-rw-r--r-- | drivers/w1/masters/ds1wm.c | 5 | ||||
-rw-r--r-- | fs/btrfs/ctree.h | 5 | ||||
-rw-r--r-- | fs/btrfs/inode.c | 12 | ||||
-rw-r--r-- | fs/btrfs/super.c | 6 | ||||
-rw-r--r-- | fs/btrfs/volumes.c | 3 | ||||
-rw-r--r-- | include/linux/mfd/ds1wm.h | 7 | ||||
-rw-r--r-- | mm/memcontrol.c | 140 | ||||
-rw-r--r-- | mm/memory.c | 1 | ||||
-rw-r--r-- | mm/nommu.c | 9 | ||||
-rw-r--r-- | mm/vmscan.c | 59 |
14 files changed, 181 insertions, 72 deletions
diff --git a/MAINTAINERS b/MAINTAINERS index 9820e89c827c..4af7ab032660 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4982,7 +4982,7 @@ F: drivers/power/power_supply* PNP SUPPORT M: Adam Belay <abelay@mit.edu> -M: Bjorn Helgaas <bjorn.helgaas@hp.com> +M: Bjorn Helgaas <bhelgaas@google.com> S: Maintained F: drivers/pnp/ diff --git a/drivers/leds/leds-pca9532.c b/drivers/leds/leds-pca9532.c index d8d3a1e910a1..a2c874623e35 100644 --- a/drivers/leds/leds-pca9532.c +++ b/drivers/leds/leds-pca9532.c @@ -88,7 +88,7 @@ static const struct pca9532_chip_info pca9532_chip_info_tbl[] = { static struct i2c_driver pca9532_driver = { .driver = { - .name = "pca953x", + .name = "leds-pca953x", }, .probe = pca9532_probe, .remove = pca9532_remove, diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c index c27fd1fc3b86..c71ae09430c5 100644 --- a/drivers/mfd/asic3.c +++ b/drivers/mfd/asic3.c @@ -619,6 +619,7 @@ static void asic3_clk_disable(struct asic3 *asic, struct asic3_clk *clk) /* MFD cells (SPI, PWM, LED, DS1WM, MMC) */ static struct ds1wm_driver_data ds1wm_pdata = { .active_high = 1, + .reset_recover_delay = 1, }; static struct resource ds1wm_resources[] = { diff --git a/drivers/mfd/htc-pasic3.c b/drivers/mfd/htc-pasic3.c index 2808bd125d13..04c7093d6499 100644 --- a/drivers/mfd/htc-pasic3.c +++ b/drivers/mfd/htc-pasic3.c @@ -99,6 +99,7 @@ static int ds1wm_disable(struct platform_device *pdev) static struct ds1wm_driver_data ds1wm_pdata = { .active_high = 0, + .reset_recover_delay = 1, }; static struct resource ds1wm_resources[] __initdata = { diff --git a/drivers/w1/masters/ds1wm.c b/drivers/w1/masters/ds1wm.c index ad57593d224a..a0c8965c1a79 100644 --- a/drivers/w1/masters/ds1wm.c +++ b/drivers/w1/masters/ds1wm.c @@ -109,6 +109,7 @@ struct ds1wm_data { /* byte to write that makes all intr disabled, */ /* considering active_state (IAS) (optimization) */ u8 int_en_reg_none; + unsigned int reset_recover_delay; /* see ds1wm.h */ }; static inline void ds1wm_write_register(struct ds1wm_data *ds1wm_data, u32 reg, @@ -187,6 +188,9 @@ static int ds1wm_reset(struct ds1wm_data *ds1wm_data) return 1; } + if (ds1wm_data->reset_recover_delay) + msleep(ds1wm_data->reset_recover_delay); + return 0; } @@ -490,6 +494,7 @@ static int ds1wm_probe(struct platform_device *pdev) } ds1wm_data->irq = res->start; ds1wm_data->int_en_reg_none = (plat->active_high ? DS1WM_INTEN_IAS : 0); + ds1wm_data->reset_recover_delay = plat->reset_recover_delay; if (res->flags & IORESOURCE_IRQ_HIGHEDGE) irq_set_irq_type(ds1wm_data->irq, IRQ_TYPE_EDGE_RISING); diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index f30ac05dbda7..3b859a3e6a0e 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1335,6 +1335,11 @@ struct btrfs_ioctl_defrag_range_args { */ #define BTRFS_STRING_ITEM_KEY 253 +/* + * Flags for mount options. + * + * Note: don't forget to add new options to btrfs_show_options() + */ #define BTRFS_MOUNT_NODATASUM (1 << 0) #define BTRFS_MOUNT_NODATACOW (1 << 1) #define BTRFS_MOUNT_NOBARRIER (1 << 2) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d340f63d8f07..3601f0aebddf 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2678,12 +2678,14 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, int ret; /* - * If root is tree root, it means this inode is used to - * store free space information. And these inodes are updated - * when committing the transaction, so they needn't delaye to - * be updated, or deadlock will occured. + * If the inode is a free space inode, we can deadlock during commit + * if we put it into the delayed code. + * + * The data relocation inode should also be directly updated + * without delay */ - if (!is_free_space_inode(root, inode)) { + if (!is_free_space_inode(root, inode) + && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) { ret = btrfs_delayed_update_inode(trans, root, inode); if (!ret) btrfs_set_inode_last_trans(trans, inode); diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 0bb4ebbb71b7..15634d4648d7 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -723,6 +723,12 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",clear_cache"); if (btrfs_test_opt(root, USER_SUBVOL_RM_ALLOWED)) seq_puts(seq, ",user_subvol_rm_allowed"); + if (btrfs_test_opt(root, ENOSPC_DEBUG)) + seq_puts(seq, ",enospc_debug"); + if (btrfs_test_opt(root, AUTO_DEFRAG)) + seq_puts(seq, ",autodefrag"); + if (btrfs_test_opt(root, INODE_MAP_CACHE)) + seq_puts(seq, ",inode_cache"); return 0; } diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 1efa56e18f9b..19450bc53632 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -2098,7 +2098,8 @@ int btrfs_balance(struct btrfs_root *dev_root) chunk_root->root_key.objectid, found_key.objectid, found_key.offset); - BUG_ON(ret && ret != -ENOSPC); + if (ret && ret != -ENOSPC) + goto error; key.offset = found_key.offset - 1; } ret = 0; diff --git a/include/linux/mfd/ds1wm.h b/include/linux/mfd/ds1wm.h index be469a357cbb..38a372a0e285 100644 --- a/include/linux/mfd/ds1wm.h +++ b/include/linux/mfd/ds1wm.h @@ -3,4 +3,11 @@ struct ds1wm_driver_data { int active_high; int clock_rate; + /* in milliseconds, the amount of time to */ + /* sleep following a reset pulse. Zero */ + /* should work if your bus devices recover*/ + /* time respects the 1-wire spec since the*/ + /* ds1wm implements the precise timings of*/ + /* a reset pulse/presence detect sequence.*/ + unsigned int reset_recover_delay; }; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index ddffc74cdebe..e013b8e57d25 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -108,10 +108,12 @@ enum mem_cgroup_events_index { enum mem_cgroup_events_target { MEM_CGROUP_TARGET_THRESH, MEM_CGROUP_TARGET_SOFTLIMIT, + MEM_CGROUP_TARGET_NUMAINFO, MEM_CGROUP_NTARGETS, }; #define THRESHOLDS_EVENTS_TARGET (128) #define SOFTLIMIT_EVENTS_TARGET (1024) +#define NUMAINFO_EVENTS_TARGET (1024) struct mem_cgroup_stat_cpu { long count[MEM_CGROUP_STAT_NSTATS]; @@ -237,7 +239,8 @@ struct mem_cgroup { int last_scanned_node; #if MAX_NUMNODES > 1 nodemask_t scan_nodes; - unsigned long next_scan_node_update; + atomic_t numainfo_events; + atomic_t numainfo_updating; #endif /* * Should the accounting and control be hierarchical, per subtree? @@ -577,15 +580,6 @@ static long mem_cgroup_read_stat(struct mem_cgroup *mem, return val; } -static long mem_cgroup_local_usage(struct mem_cgroup *mem) -{ - long ret; - - ret = mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_RSS); - ret += mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_CACHE); - return ret; -} - static void mem_cgroup_swap_statistics(struct mem_cgroup *mem, bool charge) { @@ -689,6 +683,9 @@ static void __mem_cgroup_target_update(struct mem_cgroup *mem, int target) case MEM_CGROUP_TARGET_SOFTLIMIT: next = val + SOFTLIMIT_EVENTS_TARGET; break; + case MEM_CGROUP_TARGET_NUMAINFO: + next = val + NUMAINFO_EVENTS_TARGET; + break; default: return; } @@ -707,11 +704,19 @@ static void memcg_check_events(struct mem_cgroup *mem, struct page *page) mem_cgroup_threshold(mem); __mem_cgroup_target_update(mem, MEM_CGROUP_TARGET_THRESH); if (unlikely(__memcg_event_check(mem, - MEM_CGROUP_TARGET_SOFTLIMIT))){ + MEM_CGROUP_TARGET_SOFTLIMIT))) { mem_cgroup_update_tree(mem, page); __mem_cgroup_target_update(mem, - MEM_CGROUP_TARGET_SOFTLIMIT); + MEM_CGROUP_TARGET_SOFTLIMIT); + } +#if MAX_NUMNODES > 1 + if (unlikely(__memcg_event_check(mem, + MEM_CGROUP_TARGET_NUMAINFO))) { + atomic_inc(&mem->numainfo_events); + __mem_cgroup_target_update(mem, + MEM_CGROUP_TARGET_NUMAINFO); } +#endif } } @@ -1129,7 +1134,6 @@ unsigned long mem_cgroup_zone_nr_lru_pages(struct mem_cgroup *memcg, return MEM_CGROUP_ZSTAT(mz, lru); } -#ifdef CONFIG_NUMA static unsigned long mem_cgroup_node_nr_file_lru_pages(struct mem_cgroup *memcg, int nid) { @@ -1141,6 +1145,17 @@ static unsigned long mem_cgroup_node_nr_file_lru_pages(struct mem_cgroup *memcg, return ret; } +static unsigned long mem_cgroup_node_nr_anon_lru_pages(struct mem_cgroup *memcg, + int nid) +{ + unsigned long ret; + + ret = mem_cgroup_get_zonestat_node(memcg, nid, LRU_INACTIVE_ANON) + + mem_cgroup_get_zonestat_node(memcg, nid, LRU_ACTIVE_ANON); + return ret; +} + +#if MAX_NUMNODES > 1 static unsigned long mem_cgroup_nr_file_lru_pages(struct mem_cgroup *memcg) { u64 total = 0; @@ -1152,17 +1167,6 @@ static unsigned long mem_cgroup_nr_file_lru_pages(struct mem_cgroup *memcg) return total; } -static unsigned long mem_cgroup_node_nr_anon_lru_pages(struct mem_cgroup *memcg, - int nid) -{ - unsigned long ret; - - ret = mem_cgroup_get_zonestat_node(memcg, nid, LRU_INACTIVE_ANON) + - mem_cgroup_get_zonestat_node(memcg, nid, LRU_ACTIVE_ANON); - - return ret; -} - static unsigned long mem_cgroup_nr_anon_lru_pages(struct mem_cgroup *memcg) { u64 total = 0; @@ -1559,6 +1563,28 @@ mem_cgroup_select_victim(struct mem_cgroup *root_mem) return ret; } +/** + * test_mem_cgroup_node_reclaimable + * @mem: the target memcg + * @nid: the node ID to be checked. + * @noswap : specify true here if the user wants flle only information. + * + * This function returns whether the specified memcg contains any + * reclaimable pages on a node. Returns true if there are any reclaimable + * pages in the node. + */ +static bool test_mem_cgroup_node_reclaimable(struct mem_cgroup *mem, + int nid, bool noswap) +{ + if (mem_cgroup_node_nr_file_lru_pages(mem, nid)) + return true; + if (noswap || !total_swap_pages) + return false; + if (mem_cgroup_node_nr_anon_lru_pages(mem, nid)) + return true; + return false; + +} #if MAX_NUMNODES > 1 /* @@ -1570,26 +1596,26 @@ mem_cgroup_select_victim(struct mem_cgroup *root_mem) static void mem_cgroup_may_update_nodemask(struct mem_cgroup *mem) { int nid; - - if (time_after(mem->next_scan_node_update, jiffies)) + /* + * numainfo_events > 0 means there was at least NUMAINFO_EVENTS_TARGET + * pagein/pageout changes since the last update. + */ + if (!atomic_read(&mem->numainfo_events)) + return; + if (atomic_inc_return(&mem->numainfo_updating) > 1) return; - mem->next_scan_node_update = jiffies + 10*HZ; /* make a nodemask where this memcg uses memory from */ mem->scan_nodes = node_states[N_HIGH_MEMORY]; for_each_node_mask(nid, node_states[N_HIGH_MEMORY]) { - if (mem_cgroup_get_zonestat_node(mem, nid, LRU_INACTIVE_FILE) || - mem_cgroup_get_zonestat_node(mem, nid, LRU_ACTIVE_FILE)) - continue; - - if (total_swap_pages && - (mem_cgroup_get_zonestat_node(mem, nid, LRU_INACTIVE_ANON) || - mem_cgroup_get_zonestat_node(mem, nid, LRU_ACTIVE_ANON))) - continue; - node_clear(nid, mem->scan_nodes); + if (!test_mem_cgroup_node_reclaimable(mem, nid, false)) + node_clear(nid, mem->scan_nodes); } + + atomic_set(&mem->numainfo_events, 0); + atomic_set(&mem->numainfo_updating, 0); } /* @@ -1627,11 +1653,51 @@ int mem_cgroup_select_victim_node(struct mem_cgroup *mem) return node; } +/* + * Check all nodes whether it contains reclaimable pages or not. + * For quick scan, we make use of scan_nodes. This will allow us to skip + * unused nodes. But scan_nodes is lazily updated and may not cotain + * enough new information. We need to do double check. + */ +bool mem_cgroup_reclaimable(struct mem_cgroup *mem, bool noswap) +{ + int nid; + + /* + * quick check...making use of scan_node. + * We can skip unused nodes. + */ + if (!nodes_empty(mem->scan_nodes)) { + for (nid = first_node(mem->scan_nodes); + nid < MAX_NUMNODES; + nid = next_node(nid, mem->scan_nodes)) { + + if (test_mem_cgroup_node_reclaimable(mem, nid, noswap)) + return true; + } + } + /* + * Check rest of nodes. + */ + for_each_node_state(nid, N_HIGH_MEMORY) { + if (node_isset(nid, mem->scan_nodes)) + continue; + if (test_mem_cgroup_node_reclaimable(mem, nid, noswap)) + return true; + } + return false; +} + #else int mem_cgroup_select_victim_node(struct mem_cgroup *mem) { return 0; } + +bool mem_cgroup_reclaimable(struct mem_cgroup *mem, bool noswap) +{ + return test_mem_cgroup_node_reclaimable(mem, 0, noswap); +} #endif /* @@ -1702,7 +1768,7 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem, } } } - if (!mem_cgroup_local_usage(victim)) { + if (!mem_cgroup_reclaimable(victim, noswap)) { /* this cgroup's local usage == 0 */ css_put(&victim->css); continue; diff --git a/mm/memory.c b/mm/memory.c index 40b7531ee8ba..9b8a01d941cb 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -305,6 +305,7 @@ int __tlb_remove_page(struct mmu_gather *tlb, struct page *page) if (batch->nr == batch->max) { if (!tlb_next_batch(tlb)) return 0; + batch = tlb->active; } VM_BUG_ON(batch->nr > batch->max); diff --git a/mm/nommu.c b/mm/nommu.c index 1fd0c51b10a6..9edc897a3970 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -1813,10 +1813,13 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address, return NULL; } -int remap_pfn_range(struct vm_area_struct *vma, unsigned long from, - unsigned long to, unsigned long size, pgprot_t prot) +int remap_pfn_range(struct vm_area_struct *vma, unsigned long addr, + unsigned long pfn, unsigned long size, pgprot_t prot) { - vma->vm_start = vma->vm_pgoff << PAGE_SHIFT; + if (addr != (pfn << PAGE_SHIFT)) + return -EINVAL; + + vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP; return 0; } EXPORT_SYMBOL(remap_pfn_range); diff --git a/mm/vmscan.c b/mm/vmscan.c index 4f49535d4cd3..5ed24b94c5e6 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2326,7 +2326,7 @@ static bool sleeping_prematurely(pg_data_t *pgdat, int order, long remaining, return true; /* Check the watermark levels */ - for (i = 0; i < pgdat->nr_zones; i++) { + for (i = 0; i <= classzone_idx; i++) { struct zone *zone = pgdat->node_zones + i; if (!populated_zone(zone)) @@ -2344,7 +2344,7 @@ static bool sleeping_prematurely(pg_data_t *pgdat, int order, long remaining, } if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), - classzone_idx, 0)) + i, 0)) all_zones_ok = false; else balanced += zone->present_pages; @@ -2451,7 +2451,6 @@ loop_again: if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), 0, 0)) { end_zone = i; - *classzone_idx = i; break; } } @@ -2510,18 +2509,18 @@ loop_again: KSWAPD_ZONE_BALANCE_GAP_RATIO); if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone) + balance_gap, - end_zone, 0)) + end_zone, 0)) { shrink_zone(priority, zone, &sc); - reclaim_state->reclaimed_slab = 0; - nr_slab = shrink_slab(&shrink, sc.nr_scanned, lru_pages); - sc.nr_reclaimed += reclaim_state->reclaimed_slab; - total_scanned += sc.nr_scanned; - if (zone->all_unreclaimable) - continue; - if (nr_slab == 0 && - !zone_reclaimable(zone)) - zone->all_unreclaimable = 1; + reclaim_state->reclaimed_slab = 0; + nr_slab = shrink_slab(&shrink, sc.nr_scanned, lru_pages); + sc.nr_reclaimed += reclaim_state->reclaimed_slab; + total_scanned += sc.nr_scanned; + + if (nr_slab == 0 && !zone_reclaimable(zone)) + zone->all_unreclaimable = 1; + } + /* * If we've done a decent amount of scanning and * the reclaim ratio is low, start doing writepage @@ -2531,6 +2530,12 @@ loop_again: total_scanned > sc.nr_reclaimed + sc.nr_reclaimed / 2) sc.may_writepage = 1; + if (zone->all_unreclaimable) { + if (end_zone && end_zone == i) + end_zone--; + continue; + } + if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), end_zone, 0)) { all_zones_ok = 0; @@ -2709,8 +2714,8 @@ static void kswapd_try_to_sleep(pg_data_t *pgdat, int order, int classzone_idx) */ static int kswapd(void *p) { - unsigned long order; - int classzone_idx; + unsigned long order, new_order; + int classzone_idx, new_classzone_idx; pg_data_t *pgdat = (pg_data_t*)p; struct task_struct *tsk = current; @@ -2740,17 +2745,23 @@ static int kswapd(void *p) tsk->flags |= PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD; set_freezable(); - order = 0; - classzone_idx = MAX_NR_ZONES - 1; + order = new_order = 0; + classzone_idx = new_classzone_idx = pgdat->nr_zones - 1; for ( ; ; ) { - unsigned long new_order; - int new_classzone_idx; int ret; - new_order = pgdat->kswapd_max_order; - new_classzone_idx = pgdat->classzone_idx; - pgdat->kswapd_max_order = 0; - pgdat->classzone_idx = MAX_NR_ZONES - 1; + /* + * If the last balance_pgdat was unsuccessful it's unlikely a + * new request of a similar or harder type will succeed soon + * so consider going to sleep on the basis we reclaimed at + */ + if (classzone_idx >= new_classzone_idx && order == new_order) { + new_order = pgdat->kswapd_max_order; + new_classzone_idx = pgdat->classzone_idx; + pgdat->kswapd_max_order = 0; + pgdat->classzone_idx = pgdat->nr_zones - 1; + } + if (order < new_order || classzone_idx > new_classzone_idx) { /* * Don't sleep if someone wants a larger 'order' @@ -2763,7 +2774,7 @@ static int kswapd(void *p) order = pgdat->kswapd_max_order; classzone_idx = pgdat->classzone_idx; pgdat->kswapd_max_order = 0; - pgdat->classzone_idx = MAX_NR_ZONES - 1; + pgdat->classzone_idx = pgdat->nr_zones - 1; } ret = try_to_freeze(); |