From 11f81becca04bb7d2826a9b65bb8d27b0a1bb543 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:15 -0400 Subject: [PATCH 01/85] page_writeback: revive cancel_dirty_page() in a restricted form cancel_dirty_page() had some issues and b9ea25152e56 ("page_writeback: clean up mess around cancel_dirty_page()") replaced it with account_page_cleaned() which makes the caller responsible for clearing the dirty bit; unfortunately, the planned changes for cgroup writeback support requires synchronization between dirty bit manipulation and stat updates. While we can open-code such synchronization in each account_page_cleaned() callsite, that's gonna be unnecessarily awkward and verbose. This patch revives cancel_dirty_page() but in a more restricted form. All it does is TestClearPageDirty() followed by account_page_cleaned() invocation if the page was dirty. This helper covers all account_page_cleaned() usages except for __delete_from_page_cache() which is a special case anyway and left alone. As this leaves no module user for account_page_cleaned(), EXPORT_SYMBOL() is dropped from it. This patch just revives cancel_dirty_page() as a trivial wrapper to replace equivalent usages and doesn't introduce any functional changes. Signed-off-by: Tejun Heo Cc: Konstantin Khlebnikov Signed-off-by: Jens Axboe --- .../include/linux/lustre_patchless_compat.h | 4 +-- fs/buffer.c | 4 +-- include/linux/mm.h | 1 + mm/page-writeback.c | 27 ++++++++++++++----- mm/truncate.c | 4 +-- 5 files changed, 25 insertions(+), 15 deletions(-) diff --git a/drivers/staging/lustre/lustre/include/linux/lustre_patchless_compat.h b/drivers/staging/lustre/lustre/include/linux/lustre_patchless_compat.h index d72605864b0a..14562788e4e0 100644 --- a/drivers/staging/lustre/lustre/include/linux/lustre_patchless_compat.h +++ b/drivers/staging/lustre/lustre/include/linux/lustre_patchless_compat.h @@ -55,9 +55,7 @@ truncate_complete_page(struct address_space *mapping, struct page *page) if (PagePrivate(page)) page->mapping->a_ops->invalidatepage(page, 0, PAGE_CACHE_SIZE); - if (TestClearPageDirty(page)) - account_page_cleaned(page, mapping); - + cancel_dirty_page(page); ClearPageMappedToDisk(page); ll_delete_from_page_cache(page); } diff --git a/fs/buffer.c b/fs/buffer.c index f96173ad62d9..f21327d1f673 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -3232,8 +3232,8 @@ int try_to_free_buffers(struct page *page) * to synchronise against __set_page_dirty_buffers and prevent the * dirty bit from being lost. */ - if (ret && TestClearPageDirty(page)) - account_page_cleaned(page, mapping); + if (ret) + cancel_dirty_page(page); spin_unlock(&mapping->private_lock); out: if (buffers_to_free) { diff --git a/include/linux/mm.h b/include/linux/mm.h index 0755b9fd03a7..a83cf3a6f78e 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1215,6 +1215,7 @@ void account_page_dirtied(struct page *page, struct address_space *mapping); void account_page_cleaned(struct page *page, struct address_space *mapping); int set_page_dirty(struct page *page); int set_page_dirty_lock(struct page *page); +void cancel_dirty_page(struct page *page); int clear_page_dirty_for_io(struct page *page); int get_cmdline(struct task_struct *task, char *buffer, int buflen); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 5daf5568b9e1..227b867598e1 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -2112,12 +2112,6 @@ EXPORT_SYMBOL(account_page_dirtied); /* * Helper function for deaccounting dirty page without writeback. - * - * Doing this should *normally* only ever be done when a page - * is truncated, and is not actually mapped anywhere at all. However, - * fs/buffer.c does this when it notices that somebody has cleaned - * out all the buffers on a page without actually doing it through - * the VM. Can you say "ext3 is horribly ugly"? Thought you could. */ void account_page_cleaned(struct page *page, struct address_space *mapping) { @@ -2127,7 +2121,6 @@ void account_page_cleaned(struct page *page, struct address_space *mapping) task_io_account_cancelled_write(PAGE_CACHE_SIZE); } } -EXPORT_SYMBOL(account_page_cleaned); /* * For address_spaces which do not use buffers. Just tag the page as dirty in @@ -2265,6 +2258,26 @@ int set_page_dirty_lock(struct page *page) } EXPORT_SYMBOL(set_page_dirty_lock); +/* + * This cancels just the dirty bit on the kernel page itself, it does NOT + * actually remove dirty bits on any mmap's that may be around. It also + * leaves the page tagged dirty, so any sync activity will still find it on + * the dirty lists, and in particular, clear_page_dirty_for_io() will still + * look at the dirty bits in the VM. + * + * Doing this should *normally* only ever be done when a page is truncated, + * and is not actually mapped anywhere at all. However, fs/buffer.c does + * this when it notices that somebody has cleaned out all the buffers on a + * page without actually doing it through the VM. Can you say "ext3 is + * horribly ugly"? Thought you could. + */ +void cancel_dirty_page(struct page *page) +{ + if (TestClearPageDirty(page)) + account_page_cleaned(page, page_mapping(page)); +} +EXPORT_SYMBOL(cancel_dirty_page); + /* * Clear a page's dirty flag, while caring for dirty memory accounting. * Returns true if the page was previously dirty. diff --git a/mm/truncate.c b/mm/truncate.c index 66af9031fae8..0c360259c085 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -116,9 +116,7 @@ truncate_complete_page(struct address_space *mapping, struct page *page) * the VM has canceled the dirty bit (eg ext3 journaling). * Hence dirty accounting check is placed after invalidation. */ - if (TestClearPageDirty(page)) - account_page_cleaned(page, mapping); - + cancel_dirty_page(page); ClearPageMappedToDisk(page); delete_from_page_cache(page); return 0; From c4843a7593a9df3ff5b1806084cefdfa81dd7c79 Mon Sep 17 00:00:00 2001 From: Greg Thelen Date: Fri, 22 May 2015 17:13:16 -0400 Subject: [PATCH 02/85] memcg: add per cgroup dirty page accounting When modifying PG_Dirty on cached file pages, update the new MEM_CGROUP_STAT_DIRTY counter. This is done in the same places where global NR_FILE_DIRTY is managed. The new memcg stat is visible in the per memcg memory.stat cgroupfs file. The most recent past attempt at this was http://thread.gmane.org/gmane.linux.kernel.cgroups/8632 The new accounting supports future efforts to add per cgroup dirty page throttling and writeback. It also helps an administrator break down a container's memory usage and provides evidence to understand memcg oom kills (the new dirty count is included in memcg oom kill messages). The ability to move page accounting between memcg (memory.move_charge_at_immigrate) makes this accounting more complicated than the global counter. The existing mem_cgroup_{begin,end}_page_stat() lock is used to serialize move accounting with stat updates. Typical update operation: memcg = mem_cgroup_begin_page_stat(page) if (TestSetPageDirty()) { [...] mem_cgroup_update_page_stat(memcg) } mem_cgroup_end_page_stat(memcg) Summary of mem_cgroup_end_page_stat() overhead: - Without CONFIG_MEMCG it's a no-op - With CONFIG_MEMCG and no inter memcg task movement, it's just rcu_read_lock() - With CONFIG_MEMCG and inter memcg task movement, it's rcu_read_lock() + spin_lock_irqsave() A memcg parameter is added to several routines because their callers now grab mem_cgroup_begin_page_stat() which returns the memcg later needed by for mem_cgroup_update_page_stat(). Because mem_cgroup_begin_page_stat() may disable interrupts, some adjustments are needed: - move __mark_inode_dirty() from __set_page_dirty() to its caller. __mark_inode_dirty() locking does not want interrupts disabled. - use spin_lock_irqsave(tree_lock) rather than spin_lock_irq() in __delete_from_page_cache(), replace_page_cache_page(), invalidate_complete_page2(), and __remove_mapping(). text data bss dec hex filename 8925147 1774832 1785856 12485835 be84cb vmlinux-!CONFIG_MEMCG-before 8925339 1774832 1785856 12486027 be858b vmlinux-!CONFIG_MEMCG-after +192 text bytes 8965977 1784992 1785856 12536825 bf4bf9 vmlinux-CONFIG_MEMCG-before 8966750 1784992 1785856 12537598 bf4efe vmlinux-CONFIG_MEMCG-after +773 text bytes Performance tests run on v4.0-rc1-36-g4f671fe2f952. Lower is better for all metrics, they're all wall clock or cycle counts. The read and write fault benchmarks just measure fault time, they do not include I/O time. * CONFIG_MEMCG not set: baseline patched kbuild 1m25.030000(+-0.088% 3 samples) 1m25.426667(+-0.120% 3 samples) dd write 100 MiB 0.859211561 +-15.10% 0.874162885 +-15.03% dd write 200 MiB 1.670653105 +-17.87% 1.669384764 +-11.99% dd write 1000 MiB 8.434691190 +-14.15% 8.474733215 +-14.77% read fault cycles 254.0(+-0.000% 10 samples) 253.0(+-0.000% 10 samples) write fault cycles 2021.2(+-3.070% 10 samples) 1984.5(+-1.036% 10 samples) * CONFIG_MEMCG=y root_memcg: baseline patched kbuild 1m25.716667(+-0.105% 3 samples) 1m25.686667(+-0.153% 3 samples) dd write 100 MiB 0.855650830 +-14.90% 0.887557919 +-14.90% dd write 200 MiB 1.688322953 +-12.72% 1.667682724 +-13.33% dd write 1000 MiB 8.418601605 +-14.30% 8.673532299 +-15.00% read fault cycles 266.0(+-0.000% 10 samples) 266.0(+-0.000% 10 samples) write fault cycles 2051.7(+-1.349% 10 samples) 2049.6(+-1.686% 10 samples) * CONFIG_MEMCG=y non-root_memcg: baseline patched kbuild 1m26.120000(+-0.273% 3 samples) 1m25.763333(+-0.127% 3 samples) dd write 100 MiB 0.861723964 +-15.25% 0.818129350 +-14.82% dd write 200 MiB 1.669887569 +-13.30% 1.698645885 +-13.27% dd write 1000 MiB 8.383191730 +-14.65% 8.351742280 +-14.52% read fault cycles 265.7(+-0.172% 10 samples) 267.0(+-0.000% 10 samples) write fault cycles 2070.6(+-1.512% 10 samples) 2084.4(+-2.148% 10 samples) As expected anon page faults are not affected by this patch. tj: Updated to apply on top of the recent cancel_dirty_page() changes. Signed-off-by: Sha Zhengju Signed-off-by: Greg Thelen Signed-off-by: Tejun Heo Signed-off-by: Jens Axboe --- Documentation/cgroups/memory.txt | 1 + fs/buffer.c | 34 +++++++++++++++++----- fs/xfs/xfs_aops.c | 12 ++++++-- include/linux/memcontrol.h | 1 + include/linux/mm.h | 6 ++-- include/linux/pagemap.h | 3 +- mm/filemap.c | 31 ++++++++++++++------ mm/memcontrol.c | 24 ++++++++++++++- mm/page-writeback.c | 50 +++++++++++++++++++++++++++----- mm/rmap.c | 2 ++ mm/truncate.c | 14 ++++++--- mm/vmscan.c | 17 +++++++---- 12 files changed, 156 insertions(+), 39 deletions(-) diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt index f456b4315e86..ff71e16cc752 100644 --- a/Documentation/cgroups/memory.txt +++ b/Documentation/cgroups/memory.txt @@ -493,6 +493,7 @@ pgpgin - # of charging events to the memory cgroup. The charging pgpgout - # of uncharging events to the memory cgroup. The uncharging event happens each time a page is unaccounted from the cgroup. swap - # of bytes of swap usage +dirty - # of bytes that are waiting to get written back to the disk. writeback - # of bytes of file/anon cache that are queued for syncing to disk. inactive_anon - # of bytes of anonymous and swap cache memory on inactive diff --git a/fs/buffer.c b/fs/buffer.c index f21327d1f673..23b640d5d6e9 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -623,21 +623,22 @@ EXPORT_SYMBOL(mark_buffer_dirty_inode); * * If warn is true, then emit a warning if the page is not uptodate and has * not been truncated. + * + * The caller must hold mem_cgroup_begin_page_stat() lock. */ -static void __set_page_dirty(struct page *page, - struct address_space *mapping, int warn) +static void __set_page_dirty(struct page *page, struct address_space *mapping, + struct mem_cgroup *memcg, int warn) { unsigned long flags; spin_lock_irqsave(&mapping->tree_lock, flags); if (page->mapping) { /* Race with truncate? */ WARN_ON_ONCE(warn && !PageUptodate(page)); - account_page_dirtied(page, mapping); + account_page_dirtied(page, mapping, memcg); radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); } spin_unlock_irqrestore(&mapping->tree_lock, flags); - __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); } /* @@ -668,6 +669,7 @@ static void __set_page_dirty(struct page *page, int __set_page_dirty_buffers(struct page *page) { int newly_dirty; + struct mem_cgroup *memcg; struct address_space *mapping = page_mapping(page); if (unlikely(!mapping)) @@ -683,11 +685,22 @@ int __set_page_dirty_buffers(struct page *page) bh = bh->b_this_page; } while (bh != head); } + /* + * Use mem_group_begin_page_stat() to keep PageDirty synchronized with + * per-memcg dirty page counters. + */ + memcg = mem_cgroup_begin_page_stat(page); newly_dirty = !TestSetPageDirty(page); spin_unlock(&mapping->private_lock); if (newly_dirty) - __set_page_dirty(page, mapping, 1); + __set_page_dirty(page, mapping, memcg, 1); + + mem_cgroup_end_page_stat(memcg); + + if (newly_dirty) + __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); + return newly_dirty; } EXPORT_SYMBOL(__set_page_dirty_buffers); @@ -1158,11 +1171,18 @@ void mark_buffer_dirty(struct buffer_head *bh) if (!test_set_buffer_dirty(bh)) { struct page *page = bh->b_page; + struct address_space *mapping = NULL; + struct mem_cgroup *memcg; + + memcg = mem_cgroup_begin_page_stat(page); if (!TestSetPageDirty(page)) { - struct address_space *mapping = page_mapping(page); + mapping = page_mapping(page); if (mapping) - __set_page_dirty(page, mapping, 0); + __set_page_dirty(page, mapping, memcg, 0); } + mem_cgroup_end_page_stat(memcg); + if (mapping) + __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); } } EXPORT_SYMBOL(mark_buffer_dirty); diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 095f94c2d8b5..e5099f268032 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -1873,6 +1873,7 @@ xfs_vm_set_page_dirty( loff_t end_offset; loff_t offset; int newly_dirty; + struct mem_cgroup *memcg; if (unlikely(!mapping)) return !TestSetPageDirty(page); @@ -1892,6 +1893,11 @@ xfs_vm_set_page_dirty( offset += 1 << inode->i_blkbits; } while (bh != head); } + /* + * Use mem_group_begin_page_stat() to keep PageDirty synchronized with + * per-memcg dirty page counters. + */ + memcg = mem_cgroup_begin_page_stat(page); newly_dirty = !TestSetPageDirty(page); spin_unlock(&mapping->private_lock); @@ -1902,13 +1908,15 @@ xfs_vm_set_page_dirty( spin_lock_irqsave(&mapping->tree_lock, flags); if (page->mapping) { /* Race with truncate? */ WARN_ON_ONCE(!PageUptodate(page)); - account_page_dirtied(page, mapping); + account_page_dirtied(page, mapping, memcg); radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); } spin_unlock_irqrestore(&mapping->tree_lock, flags); - __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); } + mem_cgroup_end_page_stat(memcg); + if (newly_dirty) + __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); return newly_dirty; } diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 72dff5fb0d0c..5fe6411b5e54 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -41,6 +41,7 @@ enum mem_cgroup_stat_index { MEM_CGROUP_STAT_RSS, /* # of pages charged as anon rss */ MEM_CGROUP_STAT_RSS_HUGE, /* # of pages charged as anon huge */ MEM_CGROUP_STAT_FILE_MAPPED, /* # of pages charged as file rss */ + MEM_CGROUP_STAT_DIRTY, /* # of dirty pages in page cache */ MEM_CGROUP_STAT_WRITEBACK, /* # of pages under writeback */ MEM_CGROUP_STAT_SWAP, /* # of pages, swapped out */ MEM_CGROUP_STAT_NSTATS, diff --git a/include/linux/mm.h b/include/linux/mm.h index a83cf3a6f78e..f48d979ced4b 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1211,8 +1211,10 @@ int __set_page_dirty_nobuffers(struct page *page); int __set_page_dirty_no_writeback(struct page *page); int redirty_page_for_writepage(struct writeback_control *wbc, struct page *page); -void account_page_dirtied(struct page *page, struct address_space *mapping); -void account_page_cleaned(struct page *page, struct address_space *mapping); +void account_page_dirtied(struct page *page, struct address_space *mapping, + struct mem_cgroup *memcg); +void account_page_cleaned(struct page *page, struct address_space *mapping, + struct mem_cgroup *memcg); int set_page_dirty(struct page *page); int set_page_dirty_lock(struct page *page); void cancel_dirty_page(struct page *page); diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index 4b3736f7065c..fb0814ca65c7 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -651,7 +651,8 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping, int add_to_page_cache_lru(struct page *page, struct address_space *mapping, pgoff_t index, gfp_t gfp_mask); extern void delete_from_page_cache(struct page *page); -extern void __delete_from_page_cache(struct page *page, void *shadow); +extern void __delete_from_page_cache(struct page *page, void *shadow, + struct mem_cgroup *memcg); int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask); /* diff --git a/mm/filemap.c b/mm/filemap.c index 6bf5e42d560a..7b1443dc3ad0 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -100,6 +100,7 @@ * ->tree_lock (page_remove_rmap->set_page_dirty) * bdi.wb->list_lock (page_remove_rmap->set_page_dirty) * ->inode->i_lock (page_remove_rmap->set_page_dirty) + * ->memcg->move_lock (page_remove_rmap->mem_cgroup_begin_page_stat) * bdi.wb->list_lock (zap_pte_range->set_page_dirty) * ->inode->i_lock (zap_pte_range->set_page_dirty) * ->private_lock (zap_pte_range->__set_page_dirty_buffers) @@ -174,9 +175,11 @@ static void page_cache_tree_delete(struct address_space *mapping, /* * Delete a page from the page cache and free it. Caller has to make * sure the page is locked and that nobody else uses it - or that usage - * is safe. The caller must hold the mapping's tree_lock. + * is safe. The caller must hold the mapping's tree_lock and + * mem_cgroup_begin_page_stat(). */ -void __delete_from_page_cache(struct page *page, void *shadow) +void __delete_from_page_cache(struct page *page, void *shadow, + struct mem_cgroup *memcg) { struct address_space *mapping = page->mapping; @@ -210,7 +213,7 @@ void __delete_from_page_cache(struct page *page, void *shadow) * anyway will be cleared before returning page into buddy allocator. */ if (WARN_ON_ONCE(PageDirty(page))) - account_page_cleaned(page, mapping); + account_page_cleaned(page, mapping, memcg); } /** @@ -224,14 +227,20 @@ void __delete_from_page_cache(struct page *page, void *shadow) void delete_from_page_cache(struct page *page) { struct address_space *mapping = page->mapping; + struct mem_cgroup *memcg; + unsigned long flags; + void (*freepage)(struct page *); BUG_ON(!PageLocked(page)); freepage = mapping->a_ops->freepage; - spin_lock_irq(&mapping->tree_lock); - __delete_from_page_cache(page, NULL); - spin_unlock_irq(&mapping->tree_lock); + + memcg = mem_cgroup_begin_page_stat(page); + spin_lock_irqsave(&mapping->tree_lock, flags); + __delete_from_page_cache(page, NULL, memcg); + spin_unlock_irqrestore(&mapping->tree_lock, flags); + mem_cgroup_end_page_stat(memcg); if (freepage) freepage(page); @@ -470,6 +479,8 @@ int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask) if (!error) { struct address_space *mapping = old->mapping; void (*freepage)(struct page *); + struct mem_cgroup *memcg; + unsigned long flags; pgoff_t offset = old->index; freepage = mapping->a_ops->freepage; @@ -478,15 +489,17 @@ int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask) new->mapping = mapping; new->index = offset; - spin_lock_irq(&mapping->tree_lock); - __delete_from_page_cache(old, NULL); + memcg = mem_cgroup_begin_page_stat(old); + spin_lock_irqsave(&mapping->tree_lock, flags); + __delete_from_page_cache(old, NULL, memcg); error = radix_tree_insert(&mapping->page_tree, offset, new); BUG_ON(error); mapping->nrpages++; __inc_zone_page_state(new, NR_FILE_PAGES); if (PageSwapBacked(new)) __inc_zone_page_state(new, NR_SHMEM); - spin_unlock_irq(&mapping->tree_lock); + spin_unlock_irqrestore(&mapping->tree_lock, flags); + mem_cgroup_end_page_stat(memcg); mem_cgroup_migrate(old, new, true); radix_tree_preload_end(); if (freepage) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 14c2f2017e37..c23c1a3e8e16 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -90,6 +90,7 @@ static const char * const mem_cgroup_stat_names[] = { "rss", "rss_huge", "mapped_file", + "dirty", "writeback", "swap", }; @@ -2011,6 +2012,7 @@ again: return memcg; } +EXPORT_SYMBOL(mem_cgroup_begin_page_stat); /** * mem_cgroup_end_page_stat - finish a page state statistics transaction @@ -2029,6 +2031,7 @@ void mem_cgroup_end_page_stat(struct mem_cgroup *memcg) rcu_read_unlock(); } +EXPORT_SYMBOL(mem_cgroup_end_page_stat); /** * mem_cgroup_update_page_stat - update page state statistics @@ -4746,6 +4749,7 @@ static int mem_cgroup_move_account(struct page *page, { unsigned long flags; int ret; + bool anon; VM_BUG_ON(from == to); VM_BUG_ON_PAGE(PageLRU(page), page); @@ -4771,15 +4775,33 @@ static int mem_cgroup_move_account(struct page *page, if (page->mem_cgroup != from) goto out_unlock; + anon = PageAnon(page); + spin_lock_irqsave(&from->move_lock, flags); - if (!PageAnon(page) && page_mapped(page)) { + if (!anon && page_mapped(page)) { __this_cpu_sub(from->stat->count[MEM_CGROUP_STAT_FILE_MAPPED], nr_pages); __this_cpu_add(to->stat->count[MEM_CGROUP_STAT_FILE_MAPPED], nr_pages); } + /* + * move_lock grabbed above and caller set from->moving_account, so + * mem_cgroup_update_page_stat() will serialize updates to PageDirty. + * So mapping should be stable for dirty pages. + */ + if (!anon && PageDirty(page)) { + struct address_space *mapping = page_mapping(page); + + if (mapping_cap_account_dirty(mapping)) { + __this_cpu_sub(from->stat->count[MEM_CGROUP_STAT_DIRTY], + nr_pages); + __this_cpu_add(to->stat->count[MEM_CGROUP_STAT_DIRTY], + nr_pages); + } + } + if (PageWriteback(page)) { __this_cpu_sub(from->stat->count[MEM_CGROUP_STAT_WRITEBACK], nr_pages); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 227b867598e1..bdeecad00489 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -2090,15 +2090,20 @@ int __set_page_dirty_no_writeback(struct page *page) /* * Helper function for set_page_dirty family. + * + * Caller must hold mem_cgroup_begin_page_stat(). + * * NOTE: This relies on being atomic wrt interrupts. */ -void account_page_dirtied(struct page *page, struct address_space *mapping) +void account_page_dirtied(struct page *page, struct address_space *mapping, + struct mem_cgroup *memcg) { trace_writeback_dirty_page(page, mapping); if (mapping_cap_account_dirty(mapping)) { struct backing_dev_info *bdi = inode_to_bdi(mapping->host); + mem_cgroup_inc_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); __inc_zone_page_state(page, NR_FILE_DIRTY); __inc_zone_page_state(page, NR_DIRTIED); __inc_bdi_stat(bdi, BDI_RECLAIMABLE); @@ -2112,10 +2117,14 @@ EXPORT_SYMBOL(account_page_dirtied); /* * Helper function for deaccounting dirty page without writeback. + * + * Caller must hold mem_cgroup_begin_page_stat(). */ -void account_page_cleaned(struct page *page, struct address_space *mapping) +void account_page_cleaned(struct page *page, struct address_space *mapping, + struct mem_cgroup *memcg) { if (mapping_cap_account_dirty(mapping)) { + mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); dec_zone_page_state(page, NR_FILE_DIRTY); dec_bdi_stat(inode_to_bdi(mapping->host), BDI_RECLAIMABLE); task_io_account_cancelled_write(PAGE_CACHE_SIZE); @@ -2136,26 +2145,34 @@ void account_page_cleaned(struct page *page, struct address_space *mapping) */ int __set_page_dirty_nobuffers(struct page *page) { + struct mem_cgroup *memcg; + + memcg = mem_cgroup_begin_page_stat(page); if (!TestSetPageDirty(page)) { struct address_space *mapping = page_mapping(page); unsigned long flags; - if (!mapping) + if (!mapping) { + mem_cgroup_end_page_stat(memcg); return 1; + } spin_lock_irqsave(&mapping->tree_lock, flags); BUG_ON(page_mapping(page) != mapping); WARN_ON_ONCE(!PagePrivate(page) && !PageUptodate(page)); - account_page_dirtied(page, mapping); + account_page_dirtied(page, mapping, memcg); radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); spin_unlock_irqrestore(&mapping->tree_lock, flags); + mem_cgroup_end_page_stat(memcg); + if (mapping->host) { /* !PageAnon && !swapper_space */ __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); } return 1; } + mem_cgroup_end_page_stat(memcg); return 0; } EXPORT_SYMBOL(__set_page_dirty_nobuffers); @@ -2273,8 +2290,20 @@ EXPORT_SYMBOL(set_page_dirty_lock); */ void cancel_dirty_page(struct page *page) { - if (TestClearPageDirty(page)) - account_page_cleaned(page, page_mapping(page)); + struct address_space *mapping = page_mapping(page); + + if (mapping_cap_account_dirty(mapping)) { + struct mem_cgroup *memcg; + + memcg = mem_cgroup_begin_page_stat(page); + + if (TestClearPageDirty(page)) + account_page_cleaned(page, mapping, memcg); + + mem_cgroup_end_page_stat(memcg); + } else { + ClearPageDirty(page); + } } EXPORT_SYMBOL(cancel_dirty_page); @@ -2295,6 +2324,8 @@ EXPORT_SYMBOL(cancel_dirty_page); int clear_page_dirty_for_io(struct page *page) { struct address_space *mapping = page_mapping(page); + struct mem_cgroup *memcg; + int ret = 0; BUG_ON(!PageLocked(page)); @@ -2334,13 +2365,16 @@ int clear_page_dirty_for_io(struct page *page) * always locked coming in here, so we get the desired * exclusion. */ + memcg = mem_cgroup_begin_page_stat(page); if (TestClearPageDirty(page)) { + mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); dec_zone_page_state(page, NR_FILE_DIRTY); dec_bdi_stat(inode_to_bdi(mapping->host), BDI_RECLAIMABLE); - return 1; + ret = 1; } - return 0; + mem_cgroup_end_page_stat(memcg); + return ret; } return TestClearPageDirty(page); } diff --git a/mm/rmap.c b/mm/rmap.c index 24dd3f9fee27..8fc556ce2dcb 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -30,6 +30,8 @@ * swap_lock (in swap_duplicate, swap_info_get) * mmlist_lock (in mmput, drain_mmlist and others) * mapping->private_lock (in __set_page_dirty_buffers) + * mem_cgroup_{begin,end}_page_stat (memcg->move_lock) + * mapping->tree_lock (widely used) * inode->i_lock (in set_page_dirty's __mark_inode_dirty) * bdi.wb->list_lock (in set_page_dirty's __mark_inode_dirty) * sb_lock (within inode_lock in fs/fs-writeback.c) diff --git a/mm/truncate.c b/mm/truncate.c index 0c360259c085..76e35ad97102 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -510,19 +510,24 @@ EXPORT_SYMBOL(invalidate_mapping_pages); static int invalidate_complete_page2(struct address_space *mapping, struct page *page) { + struct mem_cgroup *memcg; + unsigned long flags; + if (page->mapping != mapping) return 0; if (page_has_private(page) && !try_to_release_page(page, GFP_KERNEL)) return 0; - spin_lock_irq(&mapping->tree_lock); + memcg = mem_cgroup_begin_page_stat(page); + spin_lock_irqsave(&mapping->tree_lock, flags); if (PageDirty(page)) goto failed; BUG_ON(page_has_private(page)); - __delete_from_page_cache(page, NULL); - spin_unlock_irq(&mapping->tree_lock); + __delete_from_page_cache(page, NULL, memcg); + spin_unlock_irqrestore(&mapping->tree_lock, flags); + mem_cgroup_end_page_stat(memcg); if (mapping->a_ops->freepage) mapping->a_ops->freepage(page); @@ -530,7 +535,8 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page) page_cache_release(page); /* pagecache ref */ return 1; failed: - spin_unlock_irq(&mapping->tree_lock); + spin_unlock_irqrestore(&mapping->tree_lock, flags); + mem_cgroup_end_page_stat(memcg); return 0; } diff --git a/mm/vmscan.c b/mm/vmscan.c index 5e8eadd71bac..7582f9fcda92 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -579,10 +579,14 @@ static pageout_t pageout(struct page *page, struct address_space *mapping, static int __remove_mapping(struct address_space *mapping, struct page *page, bool reclaimed) { + unsigned long flags; + struct mem_cgroup *memcg; + BUG_ON(!PageLocked(page)); BUG_ON(mapping != page_mapping(page)); - spin_lock_irq(&mapping->tree_lock); + memcg = mem_cgroup_begin_page_stat(page); + spin_lock_irqsave(&mapping->tree_lock, flags); /* * The non racy check for a busy page. * @@ -620,7 +624,8 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, swp_entry_t swap = { .val = page_private(page) }; mem_cgroup_swapout(page, swap); __delete_from_swap_cache(page); - spin_unlock_irq(&mapping->tree_lock); + spin_unlock_irqrestore(&mapping->tree_lock, flags); + mem_cgroup_end_page_stat(memcg); swapcache_free(swap); } else { void (*freepage)(struct page *); @@ -640,8 +645,9 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, if (reclaimed && page_is_file_cache(page) && !mapping_exiting(mapping)) shadow = workingset_eviction(mapping, page); - __delete_from_page_cache(page, shadow); - spin_unlock_irq(&mapping->tree_lock); + __delete_from_page_cache(page, shadow, memcg); + spin_unlock_irqrestore(&mapping->tree_lock, flags); + mem_cgroup_end_page_stat(memcg); if (freepage != NULL) freepage(page); @@ -650,7 +656,8 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, return 1; cannot_free: - spin_unlock_irq(&mapping->tree_lock); + spin_unlock_irqrestore(&mapping->tree_lock, flags); + mem_cgroup_end_page_stat(memcg); return 0; } From eea8f41cc58849e354ecf8b95bd7f806e1d1f703 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:17 -0400 Subject: [PATCH 03/85] blkcg: move block/blk-cgroup.h to include/linux/blk-cgroup.h cgroup aware writeback support will require exposing some of blkcg details. In preprataion, move block/blk-cgroup.h to include/linux/blk-cgroup.h. This patch is pure file move. Signed-off-by: Tejun Heo Cc: Vivek Goyal Signed-off-by: Jens Axboe --- block/blk-cgroup.c | 2 +- block/blk-core.c | 2 +- block/blk-sysfs.c | 2 +- block/blk-throttle.c | 2 +- block/cfq-iosched.c | 2 +- block/elevator.c | 2 +- {block => include/linux}/blk-cgroup.h | 0 7 files changed, 6 insertions(+), 6 deletions(-) rename {block => include/linux}/blk-cgroup.h (100%) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 0ac817b750db..c3226ce8c5f6 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -19,7 +19,7 @@ #include #include #include -#include "blk-cgroup.h" +#include #include "blk.h" #define MAX_KEY_LEN 100 diff --git a/block/blk-core.c b/block/blk-core.c index aa819a58ea24..4afac14d4499 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -32,12 +32,12 @@ #include #include #include +#include #define CREATE_TRACE_POINTS #include #include "blk.h" -#include "blk-cgroup.h" #include "blk-mq.h" EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_remap); diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index faaf36ade7eb..5677eb78557d 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -8,9 +8,9 @@ #include #include #include +#include #include "blk.h" -#include "blk-cgroup.h" #include "blk-mq.h" struct queue_sysfs_entry { diff --git a/block/blk-throttle.c b/block/blk-throttle.c index 5b9c6d5c3636..b23193518ac7 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -9,7 +9,7 @@ #include #include #include -#include "blk-cgroup.h" +#include #include "blk.h" /* Max dispatch from a group in 1 round */ diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 5da8e6e9ab4b..bc8f42930773 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -14,8 +14,8 @@ #include #include #include +#include #include "blk.h" -#include "blk-cgroup.h" /* * tunables diff --git a/block/elevator.c b/block/elevator.c index 59794d0d38e3..3bbb48f430e4 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -35,11 +35,11 @@ #include #include #include +#include #include #include "blk.h" -#include "blk-cgroup.h" static DEFINE_SPINLOCK(elv_list_lock); static LIST_HEAD(elv_list); diff --git a/block/blk-cgroup.h b/include/linux/blk-cgroup.h similarity index 100% rename from block/blk-cgroup.h rename to include/linux/blk-cgroup.h From efa7d1c733d1d2c1a468b85126d70bad9fdf6ba8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:18 -0400 Subject: [PATCH 04/85] update !CONFIG_BLK_CGROUP dummies in include/linux/blk-cgroup.h The header file will be used more widely with the pending cgroup writeback support and the current set of dummy declarations aren't enough to handle different config combinations. Update as follows. * Drop the struct cgroup declaration. None of the dummy defs need it. * Define blkcg as an empty struct instead of just declaring it. * Wrap dummy function defs in CONFIG_BLOCK. Some functions use block data types and none of them are to be used w/o block enabled. Signed-off-by: Tejun Heo Signed-off-by: Jens Axboe --- include/linux/blk-cgroup.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h index c567865b5f1d..51f95b34d3f0 100644 --- a/include/linux/blk-cgroup.h +++ b/include/linux/blk-cgroup.h @@ -558,8 +558,8 @@ static inline void blkg_rwstat_merge(struct blkg_rwstat *to, #else /* CONFIG_BLK_CGROUP */ -struct cgroup; -struct blkcg; +struct blkcg { +}; struct blkg_policy_data { }; @@ -570,6 +570,8 @@ struct blkcg_gq { struct blkcg_policy { }; +#ifdef CONFIG_BLOCK + static inline struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, void *key) { return NULL; } static inline int blkcg_init_queue(struct request_queue *q) { return 0; } static inline void blkcg_drain_queue(struct request_queue *q) { } @@ -599,5 +601,6 @@ static inline struct request_list *blk_rq_rl(struct request *rq) { return &rq->q #define blk_queue_for_each_rl(rl, q) \ for ((rl) = &(q)->root_rl; (rl); (rl) = NULL) +#endif /* CONFIG_BLOCK */ #endif /* CONFIG_BLK_CGROUP */ #endif /* _BLK_CGROUP_H */ From ec13b1d6f0a0457312e615335ce8ceb07da50a11 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:19 -0400 Subject: [PATCH 05/85] blkcg: always create the blkcg_gq for the root blkcg Currently, blkcg does a minor optimization where the root blkcg is created when the first blkcg policy is activated on a queue and destroyed on the deactivation of the last. On systems where blkcg is configured but not used, this saves one blkcg_gq struct per queue. On systems where blkcg is actually used, there's no difference. The only case where this can lead to any meaninful, albeit still minute, save in memory consumption is when all blkcg policies are deactivated after being widely used in the system, which is a hihgly unlikely scenario. The conditional existence of root blkcg_gq has already created several bugs in blkcg and became an issue once again for the new per-cgroup wb_congested mechanism for cgroup writeback support leading to a NULL dereference when no blkcg policy is active. This is really not worth bothering with. This patch makes blkcg always allocate and link the root blkcg_gq and release it only on queue destruction. Signed-off-by: Tejun Heo Reported-by: Fengguang Wu Signed-off-by: Jens Axboe --- block/blk-cgroup.c | 96 ++++++++++++++++++++-------------------------- 1 file changed, 41 insertions(+), 55 deletions(-) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index c3226ce8c5f6..2a4f77f2b229 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -235,13 +235,8 @@ static struct blkcg_gq *blkg_create(struct blkcg *blkcg, blkg->online = true; spin_unlock(&blkcg->lock); - if (!ret) { - if (blkcg == &blkcg_root) { - q->root_blkg = blkg; - q->root_rl.blkg = blkg; - } + if (!ret) return blkg; - } /* @blkg failed fully initialized, use the usual release path */ blkg_put(blkg); @@ -339,15 +334,6 @@ static void blkg_destroy(struct blkcg_gq *blkg) if (rcu_access_pointer(blkcg->blkg_hint) == blkg) rcu_assign_pointer(blkcg->blkg_hint, NULL); - /* - * If root blkg is destroyed. Just clear the pointer since root_rl - * does not take reference on root blkg. - */ - if (blkcg == &blkcg_root) { - blkg->q->root_blkg = NULL; - blkg->q->root_rl.blkg = NULL; - } - /* * Put the reference taken at the time of creation so that when all * queues are gone, group can be destroyed. @@ -855,9 +841,45 @@ done: */ int blkcg_init_queue(struct request_queue *q) { - might_sleep(); + struct blkcg_gq *new_blkg, *blkg; + bool preloaded; + int ret; - return blk_throtl_init(q); + new_blkg = blkg_alloc(&blkcg_root, q, GFP_KERNEL); + if (!new_blkg) + return -ENOMEM; + + preloaded = !radix_tree_preload(GFP_KERNEL); + + /* + * Make sure the root blkg exists and count the existing blkgs. As + * @q is bypassing at this point, blkg_lookup_create() can't be + * used. Open code insertion. + */ + rcu_read_lock(); + spin_lock_irq(q->queue_lock); + blkg = blkg_create(&blkcg_root, q, new_blkg); + spin_unlock_irq(q->queue_lock); + rcu_read_unlock(); + + if (preloaded) + radix_tree_preload_end(); + + if (IS_ERR(blkg)) { + kfree(new_blkg); + return PTR_ERR(blkg); + } + + q->root_blkg = blkg; + q->root_rl.blkg = blkg; + + ret = blk_throtl_init(q); + if (ret) { + spin_lock_irq(q->queue_lock); + blkg_destroy_all(q); + spin_unlock_irq(q->queue_lock); + } + return ret; } /** @@ -958,52 +980,20 @@ int blkcg_activate_policy(struct request_queue *q, const struct blkcg_policy *pol) { LIST_HEAD(pds); - struct blkcg_gq *blkg, *new_blkg; + struct blkcg_gq *blkg; struct blkg_policy_data *pd, *n; int cnt = 0, ret; - bool preloaded; if (blkcg_policy_enabled(q, pol)) return 0; - /* preallocations for root blkg */ - new_blkg = blkg_alloc(&blkcg_root, q, GFP_KERNEL); - if (!new_blkg) - return -ENOMEM; - + /* count and allocate policy_data for all existing blkgs */ blk_queue_bypass_start(q); - - preloaded = !radix_tree_preload(GFP_KERNEL); - - /* - * Make sure the root blkg exists and count the existing blkgs. As - * @q is bypassing at this point, blkg_lookup_create() can't be - * used. Open code it. - */ spin_lock_irq(q->queue_lock); - - rcu_read_lock(); - blkg = __blkg_lookup(&blkcg_root, q, false); - if (blkg) - blkg_free(new_blkg); - else - blkg = blkg_create(&blkcg_root, q, new_blkg); - rcu_read_unlock(); - - if (preloaded) - radix_tree_preload_end(); - - if (IS_ERR(blkg)) { - ret = PTR_ERR(blkg); - goto out_unlock; - } - list_for_each_entry(blkg, &q->blkg_list, q_node) cnt++; - spin_unlock_irq(q->queue_lock); - /* allocate policy_data for all existing blkgs */ while (cnt--) { pd = kzalloc_node(pol->pd_size, GFP_KERNEL, q->node); if (!pd) { @@ -1072,10 +1062,6 @@ void blkcg_deactivate_policy(struct request_queue *q, __clear_bit(pol->plid, q->blkcg_pols); - /* if no policy is left, no need for blkgs - shoot them down */ - if (bitmap_empty(q->blkcg_pols, BLKCG_MAX_POLS)) - blkg_destroy_all(q); - list_for_each_entry(blkg, &q->blkg_list, q_node) { /* grab blkcg lock too while removing @pd from @blkg */ spin_lock(&blkg->blkcg->lock); From 56161634e4824380a67243a4cf3fa52eb1e5d836 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:20 -0400 Subject: [PATCH 06/85] memcg: add mem_cgroup_root_css Add global mem_cgroup_root_css which points to the root memcg css. This will be used by cgroup writeback support. If memcg is disabled, it's defined as ERR_PTR(-EINVAL). Signed-off-by: Tejun Heo Cc: Johannes Weiner aCc: Michal Hocko Signed-off-by: Jens Axboe --- include/linux/memcontrol.h | 4 ++++ mm/memcontrol.c | 2 ++ 2 files changed, 6 insertions(+) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 5fe6411b5e54..294498f4f6fc 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -68,6 +68,8 @@ enum mem_cgroup_events_index { }; #ifdef CONFIG_MEMCG +extern struct cgroup_subsys_state *mem_cgroup_root_css; + void mem_cgroup_events(struct mem_cgroup *memcg, enum mem_cgroup_events_index idx, unsigned int nr); @@ -196,6 +198,8 @@ void mem_cgroup_split_huge_fixup(struct page *head); #else /* CONFIG_MEMCG */ struct mem_cgroup; +#define mem_cgroup_root_css ((struct cgroup_subsys_state *)ERR_PTR(-EINVAL)) + static inline void mem_cgroup_events(struct mem_cgroup *memcg, enum mem_cgroup_events_index idx, unsigned int nr) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index c23c1a3e8e16..b22a92b37480 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -77,6 +77,7 @@ EXPORT_SYMBOL(memory_cgrp_subsys); #define MEM_CGROUP_RECLAIM_RETRIES 5 static struct mem_cgroup *root_mem_cgroup __read_mostly; +struct cgroup_subsys_state *mem_cgroup_root_css __read_mostly; /* Whether the swap controller is active */ #ifdef CONFIG_MEMCG_SWAP @@ -4441,6 +4442,7 @@ mem_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) /* root ? */ if (parent_css == NULL) { root_mem_cgroup = memcg; + mem_cgroup_root_css = &memcg->css; page_counter_init(&memcg->memory, NULL); memcg->high = PAGE_COUNTER_MAX; memcg->soft_limit = PAGE_COUNTER_MAX; From 496d5e7560dbb84399dbd92316fc33857aa83900 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:21 -0400 Subject: [PATCH 07/85] blkcg: add blkcg_root_css Add global constant blkcg_root_css which points to &blkcg_root.css. This will be used by cgroup writeback support. If blkcg is disabled, it's defined as ERR_PTR(-EINVAL). v2: The declarations moved to include/linux/blk-cgroup.h as suggested by Vivek. Signed-off-by: Tejun Heo Cc: Vivek Goyal Cc: Jens Axboe Signed-off-by: Jens Axboe --- block/blk-cgroup.c | 2 ++ include/linux/blk-cgroup.h | 3 +++ 2 files changed, 5 insertions(+) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 2a4f77f2b229..54ec1721cd3e 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -30,6 +30,8 @@ struct blkcg blkcg_root = { .cfq_weight = 2 * CFQ_WEIGHT_DEFAULT, .cfq_leaf_weight = 2 * CFQ_WEIGHT_DEFAULT, }; EXPORT_SYMBOL_GPL(blkcg_root); +struct cgroup_subsys_state * const blkcg_root_css = &blkcg_root.css; + static struct blkcg_policy *blkcg_policy[BLKCG_MAX_POLS]; static bool blkcg_policy_enabled(struct request_queue *q, diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h index 51f95b34d3f0..65f0c178fd04 100644 --- a/include/linux/blk-cgroup.h +++ b/include/linux/blk-cgroup.h @@ -134,6 +134,7 @@ struct blkcg_policy { }; extern struct blkcg blkcg_root; +extern struct cgroup_subsys_state * const blkcg_root_css; struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, struct request_queue *q); struct blkcg_gq *blkg_lookup_create(struct blkcg *blkcg, @@ -570,6 +571,8 @@ struct blkcg_gq { struct blkcg_policy { }; +#define blkcg_root_css ((struct cgroup_subsys_state *)ERR_PTR(-EINVAL)) + #ifdef CONFIG_BLOCK static inline struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, void *key) { return NULL; } From ec438699a9ae0856c2ce20a50dd39cdc7e92a732 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:22 -0400 Subject: [PATCH 08/85] cgroup, block: implement task_get_css() and use it in bio_associate_current() bio_associate_current() currently open codes task_css() and css_tryget_online() to find and pin $current's blkcg css. Abstract it into task_get_css() which is implemented from cgroup side. As a task is always associated with an online css for every subsystem except while the css_set update is propagating, task_get_css() retries till css_tryget_online() succeeds. This is a cleanup and shouldn't lead to noticeable behavior changes. Signed-off-by: Tejun Heo Cc: Li Zefan Cc: Jens Axboe Cc: Vivek Goyal Signed-off-by: Jens Axboe --- block/bio.c | 11 +---------- include/linux/cgroup.h | 25 +++++++++++++++++++++++++ 2 files changed, 26 insertions(+), 10 deletions(-) diff --git a/block/bio.c b/block/bio.c index 259197d97de1..c4f87018fb79 100644 --- a/block/bio.c +++ b/block/bio.c @@ -2004,7 +2004,6 @@ EXPORT_SYMBOL(bioset_create_nobvec); int bio_associate_current(struct bio *bio) { struct io_context *ioc; - struct cgroup_subsys_state *css; if (bio->bi_ioc) return -EBUSY; @@ -2013,17 +2012,9 @@ int bio_associate_current(struct bio *bio) if (!ioc) return -ENOENT; - /* acquire active ref on @ioc and associate */ get_io_context_active(ioc); bio->bi_ioc = ioc; - - /* associate blkcg if exists */ - rcu_read_lock(); - css = task_css(current, blkio_cgrp_id); - if (css && css_tryget_online(css)) - bio->bi_css = css; - rcu_read_unlock(); - + bio->bi_css = task_get_css(current, blkio_cgrp_id); return 0; } diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index b9cb94c3102a..e7da0aa65b2d 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -773,6 +773,31 @@ static inline struct cgroup_subsys_state *task_css(struct task_struct *task, return task_css_check(task, subsys_id, false); } +/** + * task_get_css - find and get the css for (task, subsys) + * @task: the target task + * @subsys_id: the target subsystem ID + * + * Find the css for the (@task, @subsys_id) combination, increment a + * reference on and return it. This function is guaranteed to return a + * valid css. + */ +static inline struct cgroup_subsys_state * +task_get_css(struct task_struct *task, int subsys_id) +{ + struct cgroup_subsys_state *css; + + rcu_read_lock(); + while (true) { + css = task_css(task, subsys_id); + if (likely(css_tryget_online(css))) + break; + cpu_relax(); + } + rcu_read_unlock(); + return css; +} + /** * task_css_is_root - test whether a task belongs to the root css * @task: the target task From fd383c2d3cae146337cea809de0d622b8b887e6c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:23 -0400 Subject: [PATCH 09/85] blkcg: implement task_get_blkcg_css() Implement a wrapper around task_get_css() to acquire the blkcg css for a given task. The wrapper is necessary for cgroup writeback support as there will be places outside blkcg proper trying to acquire blkcg_css and blkio_cgrp_id will be undefined when !CONFIG_BLK_CGROUP. Signed-off-by: Tejun Heo Signed-off-by: Jens Axboe --- include/linux/blk-cgroup.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h index 65f0c178fd04..4dc643f2046e 100644 --- a/include/linux/blk-cgroup.h +++ b/include/linux/blk-cgroup.h @@ -195,6 +195,12 @@ static inline struct blkcg *bio_blkcg(struct bio *bio) return task_blkcg(current); } +static inline struct cgroup_subsys_state * +task_get_blkcg_css(struct task_struct *task) +{ + return task_get_css(task, blkio_cgrp_id); +} + /** * blkcg_parent - get the parent of a blkcg * @blkcg: blkcg of interest @@ -573,6 +579,12 @@ struct blkcg_policy { #define blkcg_root_css ((struct cgroup_subsys_state *)ERR_PTR(-EINVAL)) +static inline struct cgroup_subsys_state * +task_get_blkcg_css(struct task_struct *task) +{ + return NULL; +} + #ifdef CONFIG_BLOCK static inline struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, void *key) { return NULL; } From 1d933cf096e3aea15f1aec8297657b7a846fab63 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:24 -0400 Subject: [PATCH 10/85] blkcg: implement bio_associate_blkcg() Currently, a bio can only be associated with the io_context and blkcg of %current using bio_associate_current(). This is too restrictive for cgroup writeback support. Implement bio_associate_blkcg() which associates a bio with the specified blkcg. bio_associate_blkcg() leaves the io_context unassociated. bio_associate_current() is updated so that it considers a bio as already associated if it has a blkcg_css, instead of an io_context, associated with it. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Vivek Goyal Signed-off-by: Jens Axboe --- block/bio.c | 24 +++++++++++++++++++++++- include/linux/bio.h | 3 +++ 2 files changed, 26 insertions(+), 1 deletion(-) diff --git a/block/bio.c b/block/bio.c index c4f87018fb79..2a00d349cd68 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1988,6 +1988,28 @@ struct bio_set *bioset_create_nobvec(unsigned int pool_size, unsigned int front_ EXPORT_SYMBOL(bioset_create_nobvec); #ifdef CONFIG_BLK_CGROUP + +/** + * bio_associate_blkcg - associate a bio with the specified blkcg + * @bio: target bio + * @blkcg_css: css of the blkcg to associate + * + * Associate @bio with the blkcg specified by @blkcg_css. Block layer will + * treat @bio as if it were issued by a task which belongs to the blkcg. + * + * This function takes an extra reference of @blkcg_css which will be put + * when @bio is released. The caller must own @bio and is responsible for + * synchronizing calls to this function. + */ +int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css) +{ + if (unlikely(bio->bi_css)) + return -EBUSY; + css_get(blkcg_css); + bio->bi_css = blkcg_css; + return 0; +} + /** * bio_associate_current - associate a bio with %current * @bio: target bio @@ -2005,7 +2027,7 @@ int bio_associate_current(struct bio *bio) { struct io_context *ioc; - if (bio->bi_ioc) + if (bio->bi_css) return -EBUSY; ioc = current->io_context; diff --git a/include/linux/bio.h b/include/linux/bio.h index f0291cf64cc5..5e963a6d7c14 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -482,9 +482,12 @@ extern void bvec_free(mempool_t *, struct bio_vec *, unsigned int); extern unsigned int bvec_nr_vecs(unsigned short idx); #ifdef CONFIG_BLK_CGROUP +int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css); int bio_associate_current(struct bio *bio); void bio_disassociate_task(struct bio *bio); #else /* CONFIG_BLK_CGROUP */ +static inline int bio_associate_blkcg(struct bio *bio, + struct cgroup_subsys_state *blkcg_css) { return 0; } static inline int bio_associate_current(struct bio *bio) { return -ENOENT; } static inline void bio_disassociate_task(struct bio *bio) { } #endif /* CONFIG_BLK_CGROUP */ From ad7fa852d3d2816d68a138ebc5bc8967aeb7fd86 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 27 May 2015 20:00:02 -0400 Subject: [PATCH 11/85] memcg: implement mem_cgroup_css_from_page() Implement mem_cgroup_css_from_page() which returns the cgroup_subsys_state of the memcg associated with a given page on the default hierarchy. This will be used by cgroup writeback support. This function assumes that page->mem_cgroup association doesn't change until the page is released, which is true on the default hierarchy as long as replace_page_cache_page() is not used. As the only user of replace_page_cache_page() is FUSE which won't support cgroup writeback for the time being, this works for now, and replace_page_cache_page() will soon be updated so that the invariant actually holds. Note that the RCU protected page->mem_cgroup access is consistent with other usages across memcg but ultimately incorrect. These unlocked accesses are missing required barriers. page->mem_cgroup should be made an RCU pointer and updated and accessed using RCU operations. v4: Instead of triggering WARN, return the root css on the traditional hierarchies. This makes the function a lot easier to deal with especially as there's no light way to synchronize against hierarchy rebinding. v3: s/mem_cgroup_migrate()/mem_cgroup_css_from_page()/ v2: Trigger WARN if the function is used on the traditional hierarchies and add comment about the assumed invariant. Signed-off-by: Tejun Heo Cc: Johannes Weiner Cc: Michal Hocko Signed-off-by: Jens Axboe --- include/linux/memcontrol.h | 1 + mm/memcontrol.c | 33 +++++++++++++++++++++++++++++++++ 2 files changed, 34 insertions(+) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 294498f4f6fc..637ef626008e 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -115,6 +115,7 @@ static inline bool mm_match_cgroup(struct mm_struct *mm, } extern struct cgroup_subsys_state *mem_cgroup_css(struct mem_cgroup *memcg); +extern struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page); struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *, struct mem_cgroup *, diff --git a/mm/memcontrol.c b/mm/memcontrol.c index b22a92b37480..5c270a03729c 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -598,6 +598,39 @@ struct cgroup_subsys_state *mem_cgroup_css(struct mem_cgroup *memcg) return &memcg->css; } +/** + * mem_cgroup_css_from_page - css of the memcg associated with a page + * @page: page of interest + * + * If memcg is bound to the default hierarchy, css of the memcg associated + * with @page is returned. The returned css remains associated with @page + * until it is released. + * + * If memcg is bound to a traditional hierarchy, the css of root_mem_cgroup + * is returned. + * + * XXX: The above description of behavior on the default hierarchy isn't + * strictly true yet as replace_page_cache_page() can modify the + * association before @page is released even on the default hierarchy; + * however, the current and planned usages don't mix the the two functions + * and replace_page_cache_page() will soon be updated to make the invariant + * actually true. + */ +struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page) +{ + struct mem_cgroup *memcg; + + rcu_read_lock(); + + memcg = page->mem_cgroup; + + if (!memcg || !cgroup_on_dfl(memcg->css.cgroup)) + memcg = root_mem_cgroup; + + rcu_read_unlock(); + return &memcg->css; +} + static struct mem_cgroup_per_zone * mem_cgroup_page_zoneinfo(struct mem_cgroup *memcg, struct page *page) { From 4452226ea276e74fc3e252c88d9bb7e8f8e44bf0 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:26 -0400 Subject: [PATCH 12/85] writeback: move backing_dev_info->state into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->state into wb. * enum bdi_state is renamed to wb_state and the prefix of all enums is changed from BDI_ to WB_. * Explicit zeroing of bdi->state is removed without adding zeoring of wb->state as the whole data structure is zeroed on init anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->state are mechanically replaced with bdi->wb.state introducing no behavior changes. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Cc: Wu Fengguang Cc: drbd-dev@lists.linbit.com Cc: Neil Brown Cc: Alasdair Kergon Cc: Mike Snitzer Signed-off-by: Jens Axboe --- block/blk-core.c | 1 - drivers/block/drbd/drbd_main.c | 10 +++++----- drivers/md/dm.c | 2 +- drivers/md/raid1.c | 4 ++-- drivers/md/raid10.c | 2 +- fs/fs-writeback.c | 14 +++++++------- include/linux/backing-dev.h | 24 ++++++++++++------------ mm/backing-dev.c | 20 ++++++++++---------- 8 files changed, 38 insertions(+), 39 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index 4afac14d4499..c3ba9c3b08d4 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -621,7 +621,6 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id) q->backing_dev_info.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; - q->backing_dev_info.state = 0; q->backing_dev_info.capabilities = 0; q->backing_dev_info.name = "block"; q->node = node_id; diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c index 81fde9ef7f8e..a1518539b858 100644 --- a/drivers/block/drbd/drbd_main.c +++ b/drivers/block/drbd/drbd_main.c @@ -2359,7 +2359,7 @@ static void drbd_cleanup(void) * @congested_data: User data * @bdi_bits: Bits the BDI flusher thread is currently interested in * - * Returns 1<connection->flags)) { - r |= (1 << BDI_async_congested); + r |= (1 << WB_async_congested); /* Without good local data, we would need to read from remote, * and that would need the worker thread as well, which is * currently blocked waiting for that usermode helper to * finish. */ if (!get_ldev_if_state(device, D_UP_TO_DATE)) - r |= (1 << BDI_sync_congested); + r |= (1 << WB_sync_congested); else put_ldev(device); r &= bdi_bits; @@ -2399,9 +2399,9 @@ static int drbd_congested(void *congested_data, int bdi_bits) reason = 'b'; } - if (bdi_bits & (1 << BDI_async_congested) && + if (bdi_bits & (1 << WB_async_congested) && test_bit(NET_CONGESTED, &first_peer_device(device)->connection->flags)) { - r |= (1 << BDI_async_congested); + r |= (1 << WB_async_congested); reason = reason == 'b' ? 'a' : 'n'; } diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 38837f8ea327..2161ed9329c4 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -2074,7 +2074,7 @@ static int dm_any_congested(void *congested_data, int bdi_bits) * the query about congestion status of request_queue */ if (dm_request_based(md)) - r = md->queue->backing_dev_info.state & + r = md->queue->backing_dev_info.wb.state & bdi_bits; else r = dm_table_any_congested(map, bdi_bits); diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 9157a29c8dbf..f80f1af61ce7 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -745,7 +745,7 @@ static int raid1_congested(struct mddev *mddev, int bits) struct r1conf *conf = mddev->private; int i, ret = 0; - if ((bits & (1 << BDI_async_congested)) && + if ((bits & (1 << WB_async_congested)) && conf->pending_count >= max_queued_requests) return 1; @@ -760,7 +760,7 @@ static int raid1_congested(struct mddev *mddev, int bits) /* Note the '|| 1' - when read_balance prefers * non-congested targets, it can be removed */ - if ((bits & (1<backing_dev_info, bits); else ret &= bdi_congested(&q->backing_dev_info, bits); diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index e793ab6b3570..fca825718f29 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -914,7 +914,7 @@ static int raid10_congested(struct mddev *mddev, int bits) struct r10conf *conf = mddev->private; int i, ret = 0; - if ((bits & (1 << BDI_async_congested)) && + if ((bits & (1 << WB_async_congested)) && conf->pending_count >= max_queued_requests) return 1; diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 32a8bbd7a9ad..983312cea245 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -74,7 +74,7 @@ unsigned int dirtytime_expire_interval = 12 * 60 * 60; */ int writeback_in_progress(struct backing_dev_info *bdi) { - return test_bit(BDI_writeback_running, &bdi->state); + return test_bit(WB_writeback_running, &bdi->wb.state); } EXPORT_SYMBOL(writeback_in_progress); @@ -112,7 +112,7 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(wbc_writepage); static void bdi_wakeup_thread(struct backing_dev_info *bdi) { spin_lock_bh(&bdi->wb_lock); - if (test_bit(BDI_registered, &bdi->state)) + if (test_bit(WB_registered, &bdi->wb.state)) mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); spin_unlock_bh(&bdi->wb_lock); } @@ -123,7 +123,7 @@ static void bdi_queue_work(struct backing_dev_info *bdi, trace_writeback_queue(bdi, work); spin_lock_bh(&bdi->wb_lock); - if (!test_bit(BDI_registered, &bdi->state)) { + if (!test_bit(WB_registered, &bdi->wb.state)) { if (work->done) complete(work->done); goto out_unlock; @@ -1057,7 +1057,7 @@ static long wb_do_writeback(struct bdi_writeback *wb) struct wb_writeback_work *work; long wrote = 0; - set_bit(BDI_writeback_running, &wb->bdi->state); + set_bit(WB_writeback_running, &wb->state); while ((work = get_next_work_item(bdi)) != NULL) { trace_writeback_exec(bdi, work); @@ -1079,7 +1079,7 @@ static long wb_do_writeback(struct bdi_writeback *wb) */ wrote += wb_check_old_data_flush(wb); wrote += wb_check_background_flush(wb); - clear_bit(BDI_writeback_running, &wb->bdi->state); + clear_bit(WB_writeback_running, &wb->state); return wrote; } @@ -1099,7 +1099,7 @@ void bdi_writeback_workfn(struct work_struct *work) current->flags |= PF_SWAPWRITE; if (likely(!current_is_workqueue_rescuer() || - !test_bit(BDI_registered, &bdi->state))) { + !test_bit(WB_registered, &wb->state))) { /* * The normal path. Keep writing back @bdi until its * work_list is empty. Note that this path is also taken @@ -1323,7 +1323,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) spin_unlock(&inode->i_lock); spin_lock(&bdi->wb.list_lock); if (bdi_cap_writeback_dirty(bdi)) { - WARN(!test_bit(BDI_registered, &bdi->state), + WARN(!test_bit(WB_registered, &bdi->wb.state), "bdi-%s not registered\n", bdi->name); /* diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index aff923ae8c4b..eb14f988a63e 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -25,13 +25,13 @@ struct device; struct dentry; /* - * Bits in backing_dev_info.state + * Bits in bdi_writeback.state */ -enum bdi_state { - BDI_async_congested, /* The async (write) queue is getting full */ - BDI_sync_congested, /* The sync queue is getting full */ - BDI_registered, /* bdi_register() was done */ - BDI_writeback_running, /* Writeback is in progress */ +enum wb_state { + WB_async_congested, /* The async (write) queue is getting full */ + WB_sync_congested, /* The sync queue is getting full */ + WB_registered, /* bdi_register() was done */ + WB_writeback_running, /* Writeback is in progress */ }; typedef int (congested_fn)(void *, int); @@ -49,6 +49,7 @@ enum bdi_stat_item { struct bdi_writeback { struct backing_dev_info *bdi; /* our parent bdi */ + unsigned long state; /* Always use atomic bitops on this */ unsigned long last_old_flush; /* last old data flush */ struct delayed_work dwork; /* work item used for writeback */ @@ -62,7 +63,6 @@ struct bdi_writeback { struct backing_dev_info { struct list_head bdi_list; unsigned long ra_pages; /* max readahead in PAGE_CACHE_SIZE units */ - unsigned long state; /* Always use atomic bitops on this */ unsigned int capabilities; /* Device capabilities */ congested_fn *congested_fn; /* Function pointer if device is md/dm */ void *congested_data; /* Pointer to aux data for congested func */ @@ -250,23 +250,23 @@ static inline int bdi_congested(struct backing_dev_info *bdi, int bdi_bits) { if (bdi->congested_fn) return bdi->congested_fn(bdi->congested_data, bdi_bits); - return (bdi->state & bdi_bits); + return (bdi->wb.state & bdi_bits); } static inline int bdi_read_congested(struct backing_dev_info *bdi) { - return bdi_congested(bdi, 1 << BDI_sync_congested); + return bdi_congested(bdi, 1 << WB_sync_congested); } static inline int bdi_write_congested(struct backing_dev_info *bdi) { - return bdi_congested(bdi, 1 << BDI_async_congested); + return bdi_congested(bdi, 1 << WB_async_congested); } static inline int bdi_rw_congested(struct backing_dev_info *bdi) { - return bdi_congested(bdi, (1 << BDI_sync_congested) | - (1 << BDI_async_congested)); + return bdi_congested(bdi, (1 << WB_sync_congested) | + (1 << WB_async_congested)); } enum { diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 6dc4580df2af..b23cf0ea5912 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -96,7 +96,7 @@ static int bdi_debug_stats_show(struct seq_file *m, void *v) nr_io, nr_more_io, nr_dirty_time, - !list_empty(&bdi->bdi_list), bdi->state); + !list_empty(&bdi->bdi_list), bdi->wb.state); #undef K return 0; @@ -280,7 +280,7 @@ void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi) timeout = msecs_to_jiffies(dirty_writeback_interval * 10); spin_lock_bh(&bdi->wb_lock); - if (test_bit(BDI_registered, &bdi->state)) + if (test_bit(WB_registered, &bdi->wb.state)) queue_delayed_work(bdi_wq, &bdi->wb.dwork, timeout); spin_unlock_bh(&bdi->wb_lock); } @@ -315,7 +315,7 @@ int bdi_register(struct backing_dev_info *bdi, struct device *parent, bdi->dev = dev; bdi_debug_register(bdi, dev_name(dev)); - set_bit(BDI_registered, &bdi->state); + set_bit(WB_registered, &bdi->wb.state); spin_lock_bh(&bdi_lock); list_add_tail_rcu(&bdi->bdi_list, &bdi_list); @@ -339,7 +339,7 @@ static void bdi_wb_shutdown(struct backing_dev_info *bdi) { /* Make sure nobody queues further work */ spin_lock_bh(&bdi->wb_lock); - if (!test_and_clear_bit(BDI_registered, &bdi->state)) { + if (!test_and_clear_bit(WB_registered, &bdi->wb.state)) { spin_unlock_bh(&bdi->wb_lock); return; } @@ -492,11 +492,11 @@ static atomic_t nr_bdi_congested[2]; void clear_bdi_congested(struct backing_dev_info *bdi, int sync) { - enum bdi_state bit; + enum wb_state bit; wait_queue_head_t *wqh = &congestion_wqh[sync]; - bit = sync ? BDI_sync_congested : BDI_async_congested; - if (test_and_clear_bit(bit, &bdi->state)) + bit = sync ? WB_sync_congested : WB_async_congested; + if (test_and_clear_bit(bit, &bdi->wb.state)) atomic_dec(&nr_bdi_congested[sync]); smp_mb__after_atomic(); if (waitqueue_active(wqh)) @@ -506,10 +506,10 @@ EXPORT_SYMBOL(clear_bdi_congested); void set_bdi_congested(struct backing_dev_info *bdi, int sync) { - enum bdi_state bit; + enum wb_state bit; - bit = sync ? BDI_sync_congested : BDI_async_congested; - if (!test_and_set_bit(bit, &bdi->state)) + bit = sync ? WB_sync_congested : WB_async_congested; + if (!test_and_set_bit(bit, &bdi->wb.state)) atomic_inc(&nr_bdi_congested[sync]); } EXPORT_SYMBOL(set_bdi_congested); From 93f78d882865cb90020d0f80a9523c99cf46924c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:27 -0400 Subject: [PATCH 13/85] writeback: move backing_dev_info->bdi_stat[] into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->bdi_stat[] into wb. * enum bdi_stat_item is renamed to wb_stat_item and the prefix of all enums is changed from BDI_ to WB_. * BDI_STAT_BATCH() -> WB_STAT_BATCH() * [__]{add|inc|dec|sum}_wb_stat(bdi, ...) -> [__]{add|inc}_wb_stat(wb, ...) * bdi_stat[_error]() -> wb_stat[_error]() * bdi_writeout_inc() -> wb_writeout_inc() * stat init is moved to bdi_wb_init() and bdi_wb_exit() is added and frees stat. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Cc: Wu Fengguang Cc: Miklos Szeredi Cc: Trond Myklebust Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 2 +- fs/fuse/file.c | 12 +++---- fs/nfs/internal.h | 2 +- fs/nfs/write.c | 3 +- include/linux/backing-dev.h | 68 +++++++++++++++++-------------------- mm/backing-dev.c | 60 +++++++++++++++++++------------- mm/page-writeback.c | 55 +++++++++++++++--------------- 7 files changed, 106 insertions(+), 96 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 983312cea245..8873ecd1578c 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -840,7 +840,7 @@ static bool over_bground_thresh(struct backing_dev_info *bdi) global_page_state(NR_UNSTABLE_NFS) > background_thresh) return true; - if (bdi_stat(bdi, BDI_RECLAIMABLE) > + if (wb_stat(&bdi->wb, WB_RECLAIMABLE) > bdi_dirty_limit(bdi, background_thresh)) return true; diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 5ef05b5c4cff..8c5e2fa68835 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1445,9 +1445,9 @@ static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req) list_del(&req->writepages_entry); for (i = 0; i < req->num_pages; i++) { - dec_bdi_stat(bdi, BDI_WRITEBACK); + dec_wb_stat(&bdi->wb, WB_WRITEBACK); dec_zone_page_state(req->pages[i], NR_WRITEBACK_TEMP); - bdi_writeout_inc(bdi); + wb_writeout_inc(&bdi->wb); } wake_up(&fi->page_waitq); } @@ -1634,7 +1634,7 @@ static int fuse_writepage_locked(struct page *page) req->end = fuse_writepage_end; req->inode = inode; - inc_bdi_stat(inode_to_bdi(inode), BDI_WRITEBACK); + inc_wb_stat(&inode_to_bdi(inode)->wb, WB_WRITEBACK); inc_zone_page_state(tmp_page, NR_WRITEBACK_TEMP); spin_lock(&fc->lock); @@ -1749,9 +1749,9 @@ static bool fuse_writepage_in_flight(struct fuse_req *new_req, copy_highpage(old_req->pages[0], page); spin_unlock(&fc->lock); - dec_bdi_stat(bdi, BDI_WRITEBACK); + dec_wb_stat(&bdi->wb, WB_WRITEBACK); dec_zone_page_state(page, NR_WRITEBACK_TEMP); - bdi_writeout_inc(bdi); + wb_writeout_inc(&bdi->wb); fuse_writepage_free(fc, new_req); fuse_request_free(new_req); goto out; @@ -1848,7 +1848,7 @@ static int fuse_writepages_fill(struct page *page, req->page_descs[req->num_pages].offset = 0; req->page_descs[req->num_pages].length = PAGE_SIZE; - inc_bdi_stat(inode_to_bdi(inode), BDI_WRITEBACK); + inc_wb_stat(&inode_to_bdi(inode)->wb, WB_WRITEBACK); inc_zone_page_state(tmp_page, NR_WRITEBACK_TEMP); err = 0; diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 9e6475bc5ba2..7e3c4604bea8 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -607,7 +607,7 @@ void nfs_mark_page_unstable(struct page *page) struct inode *inode = page_file_mapping(page)->host; inc_zone_page_state(page, NR_UNSTABLE_NFS); - inc_bdi_stat(inode_to_bdi(inode), BDI_RECLAIMABLE); + inc_wb_stat(&inode_to_bdi(inode)->wb, WB_RECLAIMABLE); __mark_inode_dirty(inode, I_DIRTY_DATASYNC); } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index d12a4be613a5..94c7ce01dfb1 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -853,7 +853,8 @@ static void nfs_clear_page_commit(struct page *page) { dec_zone_page_state(page, NR_UNSTABLE_NFS); - dec_bdi_stat(inode_to_bdi(page_file_mapping(page)->host), BDI_RECLAIMABLE); + dec_wb_stat(&inode_to_bdi(page_file_mapping(page)->host)->wb, + WB_RECLAIMABLE); } /* Called holding inode (/cinfo) lock */ diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index eb14f988a63e..fe7a907a4e16 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -36,15 +36,15 @@ enum wb_state { typedef int (congested_fn)(void *, int); -enum bdi_stat_item { - BDI_RECLAIMABLE, - BDI_WRITEBACK, - BDI_DIRTIED, - BDI_WRITTEN, - NR_BDI_STAT_ITEMS +enum wb_stat_item { + WB_RECLAIMABLE, + WB_WRITEBACK, + WB_DIRTIED, + WB_WRITTEN, + NR_WB_STAT_ITEMS }; -#define BDI_STAT_BATCH (8*(1+ilog2(nr_cpu_ids))) +#define WB_STAT_BATCH (8*(1+ilog2(nr_cpu_ids))) struct bdi_writeback { struct backing_dev_info *bdi; /* our parent bdi */ @@ -58,6 +58,8 @@ struct bdi_writeback { struct list_head b_more_io; /* parked for more writeback */ struct list_head b_dirty_time; /* time stamps are dirty */ spinlock_t list_lock; /* protects the b_* lists */ + + struct percpu_counter stat[NR_WB_STAT_ITEMS]; }; struct backing_dev_info { @@ -69,8 +71,6 @@ struct backing_dev_info { char *name; - struct percpu_counter bdi_stat[NR_BDI_STAT_ITEMS]; - unsigned long bw_time_stamp; /* last time write bw is updated */ unsigned long dirtied_stamp; unsigned long written_stamp; /* pages written at bw_time_stamp */ @@ -137,78 +137,74 @@ static inline int wb_has_dirty_io(struct bdi_writeback *wb) !list_empty(&wb->b_more_io); } -static inline void __add_bdi_stat(struct backing_dev_info *bdi, - enum bdi_stat_item item, s64 amount) +static inline void __add_wb_stat(struct bdi_writeback *wb, + enum wb_stat_item item, s64 amount) { - __percpu_counter_add(&bdi->bdi_stat[item], amount, BDI_STAT_BATCH); + __percpu_counter_add(&wb->stat[item], amount, WB_STAT_BATCH); } -static inline void __inc_bdi_stat(struct backing_dev_info *bdi, - enum bdi_stat_item item) +static inline void __inc_wb_stat(struct bdi_writeback *wb, + enum wb_stat_item item) { - __add_bdi_stat(bdi, item, 1); + __add_wb_stat(wb, item, 1); } -static inline void inc_bdi_stat(struct backing_dev_info *bdi, - enum bdi_stat_item item) +static inline void inc_wb_stat(struct bdi_writeback *wb, enum wb_stat_item item) { unsigned long flags; local_irq_save(flags); - __inc_bdi_stat(bdi, item); + __inc_wb_stat(wb, item); local_irq_restore(flags); } -static inline void __dec_bdi_stat(struct backing_dev_info *bdi, - enum bdi_stat_item item) +static inline void __dec_wb_stat(struct bdi_writeback *wb, + enum wb_stat_item item) { - __add_bdi_stat(bdi, item, -1); + __add_wb_stat(wb, item, -1); } -static inline void dec_bdi_stat(struct backing_dev_info *bdi, - enum bdi_stat_item item) +static inline void dec_wb_stat(struct bdi_writeback *wb, enum wb_stat_item item) { unsigned long flags; local_irq_save(flags); - __dec_bdi_stat(bdi, item); + __dec_wb_stat(wb, item); local_irq_restore(flags); } -static inline s64 bdi_stat(struct backing_dev_info *bdi, - enum bdi_stat_item item) +static inline s64 wb_stat(struct bdi_writeback *wb, enum wb_stat_item item) { - return percpu_counter_read_positive(&bdi->bdi_stat[item]); + return percpu_counter_read_positive(&wb->stat[item]); } -static inline s64 __bdi_stat_sum(struct backing_dev_info *bdi, - enum bdi_stat_item item) +static inline s64 __wb_stat_sum(struct bdi_writeback *wb, + enum wb_stat_item item) { - return percpu_counter_sum_positive(&bdi->bdi_stat[item]); + return percpu_counter_sum_positive(&wb->stat[item]); } -static inline s64 bdi_stat_sum(struct backing_dev_info *bdi, - enum bdi_stat_item item) +static inline s64 wb_stat_sum(struct bdi_writeback *wb, enum wb_stat_item item) { s64 sum; unsigned long flags; local_irq_save(flags); - sum = __bdi_stat_sum(bdi, item); + sum = __wb_stat_sum(wb, item); local_irq_restore(flags); return sum; } -extern void bdi_writeout_inc(struct backing_dev_info *bdi); +extern void wb_writeout_inc(struct bdi_writeback *wb); /* * maximal error of a stat counter. */ -static inline unsigned long bdi_stat_error(struct backing_dev_info *bdi) +static inline unsigned long wb_stat_error(struct bdi_writeback *wb) { #ifdef CONFIG_SMP - return nr_cpu_ids * BDI_STAT_BATCH; + return nr_cpu_ids * WB_STAT_BATCH; #else return 1; #endif diff --git a/mm/backing-dev.c b/mm/backing-dev.c index b23cf0ea5912..7b1d1917b658 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -84,13 +84,13 @@ static int bdi_debug_stats_show(struct seq_file *m, void *v) "b_dirty_time: %10lu\n" "bdi_list: %10u\n" "state: %10lx\n", - (unsigned long) K(bdi_stat(bdi, BDI_WRITEBACK)), - (unsigned long) K(bdi_stat(bdi, BDI_RECLAIMABLE)), + (unsigned long) K(wb_stat(wb, WB_WRITEBACK)), + (unsigned long) K(wb_stat(wb, WB_RECLAIMABLE)), K(bdi_thresh), K(dirty_thresh), K(background_thresh), - (unsigned long) K(bdi_stat(bdi, BDI_DIRTIED)), - (unsigned long) K(bdi_stat(bdi, BDI_WRITTEN)), + (unsigned long) K(wb_stat(wb, WB_DIRTIED)), + (unsigned long) K(wb_stat(wb, WB_WRITTEN)), (unsigned long) K(bdi->write_bandwidth), nr_dirty, nr_io, @@ -376,8 +376,10 @@ void bdi_unregister(struct backing_dev_info *bdi) } EXPORT_SYMBOL(bdi_unregister); -static void bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) +static int bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) { + int i, err; + memset(wb, 0, sizeof(*wb)); wb->bdi = bdi; @@ -388,6 +390,27 @@ static void bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) INIT_LIST_HEAD(&wb->b_dirty_time); spin_lock_init(&wb->list_lock); INIT_DELAYED_WORK(&wb->dwork, bdi_writeback_workfn); + + for (i = 0; i < NR_WB_STAT_ITEMS; i++) { + err = percpu_counter_init(&wb->stat[i], 0, GFP_KERNEL); + if (err) { + while (--i) + percpu_counter_destroy(&wb->stat[i]); + return err; + } + } + + return 0; +} + +static void bdi_wb_exit(struct bdi_writeback *wb) +{ + int i; + + WARN_ON(delayed_work_pending(&wb->dwork)); + + for (i = 0; i < NR_WB_STAT_ITEMS; i++) + percpu_counter_destroy(&wb->stat[i]); } /* @@ -397,7 +420,7 @@ static void bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) int bdi_init(struct backing_dev_info *bdi) { - int i, err; + int err; bdi->dev = NULL; @@ -408,13 +431,9 @@ int bdi_init(struct backing_dev_info *bdi) INIT_LIST_HEAD(&bdi->bdi_list); INIT_LIST_HEAD(&bdi->work_list); - bdi_wb_init(&bdi->wb, bdi); - - for (i = 0; i < NR_BDI_STAT_ITEMS; i++) { - err = percpu_counter_init(&bdi->bdi_stat[i], 0, GFP_KERNEL); - if (err) - goto err; - } + err = bdi_wb_init(&bdi->wb, bdi); + if (err) + return err; bdi->dirty_exceeded = 0; @@ -427,25 +446,20 @@ int bdi_init(struct backing_dev_info *bdi) bdi->avg_write_bandwidth = INIT_BW; err = fprop_local_init_percpu(&bdi->completions, GFP_KERNEL); - if (err) { -err: - while (i--) - percpu_counter_destroy(&bdi->bdi_stat[i]); + bdi_wb_exit(&bdi->wb); + return err; } - return err; + return 0; } EXPORT_SYMBOL(bdi_init); void bdi_destroy(struct backing_dev_info *bdi) { - int i; - bdi_wb_shutdown(bdi); WARN_ON(!list_empty(&bdi->work_list)); - WARN_ON(delayed_work_pending(&bdi->wb.dwork)); if (bdi->dev) { bdi_debug_unregister(bdi); @@ -453,8 +467,8 @@ void bdi_destroy(struct backing_dev_info *bdi) bdi->dev = NULL; } - for (i = 0; i < NR_BDI_STAT_ITEMS; i++) - percpu_counter_destroy(&bdi->bdi_stat[i]); + bdi_wb_exit(&bdi->wb); + fprop_local_destroy_percpu(&bdi->completions); } EXPORT_SYMBOL(bdi_destroy); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index bdeecad00489..dc673a035413 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -396,11 +396,11 @@ static unsigned long wp_next_time(unsigned long cur_time) * Increment the BDI's writeout completion count and the global writeout * completion count. Called from test_clear_page_writeback(). */ -static inline void __bdi_writeout_inc(struct backing_dev_info *bdi) +static inline void __wb_writeout_inc(struct bdi_writeback *wb) { - __inc_bdi_stat(bdi, BDI_WRITTEN); - __fprop_inc_percpu_max(&writeout_completions, &bdi->completions, - bdi->max_prop_frac); + __inc_wb_stat(wb, WB_WRITTEN); + __fprop_inc_percpu_max(&writeout_completions, &wb->bdi->completions, + wb->bdi->max_prop_frac); /* First event after period switching was turned off? */ if (!unlikely(writeout_period_time)) { /* @@ -414,15 +414,15 @@ static inline void __bdi_writeout_inc(struct backing_dev_info *bdi) } } -void bdi_writeout_inc(struct backing_dev_info *bdi) +void wb_writeout_inc(struct bdi_writeback *wb) { unsigned long flags; local_irq_save(flags); - __bdi_writeout_inc(bdi); + __wb_writeout_inc(wb); local_irq_restore(flags); } -EXPORT_SYMBOL_GPL(bdi_writeout_inc); +EXPORT_SYMBOL_GPL(wb_writeout_inc); /* * Obtain an accurate fraction of the BDI's portion. @@ -1130,8 +1130,8 @@ void __bdi_update_bandwidth(struct backing_dev_info *bdi, if (elapsed < BANDWIDTH_INTERVAL) return; - dirtied = percpu_counter_read(&bdi->bdi_stat[BDI_DIRTIED]); - written = percpu_counter_read(&bdi->bdi_stat[BDI_WRITTEN]); + dirtied = percpu_counter_read(&bdi->wb.stat[WB_DIRTIED]); + written = percpu_counter_read(&bdi->wb.stat[WB_WRITTEN]); /* * Skip quiet periods when disk bandwidth is under-utilized. @@ -1288,7 +1288,8 @@ static inline void bdi_dirty_limits(struct backing_dev_info *bdi, unsigned long *bdi_thresh, unsigned long *bdi_bg_thresh) { - unsigned long bdi_reclaimable; + struct bdi_writeback *wb = &bdi->wb; + unsigned long wb_reclaimable; /* * bdi_thresh is not treated as some limiting factor as @@ -1320,14 +1321,12 @@ static inline void bdi_dirty_limits(struct backing_dev_info *bdi, * actually dirty; with m+n sitting in the percpu * deltas. */ - if (*bdi_thresh < 2 * bdi_stat_error(bdi)) { - bdi_reclaimable = bdi_stat_sum(bdi, BDI_RECLAIMABLE); - *bdi_dirty = bdi_reclaimable + - bdi_stat_sum(bdi, BDI_WRITEBACK); + if (*bdi_thresh < 2 * wb_stat_error(wb)) { + wb_reclaimable = wb_stat_sum(wb, WB_RECLAIMABLE); + *bdi_dirty = wb_reclaimable + wb_stat_sum(wb, WB_WRITEBACK); } else { - bdi_reclaimable = bdi_stat(bdi, BDI_RECLAIMABLE); - *bdi_dirty = bdi_reclaimable + - bdi_stat(bdi, BDI_WRITEBACK); + wb_reclaimable = wb_stat(wb, WB_RECLAIMABLE); + *bdi_dirty = wb_reclaimable + wb_stat(wb, WB_WRITEBACK); } } @@ -1514,9 +1513,9 @@ pause: * In theory 1 page is enough to keep the comsumer-producer * pipe going: the flusher cleans 1 page => the task dirties 1 * more page. However bdi_dirty has accounting errors. So use - * the larger and more IO friendly bdi_stat_error. + * the larger and more IO friendly wb_stat_error. */ - if (bdi_dirty <= bdi_stat_error(bdi)) + if (bdi_dirty <= wb_stat_error(&bdi->wb)) break; if (fatal_signal_pending(current)) @@ -2106,8 +2105,8 @@ void account_page_dirtied(struct page *page, struct address_space *mapping, mem_cgroup_inc_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); __inc_zone_page_state(page, NR_FILE_DIRTY); __inc_zone_page_state(page, NR_DIRTIED); - __inc_bdi_stat(bdi, BDI_RECLAIMABLE); - __inc_bdi_stat(bdi, BDI_DIRTIED); + __inc_wb_stat(&bdi->wb, WB_RECLAIMABLE); + __inc_wb_stat(&bdi->wb, WB_DIRTIED); task_io_account_write(PAGE_CACHE_SIZE); current->nr_dirtied++; this_cpu_inc(bdp_ratelimits); @@ -2126,7 +2125,7 @@ void account_page_cleaned(struct page *page, struct address_space *mapping, if (mapping_cap_account_dirty(mapping)) { mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); dec_zone_page_state(page, NR_FILE_DIRTY); - dec_bdi_stat(inode_to_bdi(mapping->host), BDI_RECLAIMABLE); + dec_wb_stat(&inode_to_bdi(mapping->host)->wb, WB_RECLAIMABLE); task_io_account_cancelled_write(PAGE_CACHE_SIZE); } } @@ -2190,7 +2189,7 @@ void account_page_redirty(struct page *page) if (mapping && mapping_cap_account_dirty(mapping)) { current->nr_dirtied--; dec_zone_page_state(page, NR_DIRTIED); - dec_bdi_stat(inode_to_bdi(mapping->host), BDI_DIRTIED); + dec_wb_stat(&inode_to_bdi(mapping->host)->wb, WB_DIRTIED); } } EXPORT_SYMBOL(account_page_redirty); @@ -2369,8 +2368,8 @@ int clear_page_dirty_for_io(struct page *page) if (TestClearPageDirty(page)) { mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); dec_zone_page_state(page, NR_FILE_DIRTY); - dec_bdi_stat(inode_to_bdi(mapping->host), - BDI_RECLAIMABLE); + dec_wb_stat(&inode_to_bdi(mapping->host)->wb, + WB_RECLAIMABLE); ret = 1; } mem_cgroup_end_page_stat(memcg); @@ -2398,8 +2397,8 @@ int test_clear_page_writeback(struct page *page) page_index(page), PAGECACHE_TAG_WRITEBACK); if (bdi_cap_account_writeback(bdi)) { - __dec_bdi_stat(bdi, BDI_WRITEBACK); - __bdi_writeout_inc(bdi); + __dec_wb_stat(&bdi->wb, WB_WRITEBACK); + __wb_writeout_inc(&bdi->wb); } } spin_unlock_irqrestore(&mapping->tree_lock, flags); @@ -2433,7 +2432,7 @@ int __test_set_page_writeback(struct page *page, bool keep_write) page_index(page), PAGECACHE_TAG_WRITEBACK); if (bdi_cap_account_writeback(bdi)) - __inc_bdi_stat(bdi, BDI_WRITEBACK); + __inc_wb_stat(&bdi->wb, WB_WRITEBACK); } if (!PageDirty(page)) radix_tree_tag_clear(&mapping->page_tree, From a88a341a73be4ef035ca26170c849f002797da27 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:28 -0400 Subject: [PATCH 14/85] writeback: move bandwidth related fields from backing_dev_info into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bandwidth related fields from backing_dev_info into bdi_writeback. * The moved fields are: bw_time_stamp, dirtied_stamp, written_stamp, write_bandwidth, avg_write_bandwidth, dirty_ratelimit, balanced_dirty_ratelimit, completions and dirty_exceeded. * writeback_chunk_size() and over_bground_thresh() now take @wb instead of @bdi. * bdi_writeout_fraction(bdi, ...) -> wb_writeout_fraction(wb, ...) bdi_dirty_limit(bdi, ...) -> wb_dirty_limit(wb, ...) bdi_position_ration(bdi, ...) -> wb_position_ratio(wb, ...) bdi_update_writebandwidth(bdi, ...) -> wb_update_write_bandwidth(wb, ...) [__]bdi_update_bandwidth(bdi, ...) -> [__]wb_update_bandwidth(wb, ...) bdi_{max|min}_pause(bdi, ...) -> wb_{max|min}_pause(wb, ...) bdi_dirty_limits(bdi, ...) -> wb_dirty_limits(wb, ...) * Init/exits of the relocated fields are moved to bdi_wb_init/exit() respectively. Note that explicit zeroing is dropped in the process as wb's are cleared in entirety anyway. * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->stat[] are mechanically replaced with bdi->wb.stat[] introducing no behavior changes. v2: Typo in description fixed as suggested by Jan. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Cc: Wu Fengguang Cc: Jaegeuk Kim Cc: Steven Whitehouse Signed-off-by: Jens Axboe --- fs/f2fs/node.c | 4 +- fs/f2fs/segment.h | 2 +- fs/fs-writeback.c | 17 +- fs/gfs2/super.c | 2 +- include/linux/backing-dev.h | 20 +-- include/linux/writeback.h | 17 +- include/trace/events/writeback.h | 8 +- mm/backing-dev.c | 45 +++--- mm/page-writeback.c | 262 ++++++++++++++++--------------- 9 files changed, 186 insertions(+), 191 deletions(-) diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 8ab0cf1930bd..d211602e0f86 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -53,7 +53,7 @@ bool available_free_memory(struct f2fs_sb_info *sbi, int type) PAGE_CACHE_SHIFT; res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 2); } else if (type == DIRTY_DENTS) { - if (sbi->sb->s_bdi->dirty_exceeded) + if (sbi->sb->s_bdi->wb.dirty_exceeded) return false; mem_size = get_pages(sbi, F2FS_DIRTY_DENTS); res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1); @@ -70,7 +70,7 @@ bool available_free_memory(struct f2fs_sb_info *sbi, int type) sizeof(struct extent_node)) >> PAGE_CACHE_SHIFT; res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1); } else { - if (sbi->sb->s_bdi->dirty_exceeded) + if (sbi->sb->s_bdi->wb.dirty_exceeded) return false; } return res; diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 85d7fa7514b2..6408989857e0 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -713,7 +713,7 @@ static inline unsigned int max_hw_blocks(struct f2fs_sb_info *sbi) */ static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type) { - if (sbi->sb->s_bdi->dirty_exceeded) + if (sbi->sb->s_bdi->wb.dirty_exceeded) return 0; if (type == DATA) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 8873ecd1578c..1945cb9f6b97 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -624,7 +624,7 @@ out: return ret; } -static long writeback_chunk_size(struct backing_dev_info *bdi, +static long writeback_chunk_size(struct bdi_writeback *wb, struct wb_writeback_work *work) { long pages; @@ -645,7 +645,7 @@ static long writeback_chunk_size(struct backing_dev_info *bdi, if (work->sync_mode == WB_SYNC_ALL || work->tagged_writepages) pages = LONG_MAX; else { - pages = min(bdi->avg_write_bandwidth / 2, + pages = min(wb->avg_write_bandwidth / 2, global_dirty_limit / DIRTY_SCOPE); pages = min(pages, work->nr_pages); pages = round_down(pages + MIN_WRITEBACK_PAGES, @@ -743,7 +743,7 @@ static long writeback_sb_inodes(struct super_block *sb, inode->i_state |= I_SYNC; spin_unlock(&inode->i_lock); - write_chunk = writeback_chunk_size(wb->bdi, work); + write_chunk = writeback_chunk_size(wb, work); wbc.nr_to_write = write_chunk; wbc.pages_skipped = 0; @@ -830,7 +830,7 @@ static long writeback_inodes_wb(struct bdi_writeback *wb, long nr_pages, return nr_pages - work.nr_pages; } -static bool over_bground_thresh(struct backing_dev_info *bdi) +static bool over_bground_thresh(struct bdi_writeback *wb) { unsigned long background_thresh, dirty_thresh; @@ -840,8 +840,7 @@ static bool over_bground_thresh(struct backing_dev_info *bdi) global_page_state(NR_UNSTABLE_NFS) > background_thresh) return true; - if (wb_stat(&bdi->wb, WB_RECLAIMABLE) > - bdi_dirty_limit(bdi, background_thresh)) + if (wb_stat(wb, WB_RECLAIMABLE) > wb_dirty_limit(wb, background_thresh)) return true; return false; @@ -854,7 +853,7 @@ static bool over_bground_thresh(struct backing_dev_info *bdi) static void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long start_time) { - __bdi_update_bandwidth(wb->bdi, 0, 0, 0, 0, 0, start_time); + __wb_update_bandwidth(wb, 0, 0, 0, 0, 0, start_time); } /* @@ -906,7 +905,7 @@ static long wb_writeback(struct bdi_writeback *wb, * For background writeout, stop when we are below the * background dirty threshold */ - if (work->for_background && !over_bground_thresh(wb->bdi)) + if (work->for_background && !over_bground_thresh(wb)) break; /* @@ -998,7 +997,7 @@ static unsigned long get_nr_dirty_pages(void) static long wb_check_background_flush(struct bdi_writeback *wb) { - if (over_bground_thresh(wb->bdi)) { + if (over_bground_thresh(wb)) { struct wb_writeback_work work = { .nr_pages = LONG_MAX, diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 859c6edbf81a..2982445947e1 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -748,7 +748,7 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc) if (wbc->sync_mode == WB_SYNC_ALL) gfs2_log_flush(GFS2_SB(inode), ip->i_gl, NORMAL_FLUSH); - if (bdi->dirty_exceeded) + if (bdi->wb.dirty_exceeded) gfs2_ail1_flush(sdp, wbc); else filemap_fdatawrite(metamapping); diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index fe7a907a4e16..2ab06049d812 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -60,16 +60,6 @@ struct bdi_writeback { spinlock_t list_lock; /* protects the b_* lists */ struct percpu_counter stat[NR_WB_STAT_ITEMS]; -}; - -struct backing_dev_info { - struct list_head bdi_list; - unsigned long ra_pages; /* max readahead in PAGE_CACHE_SIZE units */ - unsigned int capabilities; /* Device capabilities */ - congested_fn *congested_fn; /* Function pointer if device is md/dm */ - void *congested_data; /* Pointer to aux data for congested func */ - - char *name; unsigned long bw_time_stamp; /* last time write bw is updated */ unsigned long dirtied_stamp; @@ -88,6 +78,16 @@ struct backing_dev_info { struct fprop_local_percpu completions; int dirty_exceeded; +}; + +struct backing_dev_info { + struct list_head bdi_list; + unsigned long ra_pages; /* max readahead in PAGE_CACHE_SIZE units */ + unsigned int capabilities; /* Device capabilities */ + congested_fn *congested_fn; /* Function pointer if device is md/dm */ + void *congested_data; /* Pointer to aux data for congested func */ + + char *name; unsigned int min_ratio; unsigned int max_ratio, max_prop_frac; diff --git a/include/linux/writeback.h b/include/linux/writeback.h index b2dd371ec0ca..a6b9db7fcee8 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -155,16 +155,15 @@ int dirty_writeback_centisecs_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *); void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty); -unsigned long bdi_dirty_limit(struct backing_dev_info *bdi, - unsigned long dirty); +unsigned long wb_dirty_limit(struct bdi_writeback *wb, unsigned long dirty); -void __bdi_update_bandwidth(struct backing_dev_info *bdi, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty, - unsigned long start_time); +void __wb_update_bandwidth(struct bdi_writeback *wb, + unsigned long thresh, + unsigned long bg_thresh, + unsigned long dirty, + unsigned long bdi_thresh, + unsigned long bdi_dirty, + unsigned long start_time); void page_writeback_init(void); void balance_dirty_pages_ratelimited(struct address_space *mapping); diff --git a/include/trace/events/writeback.h b/include/trace/events/writeback.h index 880dd7437172..9b876f6cc81a 100644 --- a/include/trace/events/writeback.h +++ b/include/trace/events/writeback.h @@ -400,13 +400,13 @@ TRACE_EVENT(bdi_dirty_ratelimit, TP_fast_assign( strlcpy(__entry->bdi, dev_name(bdi->dev), 32); - __entry->write_bw = KBps(bdi->write_bandwidth); - __entry->avg_write_bw = KBps(bdi->avg_write_bandwidth); + __entry->write_bw = KBps(bdi->wb.write_bandwidth); + __entry->avg_write_bw = KBps(bdi->wb.avg_write_bandwidth); __entry->dirty_rate = KBps(dirty_rate); - __entry->dirty_ratelimit = KBps(bdi->dirty_ratelimit); + __entry->dirty_ratelimit = KBps(bdi->wb.dirty_ratelimit); __entry->task_ratelimit = KBps(task_ratelimit); __entry->balanced_dirty_ratelimit = - KBps(bdi->balanced_dirty_ratelimit); + KBps(bdi->wb.balanced_dirty_ratelimit); ), TP_printk("bdi %s: " diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 7b1d1917b658..9a6c472ebd0b 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -66,7 +66,7 @@ static int bdi_debug_stats_show(struct seq_file *m, void *v) spin_unlock(&wb->list_lock); global_dirty_limits(&background_thresh, &dirty_thresh); - bdi_thresh = bdi_dirty_limit(bdi, dirty_thresh); + bdi_thresh = wb_dirty_limit(wb, dirty_thresh); #define K(x) ((x) << (PAGE_SHIFT - 10)) seq_printf(m, @@ -91,7 +91,7 @@ static int bdi_debug_stats_show(struct seq_file *m, void *v) K(background_thresh), (unsigned long) K(wb_stat(wb, WB_DIRTIED)), (unsigned long) K(wb_stat(wb, WB_WRITTEN)), - (unsigned long) K(bdi->write_bandwidth), + (unsigned long) K(wb->write_bandwidth), nr_dirty, nr_io, nr_more_io, @@ -376,6 +376,11 @@ void bdi_unregister(struct backing_dev_info *bdi) } EXPORT_SYMBOL(bdi_unregister); +/* + * Initial write bandwidth: 100 MB/s + */ +#define INIT_BW (100 << (20 - PAGE_SHIFT)) + static int bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) { int i, err; @@ -391,11 +396,22 @@ static int bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) spin_lock_init(&wb->list_lock); INIT_DELAYED_WORK(&wb->dwork, bdi_writeback_workfn); + wb->bw_time_stamp = jiffies; + wb->balanced_dirty_ratelimit = INIT_BW; + wb->dirty_ratelimit = INIT_BW; + wb->write_bandwidth = INIT_BW; + wb->avg_write_bandwidth = INIT_BW; + + err = fprop_local_init_percpu(&wb->completions, GFP_KERNEL); + if (err) + return err; + for (i = 0; i < NR_WB_STAT_ITEMS; i++) { err = percpu_counter_init(&wb->stat[i], 0, GFP_KERNEL); if (err) { while (--i) percpu_counter_destroy(&wb->stat[i]); + fprop_local_destroy_percpu(&wb->completions); return err; } } @@ -411,12 +427,9 @@ static void bdi_wb_exit(struct bdi_writeback *wb) for (i = 0; i < NR_WB_STAT_ITEMS; i++) percpu_counter_destroy(&wb->stat[i]); -} -/* - * Initial write bandwidth: 100 MB/s - */ -#define INIT_BW (100 << (20 - PAGE_SHIFT)) + fprop_local_destroy_percpu(&wb->completions); +} int bdi_init(struct backing_dev_info *bdi) { @@ -435,22 +448,6 @@ int bdi_init(struct backing_dev_info *bdi) if (err) return err; - bdi->dirty_exceeded = 0; - - bdi->bw_time_stamp = jiffies; - bdi->written_stamp = 0; - - bdi->balanced_dirty_ratelimit = INIT_BW; - bdi->dirty_ratelimit = INIT_BW; - bdi->write_bandwidth = INIT_BW; - bdi->avg_write_bandwidth = INIT_BW; - - err = fprop_local_init_percpu(&bdi->completions, GFP_KERNEL); - if (err) { - bdi_wb_exit(&bdi->wb); - return err; - } - return 0; } EXPORT_SYMBOL(bdi_init); @@ -468,8 +465,6 @@ void bdi_destroy(struct backing_dev_info *bdi) } bdi_wb_exit(&bdi->wb); - - fprop_local_destroy_percpu(&bdi->completions); } EXPORT_SYMBOL(bdi_destroy); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index dc673a035413..cd39ee91b7bb 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -399,7 +399,7 @@ static unsigned long wp_next_time(unsigned long cur_time) static inline void __wb_writeout_inc(struct bdi_writeback *wb) { __inc_wb_stat(wb, WB_WRITTEN); - __fprop_inc_percpu_max(&writeout_completions, &wb->bdi->completions, + __fprop_inc_percpu_max(&writeout_completions, &wb->completions, wb->bdi->max_prop_frac); /* First event after period switching was turned off? */ if (!unlikely(writeout_period_time)) { @@ -427,10 +427,10 @@ EXPORT_SYMBOL_GPL(wb_writeout_inc); /* * Obtain an accurate fraction of the BDI's portion. */ -static void bdi_writeout_fraction(struct backing_dev_info *bdi, - long *numerator, long *denominator) +static void wb_writeout_fraction(struct bdi_writeback *wb, + long *numerator, long *denominator) { - fprop_fraction_percpu(&writeout_completions, &bdi->completions, + fprop_fraction_percpu(&writeout_completions, &wb->completions, numerator, denominator); } @@ -516,11 +516,11 @@ static unsigned long hard_dirty_limit(unsigned long thresh) } /** - * bdi_dirty_limit - @bdi's share of dirty throttling threshold - * @bdi: the backing_dev_info to query + * wb_dirty_limit - @wb's share of dirty throttling threshold + * @wb: bdi_writeback to query * @dirty: global dirty limit in pages * - * Returns @bdi's dirty limit in pages. The term "dirty" in the context of + * Returns @wb's dirty limit in pages. The term "dirty" in the context of * dirty balancing includes all PG_dirty, PG_writeback and NFS unstable pages. * * Note that balance_dirty_pages() will only seriously take it as a hard limit @@ -528,34 +528,35 @@ static unsigned long hard_dirty_limit(unsigned long thresh) * control. For example, when the device is completely stalled due to some error * conditions, or when there are 1000 dd tasks writing to a slow 10MB/s USB key. * In the other normal situations, it acts more gently by throttling the tasks - * more (rather than completely block them) when the bdi dirty pages go high. + * more (rather than completely block them) when the wb dirty pages go high. * * It allocates high/low dirty limits to fast/slow devices, in order to prevent * - starving fast devices * - piling up dirty pages (that will take long time to sync) on slow devices * - * The bdi's share of dirty limit will be adapting to its throughput and + * The wb's share of dirty limit will be adapting to its throughput and * bounded by the bdi->min_ratio and/or bdi->max_ratio parameters, if set. */ -unsigned long bdi_dirty_limit(struct backing_dev_info *bdi, unsigned long dirty) +unsigned long wb_dirty_limit(struct bdi_writeback *wb, unsigned long dirty) { - u64 bdi_dirty; + struct backing_dev_info *bdi = wb->bdi; + u64 wb_dirty; long numerator, denominator; /* * Calculate this BDI's share of the dirty ratio. */ - bdi_writeout_fraction(bdi, &numerator, &denominator); + wb_writeout_fraction(wb, &numerator, &denominator); - bdi_dirty = (dirty * (100 - bdi_min_ratio)) / 100; - bdi_dirty *= numerator; - do_div(bdi_dirty, denominator); + wb_dirty = (dirty * (100 - bdi_min_ratio)) / 100; + wb_dirty *= numerator; + do_div(wb_dirty, denominator); - bdi_dirty += (dirty * bdi->min_ratio) / 100; - if (bdi_dirty > (dirty * bdi->max_ratio) / 100) - bdi_dirty = dirty * bdi->max_ratio / 100; + wb_dirty += (dirty * bdi->min_ratio) / 100; + if (wb_dirty > (dirty * bdi->max_ratio) / 100) + wb_dirty = dirty * bdi->max_ratio / 100; - return bdi_dirty; + return wb_dirty; } /* @@ -664,14 +665,14 @@ static long long pos_ratio_polynom(unsigned long setpoint, * card's bdi_dirty may rush to many times higher than bdi_setpoint. * - the bdi dirty thresh drops quickly due to change of JBOD workload */ -static unsigned long bdi_position_ratio(struct backing_dev_info *bdi, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty) +static unsigned long wb_position_ratio(struct bdi_writeback *wb, + unsigned long thresh, + unsigned long bg_thresh, + unsigned long dirty, + unsigned long bdi_thresh, + unsigned long bdi_dirty) { - unsigned long write_bw = bdi->avg_write_bandwidth; + unsigned long write_bw = wb->avg_write_bandwidth; unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh); unsigned long limit = hard_dirty_limit(thresh); unsigned long x_intercept; @@ -702,12 +703,12 @@ static unsigned long bdi_position_ratio(struct backing_dev_info *bdi, * consume arbitrary amount of RAM because it is accounted in * NR_WRITEBACK_TEMP which is not involved in calculating "nr_dirty". * - * Here, in bdi_position_ratio(), we calculate pos_ratio based on + * Here, in wb_position_ratio(), we calculate pos_ratio based on * two values: bdi_dirty and bdi_thresh. Let's consider an example: * total amount of RAM is 16GB, bdi->max_ratio is equal to 1%, global * limits are set by default to 10% and 20% (background and throttle). * Then bdi_thresh is 1% of 20% of 16GB. This amounts to ~8K pages. - * bdi_dirty_limit(bdi, bg_thresh) is about ~4K pages. bdi_setpoint is + * wb_dirty_limit(wb, bg_thresh) is about ~4K pages. bdi_setpoint is * about ~6K pages (as the average of background and throttle bdi * limits). The 3rd order polynomial will provide positive feedback if * bdi_dirty is under bdi_setpoint and vice versa. @@ -717,7 +718,7 @@ static unsigned long bdi_position_ratio(struct backing_dev_info *bdi, * much earlier than global "freerun" is reached (~23MB vs. ~2.3GB * in the example above). */ - if (unlikely(bdi->capabilities & BDI_CAP_STRICTLIMIT)) { + if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { long long bdi_pos_ratio; unsigned long bdi_bg_thresh; @@ -842,13 +843,13 @@ static unsigned long bdi_position_ratio(struct backing_dev_info *bdi, return pos_ratio; } -static void bdi_update_write_bandwidth(struct backing_dev_info *bdi, - unsigned long elapsed, - unsigned long written) +static void wb_update_write_bandwidth(struct bdi_writeback *wb, + unsigned long elapsed, + unsigned long written) { const unsigned long period = roundup_pow_of_two(3 * HZ); - unsigned long avg = bdi->avg_write_bandwidth; - unsigned long old = bdi->write_bandwidth; + unsigned long avg = wb->avg_write_bandwidth; + unsigned long old = wb->write_bandwidth; u64 bw; /* @@ -861,14 +862,14 @@ static void bdi_update_write_bandwidth(struct backing_dev_info *bdi, * @written may have decreased due to account_page_redirty(). * Avoid underflowing @bw calculation. */ - bw = written - min(written, bdi->written_stamp); + bw = written - min(written, wb->written_stamp); bw *= HZ; if (unlikely(elapsed > period)) { do_div(bw, elapsed); avg = bw; goto out; } - bw += (u64)bdi->write_bandwidth * (period - elapsed); + bw += (u64)wb->write_bandwidth * (period - elapsed); bw >>= ilog2(period); /* @@ -881,8 +882,8 @@ static void bdi_update_write_bandwidth(struct backing_dev_info *bdi, avg += (old - avg) >> 3; out: - bdi->write_bandwidth = bw; - bdi->avg_write_bandwidth = avg; + wb->write_bandwidth = bw; + wb->avg_write_bandwidth = avg; } /* @@ -947,20 +948,20 @@ static void global_update_bandwidth(unsigned long thresh, * Normal bdi tasks will be curbed at or below it in long term. * Obviously it should be around (write_bw / N) when there are N dd tasks. */ -static void bdi_update_dirty_ratelimit(struct backing_dev_info *bdi, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty, - unsigned long dirtied, - unsigned long elapsed) +static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, + unsigned long thresh, + unsigned long bg_thresh, + unsigned long dirty, + unsigned long bdi_thresh, + unsigned long bdi_dirty, + unsigned long dirtied, + unsigned long elapsed) { unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh); unsigned long limit = hard_dirty_limit(thresh); unsigned long setpoint = (freerun + limit) / 2; - unsigned long write_bw = bdi->avg_write_bandwidth; - unsigned long dirty_ratelimit = bdi->dirty_ratelimit; + unsigned long write_bw = wb->avg_write_bandwidth; + unsigned long dirty_ratelimit = wb->dirty_ratelimit; unsigned long dirty_rate; unsigned long task_ratelimit; unsigned long balanced_dirty_ratelimit; @@ -972,10 +973,10 @@ static void bdi_update_dirty_ratelimit(struct backing_dev_info *bdi, * The dirty rate will match the writeout rate in long term, except * when dirty pages are truncated by userspace or re-dirtied by FS. */ - dirty_rate = (dirtied - bdi->dirtied_stamp) * HZ / elapsed; + dirty_rate = (dirtied - wb->dirtied_stamp) * HZ / elapsed; - pos_ratio = bdi_position_ratio(bdi, thresh, bg_thresh, dirty, - bdi_thresh, bdi_dirty); + pos_ratio = wb_position_ratio(wb, thresh, bg_thresh, dirty, + bdi_thresh, bdi_dirty); /* * task_ratelimit reflects each dd's dirty rate for the past 200ms. */ @@ -1059,31 +1060,31 @@ static void bdi_update_dirty_ratelimit(struct backing_dev_info *bdi, /* * For strictlimit case, calculations above were based on bdi counters - * and limits (starting from pos_ratio = bdi_position_ratio() and up to + * and limits (starting from pos_ratio = wb_position_ratio() and up to * balanced_dirty_ratelimit = task_ratelimit * write_bw / dirty_rate). * Hence, to calculate "step" properly, we have to use bdi_dirty as * "dirty" and bdi_setpoint as "setpoint". * * We rampup dirty_ratelimit forcibly if bdi_dirty is low because * it's possible that bdi_thresh is close to zero due to inactivity - * of backing device (see the implementation of bdi_dirty_limit()). + * of backing device (see the implementation of wb_dirty_limit()). */ - if (unlikely(bdi->capabilities & BDI_CAP_STRICTLIMIT)) { + if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { dirty = bdi_dirty; if (bdi_dirty < 8) setpoint = bdi_dirty + 1; else setpoint = (bdi_thresh + - bdi_dirty_limit(bdi, bg_thresh)) / 2; + wb_dirty_limit(wb, bg_thresh)) / 2; } if (dirty < setpoint) { - x = min3(bdi->balanced_dirty_ratelimit, + x = min3(wb->balanced_dirty_ratelimit, balanced_dirty_ratelimit, task_ratelimit); if (dirty_ratelimit < x) step = x - dirty_ratelimit; } else { - x = max3(bdi->balanced_dirty_ratelimit, + x = max3(wb->balanced_dirty_ratelimit, balanced_dirty_ratelimit, task_ratelimit); if (dirty_ratelimit > x) step = dirty_ratelimit - x; @@ -1105,22 +1106,22 @@ static void bdi_update_dirty_ratelimit(struct backing_dev_info *bdi, else dirty_ratelimit -= step; - bdi->dirty_ratelimit = max(dirty_ratelimit, 1UL); - bdi->balanced_dirty_ratelimit = balanced_dirty_ratelimit; + wb->dirty_ratelimit = max(dirty_ratelimit, 1UL); + wb->balanced_dirty_ratelimit = balanced_dirty_ratelimit; - trace_bdi_dirty_ratelimit(bdi, dirty_rate, task_ratelimit); + trace_bdi_dirty_ratelimit(wb->bdi, dirty_rate, task_ratelimit); } -void __bdi_update_bandwidth(struct backing_dev_info *bdi, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty, - unsigned long start_time) +void __wb_update_bandwidth(struct bdi_writeback *wb, + unsigned long thresh, + unsigned long bg_thresh, + unsigned long dirty, + unsigned long bdi_thresh, + unsigned long bdi_dirty, + unsigned long start_time) { unsigned long now = jiffies; - unsigned long elapsed = now - bdi->bw_time_stamp; + unsigned long elapsed = now - wb->bw_time_stamp; unsigned long dirtied; unsigned long written; @@ -1130,44 +1131,44 @@ void __bdi_update_bandwidth(struct backing_dev_info *bdi, if (elapsed < BANDWIDTH_INTERVAL) return; - dirtied = percpu_counter_read(&bdi->wb.stat[WB_DIRTIED]); - written = percpu_counter_read(&bdi->wb.stat[WB_WRITTEN]); + dirtied = percpu_counter_read(&wb->stat[WB_DIRTIED]); + written = percpu_counter_read(&wb->stat[WB_WRITTEN]); /* * Skip quiet periods when disk bandwidth is under-utilized. * (at least 1s idle time between two flusher runs) */ - if (elapsed > HZ && time_before(bdi->bw_time_stamp, start_time)) + if (elapsed > HZ && time_before(wb->bw_time_stamp, start_time)) goto snapshot; if (thresh) { global_update_bandwidth(thresh, dirty, now); - bdi_update_dirty_ratelimit(bdi, thresh, bg_thresh, dirty, - bdi_thresh, bdi_dirty, - dirtied, elapsed); + wb_update_dirty_ratelimit(wb, thresh, bg_thresh, dirty, + bdi_thresh, bdi_dirty, + dirtied, elapsed); } - bdi_update_write_bandwidth(bdi, elapsed, written); + wb_update_write_bandwidth(wb, elapsed, written); snapshot: - bdi->dirtied_stamp = dirtied; - bdi->written_stamp = written; - bdi->bw_time_stamp = now; + wb->dirtied_stamp = dirtied; + wb->written_stamp = written; + wb->bw_time_stamp = now; } -static void bdi_update_bandwidth(struct backing_dev_info *bdi, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty, - unsigned long start_time) +static void wb_update_bandwidth(struct bdi_writeback *wb, + unsigned long thresh, + unsigned long bg_thresh, + unsigned long dirty, + unsigned long bdi_thresh, + unsigned long bdi_dirty, + unsigned long start_time) { - if (time_is_after_eq_jiffies(bdi->bw_time_stamp + BANDWIDTH_INTERVAL)) + if (time_is_after_eq_jiffies(wb->bw_time_stamp + BANDWIDTH_INTERVAL)) return; - spin_lock(&bdi->wb.list_lock); - __bdi_update_bandwidth(bdi, thresh, bg_thresh, dirty, - bdi_thresh, bdi_dirty, start_time); - spin_unlock(&bdi->wb.list_lock); + spin_lock(&wb->list_lock); + __wb_update_bandwidth(wb, thresh, bg_thresh, dirty, + bdi_thresh, bdi_dirty, start_time); + spin_unlock(&wb->list_lock); } /* @@ -1187,10 +1188,10 @@ static unsigned long dirty_poll_interval(unsigned long dirty, return 1; } -static unsigned long bdi_max_pause(struct backing_dev_info *bdi, - unsigned long bdi_dirty) +static unsigned long wb_max_pause(struct bdi_writeback *wb, + unsigned long bdi_dirty) { - unsigned long bw = bdi->avg_write_bandwidth; + unsigned long bw = wb->avg_write_bandwidth; unsigned long t; /* @@ -1206,14 +1207,14 @@ static unsigned long bdi_max_pause(struct backing_dev_info *bdi, return min_t(unsigned long, t, MAX_PAUSE); } -static long bdi_min_pause(struct backing_dev_info *bdi, - long max_pause, - unsigned long task_ratelimit, - unsigned long dirty_ratelimit, - int *nr_dirtied_pause) +static long wb_min_pause(struct bdi_writeback *wb, + long max_pause, + unsigned long task_ratelimit, + unsigned long dirty_ratelimit, + int *nr_dirtied_pause) { - long hi = ilog2(bdi->avg_write_bandwidth); - long lo = ilog2(bdi->dirty_ratelimit); + long hi = ilog2(wb->avg_write_bandwidth); + long lo = ilog2(wb->dirty_ratelimit); long t; /* target pause */ long pause; /* estimated next pause */ int pages; /* target nr_dirtied_pause */ @@ -1281,14 +1282,13 @@ static long bdi_min_pause(struct backing_dev_info *bdi, return pages >= DIRTY_POLL_THRESH ? 1 + t / 2 : t; } -static inline void bdi_dirty_limits(struct backing_dev_info *bdi, - unsigned long dirty_thresh, - unsigned long background_thresh, - unsigned long *bdi_dirty, - unsigned long *bdi_thresh, - unsigned long *bdi_bg_thresh) +static inline void wb_dirty_limits(struct bdi_writeback *wb, + unsigned long dirty_thresh, + unsigned long background_thresh, + unsigned long *bdi_dirty, + unsigned long *bdi_thresh, + unsigned long *bdi_bg_thresh) { - struct bdi_writeback *wb = &bdi->wb; unsigned long wb_reclaimable; /* @@ -1301,10 +1301,10 @@ static inline void bdi_dirty_limits(struct backing_dev_info *bdi, * In this case we don't want to hard throttle the USB key * dirtiers for 100 seconds until bdi_dirty drops under * bdi_thresh. Instead the auxiliary bdi control line in - * bdi_position_ratio() will let the dirtier task progress + * wb_position_ratio() will let the dirtier task progress * at some rate <= (write_bw / 2) for bringing down bdi_dirty. */ - *bdi_thresh = bdi_dirty_limit(bdi, dirty_thresh); + *bdi_thresh = wb_dirty_limit(wb, dirty_thresh); if (bdi_bg_thresh) *bdi_bg_thresh = dirty_thresh ? div_u64((u64)*bdi_thresh * @@ -1354,6 +1354,7 @@ static void balance_dirty_pages(struct address_space *mapping, unsigned long dirty_ratelimit; unsigned long pos_ratio; struct backing_dev_info *bdi = inode_to_bdi(mapping->host); + struct bdi_writeback *wb = &bdi->wb; bool strictlimit = bdi->capabilities & BDI_CAP_STRICTLIMIT; unsigned long start_time = jiffies; @@ -1378,8 +1379,8 @@ static void balance_dirty_pages(struct address_space *mapping, global_dirty_limits(&background_thresh, &dirty_thresh); if (unlikely(strictlimit)) { - bdi_dirty_limits(bdi, dirty_thresh, background_thresh, - &bdi_dirty, &bdi_thresh, &bg_thresh); + wb_dirty_limits(wb, dirty_thresh, background_thresh, + &bdi_dirty, &bdi_thresh, &bg_thresh); dirty = bdi_dirty; thresh = bdi_thresh; @@ -1410,28 +1411,28 @@ static void balance_dirty_pages(struct address_space *mapping, bdi_start_background_writeback(bdi); if (!strictlimit) - bdi_dirty_limits(bdi, dirty_thresh, background_thresh, - &bdi_dirty, &bdi_thresh, NULL); + wb_dirty_limits(wb, dirty_thresh, background_thresh, + &bdi_dirty, &bdi_thresh, NULL); dirty_exceeded = (bdi_dirty > bdi_thresh) && ((nr_dirty > dirty_thresh) || strictlimit); - if (dirty_exceeded && !bdi->dirty_exceeded) - bdi->dirty_exceeded = 1; + if (dirty_exceeded && !wb->dirty_exceeded) + wb->dirty_exceeded = 1; - bdi_update_bandwidth(bdi, dirty_thresh, background_thresh, - nr_dirty, bdi_thresh, bdi_dirty, - start_time); + wb_update_bandwidth(wb, dirty_thresh, background_thresh, + nr_dirty, bdi_thresh, bdi_dirty, + start_time); - dirty_ratelimit = bdi->dirty_ratelimit; - pos_ratio = bdi_position_ratio(bdi, dirty_thresh, - background_thresh, nr_dirty, - bdi_thresh, bdi_dirty); + dirty_ratelimit = wb->dirty_ratelimit; + pos_ratio = wb_position_ratio(wb, dirty_thresh, + background_thresh, nr_dirty, + bdi_thresh, bdi_dirty); task_ratelimit = ((u64)dirty_ratelimit * pos_ratio) >> RATELIMIT_CALC_SHIFT; - max_pause = bdi_max_pause(bdi, bdi_dirty); - min_pause = bdi_min_pause(bdi, max_pause, - task_ratelimit, dirty_ratelimit, - &nr_dirtied_pause); + max_pause = wb_max_pause(wb, bdi_dirty); + min_pause = wb_min_pause(wb, max_pause, + task_ratelimit, dirty_ratelimit, + &nr_dirtied_pause); if (unlikely(task_ratelimit == 0)) { period = max_pause; @@ -1515,15 +1516,15 @@ pause: * more page. However bdi_dirty has accounting errors. So use * the larger and more IO friendly wb_stat_error. */ - if (bdi_dirty <= wb_stat_error(&bdi->wb)) + if (bdi_dirty <= wb_stat_error(wb)) break; if (fatal_signal_pending(current)) break; } - if (!dirty_exceeded && bdi->dirty_exceeded) - bdi->dirty_exceeded = 0; + if (!dirty_exceeded && wb->dirty_exceeded) + wb->dirty_exceeded = 0; if (writeback_in_progress(bdi)) return; @@ -1577,6 +1578,7 @@ DEFINE_PER_CPU(int, dirty_throttle_leaks) = 0; void balance_dirty_pages_ratelimited(struct address_space *mapping) { struct backing_dev_info *bdi = inode_to_bdi(mapping->host); + struct bdi_writeback *wb = &bdi->wb; int ratelimit; int *p; @@ -1584,7 +1586,7 @@ void balance_dirty_pages_ratelimited(struct address_space *mapping) return; ratelimit = current->nr_dirtied_pause; - if (bdi->dirty_exceeded) + if (wb->dirty_exceeded) ratelimit = min(ratelimit, 32 >> (PAGE_SHIFT - 10)); preempt_disable(); From de1fff37b2781f9caae7bbb7b79fa788a9bb1115 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:29 -0400 Subject: [PATCH 15/85] writeback: s/bdi/wb/ in mm/page-writeback.c Writeback operations will now be per wb (bdi_writeback) instead of bdi. Replace the relevant bdi references in symbol names and comments with wb. This patch is purely cosmetic and doesn't make any functional changes. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Wu Fengguang Cc: Jens Axboe Signed-off-by: Jens Axboe --- mm/page-writeback.c | 270 ++++++++++++++++++++++---------------------- 1 file changed, 134 insertions(+), 136 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index cd39ee91b7bb..78ef5511a198 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -595,7 +595,7 @@ static long long pos_ratio_polynom(unsigned long setpoint, * * (o) global/bdi setpoints * - * We want the dirty pages be balanced around the global/bdi setpoints. + * We want the dirty pages be balanced around the global/wb setpoints. * When the number of dirty pages is higher/lower than the setpoint, the * dirty position control ratio (and hence task dirty ratelimit) will be * decreased/increased to bring the dirty pages back to the setpoint. @@ -605,8 +605,8 @@ static long long pos_ratio_polynom(unsigned long setpoint, * if (dirty < setpoint) scale up pos_ratio * if (dirty > setpoint) scale down pos_ratio * - * if (bdi_dirty < bdi_setpoint) scale up pos_ratio - * if (bdi_dirty > bdi_setpoint) scale down pos_ratio + * if (wb_dirty < wb_setpoint) scale up pos_ratio + * if (wb_dirty > wb_setpoint) scale down pos_ratio * * task_ratelimit = dirty_ratelimit * pos_ratio >> RATELIMIT_CALC_SHIFT * @@ -631,7 +631,7 @@ static long long pos_ratio_polynom(unsigned long setpoint, * 0 +------------.------------------.----------------------*-------------> * freerun^ setpoint^ limit^ dirty pages * - * (o) bdi control line + * (o) wb control line * * ^ pos_ratio * | @@ -657,27 +657,27 @@ static long long pos_ratio_polynom(unsigned long setpoint, * | . . * | . . * 0 +----------------------.-------------------------------.-------------> - * bdi_setpoint^ x_intercept^ + * wb_setpoint^ x_intercept^ * - * The bdi control line won't drop below pos_ratio=1/4, so that bdi_dirty can + * The wb control line won't drop below pos_ratio=1/4, so that wb_dirty can * be smoothly throttled down to normal if it starts high in situations like * - start writing to a slow SD card and a fast disk at the same time. The SD - * card's bdi_dirty may rush to many times higher than bdi_setpoint. - * - the bdi dirty thresh drops quickly due to change of JBOD workload + * card's wb_dirty may rush to many times higher than wb_setpoint. + * - the wb dirty thresh drops quickly due to change of JBOD workload */ static unsigned long wb_position_ratio(struct bdi_writeback *wb, unsigned long thresh, unsigned long bg_thresh, unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty) + unsigned long wb_thresh, + unsigned long wb_dirty) { unsigned long write_bw = wb->avg_write_bandwidth; unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh); unsigned long limit = hard_dirty_limit(thresh); unsigned long x_intercept; unsigned long setpoint; /* dirty pages' target balance point */ - unsigned long bdi_setpoint; + unsigned long wb_setpoint; unsigned long span; long long pos_ratio; /* for scaling up/down the rate limit */ long x; @@ -696,146 +696,145 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, /* * The strictlimit feature is a tool preventing mistrusted filesystems * from growing a large number of dirty pages before throttling. For - * such filesystems balance_dirty_pages always checks bdi counters - * against bdi limits. Even if global "nr_dirty" is under "freerun". + * such filesystems balance_dirty_pages always checks wb counters + * against wb limits. Even if global "nr_dirty" is under "freerun". * This is especially important for fuse which sets bdi->max_ratio to * 1% by default. Without strictlimit feature, fuse writeback may * consume arbitrary amount of RAM because it is accounted in * NR_WRITEBACK_TEMP which is not involved in calculating "nr_dirty". * * Here, in wb_position_ratio(), we calculate pos_ratio based on - * two values: bdi_dirty and bdi_thresh. Let's consider an example: + * two values: wb_dirty and wb_thresh. Let's consider an example: * total amount of RAM is 16GB, bdi->max_ratio is equal to 1%, global * limits are set by default to 10% and 20% (background and throttle). - * Then bdi_thresh is 1% of 20% of 16GB. This amounts to ~8K pages. - * wb_dirty_limit(wb, bg_thresh) is about ~4K pages. bdi_setpoint is - * about ~6K pages (as the average of background and throttle bdi + * Then wb_thresh is 1% of 20% of 16GB. This amounts to ~8K pages. + * wb_dirty_limit(wb, bg_thresh) is about ~4K pages. wb_setpoint is + * about ~6K pages (as the average of background and throttle wb * limits). The 3rd order polynomial will provide positive feedback if - * bdi_dirty is under bdi_setpoint and vice versa. + * wb_dirty is under wb_setpoint and vice versa. * * Note, that we cannot use global counters in these calculations - * because we want to throttle process writing to a strictlimit BDI + * because we want to throttle process writing to a strictlimit wb * much earlier than global "freerun" is reached (~23MB vs. ~2.3GB * in the example above). */ if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { - long long bdi_pos_ratio; - unsigned long bdi_bg_thresh; + long long wb_pos_ratio; + unsigned long wb_bg_thresh; - if (bdi_dirty < 8) + if (wb_dirty < 8) return min_t(long long, pos_ratio * 2, 2 << RATELIMIT_CALC_SHIFT); - if (bdi_dirty >= bdi_thresh) + if (wb_dirty >= wb_thresh) return 0; - bdi_bg_thresh = div_u64((u64)bdi_thresh * bg_thresh, thresh); - bdi_setpoint = dirty_freerun_ceiling(bdi_thresh, - bdi_bg_thresh); + wb_bg_thresh = div_u64((u64)wb_thresh * bg_thresh, thresh); + wb_setpoint = dirty_freerun_ceiling(wb_thresh, wb_bg_thresh); - if (bdi_setpoint == 0 || bdi_setpoint == bdi_thresh) + if (wb_setpoint == 0 || wb_setpoint == wb_thresh) return 0; - bdi_pos_ratio = pos_ratio_polynom(bdi_setpoint, bdi_dirty, - bdi_thresh); + wb_pos_ratio = pos_ratio_polynom(wb_setpoint, wb_dirty, + wb_thresh); /* - * Typically, for strictlimit case, bdi_setpoint << setpoint - * and pos_ratio >> bdi_pos_ratio. In the other words global + * Typically, for strictlimit case, wb_setpoint << setpoint + * and pos_ratio >> wb_pos_ratio. In the other words global * state ("dirty") is not limiting factor and we have to - * make decision based on bdi counters. But there is an + * make decision based on wb counters. But there is an * important case when global pos_ratio should get precedence: * global limits are exceeded (e.g. due to activities on other - * BDIs) while given strictlimit BDI is below limit. + * wb's) while given strictlimit wb is below limit. * - * "pos_ratio * bdi_pos_ratio" would work for the case above, + * "pos_ratio * wb_pos_ratio" would work for the case above, * but it would look too non-natural for the case of all - * activity in the system coming from a single strictlimit BDI + * activity in the system coming from a single strictlimit wb * with bdi->max_ratio == 100%. * * Note that min() below somewhat changes the dynamics of the * control system. Normally, pos_ratio value can be well over 3 - * (when globally we are at freerun and bdi is well below bdi + * (when globally we are at freerun and wb is well below wb * setpoint). Now the maximum pos_ratio in the same situation * is 2. We might want to tweak this if we observe the control * system is too slow to adapt. */ - return min(pos_ratio, bdi_pos_ratio); + return min(pos_ratio, wb_pos_ratio); } /* * We have computed basic pos_ratio above based on global situation. If - * the bdi is over/under its share of dirty pages, we want to scale + * the wb is over/under its share of dirty pages, we want to scale * pos_ratio further down/up. That is done by the following mechanism. */ /* - * bdi setpoint + * wb setpoint * - * f(bdi_dirty) := 1.0 + k * (bdi_dirty - bdi_setpoint) + * f(wb_dirty) := 1.0 + k * (wb_dirty - wb_setpoint) * - * x_intercept - bdi_dirty + * x_intercept - wb_dirty * := -------------------------- - * x_intercept - bdi_setpoint + * x_intercept - wb_setpoint * - * The main bdi control line is a linear function that subjects to + * The main wb control line is a linear function that subjects to * - * (1) f(bdi_setpoint) = 1.0 - * (2) k = - 1 / (8 * write_bw) (in single bdi case) - * or equally: x_intercept = bdi_setpoint + 8 * write_bw + * (1) f(wb_setpoint) = 1.0 + * (2) k = - 1 / (8 * write_bw) (in single wb case) + * or equally: x_intercept = wb_setpoint + 8 * write_bw * - * For single bdi case, the dirty pages are observed to fluctuate + * For single wb case, the dirty pages are observed to fluctuate * regularly within range - * [bdi_setpoint - write_bw/2, bdi_setpoint + write_bw/2] + * [wb_setpoint - write_bw/2, wb_setpoint + write_bw/2] * for various filesystems, where (2) can yield in a reasonable 12.5% * fluctuation range for pos_ratio. * - * For JBOD case, bdi_thresh (not bdi_dirty!) could fluctuate up to its + * For JBOD case, wb_thresh (not wb_dirty!) could fluctuate up to its * own size, so move the slope over accordingly and choose a slope that - * yields 100% pos_ratio fluctuation on suddenly doubled bdi_thresh. + * yields 100% pos_ratio fluctuation on suddenly doubled wb_thresh. */ - if (unlikely(bdi_thresh > thresh)) - bdi_thresh = thresh; + if (unlikely(wb_thresh > thresh)) + wb_thresh = thresh; /* - * It's very possible that bdi_thresh is close to 0 not because the + * It's very possible that wb_thresh is close to 0 not because the * device is slow, but that it has remained inactive for long time. * Honour such devices a reasonable good (hopefully IO efficient) * threshold, so that the occasional writes won't be blocked and active * writes can rampup the threshold quickly. */ - bdi_thresh = max(bdi_thresh, (limit - dirty) / 8); + wb_thresh = max(wb_thresh, (limit - dirty) / 8); /* - * scale global setpoint to bdi's: - * bdi_setpoint = setpoint * bdi_thresh / thresh + * scale global setpoint to wb's: + * wb_setpoint = setpoint * wb_thresh / thresh */ - x = div_u64((u64)bdi_thresh << 16, thresh + 1); - bdi_setpoint = setpoint * (u64)x >> 16; + x = div_u64((u64)wb_thresh << 16, thresh + 1); + wb_setpoint = setpoint * (u64)x >> 16; /* - * Use span=(8*write_bw) in single bdi case as indicated by - * (thresh - bdi_thresh ~= 0) and transit to bdi_thresh in JBOD case. + * Use span=(8*write_bw) in single wb case as indicated by + * (thresh - wb_thresh ~= 0) and transit to wb_thresh in JBOD case. * - * bdi_thresh thresh - bdi_thresh - * span = ---------- * (8 * write_bw) + ------------------- * bdi_thresh - * thresh thresh + * wb_thresh thresh - wb_thresh + * span = --------- * (8 * write_bw) + ------------------ * wb_thresh + * thresh thresh */ - span = (thresh - bdi_thresh + 8 * write_bw) * (u64)x >> 16; - x_intercept = bdi_setpoint + span; + span = (thresh - wb_thresh + 8 * write_bw) * (u64)x >> 16; + x_intercept = wb_setpoint + span; - if (bdi_dirty < x_intercept - span / 4) { - pos_ratio = div64_u64(pos_ratio * (x_intercept - bdi_dirty), - x_intercept - bdi_setpoint + 1); + if (wb_dirty < x_intercept - span / 4) { + pos_ratio = div64_u64(pos_ratio * (x_intercept - wb_dirty), + x_intercept - wb_setpoint + 1); } else pos_ratio /= 4; /* - * bdi reserve area, safeguard against dirty pool underrun and disk idle + * wb reserve area, safeguard against dirty pool underrun and disk idle * It may push the desired control point of global dirty pages higher * than setpoint. */ - x_intercept = bdi_thresh / 2; - if (bdi_dirty < x_intercept) { - if (bdi_dirty > x_intercept / 8) - pos_ratio = div_u64(pos_ratio * x_intercept, bdi_dirty); + x_intercept = wb_thresh / 2; + if (wb_dirty < x_intercept) { + if (wb_dirty > x_intercept / 8) + pos_ratio = div_u64(pos_ratio * x_intercept, wb_dirty); else pos_ratio *= 8; } @@ -943,17 +942,17 @@ static void global_update_bandwidth(unsigned long thresh, } /* - * Maintain bdi->dirty_ratelimit, the base dirty throttle rate. + * Maintain wb->dirty_ratelimit, the base dirty throttle rate. * - * Normal bdi tasks will be curbed at or below it in long term. + * Normal wb tasks will be curbed at or below it in long term. * Obviously it should be around (write_bw / N) when there are N dd tasks. */ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, unsigned long thresh, unsigned long bg_thresh, unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty, + unsigned long wb_thresh, + unsigned long wb_dirty, unsigned long dirtied, unsigned long elapsed) { @@ -976,7 +975,7 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, dirty_rate = (dirtied - wb->dirtied_stamp) * HZ / elapsed; pos_ratio = wb_position_ratio(wb, thresh, bg_thresh, dirty, - bdi_thresh, bdi_dirty); + wb_thresh, wb_dirty); /* * task_ratelimit reflects each dd's dirty rate for the past 200ms. */ @@ -986,7 +985,7 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, /* * A linear estimation of the "balanced" throttle rate. The theory is, - * if there are N dd tasks, each throttled at task_ratelimit, the bdi's + * if there are N dd tasks, each throttled at task_ratelimit, the wb's * dirty_rate will be measured to be (N * task_ratelimit). So the below * formula will yield the balanced rate limit (write_bw / N). * @@ -1025,7 +1024,7 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, /* * We could safely do this and return immediately: * - * bdi->dirty_ratelimit = balanced_dirty_ratelimit; + * wb->dirty_ratelimit = balanced_dirty_ratelimit; * * However to get a more stable dirty_ratelimit, the below elaborated * code makes use of task_ratelimit to filter out singular points and @@ -1059,22 +1058,22 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, step = 0; /* - * For strictlimit case, calculations above were based on bdi counters + * For strictlimit case, calculations above were based on wb counters * and limits (starting from pos_ratio = wb_position_ratio() and up to * balanced_dirty_ratelimit = task_ratelimit * write_bw / dirty_rate). - * Hence, to calculate "step" properly, we have to use bdi_dirty as - * "dirty" and bdi_setpoint as "setpoint". + * Hence, to calculate "step" properly, we have to use wb_dirty as + * "dirty" and wb_setpoint as "setpoint". * - * We rampup dirty_ratelimit forcibly if bdi_dirty is low because - * it's possible that bdi_thresh is close to zero due to inactivity + * We rampup dirty_ratelimit forcibly if wb_dirty is low because + * it's possible that wb_thresh is close to zero due to inactivity * of backing device (see the implementation of wb_dirty_limit()). */ if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { - dirty = bdi_dirty; - if (bdi_dirty < 8) - setpoint = bdi_dirty + 1; + dirty = wb_dirty; + if (wb_dirty < 8) + setpoint = wb_dirty + 1; else - setpoint = (bdi_thresh + + setpoint = (wb_thresh + wb_dirty_limit(wb, bg_thresh)) / 2; } @@ -1116,8 +1115,8 @@ void __wb_update_bandwidth(struct bdi_writeback *wb, unsigned long thresh, unsigned long bg_thresh, unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty, + unsigned long wb_thresh, + unsigned long wb_dirty, unsigned long start_time) { unsigned long now = jiffies; @@ -1144,7 +1143,7 @@ void __wb_update_bandwidth(struct bdi_writeback *wb, if (thresh) { global_update_bandwidth(thresh, dirty, now); wb_update_dirty_ratelimit(wb, thresh, bg_thresh, dirty, - bdi_thresh, bdi_dirty, + wb_thresh, wb_dirty, dirtied, elapsed); } wb_update_write_bandwidth(wb, elapsed, written); @@ -1159,15 +1158,15 @@ static void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long thresh, unsigned long bg_thresh, unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty, + unsigned long wb_thresh, + unsigned long wb_dirty, unsigned long start_time) { if (time_is_after_eq_jiffies(wb->bw_time_stamp + BANDWIDTH_INTERVAL)) return; spin_lock(&wb->list_lock); __wb_update_bandwidth(wb, thresh, bg_thresh, dirty, - bdi_thresh, bdi_dirty, start_time); + wb_thresh, wb_dirty, start_time); spin_unlock(&wb->list_lock); } @@ -1189,7 +1188,7 @@ static unsigned long dirty_poll_interval(unsigned long dirty, } static unsigned long wb_max_pause(struct bdi_writeback *wb, - unsigned long bdi_dirty) + unsigned long wb_dirty) { unsigned long bw = wb->avg_write_bandwidth; unsigned long t; @@ -1201,7 +1200,7 @@ static unsigned long wb_max_pause(struct bdi_writeback *wb, * * 8 serves as the safety ratio. */ - t = bdi_dirty / (1 + bw / roundup_pow_of_two(1 + HZ / 8)); + t = wb_dirty / (1 + bw / roundup_pow_of_two(1 + HZ / 8)); t++; return min_t(unsigned long, t, MAX_PAUSE); @@ -1285,31 +1284,31 @@ static long wb_min_pause(struct bdi_writeback *wb, static inline void wb_dirty_limits(struct bdi_writeback *wb, unsigned long dirty_thresh, unsigned long background_thresh, - unsigned long *bdi_dirty, - unsigned long *bdi_thresh, - unsigned long *bdi_bg_thresh) + unsigned long *wb_dirty, + unsigned long *wb_thresh, + unsigned long *wb_bg_thresh) { unsigned long wb_reclaimable; /* - * bdi_thresh is not treated as some limiting factor as + * wb_thresh is not treated as some limiting factor as * dirty_thresh, due to reasons - * - in JBOD setup, bdi_thresh can fluctuate a lot + * - in JBOD setup, wb_thresh can fluctuate a lot * - in a system with HDD and USB key, the USB key may somehow - * go into state (bdi_dirty >> bdi_thresh) either because - * bdi_dirty starts high, or because bdi_thresh drops low. + * go into state (wb_dirty >> wb_thresh) either because + * wb_dirty starts high, or because wb_thresh drops low. * In this case we don't want to hard throttle the USB key - * dirtiers for 100 seconds until bdi_dirty drops under - * bdi_thresh. Instead the auxiliary bdi control line in + * dirtiers for 100 seconds until wb_dirty drops under + * wb_thresh. Instead the auxiliary wb control line in * wb_position_ratio() will let the dirtier task progress - * at some rate <= (write_bw / 2) for bringing down bdi_dirty. + * at some rate <= (write_bw / 2) for bringing down wb_dirty. */ - *bdi_thresh = wb_dirty_limit(wb, dirty_thresh); + *wb_thresh = wb_dirty_limit(wb, dirty_thresh); - if (bdi_bg_thresh) - *bdi_bg_thresh = dirty_thresh ? div_u64((u64)*bdi_thresh * - background_thresh, - dirty_thresh) : 0; + if (wb_bg_thresh) + *wb_bg_thresh = dirty_thresh ? div_u64((u64)*wb_thresh * + background_thresh, + dirty_thresh) : 0; /* * In order to avoid the stacked BDI deadlock we need @@ -1321,12 +1320,12 @@ static inline void wb_dirty_limits(struct bdi_writeback *wb, * actually dirty; with m+n sitting in the percpu * deltas. */ - if (*bdi_thresh < 2 * wb_stat_error(wb)) { + if (*wb_thresh < 2 * wb_stat_error(wb)) { wb_reclaimable = wb_stat_sum(wb, WB_RECLAIMABLE); - *bdi_dirty = wb_reclaimable + wb_stat_sum(wb, WB_WRITEBACK); + *wb_dirty = wb_reclaimable + wb_stat_sum(wb, WB_WRITEBACK); } else { wb_reclaimable = wb_stat(wb, WB_RECLAIMABLE); - *bdi_dirty = wb_reclaimable + wb_stat(wb, WB_WRITEBACK); + *wb_dirty = wb_reclaimable + wb_stat(wb, WB_WRITEBACK); } } @@ -1360,9 +1359,9 @@ static void balance_dirty_pages(struct address_space *mapping, for (;;) { unsigned long now = jiffies; - unsigned long uninitialized_var(bdi_thresh); + unsigned long uninitialized_var(wb_thresh); unsigned long thresh; - unsigned long uninitialized_var(bdi_dirty); + unsigned long uninitialized_var(wb_dirty); unsigned long dirty; unsigned long bg_thresh; @@ -1380,10 +1379,10 @@ static void balance_dirty_pages(struct address_space *mapping, if (unlikely(strictlimit)) { wb_dirty_limits(wb, dirty_thresh, background_thresh, - &bdi_dirty, &bdi_thresh, &bg_thresh); + &wb_dirty, &wb_thresh, &bg_thresh); - dirty = bdi_dirty; - thresh = bdi_thresh; + dirty = wb_dirty; + thresh = wb_thresh; } else { dirty = nr_dirty; thresh = dirty_thresh; @@ -1393,10 +1392,10 @@ static void balance_dirty_pages(struct address_space *mapping, /* * Throttle it only when the background writeback cannot * catch-up. This avoids (excessively) small writeouts - * when the bdi limits are ramping up in case of !strictlimit. + * when the wb limits are ramping up in case of !strictlimit. * - * In strictlimit case make decision based on the bdi counters - * and limits. Small writeouts when the bdi limits are ramping + * In strictlimit case make decision based on the wb counters + * and limits. Small writeouts when the wb limits are ramping * up are the price we consciously pay for strictlimit-ing. */ if (dirty <= dirty_freerun_ceiling(thresh, bg_thresh)) { @@ -1412,24 +1411,23 @@ static void balance_dirty_pages(struct address_space *mapping, if (!strictlimit) wb_dirty_limits(wb, dirty_thresh, background_thresh, - &bdi_dirty, &bdi_thresh, NULL); + &wb_dirty, &wb_thresh, NULL); - dirty_exceeded = (bdi_dirty > bdi_thresh) && + dirty_exceeded = (wb_dirty > wb_thresh) && ((nr_dirty > dirty_thresh) || strictlimit); if (dirty_exceeded && !wb->dirty_exceeded) wb->dirty_exceeded = 1; wb_update_bandwidth(wb, dirty_thresh, background_thresh, - nr_dirty, bdi_thresh, bdi_dirty, - start_time); + nr_dirty, wb_thresh, wb_dirty, start_time); dirty_ratelimit = wb->dirty_ratelimit; pos_ratio = wb_position_ratio(wb, dirty_thresh, background_thresh, nr_dirty, - bdi_thresh, bdi_dirty); + wb_thresh, wb_dirty); task_ratelimit = ((u64)dirty_ratelimit * pos_ratio) >> RATELIMIT_CALC_SHIFT; - max_pause = wb_max_pause(wb, bdi_dirty); + max_pause = wb_max_pause(wb, wb_dirty); min_pause = wb_min_pause(wb, max_pause, task_ratelimit, dirty_ratelimit, &nr_dirtied_pause); @@ -1455,8 +1453,8 @@ static void balance_dirty_pages(struct address_space *mapping, dirty_thresh, background_thresh, nr_dirty, - bdi_thresh, - bdi_dirty, + wb_thresh, + wb_dirty, dirty_ratelimit, task_ratelimit, pages_dirtied, @@ -1484,8 +1482,8 @@ pause: dirty_thresh, background_thresh, nr_dirty, - bdi_thresh, - bdi_dirty, + wb_thresh, + wb_dirty, dirty_ratelimit, task_ratelimit, pages_dirtied, @@ -1508,15 +1506,15 @@ pause: /* * In the case of an unresponding NFS server and the NFS dirty - * pages exceeds dirty_thresh, give the other good bdi's a pipe + * pages exceeds dirty_thresh, give the other good wb's a pipe * to go through, so that tasks on them still remain responsive. * * In theory 1 page is enough to keep the comsumer-producer * pipe going: the flusher cleans 1 page => the task dirties 1 - * more page. However bdi_dirty has accounting errors. So use + * more page. However wb_dirty has accounting errors. So use * the larger and more IO friendly wb_stat_error. */ - if (bdi_dirty <= wb_stat_error(wb)) + if (wb_dirty <= wb_stat_error(wb)) break; if (fatal_signal_pending(current)) From f0054bb1e1f3be03cc33369df640db97f10f6172 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:30 -0400 Subject: [PATCH 16/85] writeback: move backing_dev_info->wb_lock and ->worklist into bdi_writeback Currently, a bdi (backing_dev_info) embeds single wb (bdi_writeback) and the role of the separation is unclear. For cgroup support for writeback IOs, a bdi will be updated to host multiple wb's where each wb serves writeback IOs of a different cgroup on the bdi. To achieve that, a wb should carry all states necessary for servicing writeback IOs for a cgroup independently. This patch moves bdi->wb_lock and ->worklist into wb. * The lock protects bdi->worklist and bdi->wb.dwork scheduling. While moving, rename it to wb->work_lock as wb->wb_lock is confusing. Also, move wb->dwork downwards so that it's colocated with the new ->work_lock and ->work_list fields. * bdi_writeback_workfn() -> wb_workfn() bdi_wakeup_thread_delayed(bdi) -> wb_wakeup_delayed(wb) bdi_wakeup_thread(bdi) -> wb_wakeup(wb) bdi_queue_work(bdi, ...) -> wb_queue_work(wb, ...) __bdi_start_writeback(bdi, ...) -> __wb_start_writeback(wb, ...) get_next_work_item(bdi) -> get_next_work_item(wb) * bdi_wb_shutdown() is renamed to wb_shutdown() and now takes @wb. The function contained parts which belong to the containing bdi rather than the wb itself - testing cap_writeback_dirty and bdi_remove_from_list() invocation. Those are moved to bdi_unregister(). * bdi_wb_{init|exit}() are renamed to wb_{init|exit}(). Initializations of the moved bdi->wb_lock and ->work_list are relocated from bdi_init() to wb_init(). * As there's still only one bdi_writeback per backing_dev_info, all uses of bdi->state are mechanically replaced with bdi->wb.state introducing no behavior changes. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Cc: Wu Fengguang Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 86 ++++++++++++++++++------------------- include/linux/backing-dev.h | 12 +++--- mm/backing-dev.c | 59 ++++++++++++------------- 3 files changed, 75 insertions(+), 82 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 1945cb9f6b97..a69d2e10d514 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -109,34 +109,33 @@ static inline struct inode *wb_inode(struct list_head *head) EXPORT_TRACEPOINT_SYMBOL_GPL(wbc_writepage); -static void bdi_wakeup_thread(struct backing_dev_info *bdi) +static void wb_wakeup(struct bdi_writeback *wb) { - spin_lock_bh(&bdi->wb_lock); - if (test_bit(WB_registered, &bdi->wb.state)) - mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); - spin_unlock_bh(&bdi->wb_lock); + spin_lock_bh(&wb->work_lock); + if (test_bit(WB_registered, &wb->state)) + mod_delayed_work(bdi_wq, &wb->dwork, 0); + spin_unlock_bh(&wb->work_lock); } -static void bdi_queue_work(struct backing_dev_info *bdi, - struct wb_writeback_work *work) +static void wb_queue_work(struct bdi_writeback *wb, + struct wb_writeback_work *work) { - trace_writeback_queue(bdi, work); + trace_writeback_queue(wb->bdi, work); - spin_lock_bh(&bdi->wb_lock); - if (!test_bit(WB_registered, &bdi->wb.state)) { + spin_lock_bh(&wb->work_lock); + if (!test_bit(WB_registered, &wb->state)) { if (work->done) complete(work->done); goto out_unlock; } - list_add_tail(&work->list, &bdi->work_list); - mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); + list_add_tail(&work->list, &wb->work_list); + mod_delayed_work(bdi_wq, &wb->dwork, 0); out_unlock: - spin_unlock_bh(&bdi->wb_lock); + spin_unlock_bh(&wb->work_lock); } -static void -__bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, - bool range_cyclic, enum wb_reason reason) +static void __wb_start_writeback(struct bdi_writeback *wb, long nr_pages, + bool range_cyclic, enum wb_reason reason) { struct wb_writeback_work *work; @@ -146,8 +145,8 @@ __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, */ work = kzalloc(sizeof(*work), GFP_ATOMIC); if (!work) { - trace_writeback_nowork(bdi); - bdi_wakeup_thread(bdi); + trace_writeback_nowork(wb->bdi); + wb_wakeup(wb); return; } @@ -156,7 +155,7 @@ __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, work->range_cyclic = range_cyclic; work->reason = reason; - bdi_queue_work(bdi, work); + wb_queue_work(wb, work); } /** @@ -174,7 +173,7 @@ __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, enum wb_reason reason) { - __bdi_start_writeback(bdi, nr_pages, true, reason); + __wb_start_writeback(&bdi->wb, nr_pages, true, reason); } /** @@ -194,7 +193,7 @@ void bdi_start_background_writeback(struct backing_dev_info *bdi) * writeback as soon as there is no other work to do. */ trace_writeback_wake_background(bdi); - bdi_wakeup_thread(bdi); + wb_wakeup(&bdi->wb); } /* @@ -898,7 +897,7 @@ static long wb_writeback(struct bdi_writeback *wb, * after the other works are all done. */ if ((work->for_background || work->for_kupdate) && - !list_empty(&wb->bdi->work_list)) + !list_empty(&wb->work_list)) break; /* @@ -969,18 +968,17 @@ static long wb_writeback(struct bdi_writeback *wb, /* * Return the next wb_writeback_work struct that hasn't been processed yet. */ -static struct wb_writeback_work * -get_next_work_item(struct backing_dev_info *bdi) +static struct wb_writeback_work *get_next_work_item(struct bdi_writeback *wb) { struct wb_writeback_work *work = NULL; - spin_lock_bh(&bdi->wb_lock); - if (!list_empty(&bdi->work_list)) { - work = list_entry(bdi->work_list.next, + spin_lock_bh(&wb->work_lock); + if (!list_empty(&wb->work_list)) { + work = list_entry(wb->work_list.next, struct wb_writeback_work, list); list_del_init(&work->list); } - spin_unlock_bh(&bdi->wb_lock); + spin_unlock_bh(&wb->work_lock); return work; } @@ -1052,14 +1050,13 @@ static long wb_check_old_data_flush(struct bdi_writeback *wb) */ static long wb_do_writeback(struct bdi_writeback *wb) { - struct backing_dev_info *bdi = wb->bdi; struct wb_writeback_work *work; long wrote = 0; set_bit(WB_writeback_running, &wb->state); - while ((work = get_next_work_item(bdi)) != NULL) { + while ((work = get_next_work_item(wb)) != NULL) { - trace_writeback_exec(bdi, work); + trace_writeback_exec(wb->bdi, work); wrote += wb_writeback(wb, work); @@ -1087,43 +1084,42 @@ static long wb_do_writeback(struct bdi_writeback *wb) * Handle writeback of dirty data for the device backed by this bdi. Also * reschedules periodically and does kupdated style flushing. */ -void bdi_writeback_workfn(struct work_struct *work) +void wb_workfn(struct work_struct *work) { struct bdi_writeback *wb = container_of(to_delayed_work(work), struct bdi_writeback, dwork); - struct backing_dev_info *bdi = wb->bdi; long pages_written; - set_worker_desc("flush-%s", dev_name(bdi->dev)); + set_worker_desc("flush-%s", dev_name(wb->bdi->dev)); current->flags |= PF_SWAPWRITE; if (likely(!current_is_workqueue_rescuer() || !test_bit(WB_registered, &wb->state))) { /* - * The normal path. Keep writing back @bdi until its + * The normal path. Keep writing back @wb until its * work_list is empty. Note that this path is also taken - * if @bdi is shutting down even when we're running off the + * if @wb is shutting down even when we're running off the * rescuer as work_list needs to be drained. */ do { pages_written = wb_do_writeback(wb); trace_writeback_pages_written(pages_written); - } while (!list_empty(&bdi->work_list)); + } while (!list_empty(&wb->work_list)); } else { /* * bdi_wq can't get enough workers and we're running off * the emergency worker. Don't hog it. Hopefully, 1024 is * enough for efficient IO. */ - pages_written = writeback_inodes_wb(&bdi->wb, 1024, + pages_written = writeback_inodes_wb(wb, 1024, WB_REASON_FORKER_THREAD); trace_writeback_pages_written(pages_written); } - if (!list_empty(&bdi->work_list)) + if (!list_empty(&wb->work_list)) mod_delayed_work(bdi_wq, &wb->dwork, 0); else if (wb_has_dirty_io(wb) && dirty_writeback_interval) - bdi_wakeup_thread_delayed(bdi); + wb_wakeup_delayed(wb); current->flags &= ~PF_SWAPWRITE; } @@ -1143,7 +1139,7 @@ void wakeup_flusher_threads(long nr_pages, enum wb_reason reason) list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) { if (!bdi_has_dirty_io(bdi)) continue; - __bdi_start_writeback(bdi, nr_pages, false, reason); + __wb_start_writeback(&bdi->wb, nr_pages, false, reason); } rcu_read_unlock(); } @@ -1174,7 +1170,7 @@ static void wakeup_dirtytime_writeback(struct work_struct *w) list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) { if (list_empty(&bdi->wb.b_dirty_time)) continue; - bdi_wakeup_thread(bdi); + wb_wakeup(&bdi->wb); } rcu_read_unlock(); schedule_delayed_work(&dirtytime_work, dirtytime_expire_interval * HZ); @@ -1347,7 +1343,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) trace_writeback_dirty_inode_enqueue(inode); if (wakeup_bdi) - bdi_wakeup_thread_delayed(bdi); + wb_wakeup_delayed(&bdi->wb); return; } } @@ -1437,7 +1433,7 @@ void writeback_inodes_sb_nr(struct super_block *sb, if (sb->s_bdi == &noop_backing_dev_info) return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); - bdi_queue_work(sb->s_bdi, &work); + wb_queue_work(&sb->s_bdi->wb, &work); wait_for_completion(&done); } EXPORT_SYMBOL(writeback_inodes_sb_nr); @@ -1521,7 +1517,7 @@ void sync_inodes_sb(struct super_block *sb) return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); - bdi_queue_work(sb->s_bdi, &work); + wb_queue_work(&sb->s_bdi->wb, &work); wait_for_completion(&done); wait_sb_inodes(sb); diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 2ab06049d812..d796f49ce87a 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -52,7 +52,6 @@ struct bdi_writeback { unsigned long state; /* Always use atomic bitops on this */ unsigned long last_old_flush; /* last old data flush */ - struct delayed_work dwork; /* work item used for writeback */ struct list_head b_dirty; /* dirty inodes */ struct list_head b_io; /* parked for writeback */ struct list_head b_more_io; /* parked for more writeback */ @@ -78,6 +77,10 @@ struct bdi_writeback { struct fprop_local_percpu completions; int dirty_exceeded; + + spinlock_t work_lock; /* protects work_list & dwork scheduling */ + struct list_head work_list; + struct delayed_work dwork; /* work item used for writeback */ }; struct backing_dev_info { @@ -93,9 +96,6 @@ struct backing_dev_info { unsigned int max_ratio, max_prop_frac; struct bdi_writeback wb; /* default writeback info for this bdi */ - spinlock_t wb_lock; /* protects work_list & wb.dwork scheduling */ - - struct list_head work_list; struct device *dev; @@ -121,9 +121,9 @@ int __must_check bdi_setup_and_register(struct backing_dev_info *, char *); void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, enum wb_reason reason); void bdi_start_background_writeback(struct backing_dev_info *bdi); -void bdi_writeback_workfn(struct work_struct *work); +void wb_workfn(struct work_struct *work); int bdi_has_dirty_io(struct backing_dev_info *bdi); -void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi); +void wb_wakeup_delayed(struct bdi_writeback *wb); extern spinlock_t bdi_lock; extern struct list_head bdi_list; diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 9a6c472ebd0b..597f0cec8405 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -261,7 +261,7 @@ int bdi_has_dirty_io(struct backing_dev_info *bdi) } /* - * This function is used when the first inode for this bdi is marked dirty. It + * This function is used when the first inode for this wb is marked dirty. It * wakes-up the corresponding bdi thread which should then take care of the * periodic background write-out of dirty inodes. Since the write-out would * starts only 'dirty_writeback_interval' centisecs from now anyway, we just @@ -274,15 +274,15 @@ int bdi_has_dirty_io(struct backing_dev_info *bdi) * We have to be careful not to postpone flush work if it is scheduled for * earlier. Thus we use queue_delayed_work(). */ -void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi) +void wb_wakeup_delayed(struct bdi_writeback *wb) { unsigned long timeout; timeout = msecs_to_jiffies(dirty_writeback_interval * 10); - spin_lock_bh(&bdi->wb_lock); - if (test_bit(WB_registered, &bdi->wb.state)) - queue_delayed_work(bdi_wq, &bdi->wb.dwork, timeout); - spin_unlock_bh(&bdi->wb_lock); + spin_lock_bh(&wb->work_lock); + if (test_bit(WB_registered, &wb->state)) + queue_delayed_work(bdi_wq, &wb->dwork, timeout); + spin_unlock_bh(&wb->work_lock); } /* @@ -335,28 +335,24 @@ EXPORT_SYMBOL(bdi_register_dev); /* * Remove bdi from the global list and shutdown any threads we have running */ -static void bdi_wb_shutdown(struct backing_dev_info *bdi) +static void wb_shutdown(struct bdi_writeback *wb) { /* Make sure nobody queues further work */ - spin_lock_bh(&bdi->wb_lock); - if (!test_and_clear_bit(WB_registered, &bdi->wb.state)) { - spin_unlock_bh(&bdi->wb_lock); + spin_lock_bh(&wb->work_lock); + if (!test_and_clear_bit(WB_registered, &wb->state)) { + spin_unlock_bh(&wb->work_lock); return; } - spin_unlock_bh(&bdi->wb_lock); + spin_unlock_bh(&wb->work_lock); /* - * Make sure nobody finds us on the bdi_list anymore + * Drain work list and shutdown the delayed_work. !WB_registered + * tells wb_workfn() that @wb is dying and its work_list needs to + * be drained no matter what. */ - bdi_remove_from_list(bdi); - - /* - * Drain work list and shutdown the delayed_work. At this point, - * @bdi->bdi_list is empty telling bdi_Writeback_workfn() that @bdi - * is dying and its work_list needs to be drained no matter what. - */ - mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); - flush_delayed_work(&bdi->wb.dwork); + mod_delayed_work(bdi_wq, &wb->dwork, 0); + flush_delayed_work(&wb->dwork); + WARN_ON(!list_empty(&wb->work_list)); } /* @@ -381,7 +377,7 @@ EXPORT_SYMBOL(bdi_unregister); */ #define INIT_BW (100 << (20 - PAGE_SHIFT)) -static int bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) +static int wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) { int i, err; @@ -394,7 +390,6 @@ static int bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) INIT_LIST_HEAD(&wb->b_more_io); INIT_LIST_HEAD(&wb->b_dirty_time); spin_lock_init(&wb->list_lock); - INIT_DELAYED_WORK(&wb->dwork, bdi_writeback_workfn); wb->bw_time_stamp = jiffies; wb->balanced_dirty_ratelimit = INIT_BW; @@ -402,6 +397,10 @@ static int bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) wb->write_bandwidth = INIT_BW; wb->avg_write_bandwidth = INIT_BW; + spin_lock_init(&wb->work_lock); + INIT_LIST_HEAD(&wb->work_list); + INIT_DELAYED_WORK(&wb->dwork, wb_workfn); + err = fprop_local_init_percpu(&wb->completions, GFP_KERNEL); if (err) return err; @@ -419,7 +418,7 @@ static int bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) return 0; } -static void bdi_wb_exit(struct bdi_writeback *wb) +static void wb_exit(struct bdi_writeback *wb) { int i; @@ -440,11 +439,9 @@ int bdi_init(struct backing_dev_info *bdi) bdi->min_ratio = 0; bdi->max_ratio = 100; bdi->max_prop_frac = FPROP_FRAC_BASE; - spin_lock_init(&bdi->wb_lock); INIT_LIST_HEAD(&bdi->bdi_list); - INIT_LIST_HEAD(&bdi->work_list); - err = bdi_wb_init(&bdi->wb, bdi); + err = wb_init(&bdi->wb, bdi); if (err) return err; @@ -454,9 +451,9 @@ EXPORT_SYMBOL(bdi_init); void bdi_destroy(struct backing_dev_info *bdi) { - bdi_wb_shutdown(bdi); - - WARN_ON(!list_empty(&bdi->work_list)); + /* make sure nobody finds us on the bdi_list anymore */ + bdi_remove_from_list(bdi); + wb_shutdown(&bdi->wb); if (bdi->dev) { bdi_debug_unregister(bdi); @@ -464,7 +461,7 @@ void bdi_destroy(struct backing_dev_info *bdi) bdi->dev = NULL; } - bdi_wb_exit(&bdi->wb); + wb_exit(&bdi->wb); } EXPORT_SYMBOL(bdi_destroy); From 4610007142823307d930ac890d822633a05ce08c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:31 -0400 Subject: [PATCH 17/85] writeback: reorganize mm/backing-dev.c Move wb_shutdown(), bdi_register(), bdi_register_dev(), bdi_prune_sb(), bdi_remove_from_list() and bdi_unregister() so that init / exit functions are grouped together. This will make updating init / exit paths for cgroup writeback support easier. This is pure source file reorganization. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Cc: Wu Fengguang Signed-off-by: Jens Axboe --- mm/backing-dev.c | 174 +++++++++++++++++++++++------------------------ 1 file changed, 87 insertions(+), 87 deletions(-) diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 597f0cec8405..ff85ecb7d6a1 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -285,93 +285,6 @@ void wb_wakeup_delayed(struct bdi_writeback *wb) spin_unlock_bh(&wb->work_lock); } -/* - * Remove bdi from bdi_list, and ensure that it is no longer visible - */ -static void bdi_remove_from_list(struct backing_dev_info *bdi) -{ - spin_lock_bh(&bdi_lock); - list_del_rcu(&bdi->bdi_list); - spin_unlock_bh(&bdi_lock); - - synchronize_rcu_expedited(); -} - -int bdi_register(struct backing_dev_info *bdi, struct device *parent, - const char *fmt, ...) -{ - va_list args; - struct device *dev; - - if (bdi->dev) /* The driver needs to use separate queues per device */ - return 0; - - va_start(args, fmt); - dev = device_create_vargs(bdi_class, parent, MKDEV(0, 0), bdi, fmt, args); - va_end(args); - if (IS_ERR(dev)) - return PTR_ERR(dev); - - bdi->dev = dev; - - bdi_debug_register(bdi, dev_name(dev)); - set_bit(WB_registered, &bdi->wb.state); - - spin_lock_bh(&bdi_lock); - list_add_tail_rcu(&bdi->bdi_list, &bdi_list); - spin_unlock_bh(&bdi_lock); - - trace_writeback_bdi_register(bdi); - return 0; -} -EXPORT_SYMBOL(bdi_register); - -int bdi_register_dev(struct backing_dev_info *bdi, dev_t dev) -{ - return bdi_register(bdi, NULL, "%u:%u", MAJOR(dev), MINOR(dev)); -} -EXPORT_SYMBOL(bdi_register_dev); - -/* - * Remove bdi from the global list and shutdown any threads we have running - */ -static void wb_shutdown(struct bdi_writeback *wb) -{ - /* Make sure nobody queues further work */ - spin_lock_bh(&wb->work_lock); - if (!test_and_clear_bit(WB_registered, &wb->state)) { - spin_unlock_bh(&wb->work_lock); - return; - } - spin_unlock_bh(&wb->work_lock); - - /* - * Drain work list and shutdown the delayed_work. !WB_registered - * tells wb_workfn() that @wb is dying and its work_list needs to - * be drained no matter what. - */ - mod_delayed_work(bdi_wq, &wb->dwork, 0); - flush_delayed_work(&wb->dwork); - WARN_ON(!list_empty(&wb->work_list)); -} - -/* - * Called when the device behind @bdi has been removed or ejected. - * - * We can't really do much here except for reducing the dirty ratio at - * the moment. In the future we should be able to set a flag so that - * the filesystem can handle errors at mark_inode_dirty time instead - * of only at writeback time. - */ -void bdi_unregister(struct backing_dev_info *bdi) -{ - if (WARN_ON_ONCE(!bdi->dev)) - return; - - bdi_set_min_ratio(bdi, 0); -} -EXPORT_SYMBOL(bdi_unregister); - /* * Initial write bandwidth: 100 MB/s */ @@ -418,6 +331,29 @@ static int wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) return 0; } +/* + * Remove bdi from the global list and shutdown any threads we have running + */ +static void wb_shutdown(struct bdi_writeback *wb) +{ + /* Make sure nobody queues further work */ + spin_lock_bh(&wb->work_lock); + if (!test_and_clear_bit(WB_registered, &wb->state)) { + spin_unlock_bh(&wb->work_lock); + return; + } + spin_unlock_bh(&wb->work_lock); + + /* + * Drain work list and shutdown the delayed_work. !WB_registered + * tells wb_workfn() that @wb is dying and its work_list needs to + * be drained no matter what. + */ + mod_delayed_work(bdi_wq, &wb->dwork, 0); + flush_delayed_work(&wb->dwork); + WARN_ON(!list_empty(&wb->work_list)); +} + static void wb_exit(struct bdi_writeback *wb) { int i; @@ -449,6 +385,70 @@ int bdi_init(struct backing_dev_info *bdi) } EXPORT_SYMBOL(bdi_init); +int bdi_register(struct backing_dev_info *bdi, struct device *parent, + const char *fmt, ...) +{ + va_list args; + struct device *dev; + + if (bdi->dev) /* The driver needs to use separate queues per device */ + return 0; + + va_start(args, fmt); + dev = device_create_vargs(bdi_class, parent, MKDEV(0, 0), bdi, fmt, args); + va_end(args); + if (IS_ERR(dev)) + return PTR_ERR(dev); + + bdi->dev = dev; + + bdi_debug_register(bdi, dev_name(dev)); + set_bit(WB_registered, &bdi->wb.state); + + spin_lock_bh(&bdi_lock); + list_add_tail_rcu(&bdi->bdi_list, &bdi_list); + spin_unlock_bh(&bdi_lock); + + trace_writeback_bdi_register(bdi); + return 0; +} +EXPORT_SYMBOL(bdi_register); + +int bdi_register_dev(struct backing_dev_info *bdi, dev_t dev) +{ + return bdi_register(bdi, NULL, "%u:%u", MAJOR(dev), MINOR(dev)); +} +EXPORT_SYMBOL(bdi_register_dev); + +/* + * Remove bdi from bdi_list, and ensure that it is no longer visible + */ +static void bdi_remove_from_list(struct backing_dev_info *bdi) +{ + spin_lock_bh(&bdi_lock); + list_del_rcu(&bdi->bdi_list); + spin_unlock_bh(&bdi_lock); + + synchronize_rcu_expedited(); +} + +/* + * Called when the device behind @bdi has been removed or ejected. + * + * We can't really do much here except for reducing the dirty ratio at + * the moment. In the future we should be able to set a flag so that + * the filesystem can handle errors at mark_inode_dirty time instead + * of only at writeback time. + */ +void bdi_unregister(struct backing_dev_info *bdi) +{ + if (WARN_ON_ONCE(!bdi->dev)) + return; + + bdi_set_min_ratio(bdi, 0); +} +EXPORT_SYMBOL(bdi_unregister); + void bdi_destroy(struct backing_dev_info *bdi) { /* make sure nobody finds us on the bdi_list anymore */ From 66114cad64bf76a155fec1f0fff0de771cf909d5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:32 -0400 Subject: [PATCH 18/85] writeback: separate out include/linux/backing-dev-defs.h With the planned cgroup writeback support, backing-dev related declarations will be more widely used across block and cgroup; unfortunately, including backing-dev.h from include/linux/blkdev.h makes cyclic include dependency quite likely. This patch separates out backing-dev-defs.h which only has the essential definitions and updates blkdev.h to include it. c files which need access to more backing-dev details now include backing-dev.h directly. This takes backing-dev.h off the common include dependency chain making it a lot easier to use it across block and cgroup. v2: fs/fat build failure fixed. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Signed-off-by: Jens Axboe --- block/blk-integrity.c | 1 + block/blk-sysfs.c | 1 + block/bounce.c | 1 + block/genhd.c | 1 + drivers/block/drbd/drbd_int.h | 1 + drivers/block/pktcdvd.c | 1 + drivers/char/raw.c | 1 + drivers/md/bcache/request.c | 1 + drivers/md/dm.h | 1 + drivers/md/md.h | 1 + drivers/mtd/devices/block2mtd.c | 1 + fs/block_dev.c | 1 + fs/ext4/extents.c | 1 + fs/ext4/mballoc.c | 1 + fs/ext4/super.c | 1 + fs/f2fs/segment.h | 1 + fs/fat/file.c | 1 + fs/fat/inode.c | 1 + fs/hfs/super.c | 1 + fs/hfsplus/super.c | 1 + fs/nfs/filelayout/filelayout.c | 1 + fs/ocfs2/file.c | 1 + fs/reiserfs/super.c | 1 + fs/ufs/super.c | 1 + fs/xfs/xfs_file.c | 1 + include/linux/backing-dev-defs.h | 106 +++++++++++++++++++++++++++++++ include/linux/backing-dev.h | 102 +---------------------------- include/linux/blkdev.h | 2 +- mm/madvise.c | 1 + 29 files changed, 134 insertions(+), 102 deletions(-) create mode 100644 include/linux/backing-dev-defs.h diff --git a/block/blk-integrity.c b/block/blk-integrity.c index 79ffb4855af0..f548b64be092 100644 --- a/block/blk-integrity.c +++ b/block/blk-integrity.c @@ -21,6 +21,7 @@ */ #include +#include #include #include #include diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 5677eb78557d..1b60941dc4c6 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include diff --git a/block/bounce.c b/block/bounce.c index 4bac72579c1f..072280b3dd13 100644 --- a/block/bounce.c +++ b/block/bounce.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/block/genhd.c b/block/genhd.c index 0a536dc05f3b..d46ba566d62f 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h index b905e9888b88..efd19c2da9c2 100644 --- a/drivers/block/drbd/drbd_int.h +++ b/drivers/block/drbd/drbd_int.h @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index 09e628dafd9d..4c20c228184c 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -61,6 +61,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/char/raw.c b/drivers/char/raw.c index 5fc291c6157e..60316fbaf295 100644 --- a/drivers/char/raw.c +++ b/drivers/char/raw.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index 1616f668a4cb..4afb2d26b148 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -15,6 +15,7 @@ #include #include #include +#include #include diff --git a/drivers/md/dm.h b/drivers/md/dm.h index e6e66d087b26..7fff744f0865 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/md/md.h b/drivers/md/md.h index 4046a6c6f223..7da6e9c3cb53 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -16,6 +16,7 @@ #define _MD_MD_H #include +#include #include #include #include diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c index b16f3cda97ff..e2c0057737e6 100644 --- a/drivers/mtd/devices/block2mtd.c +++ b/drivers/mtd/devices/block2mtd.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/block_dev.c b/fs/block_dev.c index c7e4163ede87..e545cbfbe5b2 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index d74e08029643..e8b5866ffa07 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -39,6 +39,7 @@ #include #include #include +#include #include "ext4_jbd2.h" #include "ext4_extents.h" #include "xattr.h" diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 8d1e60214ef0..440987c8ba9e 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #ifdef CONFIG_EXT4_DEBUG diff --git a/fs/ext4/super.c b/fs/ext4/super.c index f06d0589ddba..56b8bb75c3fc 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 6408989857e0..aba72f7a8ac4 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -9,6 +9,7 @@ * published by the Free Software Foundation. */ #include +#include /* constant macro */ #define NULL_SEGNO ((unsigned int)(~0)) diff --git a/fs/fat/file.c b/fs/fat/file.c index 442d50a0e33e..a08f1039909a 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include "fat.h" diff --git a/fs/fat/inode.c b/fs/fat/inode.c index c06774658345..509411dd3698 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include "fat.h" diff --git a/fs/hfs/super.c b/fs/hfs/super.c index eee7206c38d1..55c03b9e9070 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -14,6 +14,7 @@ #include #include +#include #include #include #include diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 593af2fdcc2d..7302d96ae8bf 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c index a46bf6de9ce4..b34f2e228601 100644 --- a/fs/nfs/filelayout/filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -32,6 +32,7 @@ #include #include #include +#include #include diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index d8b670cbd909..8f1feca89fb0 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -37,6 +37,7 @@ #include #include #include +#include #include diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 0111ad0466ed..3e0af317fcc3 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -21,6 +21,7 @@ #include "xattr.h" #include #include +#include #include #include #include diff --git a/fs/ufs/super.c b/fs/ufs/super.c index b3bc3e7ae79d..098508a93c7b 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -80,6 +80,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 8121e75352ee..4e00b38efbe0 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -41,6 +41,7 @@ #include #include #include +#include static const struct vm_operations_struct xfs_file_vm_ops; diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h new file mode 100644 index 000000000000..aa18c4bd43c1 --- /dev/null +++ b/include/linux/backing-dev-defs.h @@ -0,0 +1,106 @@ +#ifndef __LINUX_BACKING_DEV_DEFS_H +#define __LINUX_BACKING_DEV_DEFS_H + +#include +#include +#include +#include +#include +#include + +struct page; +struct device; +struct dentry; + +/* + * Bits in bdi_writeback.state + */ +enum wb_state { + WB_async_congested, /* The async (write) queue is getting full */ + WB_sync_congested, /* The sync queue is getting full */ + WB_registered, /* bdi_register() was done */ + WB_writeback_running, /* Writeback is in progress */ +}; + +typedef int (congested_fn)(void *, int); + +enum wb_stat_item { + WB_RECLAIMABLE, + WB_WRITEBACK, + WB_DIRTIED, + WB_WRITTEN, + NR_WB_STAT_ITEMS +}; + +#define WB_STAT_BATCH (8*(1+ilog2(nr_cpu_ids))) + +struct bdi_writeback { + struct backing_dev_info *bdi; /* our parent bdi */ + + unsigned long state; /* Always use atomic bitops on this */ + unsigned long last_old_flush; /* last old data flush */ + + struct list_head b_dirty; /* dirty inodes */ + struct list_head b_io; /* parked for writeback */ + struct list_head b_more_io; /* parked for more writeback */ + struct list_head b_dirty_time; /* time stamps are dirty */ + spinlock_t list_lock; /* protects the b_* lists */ + + struct percpu_counter stat[NR_WB_STAT_ITEMS]; + + unsigned long bw_time_stamp; /* last time write bw is updated */ + unsigned long dirtied_stamp; + unsigned long written_stamp; /* pages written at bw_time_stamp */ + unsigned long write_bandwidth; /* the estimated write bandwidth */ + unsigned long avg_write_bandwidth; /* further smoothed write bw */ + + /* + * The base dirty throttle rate, re-calculated on every 200ms. + * All the bdi tasks' dirty rate will be curbed under it. + * @dirty_ratelimit tracks the estimated @balanced_dirty_ratelimit + * in small steps and is much more smooth/stable than the latter. + */ + unsigned long dirty_ratelimit; + unsigned long balanced_dirty_ratelimit; + + struct fprop_local_percpu completions; + int dirty_exceeded; + + spinlock_t work_lock; /* protects work_list & dwork scheduling */ + struct list_head work_list; + struct delayed_work dwork; /* work item used for writeback */ +}; + +struct backing_dev_info { + struct list_head bdi_list; + unsigned long ra_pages; /* max readahead in PAGE_CACHE_SIZE units */ + unsigned int capabilities; /* Device capabilities */ + congested_fn *congested_fn; /* Function pointer if device is md/dm */ + void *congested_data; /* Pointer to aux data for congested func */ + + char *name; + + unsigned int min_ratio; + unsigned int max_ratio, max_prop_frac; + + struct bdi_writeback wb; /* default writeback info for this bdi */ + + struct device *dev; + + struct timer_list laptop_mode_wb_timer; + +#ifdef CONFIG_DEBUG_FS + struct dentry *debug_dir; + struct dentry *debug_stats; +#endif +}; + +enum { + BLK_RW_ASYNC = 0, + BLK_RW_SYNC = 1, +}; + +void clear_bdi_congested(struct backing_dev_info *bdi, int sync); +void set_bdi_congested(struct backing_dev_info *bdi, int sync); + +#endif /* __LINUX_BACKING_DEV_DEFS_H */ diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index d796f49ce87a..5e39f7a8efed 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -8,104 +8,11 @@ #ifndef _LINUX_BACKING_DEV_H #define _LINUX_BACKING_DEV_H -#include -#include -#include #include #include #include -#include #include -#include -#include -#include - -struct page; -struct device; -struct dentry; - -/* - * Bits in bdi_writeback.state - */ -enum wb_state { - WB_async_congested, /* The async (write) queue is getting full */ - WB_sync_congested, /* The sync queue is getting full */ - WB_registered, /* bdi_register() was done */ - WB_writeback_running, /* Writeback is in progress */ -}; - -typedef int (congested_fn)(void *, int); - -enum wb_stat_item { - WB_RECLAIMABLE, - WB_WRITEBACK, - WB_DIRTIED, - WB_WRITTEN, - NR_WB_STAT_ITEMS -}; - -#define WB_STAT_BATCH (8*(1+ilog2(nr_cpu_ids))) - -struct bdi_writeback { - struct backing_dev_info *bdi; /* our parent bdi */ - - unsigned long state; /* Always use atomic bitops on this */ - unsigned long last_old_flush; /* last old data flush */ - - struct list_head b_dirty; /* dirty inodes */ - struct list_head b_io; /* parked for writeback */ - struct list_head b_more_io; /* parked for more writeback */ - struct list_head b_dirty_time; /* time stamps are dirty */ - spinlock_t list_lock; /* protects the b_* lists */ - - struct percpu_counter stat[NR_WB_STAT_ITEMS]; - - unsigned long bw_time_stamp; /* last time write bw is updated */ - unsigned long dirtied_stamp; - unsigned long written_stamp; /* pages written at bw_time_stamp */ - unsigned long write_bandwidth; /* the estimated write bandwidth */ - unsigned long avg_write_bandwidth; /* further smoothed write bw */ - - /* - * The base dirty throttle rate, re-calculated on every 200ms. - * All the bdi tasks' dirty rate will be curbed under it. - * @dirty_ratelimit tracks the estimated @balanced_dirty_ratelimit - * in small steps and is much more smooth/stable than the latter. - */ - unsigned long dirty_ratelimit; - unsigned long balanced_dirty_ratelimit; - - struct fprop_local_percpu completions; - int dirty_exceeded; - - spinlock_t work_lock; /* protects work_list & dwork scheduling */ - struct list_head work_list; - struct delayed_work dwork; /* work item used for writeback */ -}; - -struct backing_dev_info { - struct list_head bdi_list; - unsigned long ra_pages; /* max readahead in PAGE_CACHE_SIZE units */ - unsigned int capabilities; /* Device capabilities */ - congested_fn *congested_fn; /* Function pointer if device is md/dm */ - void *congested_data; /* Pointer to aux data for congested func */ - - char *name; - - unsigned int min_ratio; - unsigned int max_ratio, max_prop_frac; - - struct bdi_writeback wb; /* default writeback info for this bdi */ - - struct device *dev; - - struct timer_list laptop_mode_wb_timer; - -#ifdef CONFIG_DEBUG_FS - struct dentry *debug_dir; - struct dentry *debug_stats; -#endif -}; +#include struct backing_dev_info *inode_to_bdi(struct inode *inode); @@ -265,13 +172,6 @@ static inline int bdi_rw_congested(struct backing_dev_info *bdi) (1 << WB_async_congested)); } -enum { - BLK_RW_ASYNC = 0, - BLK_RW_SYNC = 1, -}; - -void clear_bdi_congested(struct backing_dev_info *bdi, int sync); -void set_bdi_congested(struct backing_dev_info *bdi, int sync); long congestion_wait(int sync, long timeout); long wait_iff_congested(struct zone *zone, int sync, long timeout); int pdflush_proc_obsolete(struct ctl_table *table, int write, diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index ccaa9aecd593..60d2726a6b62 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/mm/madvise.c b/mm/madvise.c index d551475517bf..64bb8a22110c 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include From a212b105b07d75b48b1a166378282e8a77fbf53d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:33 -0400 Subject: [PATCH 19/85] bdi: make inode_to_bdi() inline Now that bdi definitions are moved to backing-dev-defs.h, backing-dev.h can include blkdev.h and inline inode_to_bdi() without worrying about introducing circular include dependency. The function gets called from hot paths and fairly trivial. This patch makes inode_to_bdi() and sb_is_blkdev_sb() that the function calls inline. blockdev_superblock and noop_backing_dev_info are EXPORT_GPL'd to allow the inline functions to be used from modules. While at it, make sb_is_blkdev_sb() return bool instead of int. v2: Fixed typo in description as suggested by Jan. Signed-off-by: Tejun Heo Reviewed-by: Jens Axboe Cc: Christoph Hellwig Signed-off-by: Jens Axboe --- fs/block_dev.c | 8 ++------ fs/fs-writeback.c | 16 ---------------- include/linux/backing-dev.h | 18 ++++++++++++++++-- include/linux/fs.h | 8 +++++++- mm/backing-dev.c | 1 + 5 files changed, 26 insertions(+), 25 deletions(-) diff --git a/fs/block_dev.c b/fs/block_dev.c index e545cbfbe5b2..f04c873a7365 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -547,7 +547,8 @@ static struct file_system_type bd_type = { .kill_sb = kill_anon_super, }; -static struct super_block *blockdev_superblock __read_mostly; +struct super_block *blockdev_superblock __read_mostly; +EXPORT_SYMBOL_GPL(blockdev_superblock); void __init bdev_cache_init(void) { @@ -688,11 +689,6 @@ static struct block_device *bd_acquire(struct inode *inode) return bdev; } -int sb_is_blkdev_sb(struct super_block *sb) -{ - return sb == blockdev_superblock; -} - /* Call when you free inode */ void bd_forget(struct inode *inode) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index a69d2e10d514..34d1cb854fc1 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -78,22 +78,6 @@ int writeback_in_progress(struct backing_dev_info *bdi) } EXPORT_SYMBOL(writeback_in_progress); -struct backing_dev_info *inode_to_bdi(struct inode *inode) -{ - struct super_block *sb; - - if (!inode) - return &noop_backing_dev_info; - - sb = inode->i_sb; -#ifdef CONFIG_BLOCK - if (sb_is_blkdev_sb(sb)) - return blk_get_backing_dev_info(I_BDEV(inode)); -#endif - return sb->s_bdi; -} -EXPORT_SYMBOL_GPL(inode_to_bdi); - static inline struct inode *wb_inode(struct list_head *head) { return list_entry(head, struct inode, i_wb_list); diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 5e39f7a8efed..785782034e86 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -11,11 +11,10 @@ #include #include #include +#include #include #include -struct backing_dev_info *inode_to_bdi(struct inode *inode); - int __must_check bdi_init(struct backing_dev_info *bdi); void bdi_destroy(struct backing_dev_info *bdi); @@ -149,6 +148,21 @@ extern struct backing_dev_info noop_backing_dev_info; int writeback_in_progress(struct backing_dev_info *bdi); +static inline struct backing_dev_info *inode_to_bdi(struct inode *inode) +{ + struct super_block *sb; + + if (!inode) + return &noop_backing_dev_info; + + sb = inode->i_sb; +#ifdef CONFIG_BLOCK + if (sb_is_blkdev_sb(sb)) + return blk_get_backing_dev_info(I_BDEV(inode)); +#endif + return sb->s_bdi; +} + static inline int bdi_congested(struct backing_dev_info *bdi, int bdi_bits) { if (bdi->congested_fn) diff --git a/include/linux/fs.h b/include/linux/fs.h index 1ef63900243c..ce100b87fba3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2240,7 +2240,13 @@ extern struct super_block *freeze_bdev(struct block_device *); extern void emergency_thaw_all(void); extern int thaw_bdev(struct block_device *bdev, struct super_block *sb); extern int fsync_bdev(struct block_device *); -extern int sb_is_blkdev_sb(struct super_block *sb); + +extern struct super_block *blockdev_superblock; + +static inline bool sb_is_blkdev_sb(struct super_block *sb) +{ + return sb == blockdev_superblock; +} #else static inline void bd_forget(struct inode *inode) {} static inline int sync_blockdev(struct block_device *bdev) { return 0; } diff --git a/mm/backing-dev.c b/mm/backing-dev.c index ff85ecb7d6a1..b0707d1b1d38 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -18,6 +18,7 @@ struct backing_dev_info noop_backing_dev_info = { .name = "noop", .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; +EXPORT_SYMBOL_GPL(noop_backing_dev_info); static struct class *bdi_class; From 8395cd9f813d5d7ed9870e642230acfcfc1e8a0a Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:34 -0400 Subject: [PATCH 20/85] writeback: add @gfp to wb_init() wb_init() currently always uses GFP_KERNEL but the planned cgroup writeback support needs using other allocation masks. Add @gfp to wb_init(). This patch doesn't introduce any behavior changes. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Signed-off-by: Jens Axboe --- mm/backing-dev.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/mm/backing-dev.c b/mm/backing-dev.c index b0707d1b1d38..805b2870ca4e 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -291,7 +291,8 @@ void wb_wakeup_delayed(struct bdi_writeback *wb) */ #define INIT_BW (100 << (20 - PAGE_SHIFT)) -static int wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) +static int wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi, + gfp_t gfp) { int i, err; @@ -315,12 +316,12 @@ static int wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) INIT_LIST_HEAD(&wb->work_list); INIT_DELAYED_WORK(&wb->dwork, wb_workfn); - err = fprop_local_init_percpu(&wb->completions, GFP_KERNEL); + err = fprop_local_init_percpu(&wb->completions, gfp); if (err) return err; for (i = 0; i < NR_WB_STAT_ITEMS; i++) { - err = percpu_counter_init(&wb->stat[i], 0, GFP_KERNEL); + err = percpu_counter_init(&wb->stat[i], 0, gfp); if (err) { while (--i) percpu_counter_destroy(&wb->stat[i]); @@ -378,7 +379,7 @@ int bdi_init(struct backing_dev_info *bdi) bdi->max_prop_frac = FPROP_FRAC_BASE; INIT_LIST_HEAD(&bdi->bdi_list); - err = wb_init(&bdi->wb, bdi); + err = wb_init(&bdi->wb, bdi, GFP_KERNEL); if (err) return err; From 4aa9c692e052cf6db99db62a8fe0543e5c455da7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:35 -0400 Subject: [PATCH 21/85] bdi: separate out congested state into a separate struct Currently, a wb's (bdi_writeback) congestion state is carried in its ->state field; however, cgroup writeback support will require multiple wb's sharing the same congestion state. This patch separates out congestion state into its own struct - struct bdi_writeback_congested. A new field wb field, wb_congested, points to its associated congested struct. The default wb, bdi->wb, always points to bdi->wb_congested. While this patch adds a layer of indirection, it doesn't introduce any behavior changes. Signed-off-by: Tejun Heo Signed-off-by: Jens Axboe --- include/linux/backing-dev-defs.h | 14 ++++++++++++-- include/linux/backing-dev.h | 2 +- mm/backing-dev.c | 7 +++++-- 3 files changed, 18 insertions(+), 5 deletions(-) diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index aa18c4bd43c1..9e9eafa5f5aa 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -16,12 +16,15 @@ struct dentry; * Bits in bdi_writeback.state */ enum wb_state { - WB_async_congested, /* The async (write) queue is getting full */ - WB_sync_congested, /* The sync queue is getting full */ WB_registered, /* bdi_register() was done */ WB_writeback_running, /* Writeback is in progress */ }; +enum wb_congested_state { + WB_async_congested, /* The async (write) queue is getting full */ + WB_sync_congested, /* The sync queue is getting full */ +}; + typedef int (congested_fn)(void *, int); enum wb_stat_item { @@ -34,6 +37,10 @@ enum wb_stat_item { #define WB_STAT_BATCH (8*(1+ilog2(nr_cpu_ids))) +struct bdi_writeback_congested { + unsigned long state; /* WB_[a]sync_congested flags */ +}; + struct bdi_writeback { struct backing_dev_info *bdi; /* our parent bdi */ @@ -48,6 +55,8 @@ struct bdi_writeback { struct percpu_counter stat[NR_WB_STAT_ITEMS]; + struct bdi_writeback_congested *congested; + unsigned long bw_time_stamp; /* last time write bw is updated */ unsigned long dirtied_stamp; unsigned long written_stamp; /* pages written at bw_time_stamp */ @@ -84,6 +93,7 @@ struct backing_dev_info { unsigned int max_ratio, max_prop_frac; struct bdi_writeback wb; /* default writeback info for this bdi */ + struct bdi_writeback_congested wb_congested; struct device *dev; diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 785782034e86..bfdaa18ba0a1 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -167,7 +167,7 @@ static inline int bdi_congested(struct backing_dev_info *bdi, int bdi_bits) { if (bdi->congested_fn) return bdi->congested_fn(bdi->congested_data, bdi_bits); - return (bdi->wb.state & bdi_bits); + return (bdi->wb.congested->state & bdi_bits); } static inline int bdi_read_congested(struct backing_dev_info *bdi) diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 805b2870ca4e..5ec7658cb984 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -383,6 +383,9 @@ int bdi_init(struct backing_dev_info *bdi) if (err) return err; + bdi->wb_congested.state = 0; + bdi->wb.congested = &bdi->wb_congested; + return 0; } EXPORT_SYMBOL(bdi_init); @@ -504,7 +507,7 @@ void clear_bdi_congested(struct backing_dev_info *bdi, int sync) wait_queue_head_t *wqh = &congestion_wqh[sync]; bit = sync ? WB_sync_congested : WB_async_congested; - if (test_and_clear_bit(bit, &bdi->wb.state)) + if (test_and_clear_bit(bit, &bdi->wb.congested->state)) atomic_dec(&nr_bdi_congested[sync]); smp_mb__after_atomic(); if (waitqueue_active(wqh)) @@ -517,7 +520,7 @@ void set_bdi_congested(struct backing_dev_info *bdi, int sync) enum wb_state bit; bit = sync ? WB_sync_congested : WB_async_congested; - if (!test_and_set_bit(bit, &bdi->wb.state)) + if (!test_and_set_bit(bit, &bdi->wb.congested->state)) atomic_inc(&nr_bdi_congested[sync]); } EXPORT_SYMBOL(set_bdi_congested); From 89e9b9e07a390c50980d10aa37a04631db5a23ab Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:36 -0400 Subject: [PATCH 22/85] writeback: add {CONFIG|BDI_CAP|FS}_CGROUP_WRITEBACK cgroup writeback requires support from both bdi and filesystem sides. Add BDI_CAP_CGROUP_WRITEBACK and FS_CGROUP_WRITEBACK to indicate support and enable BDI_CAP_CGROUP_WRITEBACK on block based bdi's by default. Also, define CONFIG_CGROUP_WRITEBACK which is enabled if both MEMCG and BLK_CGROUP are enabled. inode_cgwb_enabled() which determines whether a given inode's both bdi and fs support cgroup writeback is added. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- block/blk-core.c | 2 +- include/linux/backing-dev.h | 32 +++++++++++++++++++++++++++++++- include/linux/fs.h | 1 + init/Kconfig | 5 +++++ 4 files changed, 38 insertions(+), 2 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index c3ba9c3b08d4..c114a61590bc 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -621,7 +621,7 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id) q->backing_dev_info.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; - q->backing_dev_info.capabilities = 0; + q->backing_dev_info.capabilities = BDI_CAP_CGROUP_WRITEBACK; q->backing_dev_info.name = "block"; q->node = node_id; diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index bfdaa18ba0a1..6bb31234e6a9 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -134,12 +134,15 @@ int bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ratio); * BDI_CAP_NO_WRITEBACK: Don't write pages back * BDI_CAP_NO_ACCT_WB: Don't automatically account writeback pages * BDI_CAP_STRICTLIMIT: Keep number of dirty pages below bdi threshold. + * + * BDI_CAP_CGROUP_WRITEBACK: Supports cgroup-aware writeback. */ #define BDI_CAP_NO_ACCT_DIRTY 0x00000001 #define BDI_CAP_NO_WRITEBACK 0x00000002 #define BDI_CAP_NO_ACCT_WB 0x00000004 #define BDI_CAP_STABLE_WRITES 0x00000008 #define BDI_CAP_STRICTLIMIT 0x00000010 +#define BDI_CAP_CGROUP_WRITEBACK 0x00000020 #define BDI_CAP_NO_ACCT_AND_WRITEBACK \ (BDI_CAP_NO_WRITEBACK | BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_ACCT_WB) @@ -229,4 +232,31 @@ static inline int bdi_sched_wait(void *word) return 0; } -#endif /* _LINUX_BACKING_DEV_H */ +#ifdef CONFIG_CGROUP_WRITEBACK + +/** + * inode_cgwb_enabled - test whether cgroup writeback is enabled on an inode + * @inode: inode of interest + * + * cgroup writeback requires support from both the bdi and filesystem. + * Test whether @inode has both. + */ +static inline bool inode_cgwb_enabled(struct inode *inode) +{ + struct backing_dev_info *bdi = inode_to_bdi(inode); + + return bdi_cap_account_dirty(bdi) && + (bdi->capabilities & BDI_CAP_CGROUP_WRITEBACK) && + (inode->i_sb->s_type->fs_flags & FS_CGROUP_WRITEBACK); +} + +#else /* CONFIG_CGROUP_WRITEBACK */ + +static inline bool inode_cgwb_enabled(struct inode *inode) +{ + return false; +} + +#endif /* CONFIG_CGROUP_WRITEBACK */ + +#endif /* _LINUX_BACKING_DEV_H */ diff --git a/include/linux/fs.h b/include/linux/fs.h index ce100b87fba3..74e0ae0626a8 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1897,6 +1897,7 @@ struct file_system_type { #define FS_HAS_SUBTYPE 4 #define FS_USERNS_MOUNT 8 /* Can be mounted by userns root */ #define FS_USERNS_DEV_MOUNT 16 /* A userns mount does not imply MNT_NODEV */ +#define FS_CGROUP_WRITEBACK 32 /* Supports cgroup-aware writeback */ #define FS_RENAME_DOES_D_MOVE 32768 /* FS will handle d_move() during rename() internally. */ struct dentry *(*mount) (struct file_system_type *, int, const char *, void *); diff --git a/init/Kconfig b/init/Kconfig index dc24dec60232..d4f763332f9f 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1141,6 +1141,11 @@ config DEBUG_BLK_CGROUP Enable some debugging help. Currently it exports additional stat files in a cgroup which can be useful for debugging. +config CGROUP_WRITEBACK + bool + depends on MEMCG && BLK_CGROUP + default y + endif # CGROUPS config CHECKPOINT_RESTORE From 52ebea749aaed195245701a8f90a23d672c7a933 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:37 -0400 Subject: [PATCH 23/85] writeback: make backing_dev_info host cgroup-specific bdi_writebacks For the planned cgroup writeback support, on each bdi (backing_dev_info), each memcg will be served by a separate wb (bdi_writeback). This patch updates bdi so that a bdi can host multiple wbs (bdi_writebacks). On the default hierarchy, blkcg implicitly enables memcg. This allows using memcg's page ownership for attributing writeback IOs, and every memcg - blkcg combination can be served by its own wb by assigning a dedicated wb to each memcg. This means that there may be multiple wb's of a bdi mapped to the same blkcg. As congested state is per blkcg - bdi combination, those wb's should share the same congested state. This is achieved by tracking congested state via bdi_writeback_congested structs which are keyed by blkcg. bdi->wb remains unchanged and will keep serving the root cgroup. cgwb's (cgroup wb's) for non-root cgroups are created on-demand or looked up while dirtying an inode according to the memcg of the page being dirtied or current task. Each cgwb is indexed on bdi->cgwb_tree by its memcg id. Once an inode is associated with its wb, it can be retrieved using inode_to_wb(). Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being associated with bdi->wb. v3: inode_attach_wb() in account_page_dirtied() moved inside mapping_cap_account_dirty() block where it's known to be !NULL. Also, an unnecessary NULL check before kfree() removed. Both detected by the kbuild bot. v2: Updated so that wb association is per inode and wb is per memcg rather than blkcg. Signed-off-by: Tejun Heo Cc: kbuild test robot Cc: Dan Carpenter Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- block/blk-cgroup.c | 7 +- fs/fs-writeback.c | 8 +- fs/inode.c | 1 + include/linux/backing-dev-defs.h | 59 ++++- include/linux/backing-dev.h | 195 +++++++++++++++ include/linux/blk-cgroup.h | 4 + include/linux/fs.h | 4 + include/linux/memcontrol.h | 4 + mm/backing-dev.c | 397 +++++++++++++++++++++++++++++++ mm/memcontrol.c | 19 +- mm/page-writeback.c | 11 +- 11 files changed, 698 insertions(+), 11 deletions(-) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 54ec1721cd3e..979cfdbb94e0 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include @@ -797,6 +798,8 @@ static void blkcg_css_offline(struct cgroup_subsys_state *css) } spin_unlock_irq(&blkcg->lock); + + wb_blkcg_offline(blkcg); } static void blkcg_css_free(struct cgroup_subsys_state *css) @@ -827,7 +830,9 @@ done: spin_lock_init(&blkcg->lock); INIT_RADIX_TREE(&blkcg->blkg_tree, GFP_ATOMIC); INIT_HLIST_HEAD(&blkcg->blkg_list); - +#ifdef CONFIG_CGROUP_WRITEBACK + INIT_LIST_HEAD(&blkcg->cgwb_list); +#endif return &blkcg->css; } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 34d1cb854fc1..99a2440c5588 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -185,11 +185,11 @@ void bdi_start_background_writeback(struct backing_dev_info *bdi) */ void inode_wb_list_del(struct inode *inode) { - struct backing_dev_info *bdi = inode_to_bdi(inode); + struct bdi_writeback *wb = inode_to_wb(inode); - spin_lock(&bdi->wb.list_lock); + spin_lock(&wb->list_lock); list_del_init(&inode->i_wb_list); - spin_unlock(&bdi->wb.list_lock); + spin_unlock(&wb->list_lock); } /* @@ -1268,6 +1268,8 @@ void __mark_inode_dirty(struct inode *inode, int flags) if ((inode->i_state & flags) != flags) { const int was_dirty = inode->i_state & I_DIRTY; + inode_attach_wb(inode, NULL); + if (flags & I_DIRTY_INODE) inode->i_state &= ~I_DIRTY_TIME; inode->i_state |= flags; diff --git a/fs/inode.c b/fs/inode.c index ea37cd17b53f..efc9edacfb9b 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -223,6 +223,7 @@ EXPORT_SYMBOL(free_inode_nonrcu); void __destroy_inode(struct inode *inode) { BUG_ON(inode_has_buffers(inode)); + inode_detach_wb(inode); security_inode_free(inode); fsnotify_inode_delete(inode); locks_free_lock_context(inode->i_flctx); diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index 9e9eafa5f5aa..a1e9c407a59a 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -2,8 +2,11 @@ #define __LINUX_BACKING_DEV_DEFS_H #include +#include +#include #include #include +#include #include #include #include @@ -37,10 +40,43 @@ enum wb_stat_item { #define WB_STAT_BATCH (8*(1+ilog2(nr_cpu_ids))) +/* + * For cgroup writeback, multiple wb's may map to the same blkcg. Those + * wb's can operate mostly independently but should share the congested + * state. To facilitate such sharing, the congested state is tracked using + * the following struct which is created on demand, indexed by blkcg ID on + * its bdi, and refcounted. + */ struct bdi_writeback_congested { unsigned long state; /* WB_[a]sync_congested flags */ + +#ifdef CONFIG_CGROUP_WRITEBACK + struct backing_dev_info *bdi; /* the associated bdi */ + atomic_t refcnt; /* nr of attached wb's and blkg */ + int blkcg_id; /* ID of the associated blkcg */ + struct rb_node rb_node; /* on bdi->cgwb_congestion_tree */ +#endif }; +/* + * Each wb (bdi_writeback) can perform writeback operations, is measured + * and throttled, independently. Without cgroup writeback, each bdi + * (bdi_writeback) is served by its embedded bdi->wb. + * + * On the default hierarchy, blkcg implicitly enables memcg. This allows + * using memcg's page ownership for attributing writeback IOs, and every + * memcg - blkcg combination can be served by its own wb by assigning a + * dedicated wb to each memcg, which enables isolation across different + * cgroups and propagation of IO back pressure down from the IO layer upto + * the tasks which are generating the dirty pages to be written back. + * + * A cgroup wb is indexed on its bdi by the ID of the associated memcg, + * refcounted with the number of inodes attached to it, and pins the memcg + * and the corresponding blkcg. As the corresponding blkcg for a memcg may + * change as blkcg is disabled and enabled higher up in the hierarchy, a wb + * is tested for blkcg after lookup and removed from index on mismatch so + * that a new wb for the combination can be created. + */ struct bdi_writeback { struct backing_dev_info *bdi; /* our parent bdi */ @@ -78,6 +114,19 @@ struct bdi_writeback { spinlock_t work_lock; /* protects work_list & dwork scheduling */ struct list_head work_list; struct delayed_work dwork; /* work item used for writeback */ + +#ifdef CONFIG_CGROUP_WRITEBACK + struct percpu_ref refcnt; /* used only for !root wb's */ + struct cgroup_subsys_state *memcg_css; /* the associated memcg */ + struct cgroup_subsys_state *blkcg_css; /* and blkcg */ + struct list_head memcg_node; /* anchored at memcg->cgwb_list */ + struct list_head blkcg_node; /* anchored at blkcg->cgwb_list */ + + union { + struct work_struct release_work; + struct rcu_head rcu; + }; +#endif }; struct backing_dev_info { @@ -92,9 +141,13 @@ struct backing_dev_info { unsigned int min_ratio; unsigned int max_ratio, max_prop_frac; - struct bdi_writeback wb; /* default writeback info for this bdi */ - struct bdi_writeback_congested wb_congested; - + struct bdi_writeback wb; /* the root writeback info for this bdi */ + struct bdi_writeback_congested wb_congested; /* its congested state */ +#ifdef CONFIG_CGROUP_WRITEBACK + struct radix_tree_root cgwb_tree; /* radix tree of active cgroup wbs */ + struct rb_root cgwb_congested_tree; /* their congested states */ + atomic_t usage_cnt; /* counts both cgwbs and cgwb_contested's */ +#endif struct device *dev; struct timer_list laptop_mode_wb_timer; diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 6bb31234e6a9..8ae59df2e3d1 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -13,6 +13,7 @@ #include #include #include +#include #include int __must_check bdi_init(struct backing_dev_info *bdi); @@ -234,6 +235,16 @@ static inline int bdi_sched_wait(void *word) #ifdef CONFIG_CGROUP_WRITEBACK +struct bdi_writeback_congested * +wb_congested_get_create(struct backing_dev_info *bdi, int blkcg_id, gfp_t gfp); +void wb_congested_put(struct bdi_writeback_congested *congested); +struct bdi_writeback *wb_get_create(struct backing_dev_info *bdi, + struct cgroup_subsys_state *memcg_css, + gfp_t gfp); +void __inode_attach_wb(struct inode *inode, struct page *page); +void wb_memcg_offline(struct mem_cgroup *memcg); +void wb_blkcg_offline(struct blkcg *blkcg); + /** * inode_cgwb_enabled - test whether cgroup writeback is enabled on an inode * @inode: inode of interest @@ -250,6 +261,135 @@ static inline bool inode_cgwb_enabled(struct inode *inode) (inode->i_sb->s_type->fs_flags & FS_CGROUP_WRITEBACK); } +/** + * wb_tryget - try to increment a wb's refcount + * @wb: bdi_writeback to get + */ +static inline bool wb_tryget(struct bdi_writeback *wb) +{ + if (wb != &wb->bdi->wb) + return percpu_ref_tryget(&wb->refcnt); + return true; +} + +/** + * wb_get - increment a wb's refcount + * @wb: bdi_writeback to get + */ +static inline void wb_get(struct bdi_writeback *wb) +{ + if (wb != &wb->bdi->wb) + percpu_ref_get(&wb->refcnt); +} + +/** + * wb_put - decrement a wb's refcount + * @wb: bdi_writeback to put + */ +static inline void wb_put(struct bdi_writeback *wb) +{ + if (wb != &wb->bdi->wb) + percpu_ref_put(&wb->refcnt); +} + +/** + * wb_find_current - find wb for %current on a bdi + * @bdi: bdi of interest + * + * Find the wb of @bdi which matches both the memcg and blkcg of %current. + * Must be called under rcu_read_lock() which protects the returend wb. + * NULL if not found. + */ +static inline struct bdi_writeback *wb_find_current(struct backing_dev_info *bdi) +{ + struct cgroup_subsys_state *memcg_css; + struct bdi_writeback *wb; + + memcg_css = task_css(current, memory_cgrp_id); + if (!memcg_css->parent) + return &bdi->wb; + + wb = radix_tree_lookup(&bdi->cgwb_tree, memcg_css->id); + + /* + * %current's blkcg equals the effective blkcg of its memcg. No + * need to use the relatively expensive cgroup_get_e_css(). + */ + if (likely(wb && wb->blkcg_css == task_css(current, blkio_cgrp_id))) + return wb; + return NULL; +} + +/** + * wb_get_create_current - get or create wb for %current on a bdi + * @bdi: bdi of interest + * @gfp: allocation mask + * + * Equivalent to wb_get_create() on %current's memcg. This function is + * called from a relatively hot path and optimizes the common cases using + * wb_find_current(). + */ +static inline struct bdi_writeback * +wb_get_create_current(struct backing_dev_info *bdi, gfp_t gfp) +{ + struct bdi_writeback *wb; + + rcu_read_lock(); + wb = wb_find_current(bdi); + if (wb && unlikely(!wb_tryget(wb))) + wb = NULL; + rcu_read_unlock(); + + if (unlikely(!wb)) { + struct cgroup_subsys_state *memcg_css; + + memcg_css = task_get_css(current, memory_cgrp_id); + wb = wb_get_create(bdi, memcg_css, gfp); + css_put(memcg_css); + } + return wb; +} + +/** + * inode_attach_wb - associate an inode with its wb + * @inode: inode of interest + * @page: page being dirtied (may be NULL) + * + * If @inode doesn't have its wb, associate it with the wb matching the + * memcg of @page or, if @page is NULL, %current. May be called w/ or w/o + * @inode->i_lock. + */ +static inline void inode_attach_wb(struct inode *inode, struct page *page) +{ + if (!inode->i_wb) + __inode_attach_wb(inode, page); +} + +/** + * inode_detach_wb - disassociate an inode from its wb + * @inode: inode of interest + * + * @inode is being freed. Detach from its wb. + */ +static inline void inode_detach_wb(struct inode *inode) +{ + if (inode->i_wb) { + wb_put(inode->i_wb); + inode->i_wb = NULL; + } +} + +/** + * inode_to_wb - determine the wb of an inode + * @inode: inode of interest + * + * Returns the wb @inode is currently associated with. + */ +static inline struct bdi_writeback *inode_to_wb(struct inode *inode) +{ + return inode->i_wb; +} + #else /* CONFIG_CGROUP_WRITEBACK */ static inline bool inode_cgwb_enabled(struct inode *inode) @@ -257,6 +397,61 @@ static inline bool inode_cgwb_enabled(struct inode *inode) return false; } +static inline struct bdi_writeback_congested * +wb_congested_get_create(struct backing_dev_info *bdi, int blkcg_id, gfp_t gfp) +{ + return bdi->wb.congested; +} + +static inline void wb_congested_put(struct bdi_writeback_congested *congested) +{ +} + +static inline bool wb_tryget(struct bdi_writeback *wb) +{ + return true; +} + +static inline void wb_get(struct bdi_writeback *wb) +{ +} + +static inline void wb_put(struct bdi_writeback *wb) +{ +} + +static inline struct bdi_writeback *wb_find_current(struct backing_dev_info *bdi) +{ + return &bdi->wb; +} + +static inline struct bdi_writeback * +wb_get_create_current(struct backing_dev_info *bdi, gfp_t gfp) +{ + return &bdi->wb; +} + +static inline void inode_attach_wb(struct inode *inode, struct page *page) +{ +} + +static inline void inode_detach_wb(struct inode *inode) +{ +} + +static inline struct bdi_writeback *inode_to_wb(struct inode *inode) +{ + return &inode_to_bdi(inode)->wb; +} + +static inline void wb_memcg_offline(struct mem_cgroup *memcg) +{ +} + +static inline void wb_blkcg_offline(struct blkcg *blkcg) +{ +} + #endif /* CONFIG_CGROUP_WRITEBACK */ #endif /* _LINUX_BACKING_DEV_H */ diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h index 4dc643f2046e..3033eb173eb4 100644 --- a/include/linux/blk-cgroup.h +++ b/include/linux/blk-cgroup.h @@ -53,6 +53,10 @@ struct blkcg { /* TODO: per-policy storage in blkcg */ unsigned int cfq_weight; /* belongs to cfq */ unsigned int cfq_leaf_weight; + +#ifdef CONFIG_CGROUP_WRITEBACK + struct list_head cgwb_list; +#endif }; struct blkg_stat { diff --git a/include/linux/fs.h b/include/linux/fs.h index 74e0ae0626a8..67a42ec95065 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -35,6 +35,7 @@ #include struct backing_dev_info; +struct bdi_writeback; struct export_operations; struct hd_geometry; struct iovec; @@ -635,6 +636,9 @@ struct inode { struct hlist_node i_hash; struct list_head i_wb_list; /* backing dev IO list */ +#ifdef CONFIG_CGROUP_WRITEBACK + struct bdi_writeback *i_wb; /* the associated cgroup wb */ +#endif struct list_head i_lru; /* inode LRU list */ struct list_head i_sb_list; union { diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 637ef626008e..662a953ea8ad 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -388,6 +388,10 @@ enum { OVER_LIMIT, }; +#ifdef CONFIG_CGROUP_WRITEBACK +struct list_head *mem_cgroup_cgwb_list(struct mem_cgroup *memcg); +#endif + struct sock; #if defined(CONFIG_INET) && defined(CONFIG_MEMCG_KMEM) void sock_update_memcg(struct sock *sk); diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 5ec7658cb984..4c9386c98ec1 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -368,6 +368,401 @@ static void wb_exit(struct bdi_writeback *wb) fprop_local_destroy_percpu(&wb->completions); } +#ifdef CONFIG_CGROUP_WRITEBACK + +#include + +/* + * cgwb_lock protects bdi->cgwb_tree, bdi->cgwb_congested_tree, + * blkcg->cgwb_list, and memcg->cgwb_list. bdi->cgwb_tree is also RCU + * protected. cgwb_release_wait is used to wait for the completion of cgwb + * releases from bdi destruction path. + */ +static DEFINE_SPINLOCK(cgwb_lock); +static DECLARE_WAIT_QUEUE_HEAD(cgwb_release_wait); + +/** + * wb_congested_get_create - get or create a wb_congested + * @bdi: associated bdi + * @blkcg_id: ID of the associated blkcg + * @gfp: allocation mask + * + * Look up the wb_congested for @blkcg_id on @bdi. If missing, create one. + * The returned wb_congested has its reference count incremented. Returns + * NULL on failure. + */ +struct bdi_writeback_congested * +wb_congested_get_create(struct backing_dev_info *bdi, int blkcg_id, gfp_t gfp) +{ + struct bdi_writeback_congested *new_congested = NULL, *congested; + struct rb_node **node, *parent; + unsigned long flags; + + if (blkcg_id == 1) + return &bdi->wb_congested; +retry: + spin_lock_irqsave(&cgwb_lock, flags); + + node = &bdi->cgwb_congested_tree.rb_node; + parent = NULL; + + while (*node != NULL) { + parent = *node; + congested = container_of(parent, struct bdi_writeback_congested, + rb_node); + if (congested->blkcg_id < blkcg_id) + node = &parent->rb_left; + else if (congested->blkcg_id > blkcg_id) + node = &parent->rb_right; + else + goto found; + } + + if (new_congested) { + /* !found and storage for new one already allocated, insert */ + congested = new_congested; + new_congested = NULL; + rb_link_node(&congested->rb_node, parent, node); + rb_insert_color(&congested->rb_node, &bdi->cgwb_congested_tree); + atomic_inc(&bdi->usage_cnt); + goto found; + } + + spin_unlock_irqrestore(&cgwb_lock, flags); + + /* allocate storage for new one and retry */ + new_congested = kzalloc(sizeof(*new_congested), gfp); + if (!new_congested) + return NULL; + + atomic_set(&new_congested->refcnt, 0); + new_congested->bdi = bdi; + new_congested->blkcg_id = blkcg_id; + goto retry; + +found: + atomic_inc(&congested->refcnt); + spin_unlock_irqrestore(&cgwb_lock, flags); + kfree(new_congested); + return congested; +} + +/** + * wb_congested_put - put a wb_congested + * @congested: wb_congested to put + * + * Put @congested and destroy it if the refcnt reaches zero. + */ +void wb_congested_put(struct bdi_writeback_congested *congested) +{ + struct backing_dev_info *bdi = congested->bdi; + unsigned long flags; + + if (congested->blkcg_id == 1) + return; + + local_irq_save(flags); + if (!atomic_dec_and_lock(&congested->refcnt, &cgwb_lock)) { + local_irq_restore(flags); + return; + } + + rb_erase(&congested->rb_node, &congested->bdi->cgwb_congested_tree); + spin_unlock_irqrestore(&cgwb_lock, flags); + kfree(congested); + + if (atomic_dec_and_test(&bdi->usage_cnt)) + wake_up_all(&cgwb_release_wait); +} + +static void cgwb_release_workfn(struct work_struct *work) +{ + struct bdi_writeback *wb = container_of(work, struct bdi_writeback, + release_work); + struct backing_dev_info *bdi = wb->bdi; + + wb_shutdown(wb); + + css_put(wb->memcg_css); + css_put(wb->blkcg_css); + wb_congested_put(wb->congested); + + percpu_ref_exit(&wb->refcnt); + wb_exit(wb); + kfree_rcu(wb, rcu); + + if (atomic_dec_and_test(&bdi->usage_cnt)) + wake_up_all(&cgwb_release_wait); +} + +static void cgwb_release(struct percpu_ref *refcnt) +{ + struct bdi_writeback *wb = container_of(refcnt, struct bdi_writeback, + refcnt); + schedule_work(&wb->release_work); +} + +static void cgwb_kill(struct bdi_writeback *wb) +{ + lockdep_assert_held(&cgwb_lock); + + WARN_ON(!radix_tree_delete(&wb->bdi->cgwb_tree, wb->memcg_css->id)); + list_del(&wb->memcg_node); + list_del(&wb->blkcg_node); + percpu_ref_kill(&wb->refcnt); +} + +static int cgwb_create(struct backing_dev_info *bdi, + struct cgroup_subsys_state *memcg_css, gfp_t gfp) +{ + struct mem_cgroup *memcg; + struct cgroup_subsys_state *blkcg_css; + struct blkcg *blkcg; + struct list_head *memcg_cgwb_list, *blkcg_cgwb_list; + struct bdi_writeback *wb; + unsigned long flags; + int ret = 0; + + memcg = mem_cgroup_from_css(memcg_css); + blkcg_css = cgroup_get_e_css(memcg_css->cgroup, &blkio_cgrp_subsys); + blkcg = css_to_blkcg(blkcg_css); + memcg_cgwb_list = mem_cgroup_cgwb_list(memcg); + blkcg_cgwb_list = &blkcg->cgwb_list; + + /* look up again under lock and discard on blkcg mismatch */ + spin_lock_irqsave(&cgwb_lock, flags); + wb = radix_tree_lookup(&bdi->cgwb_tree, memcg_css->id); + if (wb && wb->blkcg_css != blkcg_css) { + cgwb_kill(wb); + wb = NULL; + } + spin_unlock_irqrestore(&cgwb_lock, flags); + if (wb) + goto out_put; + + /* need to create a new one */ + wb = kmalloc(sizeof(*wb), gfp); + if (!wb) + return -ENOMEM; + + ret = wb_init(wb, bdi, gfp); + if (ret) + goto err_free; + + ret = percpu_ref_init(&wb->refcnt, cgwb_release, 0, gfp); + if (ret) + goto err_wb_exit; + + wb->congested = wb_congested_get_create(bdi, blkcg_css->id, gfp); + if (!wb->congested) + goto err_ref_exit; + + wb->memcg_css = memcg_css; + wb->blkcg_css = blkcg_css; + INIT_WORK(&wb->release_work, cgwb_release_workfn); + set_bit(WB_registered, &wb->state); + + /* + * The root wb determines the registered state of the whole bdi and + * memcg_cgwb_list and blkcg_cgwb_list's next pointers indicate + * whether they're still online. Don't link @wb if any is dead. + * See wb_memcg_offline() and wb_blkcg_offline(). + */ + ret = -ENODEV; + spin_lock_irqsave(&cgwb_lock, flags); + if (test_bit(WB_registered, &bdi->wb.state) && + blkcg_cgwb_list->next && memcg_cgwb_list->next) { + /* we might have raced another instance of this function */ + ret = radix_tree_insert(&bdi->cgwb_tree, memcg_css->id, wb); + if (!ret) { + atomic_inc(&bdi->usage_cnt); + list_add(&wb->memcg_node, memcg_cgwb_list); + list_add(&wb->blkcg_node, blkcg_cgwb_list); + css_get(memcg_css); + css_get(blkcg_css); + } + } + spin_unlock_irqrestore(&cgwb_lock, flags); + if (ret) { + if (ret == -EEXIST) + ret = 0; + goto err_put_congested; + } + goto out_put; + +err_put_congested: + wb_congested_put(wb->congested); +err_ref_exit: + percpu_ref_exit(&wb->refcnt); +err_wb_exit: + wb_exit(wb); +err_free: + kfree(wb); +out_put: + css_put(blkcg_css); + return ret; +} + +/** + * wb_get_create - get wb for a given memcg, create if necessary + * @bdi: target bdi + * @memcg_css: cgroup_subsys_state of the target memcg (must have positive ref) + * @gfp: allocation mask to use + * + * Try to get the wb for @memcg_css on @bdi. If it doesn't exist, try to + * create one. The returned wb has its refcount incremented. + * + * This function uses css_get() on @memcg_css and thus expects its refcnt + * to be positive on invocation. IOW, rcu_read_lock() protection on + * @memcg_css isn't enough. try_get it before calling this function. + * + * A wb is keyed by its associated memcg. As blkcg implicitly enables + * memcg on the default hierarchy, memcg association is guaranteed to be + * more specific (equal or descendant to the associated blkcg) and thus can + * identify both the memcg and blkcg associations. + * + * Because the blkcg associated with a memcg may change as blkcg is enabled + * and disabled closer to root in the hierarchy, each wb keeps track of + * both the memcg and blkcg associated with it and verifies the blkcg on + * each lookup. On mismatch, the existing wb is discarded and a new one is + * created. + */ +struct bdi_writeback *wb_get_create(struct backing_dev_info *bdi, + struct cgroup_subsys_state *memcg_css, + gfp_t gfp) +{ + struct bdi_writeback *wb; + + might_sleep_if(gfp & __GFP_WAIT); + + if (!memcg_css->parent) + return &bdi->wb; + + do { + rcu_read_lock(); + wb = radix_tree_lookup(&bdi->cgwb_tree, memcg_css->id); + if (wb) { + struct cgroup_subsys_state *blkcg_css; + + /* see whether the blkcg association has changed */ + blkcg_css = cgroup_get_e_css(memcg_css->cgroup, + &blkio_cgrp_subsys); + if (unlikely(wb->blkcg_css != blkcg_css || + !wb_tryget(wb))) + wb = NULL; + css_put(blkcg_css); + } + rcu_read_unlock(); + } while (!wb && !cgwb_create(bdi, memcg_css, gfp)); + + return wb; +} + +void __inode_attach_wb(struct inode *inode, struct page *page) +{ + struct backing_dev_info *bdi = inode_to_bdi(inode); + struct bdi_writeback *wb = NULL; + + if (inode_cgwb_enabled(inode)) { + struct cgroup_subsys_state *memcg_css; + + if (page) { + memcg_css = mem_cgroup_css_from_page(page); + wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); + } else { + /* must pin memcg_css, see wb_get_create() */ + memcg_css = task_get_css(current, memory_cgrp_id); + wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); + css_put(memcg_css); + } + } + + if (!wb) + wb = &bdi->wb; + + /* + * There may be multiple instances of this function racing to + * update the same inode. Use cmpxchg() to tell the winner. + */ + if (unlikely(cmpxchg(&inode->i_wb, NULL, wb))) + wb_put(wb); +} + +static void cgwb_bdi_init(struct backing_dev_info *bdi) +{ + bdi->wb.memcg_css = mem_cgroup_root_css; + bdi->wb.blkcg_css = blkcg_root_css; + bdi->wb_congested.blkcg_id = 1; + INIT_RADIX_TREE(&bdi->cgwb_tree, GFP_ATOMIC); + bdi->cgwb_congested_tree = RB_ROOT; + atomic_set(&bdi->usage_cnt, 1); +} + +static void cgwb_bdi_destroy(struct backing_dev_info *bdi) +{ + struct radix_tree_iter iter; + void **slot; + + WARN_ON(test_bit(WB_registered, &bdi->wb.state)); + + spin_lock_irq(&cgwb_lock); + radix_tree_for_each_slot(slot, &bdi->cgwb_tree, &iter, 0) + cgwb_kill(*slot); + spin_unlock_irq(&cgwb_lock); + + /* + * All cgwb's and their congested states must be shutdown and + * released before returning. Drain the usage counter to wait for + * all cgwb's and cgwb_congested's ever created on @bdi. + */ + atomic_dec(&bdi->usage_cnt); + wait_event(cgwb_release_wait, !atomic_read(&bdi->usage_cnt)); +} + +/** + * wb_memcg_offline - kill all wb's associated with a memcg being offlined + * @memcg: memcg being offlined + * + * Also prevents creation of any new wb's associated with @memcg. + */ +void wb_memcg_offline(struct mem_cgroup *memcg) +{ + LIST_HEAD(to_destroy); + struct list_head *memcg_cgwb_list = mem_cgroup_cgwb_list(memcg); + struct bdi_writeback *wb, *next; + + spin_lock_irq(&cgwb_lock); + list_for_each_entry_safe(wb, next, memcg_cgwb_list, memcg_node) + cgwb_kill(wb); + memcg_cgwb_list->next = NULL; /* prevent new wb's */ + spin_unlock_irq(&cgwb_lock); +} + +/** + * wb_blkcg_offline - kill all wb's associated with a blkcg being offlined + * @blkcg: blkcg being offlined + * + * Also prevents creation of any new wb's associated with @blkcg. + */ +void wb_blkcg_offline(struct blkcg *blkcg) +{ + LIST_HEAD(to_destroy); + struct bdi_writeback *wb, *next; + + spin_lock_irq(&cgwb_lock); + list_for_each_entry_safe(wb, next, &blkcg->cgwb_list, blkcg_node) + cgwb_kill(wb); + blkcg->cgwb_list.next = NULL; /* prevent new wb's */ + spin_unlock_irq(&cgwb_lock); +} + +#else /* CONFIG_CGROUP_WRITEBACK */ + +static void cgwb_bdi_init(struct backing_dev_info *bdi) { } +static void cgwb_bdi_destroy(struct backing_dev_info *bdi) { } + +#endif /* CONFIG_CGROUP_WRITEBACK */ + int bdi_init(struct backing_dev_info *bdi) { int err; @@ -386,6 +781,7 @@ int bdi_init(struct backing_dev_info *bdi) bdi->wb_congested.state = 0; bdi->wb.congested = &bdi->wb_congested; + cgwb_bdi_init(bdi); return 0; } EXPORT_SYMBOL(bdi_init); @@ -459,6 +855,7 @@ void bdi_destroy(struct backing_dev_info *bdi) /* make sure nobody finds us on the bdi_list anymore */ bdi_remove_from_list(bdi); wb_shutdown(&bdi->wb); + cgwb_bdi_destroy(bdi); if (bdi->dev) { bdi_debug_unregister(bdi); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 5c270a03729c..49e5aa6abca6 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -348,6 +348,10 @@ struct mem_cgroup { atomic_t numainfo_updating; #endif +#ifdef CONFIG_CGROUP_WRITEBACK + struct list_head cgwb_list; +#endif + /* List of events which userspace want to receive */ struct list_head event_list; spinlock_t event_list_lock; @@ -4030,6 +4034,15 @@ static void memcg_destroy_kmem(struct mem_cgroup *memcg) } #endif +#ifdef CONFIG_CGROUP_WRITEBACK + +struct list_head *mem_cgroup_cgwb_list(struct mem_cgroup *memcg) +{ + return &memcg->cgwb_list; +} + +#endif /* CONFIG_CGROUP_WRITEBACK */ + /* * DO NOT USE IN NEW FILES. * @@ -4494,7 +4507,9 @@ mem_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) #ifdef CONFIG_MEMCG_KMEM memcg->kmemcg_id = -1; #endif - +#ifdef CONFIG_CGROUP_WRITEBACK + INIT_LIST_HEAD(&memcg->cgwb_list); +#endif return &memcg->css; free_out: @@ -4582,6 +4597,8 @@ static void mem_cgroup_css_offline(struct cgroup_subsys_state *css) vmpressure_cleanup(&memcg->vmpressure); memcg_deactivate_kmem(memcg); + + wb_memcg_offline(memcg); } static void mem_cgroup_css_free(struct cgroup_subsys_state *css) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 78ef5511a198..9b95cf80b407 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -2097,16 +2097,21 @@ int __set_page_dirty_no_writeback(struct page *page) void account_page_dirtied(struct page *page, struct address_space *mapping, struct mem_cgroup *memcg) { + struct inode *inode = mapping->host; + trace_writeback_dirty_page(page, mapping); if (mapping_cap_account_dirty(mapping)) { - struct backing_dev_info *bdi = inode_to_bdi(mapping->host); + struct bdi_writeback *wb; + + inode_attach_wb(inode, page); + wb = inode_to_wb(inode); mem_cgroup_inc_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); __inc_zone_page_state(page, NR_FILE_DIRTY); __inc_zone_page_state(page, NR_DIRTIED); - __inc_wb_stat(&bdi->wb, WB_RECLAIMABLE); - __inc_wb_stat(&bdi->wb, WB_DIRTIED); + __inc_wb_stat(wb, WB_RECLAIMABLE); + __inc_wb_stat(wb, WB_DIRTIED); task_io_account_write(PAGE_CACHE_SIZE); current->nr_dirtied++; this_cpu_inc(bdp_ratelimits); From ce7acfeaf0363c8b75810908448f61af04d38f91 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:38 -0400 Subject: [PATCH 24/85] writeback, blkcg: associate each blkcg_gq with the corresponding bdi_writeback_congested A blkg (blkcg_gq) can be congested and decongested independently from other blkgs on the same request_queue. Accordingly, for cgroup writeback support, the congestion status at bdi (backing_dev_info) should be split and updated separately from matching blkg's. This patch prepares by adding blkg->wb_congested and associating a blkg with its matching per-blkcg bdi_writeback_congested on creation. v2: Updated to associate bdi_writeback_congested instead of bdi_writeback. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Vivek Goyal Signed-off-by: Jens Axboe --- block/blk-cgroup.c | 17 +++++++++++++++-- include/linux/blk-cgroup.h | 6 ++++++ 2 files changed, 21 insertions(+), 2 deletions(-) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 979cfdbb94e0..31610ae0ebff 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -182,6 +182,7 @@ static struct blkcg_gq *blkg_create(struct blkcg *blkcg, struct blkcg_gq *new_blkg) { struct blkcg_gq *blkg; + struct bdi_writeback_congested *wb_congested; int i, ret; WARN_ON_ONCE(!rcu_read_lock_held()); @@ -193,22 +194,30 @@ static struct blkcg_gq *blkg_create(struct blkcg *blkcg, goto err_free_blkg; } + wb_congested = wb_congested_get_create(&q->backing_dev_info, + blkcg->css.id, GFP_ATOMIC); + if (!wb_congested) { + ret = -ENOMEM; + goto err_put_css; + } + /* allocate */ if (!new_blkg) { new_blkg = blkg_alloc(blkcg, q, GFP_ATOMIC); if (unlikely(!new_blkg)) { ret = -ENOMEM; - goto err_put_css; + goto err_put_congested; } } blkg = new_blkg; + blkg->wb_congested = wb_congested; /* link parent */ if (blkcg_parent(blkcg)) { blkg->parent = __blkg_lookup(blkcg_parent(blkcg), q, false); if (WARN_ON_ONCE(!blkg->parent)) { ret = -EINVAL; - goto err_put_css; + goto err_put_congested; } blkg_get(blkg->parent); } @@ -245,6 +254,8 @@ static struct blkcg_gq *blkg_create(struct blkcg *blkcg, blkg_put(blkg); return ERR_PTR(ret); +err_put_congested: + wb_congested_put(wb_congested); err_put_css: css_put(&blkcg->css); err_free_blkg: @@ -391,6 +402,8 @@ void __blkg_release_rcu(struct rcu_head *rcu_head) if (blkg->parent) blkg_put(blkg->parent); + wb_congested_put(blkg->wb_congested); + blkg_free(blkg); } EXPORT_SYMBOL_GPL(__blkg_release_rcu); diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h index 3033eb173eb4..07a32b813ed8 100644 --- a/include/linux/blk-cgroup.h +++ b/include/linux/blk-cgroup.h @@ -99,6 +99,12 @@ struct blkcg_gq { struct hlist_node blkcg_node; struct blkcg *blkcg; + /* + * Each blkg gets congested separately and the congestion state is + * propagated to the matching bdi_writeback_congested. + */ + struct bdi_writeback_congested *wb_congested; + /* all non-root blkcg_gq's are guaranteed to have access to parent */ struct blkcg_gq *parent; From 910181343774cd5fed95900d9fd2cb4ff7758162 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:39 -0400 Subject: [PATCH 25/85] writeback: attribute stats to the matching per-cgroup bdi_writeback Until now, all WB_* stats were accounted against the root wb (bdi_writeback), now that multiple wb (bdi_writeback) support is in place, let's attributes the stats to the respective per-cgroup wb's. As no filesystem has FS_CGROUP_WRITEBACK yet, this doesn't lead to visible behavior differences. v2: Updated for per-inode wb association. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- mm/page-writeback.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 9b95cf80b407..4d0a9da2de6b 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -2130,7 +2130,7 @@ void account_page_cleaned(struct page *page, struct address_space *mapping, if (mapping_cap_account_dirty(mapping)) { mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); dec_zone_page_state(page, NR_FILE_DIRTY); - dec_wb_stat(&inode_to_bdi(mapping->host)->wb, WB_RECLAIMABLE); + dec_wb_stat(inode_to_wb(mapping->host), WB_RECLAIMABLE); task_io_account_cancelled_write(PAGE_CACHE_SIZE); } } @@ -2191,10 +2191,13 @@ EXPORT_SYMBOL(__set_page_dirty_nobuffers); void account_page_redirty(struct page *page) { struct address_space *mapping = page->mapping; + if (mapping && mapping_cap_account_dirty(mapping)) { + struct bdi_writeback *wb = inode_to_wb(mapping->host); + current->nr_dirtied--; dec_zone_page_state(page, NR_DIRTIED); - dec_wb_stat(&inode_to_bdi(mapping->host)->wb, WB_DIRTIED); + dec_wb_stat(wb, WB_DIRTIED); } } EXPORT_SYMBOL(account_page_redirty); @@ -2373,8 +2376,7 @@ int clear_page_dirty_for_io(struct page *page) if (TestClearPageDirty(page)) { mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); dec_zone_page_state(page, NR_FILE_DIRTY); - dec_wb_stat(&inode_to_bdi(mapping->host)->wb, - WB_RECLAIMABLE); + dec_wb_stat(inode_to_wb(mapping->host), WB_RECLAIMABLE); ret = 1; } mem_cgroup_end_page_stat(memcg); @@ -2392,7 +2394,8 @@ int test_clear_page_writeback(struct page *page) memcg = mem_cgroup_begin_page_stat(page); if (mapping) { - struct backing_dev_info *bdi = inode_to_bdi(mapping->host); + struct inode *inode = mapping->host; + struct backing_dev_info *bdi = inode_to_bdi(inode); unsigned long flags; spin_lock_irqsave(&mapping->tree_lock, flags); @@ -2402,8 +2405,10 @@ int test_clear_page_writeback(struct page *page) page_index(page), PAGECACHE_TAG_WRITEBACK); if (bdi_cap_account_writeback(bdi)) { - __dec_wb_stat(&bdi->wb, WB_WRITEBACK); - __wb_writeout_inc(&bdi->wb); + struct bdi_writeback *wb = inode_to_wb(inode); + + __dec_wb_stat(wb, WB_WRITEBACK); + __wb_writeout_inc(wb); } } spin_unlock_irqrestore(&mapping->tree_lock, flags); @@ -2427,7 +2432,8 @@ int __test_set_page_writeback(struct page *page, bool keep_write) memcg = mem_cgroup_begin_page_stat(page); if (mapping) { - struct backing_dev_info *bdi = inode_to_bdi(mapping->host); + struct inode *inode = mapping->host; + struct backing_dev_info *bdi = inode_to_bdi(inode); unsigned long flags; spin_lock_irqsave(&mapping->tree_lock, flags); @@ -2437,7 +2443,7 @@ int __test_set_page_writeback(struct page *page, bool keep_write) page_index(page), PAGECACHE_TAG_WRITEBACK); if (bdi_cap_account_writeback(bdi)) - __inc_wb_stat(&bdi->wb, WB_WRITEBACK); + __inc_wb_stat(inode_to_wb(inode), WB_WRITEBACK); } if (!PageDirty(page)) radix_tree_tag_clear(&mapping->page_tree, From dfb8ae567835425d27db8acc6c9fc5db88d38e2b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:40 -0400 Subject: [PATCH 26/85] writeback: let balance_dirty_pages() work on the matching cgroup bdi_writeback Currently, balance_dirty_pages() always work on bdi->wb. This patch updates it to work on the wb (bdi_writeback) matching memcg and blkcg of the current task as that's what the inode is being dirtied against. balance_dirty_pages_ratelimited() now pins the current wb and passes it to balance_dirty_pages(). As no filesystem has FS_CGROUP_WRITEBACK yet, this doesn't lead to visible behavior differences. v2: Updated for per-inode wb association. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- mm/page-writeback.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 4d0a9da2de6b..e31dea94d116 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1337,6 +1337,7 @@ static inline void wb_dirty_limits(struct bdi_writeback *wb, * perform some writeout. */ static void balance_dirty_pages(struct address_space *mapping, + struct bdi_writeback *wb, unsigned long pages_dirtied) { unsigned long nr_reclaimable; /* = file_dirty + unstable_nfs */ @@ -1352,8 +1353,7 @@ static void balance_dirty_pages(struct address_space *mapping, unsigned long task_ratelimit; unsigned long dirty_ratelimit; unsigned long pos_ratio; - struct backing_dev_info *bdi = inode_to_bdi(mapping->host); - struct bdi_writeback *wb = &bdi->wb; + struct backing_dev_info *bdi = wb->bdi; bool strictlimit = bdi->capabilities & BDI_CAP_STRICTLIMIT; unsigned long start_time = jiffies; @@ -1575,14 +1575,20 @@ DEFINE_PER_CPU(int, dirty_throttle_leaks) = 0; */ void balance_dirty_pages_ratelimited(struct address_space *mapping) { - struct backing_dev_info *bdi = inode_to_bdi(mapping->host); - struct bdi_writeback *wb = &bdi->wb; + struct inode *inode = mapping->host; + struct backing_dev_info *bdi = inode_to_bdi(inode); + struct bdi_writeback *wb = NULL; int ratelimit; int *p; if (!bdi_cap_account_dirty(bdi)) return; + if (inode_cgwb_enabled(inode)) + wb = wb_get_create_current(bdi, GFP_KERNEL); + if (!wb) + wb = &bdi->wb; + ratelimit = current->nr_dirtied_pause; if (wb->dirty_exceeded) ratelimit = min(ratelimit, 32 >> (PAGE_SHIFT - 10)); @@ -1616,7 +1622,9 @@ void balance_dirty_pages_ratelimited(struct address_space *mapping) preempt_enable(); if (unlikely(current->nr_dirtied >= ratelimit)) - balance_dirty_pages(mapping, current->nr_dirtied); + balance_dirty_pages(mapping, wb, current->nr_dirtied); + + wb_put(wb); } EXPORT_SYMBOL(balance_dirty_pages_ratelimited); From ec8a6f2643923ee5b74d24fa8d134240379f436b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:41 -0400 Subject: [PATCH 27/85] writeback: make congestion functions per bdi_writeback Currently, all congestion functions take bdi (backing_dev_info) and always operate on the root wb (bdi->wb) and the congestion state from the block layer is propagated only for the root blkcg. This patch introduces {set|clear}_wb_congested() and wb_congested() which take a bdi_writeback_congested and bdi_writeback respectively. The bdi counteparts are now wrappers invoking the wb based functions on @bdi->wb. While converting clear_bdi_congested() to clear_wb_congested(), the local variable declaration order between @wqh and @bit is swapped for cosmetic reason. This patch just adds the new wb based functions. The following patches will apply them. v2: Updated for bdi_writeback_congested. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Signed-off-by: Jens Axboe --- include/linux/backing-dev-defs.h | 14 ++++++++-- include/linux/backing-dev.h | 45 ++++++++++++++++++-------------- mm/backing-dev.c | 22 ++++++++-------- 3 files changed, 49 insertions(+), 32 deletions(-) diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index a1e9c407a59a..eb386766b5f3 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -163,7 +163,17 @@ enum { BLK_RW_SYNC = 1, }; -void clear_bdi_congested(struct backing_dev_info *bdi, int sync); -void set_bdi_congested(struct backing_dev_info *bdi, int sync); +void clear_wb_congested(struct bdi_writeback_congested *congested, int sync); +void set_wb_congested(struct bdi_writeback_congested *congested, int sync); + +static inline void clear_bdi_congested(struct backing_dev_info *bdi, int sync) +{ + clear_wb_congested(bdi->wb.congested, sync); +} + +static inline void set_bdi_congested(struct backing_dev_info *bdi, int sync) +{ + set_wb_congested(bdi->wb.congested, sync); +} #endif /* __LINUX_BACKING_DEV_DEFS_H */ diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 8ae59df2e3d1..2c498a2a8268 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -167,27 +167,13 @@ static inline struct backing_dev_info *inode_to_bdi(struct inode *inode) return sb->s_bdi; } -static inline int bdi_congested(struct backing_dev_info *bdi, int bdi_bits) +static inline int wb_congested(struct bdi_writeback *wb, int cong_bits) { + struct backing_dev_info *bdi = wb->bdi; + if (bdi->congested_fn) - return bdi->congested_fn(bdi->congested_data, bdi_bits); - return (bdi->wb.congested->state & bdi_bits); -} - -static inline int bdi_read_congested(struct backing_dev_info *bdi) -{ - return bdi_congested(bdi, 1 << WB_sync_congested); -} - -static inline int bdi_write_congested(struct backing_dev_info *bdi) -{ - return bdi_congested(bdi, 1 << WB_async_congested); -} - -static inline int bdi_rw_congested(struct backing_dev_info *bdi) -{ - return bdi_congested(bdi, (1 << WB_sync_congested) | - (1 << WB_async_congested)); + return bdi->congested_fn(bdi->congested_data, cong_bits); + return wb->congested->state & cong_bits; } long congestion_wait(int sync, long timeout); @@ -454,4 +440,25 @@ static inline void wb_blkcg_offline(struct blkcg *blkcg) #endif /* CONFIG_CGROUP_WRITEBACK */ +static inline int bdi_congested(struct backing_dev_info *bdi, int cong_bits) +{ + return wb_congested(&bdi->wb, cong_bits); +} + +static inline int bdi_read_congested(struct backing_dev_info *bdi) +{ + return bdi_congested(bdi, 1 << WB_sync_congested); +} + +static inline int bdi_write_congested(struct backing_dev_info *bdi) +{ + return bdi_congested(bdi, 1 << WB_async_congested); +} + +static inline int bdi_rw_congested(struct backing_dev_info *bdi) +{ + return bdi_congested(bdi, (1 << WB_sync_congested) | + (1 << WB_async_congested)); +} + #endif /* _LINUX_BACKING_DEV_H */ diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 4c9386c98ec1..5029c4ad2f36 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -896,31 +896,31 @@ static wait_queue_head_t congestion_wqh[2] = { __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) }; -static atomic_t nr_bdi_congested[2]; +static atomic_t nr_wb_congested[2]; -void clear_bdi_congested(struct backing_dev_info *bdi, int sync) +void clear_wb_congested(struct bdi_writeback_congested *congested, int sync) { - enum wb_state bit; wait_queue_head_t *wqh = &congestion_wqh[sync]; + enum wb_state bit; bit = sync ? WB_sync_congested : WB_async_congested; - if (test_and_clear_bit(bit, &bdi->wb.congested->state)) - atomic_dec(&nr_bdi_congested[sync]); + if (test_and_clear_bit(bit, &congested->state)) + atomic_dec(&nr_wb_congested[sync]); smp_mb__after_atomic(); if (waitqueue_active(wqh)) wake_up(wqh); } -EXPORT_SYMBOL(clear_bdi_congested); +EXPORT_SYMBOL(clear_wb_congested); -void set_bdi_congested(struct backing_dev_info *bdi, int sync) +void set_wb_congested(struct bdi_writeback_congested *congested, int sync) { enum wb_state bit; bit = sync ? WB_sync_congested : WB_async_congested; - if (!test_and_set_bit(bit, &bdi->wb.congested->state)) - atomic_inc(&nr_bdi_congested[sync]); + if (!test_and_set_bit(bit, &congested->state)) + atomic_inc(&nr_wb_congested[sync]); } -EXPORT_SYMBOL(set_bdi_congested); +EXPORT_SYMBOL(set_wb_congested); /** * congestion_wait - wait for a backing_dev to become uncongested @@ -979,7 +979,7 @@ long wait_iff_congested(struct zone *zone, int sync, long timeout) * encountered in the current zone, yield if necessary instead * of sleeping on the congestion queue */ - if (atomic_read(&nr_bdi_congested[sync]) == 0 || + if (atomic_read(&nr_wb_congested[sync]) == 0 || !test_bit(ZONE_CONGESTED, &zone->flags)) { cond_resched(); From d40f75a06dd675808eed385d490ba9468200b23f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:42 -0400 Subject: [PATCH 28/85] writeback, blkcg: restructure blk_{set|clear}_queue_congested() blk_{set|clear}_queue_congested() take @q and set or clear, respectively, the congestion state of its bdi's root wb. Because bdi used to be able to handle congestion state only on the root wb, the callers of those functions tested whether the congestion is on the root blkcg and skipped if not. This is cumbersome and makes implementation of per cgroup bdi_writeback congestion state propagation difficult. This patch renames blk_{set|clear}_queue_congested() to blk_{set|clear}_congested(), and makes them take request_list instead of request_queue and test whether the specified request_list is the root one before updating bdi_writeback congestion state. This makes the tests in the callers unnecessary and simplifies them. As there are no external users of these functions, the definitions are moved from include/linux/blkdev.h to block/blk-core.c. This patch doesn't introduce any noticeable behavior difference. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Vivek Goyal Signed-off-by: Jens Axboe --- block/blk-core.c | 64 +++++++++++++++++++++++++----------------- include/linux/blkdev.h | 19 ------------- 2 files changed, 38 insertions(+), 45 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index c114a61590bc..f46a43ef1a94 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -63,6 +63,28 @@ struct kmem_cache *blk_requestq_cachep; */ static struct workqueue_struct *kblockd_workqueue; +static void blk_clear_congested(struct request_list *rl, int sync) +{ + if (rl != &rl->q->root_rl) + return; +#ifdef CONFIG_CGROUP_WRITEBACK + clear_wb_congested(rl->blkg->wb_congested, sync); +#else + clear_wb_congested(rl->q->backing_dev_info.wb.congested, sync); +#endif +} + +static void blk_set_congested(struct request_list *rl, int sync) +{ + if (rl != &rl->q->root_rl) + return; +#ifdef CONFIG_CGROUP_WRITEBACK + set_wb_congested(rl->blkg->wb_congested, sync); +#else + set_wb_congested(rl->q->backing_dev_info.wb.congested, sync); +#endif +} + void blk_queue_congestion_threshold(struct request_queue *q) { int nr; @@ -842,13 +864,8 @@ static void __freed_request(struct request_list *rl, int sync) { struct request_queue *q = rl->q; - /* - * bdi isn't aware of blkcg yet. As all async IOs end up root - * blkcg anyway, just use root blkcg state. - */ - if (rl == &q->root_rl && - rl->count[sync] < queue_congestion_off_threshold(q)) - blk_clear_queue_congested(q, sync); + if (rl->count[sync] < queue_congestion_off_threshold(q)) + blk_clear_congested(rl, sync); if (rl->count[sync] + 1 <= q->nr_requests) { if (waitqueue_active(&rl->wait[sync])) @@ -881,25 +898,25 @@ static void freed_request(struct request_list *rl, unsigned int flags) int blk_update_nr_requests(struct request_queue *q, unsigned int nr) { struct request_list *rl; + int on_thresh, off_thresh; spin_lock_irq(q->queue_lock); q->nr_requests = nr; blk_queue_congestion_threshold(q); - - /* congestion isn't cgroup aware and follows root blkcg for now */ - rl = &q->root_rl; - - if (rl->count[BLK_RW_SYNC] >= queue_congestion_on_threshold(q)) - blk_set_queue_congested(q, BLK_RW_SYNC); - else if (rl->count[BLK_RW_SYNC] < queue_congestion_off_threshold(q)) - blk_clear_queue_congested(q, BLK_RW_SYNC); - - if (rl->count[BLK_RW_ASYNC] >= queue_congestion_on_threshold(q)) - blk_set_queue_congested(q, BLK_RW_ASYNC); - else if (rl->count[BLK_RW_ASYNC] < queue_congestion_off_threshold(q)) - blk_clear_queue_congested(q, BLK_RW_ASYNC); + on_thresh = queue_congestion_on_threshold(q); + off_thresh = queue_congestion_off_threshold(q); blk_queue_for_each_rl(rl, q) { + if (rl->count[BLK_RW_SYNC] >= on_thresh) + blk_set_congested(rl, BLK_RW_SYNC); + else if (rl->count[BLK_RW_SYNC] < off_thresh) + blk_clear_congested(rl, BLK_RW_SYNC); + + if (rl->count[BLK_RW_ASYNC] >= on_thresh) + blk_set_congested(rl, BLK_RW_ASYNC); + else if (rl->count[BLK_RW_ASYNC] < off_thresh) + blk_clear_congested(rl, BLK_RW_ASYNC); + if (rl->count[BLK_RW_SYNC] >= q->nr_requests) { blk_set_rl_full(rl, BLK_RW_SYNC); } else { @@ -1009,12 +1026,7 @@ static struct request *__get_request(struct request_list *rl, int rw_flags, } } } - /* - * bdi isn't aware of blkcg yet. As all async IOs end up - * root blkcg anyway, just use root blkcg state. - */ - if (rl == &q->root_rl) - blk_set_queue_congested(q, is_sync); + blk_set_congested(rl, is_sync); } /* diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 60d2726a6b62..ab4a27852f1b 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -790,25 +790,6 @@ extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t, extern void blk_queue_bio(struct request_queue *q, struct bio *bio); -/* - * A queue has just exitted congestion. Note this in the global counter of - * congested queues, and wake up anyone who was waiting for requests to be - * put back. - */ -static inline void blk_clear_queue_congested(struct request_queue *q, int sync) -{ - clear_bdi_congested(&q->backing_dev_info, sync); -} - -/* - * A queue has just entered congestion. Flag that in the queue's VM-visible - * state flags and increment the global gounter of congested queues. - */ -static inline void blk_set_queue_congested(struct request_queue *q, int sync) -{ - set_bdi_congested(&q->backing_dev_info, sync); -} - extern void blk_start_queue(struct request_queue *q); extern void blk_stop_queue(struct request_queue *q); extern void blk_sync_queue(struct request_queue *q); From 482cf79cdf6669667a914ffd4cbc57a762b55fef Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:43 -0400 Subject: [PATCH 29/85] writeback, blkcg: propagate non-root blkcg congestion state Now that bdi layer can handle per-blkcg bdi_writeback_congested state, blk_{set|clear}_congested() can propagate non-root blkcg congestion state to them. This can be easily achieved by disabling the root_rl tests in blk_{set|clear}_congested(). Note that we still need those tests when !CONFIG_CGROUP_WRITEBACK as otherwise we'll end up flipping root blkcg wb's congestion state for events happening on other blkcgs. v2: Updated for bdi_writeback_congested. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Vivek Goyal Signed-off-by: Jens Axboe --- block/blk-core.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index f46a43ef1a94..a4a2dbe46fe3 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -65,23 +65,26 @@ static struct workqueue_struct *kblockd_workqueue; static void blk_clear_congested(struct request_list *rl, int sync) { - if (rl != &rl->q->root_rl) - return; #ifdef CONFIG_CGROUP_WRITEBACK clear_wb_congested(rl->blkg->wb_congested, sync); #else - clear_wb_congested(rl->q->backing_dev_info.wb.congested, sync); + /* + * If !CGROUP_WRITEBACK, all blkg's map to bdi->wb and we shouldn't + * flip its congestion state for events on other blkcgs. + */ + if (rl == &rl->q->root_rl) + clear_wb_congested(rl->q->backing_dev_info.wb.congested, sync); #endif } static void blk_set_congested(struct request_list *rl, int sync) { - if (rl != &rl->q->root_rl) - return; #ifdef CONFIG_CGROUP_WRITEBACK set_wb_congested(rl->blkg->wb_congested, sync); #else - set_wb_congested(rl->q->backing_dev_info.wb.congested, sync); + /* see blk_clear_congested() */ + if (rl == &rl->q->root_rl) + set_wb_congested(rl->q->backing_dev_info.wb.congested, sync); #endif } From 703c270887bb5106c4c46a00cc7477d30d5e04f5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:44 -0400 Subject: [PATCH 30/85] writeback: implement and use inode_congested() In several places, bdi_congested() and its wrappers are used to determine whether more IOs should be issued. With cgroup writeback support, this question can't be answered solely based on the bdi (backing_dev_info). It's dependent on whether the filesystem and bdi support cgroup writeback and the blkcg the inode is associated with. This patch implements inode_congested() and its wrappers which take @inode and determines the congestion state considering cgroup writeback. The new functions replace bdi_*congested() calls in places where the query is about specific inode and task. There are several filesystem users which also fit this criteria but they should be updated when each filesystem implements cgroup writeback support. v2: Now that a given inode is associated with only one wb, congestion state can be determined independent from the asking task. Drop @task. Spotted by Vivek. Also, converted to take @inode instead of @mapping and renamed to inode_congested(). Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Vivek Goyal Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 29 +++++++++++++++++++++++++++++ include/linux/backing-dev.h | 22 ++++++++++++++++++++++ mm/fadvise.c | 2 +- mm/readahead.c | 2 +- mm/vmscan.c | 11 +++++------ 5 files changed, 58 insertions(+), 8 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 99a2440c5588..7ec491b1be04 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -142,6 +142,35 @@ static void __wb_start_writeback(struct bdi_writeback *wb, long nr_pages, wb_queue_work(wb, work); } +#ifdef CONFIG_CGROUP_WRITEBACK + +/** + * inode_congested - test whether an inode is congested + * @inode: inode to test for congestion + * @cong_bits: mask of WB_[a]sync_congested bits to test + * + * Tests whether @inode is congested. @cong_bits is the mask of congestion + * bits to test and the return value is the mask of set bits. + * + * If cgroup writeback is enabled for @inode, the congestion state is + * determined by whether the cgwb (cgroup bdi_writeback) for the blkcg + * associated with @inode is congested; otherwise, the root wb's congestion + * state is used. + */ +int inode_congested(struct inode *inode, int cong_bits) +{ + if (inode) { + struct bdi_writeback *wb = inode_to_wb(inode); + if (wb) + return wb_congested(wb, cong_bits); + } + + return wb_congested(&inode_to_bdi(inode)->wb, cong_bits); +} +EXPORT_SYMBOL_GPL(inode_congested); + +#endif /* CONFIG_CGROUP_WRITEBACK */ + /** * bdi_start_writeback - start writeback * @bdi: the backing device to write from diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 2c498a2a8268..6f0882105f95 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -230,6 +230,7 @@ struct bdi_writeback *wb_get_create(struct backing_dev_info *bdi, void __inode_attach_wb(struct inode *inode, struct page *page); void wb_memcg_offline(struct mem_cgroup *memcg); void wb_blkcg_offline(struct blkcg *blkcg); +int inode_congested(struct inode *inode, int cong_bits); /** * inode_cgwb_enabled - test whether cgroup writeback is enabled on an inode @@ -438,8 +439,29 @@ static inline void wb_blkcg_offline(struct blkcg *blkcg) { } +static inline int inode_congested(struct inode *inode, int cong_bits) +{ + return wb_congested(&inode_to_bdi(inode)->wb, cong_bits); +} + #endif /* CONFIG_CGROUP_WRITEBACK */ +static inline int inode_read_congested(struct inode *inode) +{ + return inode_congested(inode, 1 << WB_sync_congested); +} + +static inline int inode_write_congested(struct inode *inode) +{ + return inode_congested(inode, 1 << WB_async_congested); +} + +static inline int inode_rw_congested(struct inode *inode) +{ + return inode_congested(inode, (1 << WB_sync_congested) | + (1 << WB_async_congested)); +} + static inline int bdi_congested(struct backing_dev_info *bdi, int cong_bits) { return wb_congested(&bdi->wb, cong_bits); diff --git a/mm/fadvise.c b/mm/fadvise.c index 4a3907cf79f8..b8a5bc66b0c0 100644 --- a/mm/fadvise.c +++ b/mm/fadvise.c @@ -115,7 +115,7 @@ SYSCALL_DEFINE4(fadvise64_64, int, fd, loff_t, offset, loff_t, len, int, advice) case POSIX_FADV_NOREUSE: break; case POSIX_FADV_DONTNEED: - if (!bdi_write_congested(bdi)) + if (!inode_write_congested(mapping->host)) __filemap_fdatawrite_range(mapping, offset, endbyte, WB_SYNC_NONE); diff --git a/mm/readahead.c b/mm/readahead.c index 935675844b2e..60cd846a9a44 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -541,7 +541,7 @@ page_cache_async_readahead(struct address_space *mapping, /* * Defer asynchronous read-ahead on IO congestion. */ - if (bdi_read_congested(inode_to_bdi(mapping->host))) + if (inode_read_congested(mapping->host)) return; /* do read-ahead */ diff --git a/mm/vmscan.c b/mm/vmscan.c index 7582f9fcda92..f46339870147 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -452,14 +452,13 @@ static inline int is_page_cache_freeable(struct page *page) return page_count(page) - page_has_private(page) == 2; } -static int may_write_to_queue(struct backing_dev_info *bdi, - struct scan_control *sc) +static int may_write_to_inode(struct inode *inode, struct scan_control *sc) { if (current->flags & PF_SWAPWRITE) return 1; - if (!bdi_write_congested(bdi)) + if (!inode_write_congested(inode)) return 1; - if (bdi == current->backing_dev_info) + if (inode_to_bdi(inode) == current->backing_dev_info) return 1; return 0; } @@ -538,7 +537,7 @@ static pageout_t pageout(struct page *page, struct address_space *mapping, } if (mapping->a_ops->writepage == NULL) return PAGE_ACTIVATE; - if (!may_write_to_queue(inode_to_bdi(mapping->host), sc)) + if (!may_write_to_inode(mapping->host, sc)) return PAGE_KEEP; if (clear_page_dirty_for_io(page)) { @@ -924,7 +923,7 @@ static unsigned long shrink_page_list(struct list_head *page_list, */ mapping = page_mapping(page); if (((dirty || writeback) && mapping && - bdi_write_congested(inode_to_bdi(mapping->host))) || + inode_write_congested(mapping->host)) || (writeback && PageReclaim(page))) nr_congested++; From d6c10f1fc8626dc55946f4768ae322b4c57b07dd Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:45 -0400 Subject: [PATCH 31/85] writeback: implement WB_has_dirty_io wb_state flag Currently, wb_has_dirty_io() determines whether a wb (bdi_writeback) has any dirty inode by testing all three IO lists on each invocation without actively keeping track. For cgroup writeback support, a single bdi will host multiple wb's each of which will host dirty inodes separately and we'll need to make bdi_has_dirty_io(), which currently only represents the root wb, aggregate has_dirty_io from all member wb's, which requires tracking transitions in has_dirty_io state on each wb. This patch introduces inode_wb_list_{move|del}_locked() to consolidate IO list operations leaving queue_io() the only other function which directly manipulates IO lists (via move_expired_inodes()). All three functions are updated to call wb_io_lists_[de]populated() which keep track of whether the wb has dirty inodes or not and record it using the new WB_has_dirty_io flag. inode_wb_list_moved_locked()'s return value indicates whether the wb had no dirty inodes before. mark_inode_dirty() is restructured so that the return value of inode_wb_list_move_locked() can be used for deciding whether to wake up the wb. While at it, change {bdi|wb}_has_dirty_io()'s return values to bool. These functions were returning 0 and 1 before. Also, add a comment explaining the synchronization of wb_state flags. v2: Updated to accommodate b_dirty_time. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 110 ++++++++++++++++++++++++------- include/linux/backing-dev-defs.h | 1 + include/linux/backing-dev.h | 8 +-- mm/backing-dev.c | 2 +- 4 files changed, 91 insertions(+), 30 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 7ec491b1be04..0a90dc557748 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -93,6 +93,66 @@ static inline struct inode *wb_inode(struct list_head *head) EXPORT_TRACEPOINT_SYMBOL_GPL(wbc_writepage); +static bool wb_io_lists_populated(struct bdi_writeback *wb) +{ + if (wb_has_dirty_io(wb)) { + return false; + } else { + set_bit(WB_has_dirty_io, &wb->state); + return true; + } +} + +static void wb_io_lists_depopulated(struct bdi_writeback *wb) +{ + if (wb_has_dirty_io(wb) && list_empty(&wb->b_dirty) && + list_empty(&wb->b_io) && list_empty(&wb->b_more_io)) + clear_bit(WB_has_dirty_io, &wb->state); +} + +/** + * inode_wb_list_move_locked - move an inode onto a bdi_writeback IO list + * @inode: inode to be moved + * @wb: target bdi_writeback + * @head: one of @wb->b_{dirty|io|more_io} + * + * Move @inode->i_wb_list to @list of @wb and set %WB_has_dirty_io. + * Returns %true if @inode is the first occupant of the !dirty_time IO + * lists; otherwise, %false. + */ +static bool inode_wb_list_move_locked(struct inode *inode, + struct bdi_writeback *wb, + struct list_head *head) +{ + assert_spin_locked(&wb->list_lock); + + list_move(&inode->i_wb_list, head); + + /* dirty_time doesn't count as dirty_io until expiration */ + if (head != &wb->b_dirty_time) + return wb_io_lists_populated(wb); + + wb_io_lists_depopulated(wb); + return false; +} + +/** + * inode_wb_list_del_locked - remove an inode from its bdi_writeback IO list + * @inode: inode to be removed + * @wb: bdi_writeback @inode is being removed from + * + * Remove @inode which may be on one of @wb->b_{dirty|io|more_io} lists and + * clear %WB_has_dirty_io if all are empty afterwards. + */ +static void inode_wb_list_del_locked(struct inode *inode, + struct bdi_writeback *wb) +{ + assert_spin_locked(&wb->list_lock); + + list_del_init(&inode->i_wb_list); + wb_io_lists_depopulated(wb); +} + static void wb_wakeup(struct bdi_writeback *wb) { spin_lock_bh(&wb->work_lock); @@ -217,7 +277,7 @@ void inode_wb_list_del(struct inode *inode) struct bdi_writeback *wb = inode_to_wb(inode); spin_lock(&wb->list_lock); - list_del_init(&inode->i_wb_list); + inode_wb_list_del_locked(inode, wb); spin_unlock(&wb->list_lock); } @@ -232,7 +292,6 @@ void inode_wb_list_del(struct inode *inode) */ static void redirty_tail(struct inode *inode, struct bdi_writeback *wb) { - assert_spin_locked(&wb->list_lock); if (!list_empty(&wb->b_dirty)) { struct inode *tail; @@ -240,7 +299,7 @@ static void redirty_tail(struct inode *inode, struct bdi_writeback *wb) if (time_before(inode->dirtied_when, tail->dirtied_when)) inode->dirtied_when = jiffies; } - list_move(&inode->i_wb_list, &wb->b_dirty); + inode_wb_list_move_locked(inode, wb, &wb->b_dirty); } /* @@ -248,8 +307,7 @@ static void redirty_tail(struct inode *inode, struct bdi_writeback *wb) */ static void requeue_io(struct inode *inode, struct bdi_writeback *wb) { - assert_spin_locked(&wb->list_lock); - list_move(&inode->i_wb_list, &wb->b_more_io); + inode_wb_list_move_locked(inode, wb, &wb->b_more_io); } static void inode_sync_complete(struct inode *inode) @@ -358,6 +416,8 @@ static void queue_io(struct bdi_writeback *wb, struct wb_writeback_work *work) moved = move_expired_inodes(&wb->b_dirty, &wb->b_io, 0, work); moved += move_expired_inodes(&wb->b_dirty_time, &wb->b_io, EXPIRE_DIRTY_ATIME, work); + if (moved) + wb_io_lists_populated(wb); trace_writeback_queue_io(wb, work, moved); } @@ -483,10 +543,10 @@ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, redirty_tail(inode, wb); } else if (inode->i_state & I_DIRTY_TIME) { inode->dirtied_when = jiffies; - list_move(&inode->i_wb_list, &wb->b_dirty_time); + inode_wb_list_move_locked(inode, wb, &wb->b_dirty_time); } else { /* The inode is clean. Remove from writeback lists. */ - list_del_init(&inode->i_wb_list); + inode_wb_list_del_locked(inode, wb); } } @@ -628,7 +688,7 @@ writeback_single_inode(struct inode *inode, struct bdi_writeback *wb, * touch it. See comment above for explanation. */ if (!(inode->i_state & I_DIRTY_ALL)) - list_del_init(&inode->i_wb_list); + inode_wb_list_del_locked(inode, wb); spin_unlock(&wb->list_lock); inode_sync_complete(inode); out: @@ -1327,37 +1387,39 @@ void __mark_inode_dirty(struct inode *inode, int flags) * reposition it (that would break b_dirty time-ordering). */ if (!was_dirty) { + struct list_head *dirty_list; bool wakeup_bdi = false; bdi = inode_to_bdi(inode); spin_unlock(&inode->i_lock); spin_lock(&bdi->wb.list_lock); - if (bdi_cap_writeback_dirty(bdi)) { - WARN(!test_bit(WB_registered, &bdi->wb.state), - "bdi-%s not registered\n", bdi->name); - /* - * If this is the first dirty inode for this - * bdi, we have to wake-up the corresponding - * bdi thread to make sure background - * write-back happens later. - */ - if (!wb_has_dirty_io(&bdi->wb)) - wakeup_bdi = true; - } + WARN(bdi_cap_writeback_dirty(bdi) && + !test_bit(WB_registered, &bdi->wb.state), + "bdi-%s not registered\n", bdi->name); inode->dirtied_when = jiffies; if (dirtytime) inode->dirtied_time_when = jiffies; + if (inode->i_state & (I_DIRTY_INODE | I_DIRTY_PAGES)) - list_move(&inode->i_wb_list, &bdi->wb.b_dirty); + dirty_list = &bdi->wb.b_dirty; else - list_move(&inode->i_wb_list, - &bdi->wb.b_dirty_time); + dirty_list = &bdi->wb.b_dirty_time; + + wakeup_bdi = inode_wb_list_move_locked(inode, &bdi->wb, + dirty_list); + spin_unlock(&bdi->wb.list_lock); trace_writeback_dirty_inode_enqueue(inode); - if (wakeup_bdi) + /* + * If this is the first dirty inode for this bdi, + * we have to wake-up the corresponding bdi thread + * to make sure background write-back happens + * later. + */ + if (bdi_cap_writeback_dirty(bdi) && wakeup_bdi) wb_wakeup_delayed(&bdi->wb); return; } diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index eb386766b5f3..7a94b7850b7c 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -21,6 +21,7 @@ struct dentry; enum wb_state { WB_registered, /* bdi_register() was done */ WB_writeback_running, /* Writeback is in progress */ + WB_has_dirty_io, /* Dirty inodes on ->b_{dirty|io|more_io} */ }; enum wb_congested_state { diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 6f0882105f95..3c8403c012ce 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -29,7 +29,7 @@ void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, enum wb_reason reason); void bdi_start_background_writeback(struct backing_dev_info *bdi); void wb_workfn(struct work_struct *work); -int bdi_has_dirty_io(struct backing_dev_info *bdi); +bool bdi_has_dirty_io(struct backing_dev_info *bdi); void wb_wakeup_delayed(struct bdi_writeback *wb); extern spinlock_t bdi_lock; @@ -37,11 +37,9 @@ extern struct list_head bdi_list; extern struct workqueue_struct *bdi_wq; -static inline int wb_has_dirty_io(struct bdi_writeback *wb) +static inline bool wb_has_dirty_io(struct bdi_writeback *wb) { - return !list_empty(&wb->b_dirty) || - !list_empty(&wb->b_io) || - !list_empty(&wb->b_more_io); + return test_bit(WB_has_dirty_io, &wb->state); } static inline void __add_wb_stat(struct bdi_writeback *wb, diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 5029c4ad2f36..161ddf12b862 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -256,7 +256,7 @@ static int __init default_bdi_init(void) } subsys_initcall(default_bdi_init); -int bdi_has_dirty_io(struct backing_dev_info *bdi) +bool bdi_has_dirty_io(struct backing_dev_info *bdi) { return wb_has_dirty_io(&bdi->wb); } From 766a9d6e60578f1ef6de71f89f022084f8bffc82 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:46 -0400 Subject: [PATCH 32/85] writeback: implement backing_dev_info->tot_write_bandwidth cgroup writeback support needs to keep track of the sum of avg_write_bandwidth of all wb's (bdi_writeback's) with dirty inodes to distribute write workload. This patch adds bdi->tot_write_bandwidth and updates inode_wb_list_move_locked(), inode_wb_list_del_locked() and wb_update_write_bandwidth() to adjust it as wb's gain and lose dirty inodes and its avg_write_bandwidth gets updated. As the update events are not synchronized with each other, bdi->tot_write_bandwidth is an atomic_long_t. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 7 ++++++- include/linux/backing-dev-defs.h | 2 ++ mm/page-writeback.c | 3 +++ 3 files changed, 11 insertions(+), 1 deletion(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 0a90dc557748..bbccf68b38db 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -99,6 +99,8 @@ static bool wb_io_lists_populated(struct bdi_writeback *wb) return false; } else { set_bit(WB_has_dirty_io, &wb->state); + atomic_long_add(wb->avg_write_bandwidth, + &wb->bdi->tot_write_bandwidth); return true; } } @@ -106,8 +108,11 @@ static bool wb_io_lists_populated(struct bdi_writeback *wb) static void wb_io_lists_depopulated(struct bdi_writeback *wb) { if (wb_has_dirty_io(wb) && list_empty(&wb->b_dirty) && - list_empty(&wb->b_io) && list_empty(&wb->b_more_io)) + list_empty(&wb->b_io) && list_empty(&wb->b_more_io)) { clear_bit(WB_has_dirty_io, &wb->state); + atomic_long_sub(wb->avg_write_bandwidth, + &wb->bdi->tot_write_bandwidth); + } } /** diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index 7a94b7850b7c..d631a61f4023 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -142,6 +142,8 @@ struct backing_dev_info { unsigned int min_ratio; unsigned int max_ratio, max_prop_frac; + atomic_long_t tot_write_bandwidth; /* sum of active avg_write_bw */ + struct bdi_writeback wb; /* the root writeback info for this bdi */ struct bdi_writeback_congested wb_congested; /* its congested state */ #ifdef CONFIG_CGROUP_WRITEBACK diff --git a/mm/page-writeback.c b/mm/page-writeback.c index e31dea94d116..c95eb24dd652 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -881,6 +881,9 @@ static void wb_update_write_bandwidth(struct bdi_writeback *wb, avg += (old - avg) >> 3; out: + if (wb_has_dirty_io(wb)) + atomic_long_add(avg - wb->avg_write_bandwidth, + &wb->bdi->tot_write_bandwidth); wb->write_bandwidth = bw; wb->avg_write_bandwidth = avg; } From 95a46c65e3c09edb9f17dabf2dc16670cd328739 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:47 -0400 Subject: [PATCH 33/85] writeback: make bdi_has_dirty_io() take multiple bdi_writeback's into account bdi_has_dirty_io() used to only reflect whether the root wb (bdi_writeback) has dirty inodes. For cgroup writeback support, it needs to take all active wb's into account. If any wb on the bdi has dirty inodes, bdi_has_dirty_io() should return true. To achieve that, as inode_wb_list_{move|del}_locked() now keep track of the dirty state transition of each wb, the number of dirty wbs can be counted in the bdi; however, bdi is already aggregating wb->avg_write_bandwidth which can easily be guaranteed to be > 0 when there are any dirty inodes by ensuring wb->avg_write_bandwidth can't dip below 1. bdi_has_dirty_io() can simply test whether bdi->tot_write_bandwidth is zero or not. While this bumps the value of wb->avg_write_bandwidth to one when it used to be zero, this shouldn't cause any meaningful behavior difference. bdi_has_dirty_io() is made an inline function which tests whether ->tot_write_bandwidth is non-zero. Also, WARN_ON_ONCE()'s on its value are added to inode_wb_list_{move|del}_locked(). Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 5 +++-- include/linux/backing-dev-defs.h | 8 ++++++-- include/linux/backing-dev.h | 10 +++++++++- mm/backing-dev.c | 5 ----- mm/page-writeback.c | 10 +++++++--- 5 files changed, 25 insertions(+), 13 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index bbccf68b38db..c98d3923d9a9 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -99,6 +99,7 @@ static bool wb_io_lists_populated(struct bdi_writeback *wb) return false; } else { set_bit(WB_has_dirty_io, &wb->state); + WARN_ON_ONCE(!wb->avg_write_bandwidth); atomic_long_add(wb->avg_write_bandwidth, &wb->bdi->tot_write_bandwidth); return true; @@ -110,8 +111,8 @@ static void wb_io_lists_depopulated(struct bdi_writeback *wb) if (wb_has_dirty_io(wb) && list_empty(&wb->b_dirty) && list_empty(&wb->b_io) && list_empty(&wb->b_more_io)) { clear_bit(WB_has_dirty_io, &wb->state); - atomic_long_sub(wb->avg_write_bandwidth, - &wb->bdi->tot_write_bandwidth); + WARN_ON_ONCE(atomic_long_sub_return(wb->avg_write_bandwidth, + &wb->bdi->tot_write_bandwidth) < 0); } } diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index d631a61f4023..8c857d723023 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -98,7 +98,7 @@ struct bdi_writeback { unsigned long dirtied_stamp; unsigned long written_stamp; /* pages written at bw_time_stamp */ unsigned long write_bandwidth; /* the estimated write bandwidth */ - unsigned long avg_write_bandwidth; /* further smoothed write bw */ + unsigned long avg_write_bandwidth; /* further smoothed write bw, > 0 */ /* * The base dirty throttle rate, re-calculated on every 200ms. @@ -142,7 +142,11 @@ struct backing_dev_info { unsigned int min_ratio; unsigned int max_ratio, max_prop_frac; - atomic_long_t tot_write_bandwidth; /* sum of active avg_write_bw */ + /* + * Sum of avg_write_bw of wbs with dirty inodes. > 0 if there are + * any dirty wbs, which is depended upon by bdi_has_dirty(). + */ + atomic_long_t tot_write_bandwidth; struct bdi_writeback wb; /* the root writeback info for this bdi */ struct bdi_writeback_congested wb_congested; /* its congested state */ diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 3c8403c012ce..0839e44105bd 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -29,7 +29,6 @@ void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, enum wb_reason reason); void bdi_start_background_writeback(struct backing_dev_info *bdi); void wb_workfn(struct work_struct *work); -bool bdi_has_dirty_io(struct backing_dev_info *bdi); void wb_wakeup_delayed(struct bdi_writeback *wb); extern spinlock_t bdi_lock; @@ -42,6 +41,15 @@ static inline bool wb_has_dirty_io(struct bdi_writeback *wb) return test_bit(WB_has_dirty_io, &wb->state); } +static inline bool bdi_has_dirty_io(struct backing_dev_info *bdi) +{ + /* + * @bdi->tot_write_bandwidth is guaranteed to be > 0 if there are + * any dirty wbs. See wb_update_write_bandwidth(). + */ + return atomic_long_read(&bdi->tot_write_bandwidth); +} + static inline void __add_wb_stat(struct bdi_writeback *wb, enum wb_stat_item item, s64 amount) { diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 161ddf12b862..d2f16fc9069a 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -256,11 +256,6 @@ static int __init default_bdi_init(void) } subsys_initcall(default_bdi_init); -bool bdi_has_dirty_io(struct backing_dev_info *bdi) -{ - return wb_has_dirty_io(&bdi->wb); -} - /* * This function is used when the first inode for this wb is marked dirty. It * wakes-up the corresponding bdi thread which should then take care of the diff --git a/mm/page-writeback.c b/mm/page-writeback.c index c95eb24dd652..99b88465096e 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -881,9 +881,13 @@ static void wb_update_write_bandwidth(struct bdi_writeback *wb, avg += (old - avg) >> 3; out: - if (wb_has_dirty_io(wb)) - atomic_long_add(avg - wb->avg_write_bandwidth, - &wb->bdi->tot_write_bandwidth); + /* keep avg > 0 to guarantee that tot > 0 if there are dirty wbs */ + avg = max(avg, 1LU); + if (wb_has_dirty_io(wb)) { + long delta = avg - wb->avg_write_bandwidth; + WARN_ON_ONCE(atomic_long_add_return(delta, + &wb->bdi->tot_write_bandwidth) <= 0); + } wb->write_bandwidth = bw; wb->avg_write_bandwidth = avg; } From e79729123f6392b36450113c6c52074b7d389c85 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:48 -0400 Subject: [PATCH 34/85] writeback: don't issue wb_writeback_work if clean There are several places in fs/fs-writeback.c which queues wb_writeback_work without checking whether the target wb (bdi_writeback) has dirty inodes or not. The only thing wb_writeback_work does is writing back the dirty inodes for the target wb and queueing a work item for a clean wb is essentially noop. There are some side effects such as bandwidth stats being updated and triggering tracepoints but these don't affect the operation in any meaningful way. This patch makes all writeback_inodes_sb_nr() and sync_inodes_sb() skip wb_queue_work() if the target bdi is clean. Also, it moves dirtiness check from wakeup_flusher_threads() to __wb_start_writeback() so that all its callers benefit from the check. While the overhead incurred by scheduling a noop work isn't currently significant, the overhead may be higher with cgroup writeback support as we may end up issuing noop work items to a lot of clean wb's. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index c98d3923d9a9..921a9e43b1db 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -189,6 +189,9 @@ static void __wb_start_writeback(struct bdi_writeback *wb, long nr_pages, { struct wb_writeback_work *work; + if (!wb_has_dirty_io(wb)) + return; + /* * This is WB_SYNC_NONE writeback, so if allocation fails just * wakeup the thread for old dirty data writeback @@ -1215,11 +1218,8 @@ void wakeup_flusher_threads(long nr_pages, enum wb_reason reason) nr_pages = get_nr_dirty_pages(); rcu_read_lock(); - list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) { - if (!bdi_has_dirty_io(bdi)) - continue; + list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) __wb_start_writeback(&bdi->wb, nr_pages, false, reason); - } rcu_read_unlock(); } @@ -1512,11 +1512,12 @@ void writeback_inodes_sb_nr(struct super_block *sb, .nr_pages = nr, .reason = reason, }; + struct backing_dev_info *bdi = sb->s_bdi; - if (sb->s_bdi == &noop_backing_dev_info) + if (!bdi_has_dirty_io(bdi) || bdi == &noop_backing_dev_info) return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); - wb_queue_work(&sb->s_bdi->wb, &work); + wb_queue_work(&bdi->wb, &work); wait_for_completion(&done); } EXPORT_SYMBOL(writeback_inodes_sb_nr); @@ -1594,13 +1595,14 @@ void sync_inodes_sb(struct super_block *sb) .reason = WB_REASON_SYNC, .for_sync = 1, }; + struct backing_dev_info *bdi = sb->s_bdi; /* Nothing to do? */ - if (sb->s_bdi == &noop_backing_dev_info) + if (!bdi_has_dirty_io(bdi) || bdi == &noop_backing_dev_info) return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); - wb_queue_work(&sb->s_bdi->wb, &work); + wb_queue_work(&bdi->wb, &work); wait_for_completion(&done); wait_sb_inodes(sb); From 693108a8a6672cec88265d83f7187dc83ba1d6a3 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:49 -0400 Subject: [PATCH 35/85] writeback: make bdi->min/max_ratio handling cgroup writeback aware bdi->min/max_ratio are user-configurable per-bdi knobs which regulate dirty limit of each bdi. For cgroup writeback, they need to be further distributed across wb's (bdi_writeback's) belonging to the configured bdi. This patch introduces wb_min_max_ratio() which distributes bdi->min/max_ratio according to a wb's proportion in the total active bandwidth of its bdi. v2: Update wb_min_max_ratio() to fix a bug where both min and max were assigned the min value and avoid calculations when possible. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- mm/page-writeback.c | 50 +++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 46 insertions(+), 4 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 99b88465096e..9b55f12040e6 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -155,6 +155,46 @@ static unsigned long writeout_period_time = 0; */ #define VM_COMPLETIONS_PERIOD_LEN (3*HZ) +#ifdef CONFIG_CGROUP_WRITEBACK + +static void wb_min_max_ratio(struct bdi_writeback *wb, + unsigned long *minp, unsigned long *maxp) +{ + unsigned long this_bw = wb->avg_write_bandwidth; + unsigned long tot_bw = atomic_long_read(&wb->bdi->tot_write_bandwidth); + unsigned long long min = wb->bdi->min_ratio; + unsigned long long max = wb->bdi->max_ratio; + + /* + * @wb may already be clean by the time control reaches here and + * the total may not include its bw. + */ + if (this_bw < tot_bw) { + if (min) { + min *= this_bw; + do_div(min, tot_bw); + } + if (max < 100) { + max *= this_bw; + do_div(max, tot_bw); + } + } + + *minp = min; + *maxp = max; +} + +#else /* CONFIG_CGROUP_WRITEBACK */ + +static void wb_min_max_ratio(struct bdi_writeback *wb, + unsigned long *minp, unsigned long *maxp) +{ + *minp = wb->bdi->min_ratio; + *maxp = wb->bdi->max_ratio; +} + +#endif /* CONFIG_CGROUP_WRITEBACK */ + /* * In a memory zone, there is a certain amount of pages we consider * available for the page cache, which is essentially the number of @@ -539,9 +579,9 @@ static unsigned long hard_dirty_limit(unsigned long thresh) */ unsigned long wb_dirty_limit(struct bdi_writeback *wb, unsigned long dirty) { - struct backing_dev_info *bdi = wb->bdi; u64 wb_dirty; long numerator, denominator; + unsigned long wb_min_ratio, wb_max_ratio; /* * Calculate this BDI's share of the dirty ratio. @@ -552,9 +592,11 @@ unsigned long wb_dirty_limit(struct bdi_writeback *wb, unsigned long dirty) wb_dirty *= numerator; do_div(wb_dirty, denominator); - wb_dirty += (dirty * bdi->min_ratio) / 100; - if (wb_dirty > (dirty * bdi->max_ratio) / 100) - wb_dirty = dirty * bdi->max_ratio / 100; + wb_min_max_ratio(wb, &wb_min_ratio, &wb_max_ratio); + + wb_dirty += (dirty * wb_min_ratio) / 100; + if (wb_dirty > (dirty * wb_max_ratio) / 100) + wb_dirty = dirty * wb_max_ratio / 100; return wb_dirty; } From ebe41ab0c79d5633123f6faa3265a1a63c5f22d8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:50 -0400 Subject: [PATCH 36/85] writeback: implement bdi_for_each_wb() This will be used to implement bdi-wide operations which should be distributed across all its cgroup bdi_writebacks. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- include/linux/backing-dev.h | 63 +++++++++++++++++++++++++++++++++++++ 1 file changed, 63 insertions(+) diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 0839e44105bd..c7979806baee 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -383,6 +383,61 @@ static inline struct bdi_writeback *inode_to_wb(struct inode *inode) return inode->i_wb; } +struct wb_iter { + int start_blkcg_id; + struct radix_tree_iter tree_iter; + void **slot; +}; + +static inline struct bdi_writeback *__wb_iter_next(struct wb_iter *iter, + struct backing_dev_info *bdi) +{ + struct radix_tree_iter *titer = &iter->tree_iter; + + WARN_ON_ONCE(!rcu_read_lock_held()); + + if (iter->start_blkcg_id >= 0) { + iter->slot = radix_tree_iter_init(titer, iter->start_blkcg_id); + iter->start_blkcg_id = -1; + } else { + iter->slot = radix_tree_next_slot(iter->slot, titer, 0); + } + + if (!iter->slot) + iter->slot = radix_tree_next_chunk(&bdi->cgwb_tree, titer, 0); + if (iter->slot) + return *iter->slot; + return NULL; +} + +static inline struct bdi_writeback *__wb_iter_init(struct wb_iter *iter, + struct backing_dev_info *bdi, + int start_blkcg_id) +{ + iter->start_blkcg_id = start_blkcg_id; + + if (start_blkcg_id) + return __wb_iter_next(iter, bdi); + else + return &bdi->wb; +} + +/** + * bdi_for_each_wb - walk all wb's of a bdi in ascending blkcg ID order + * @wb_cur: cursor struct bdi_writeback pointer + * @bdi: bdi to walk wb's of + * @iter: pointer to struct wb_iter to be used as iteration buffer + * @start_blkcg_id: blkcg ID to start iteration from + * + * Iterate @wb_cur through the wb's (bdi_writeback's) of @bdi in ascending + * blkcg ID order starting from @start_blkcg_id. @iter is struct wb_iter + * to be used as temp storage during iteration. rcu_read_lock() must be + * held throughout iteration. + */ +#define bdi_for_each_wb(wb_cur, bdi, iter, start_blkcg_id) \ + for ((wb_cur) = __wb_iter_init(iter, bdi, start_blkcg_id); \ + (wb_cur); (wb_cur) = __wb_iter_next(iter, bdi)) + #else /* CONFIG_CGROUP_WRITEBACK */ static inline bool inode_cgwb_enabled(struct inode *inode) @@ -445,6 +500,14 @@ static inline void wb_blkcg_offline(struct blkcg *blkcg) { } +struct wb_iter { + int next_id; +}; + +#define bdi_for_each_wb(wb_cur, bdi, iter, start_blkcg_id) \ + for ((iter)->next_id = (start_blkcg_id); \ + ({ (wb_cur) = !(iter)->next_id++ ? &(bdi)->wb : NULL; }); ) + static inline int inode_congested(struct inode *inode, int cong_bits) { return wb_congested(&inode_to_bdi(inode)->wb, cong_bits); From c00ddad39f512b1a81e25b7892217ce10efab0f1 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:51 -0400 Subject: [PATCH 37/85] writeback: remove bdi_start_writeback() bdi_start_writeback() is a thin wrapper on top of __wb_start_writeback() which is used only by laptop_mode_timer_fn(). This patches removes bdi_start_writeback(), renames __wb_start_writeback() to wb_start_writeback() and makes laptop_mode_timer_fn() use it instead. This doesn't cause any functional difference and will ease making laptop_mode_timer_fn() cgroup writeback aware. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 68 ++++++++++++++----------------------- include/linux/backing-dev.h | 4 +-- mm/page-writeback.c | 4 +-- 3 files changed, 29 insertions(+), 47 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 921a9e43b1db..79f11af67357 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -184,33 +184,6 @@ out_unlock: spin_unlock_bh(&wb->work_lock); } -static void __wb_start_writeback(struct bdi_writeback *wb, long nr_pages, - bool range_cyclic, enum wb_reason reason) -{ - struct wb_writeback_work *work; - - if (!wb_has_dirty_io(wb)) - return; - - /* - * This is WB_SYNC_NONE writeback, so if allocation fails just - * wakeup the thread for old dirty data writeback - */ - work = kzalloc(sizeof(*work), GFP_ATOMIC); - if (!work) { - trace_writeback_nowork(wb->bdi); - wb_wakeup(wb); - return; - } - - work->sync_mode = WB_SYNC_NONE; - work->nr_pages = nr_pages; - work->range_cyclic = range_cyclic; - work->reason = reason; - - wb_queue_work(wb, work); -} - #ifdef CONFIG_CGROUP_WRITEBACK /** @@ -240,22 +213,31 @@ EXPORT_SYMBOL_GPL(inode_congested); #endif /* CONFIG_CGROUP_WRITEBACK */ -/** - * bdi_start_writeback - start writeback - * @bdi: the backing device to write from - * @nr_pages: the number of pages to write - * @reason: reason why some writeback work was initiated - * - * Description: - * This does WB_SYNC_NONE opportunistic writeback. The IO is only - * started when this function returns, we make no guarantees on - * completion. Caller need not hold sb s_umount semaphore. - * - */ -void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, - enum wb_reason reason) +void wb_start_writeback(struct bdi_writeback *wb, long nr_pages, + bool range_cyclic, enum wb_reason reason) { - __wb_start_writeback(&bdi->wb, nr_pages, true, reason); + struct wb_writeback_work *work; + + if (!wb_has_dirty_io(wb)) + return; + + /* + * This is WB_SYNC_NONE writeback, so if allocation fails just + * wakeup the thread for old dirty data writeback + */ + work = kzalloc(sizeof(*work), GFP_ATOMIC); + if (!work) { + trace_writeback_nowork(wb->bdi); + wb_wakeup(wb); + return; + } + + work->sync_mode = WB_SYNC_NONE; + work->nr_pages = nr_pages; + work->range_cyclic = range_cyclic; + work->reason = reason; + + wb_queue_work(wb, work); } /** @@ -1219,7 +1201,7 @@ void wakeup_flusher_threads(long nr_pages, enum wb_reason reason) rcu_read_lock(); list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) - __wb_start_writeback(&bdi->wb, nr_pages, false, reason); + wb_start_writeback(&bdi->wb, nr_pages, false, reason); rcu_read_unlock(); } diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index c7979806baee..0ff40c228bee 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -25,8 +25,8 @@ int bdi_register(struct backing_dev_info *bdi, struct device *parent, int bdi_register_dev(struct backing_dev_info *bdi, dev_t dev); void bdi_unregister(struct backing_dev_info *bdi); int __must_check bdi_setup_and_register(struct backing_dev_info *, char *); -void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, - enum wb_reason reason); +void wb_start_writeback(struct bdi_writeback *wb, long nr_pages, + bool range_cyclic, enum wb_reason reason); void bdi_start_background_writeback(struct backing_dev_info *bdi); void wb_workfn(struct work_struct *work); void wb_wakeup_delayed(struct bdi_writeback *wb); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 9b55f12040e6..6301af28da21 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1729,8 +1729,8 @@ void laptop_mode_timer_fn(unsigned long data) * threshold */ if (bdi_has_dirty_io(&q->backing_dev_info)) - bdi_start_writeback(&q->backing_dev_info, nr_pages, - WB_REASON_LAPTOP_TIMER); + wb_start_writeback(&q->backing_dev_info.wb, nr_pages, true, + WB_REASON_LAPTOP_TIMER); } /* From a06fd6b102286e3b727ed42b8fb37825fa7127a2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:52 -0400 Subject: [PATCH 38/85] writeback: make laptop_mode_timer_fn() handle multiple bdi_writeback's For cgroup writeback support, all bdi-wide operations should be distributed to all its wb's (bdi_writeback's). This patch updates laptop_mode_timer_fn() so that it invokes wb_start_writeback() on all wb's rather than just the root one. As the intent is writing out all dirty data, there's no reason to split the number of pages to write. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- mm/page-writeback.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 6301af28da21..682e3a6a8b2e 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1723,14 +1723,20 @@ void laptop_mode_timer_fn(unsigned long data) struct request_queue *q = (struct request_queue *)data; int nr_pages = global_page_state(NR_FILE_DIRTY) + global_page_state(NR_UNSTABLE_NFS); + struct bdi_writeback *wb; + struct wb_iter iter; /* * We want to write everything out, not just down to the dirty * threshold */ - if (bdi_has_dirty_io(&q->backing_dev_info)) - wb_start_writeback(&q->backing_dev_info.wb, nr_pages, true, - WB_REASON_LAPTOP_TIMER); + if (!bdi_has_dirty_io(&q->backing_dev_info)) + return; + + bdi_for_each_wb(wb, &q->backing_dev_info, &iter, 0) + if (wb_has_dirty_io(wb)) + wb_start_writeback(wb, nr_pages, true, + WB_REASON_LAPTOP_TIMER); } /* From bc05873dccd27d75d6acdf812c3edfb181f1ba17 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:53 -0400 Subject: [PATCH 39/85] writeback: make writeback_in_progress() take bdi_writeback instead of backing_dev_info writeback_in_progress() currently takes @bdi and returns whether writeback is in progress on its root wb (bdi_writeback). In preparation for cgroup writeback support, make it take wb instead. While at it, make it an inline function. This patch doesn't make any functional difference. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 15 +-------------- include/linux/backing-dev.h | 12 +++++++++++- mm/page-writeback.c | 4 ++-- 3 files changed, 14 insertions(+), 17 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 79f11af67357..45baf6c89b99 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -65,19 +65,6 @@ struct wb_writeback_work { */ unsigned int dirtytime_expire_interval = 12 * 60 * 60; -/** - * writeback_in_progress - determine whether there is writeback in progress - * @bdi: the device's backing_dev_info structure. - * - * Determine whether there is writeback waiting to be handled against a - * backing device. - */ -int writeback_in_progress(struct backing_dev_info *bdi) -{ - return test_bit(WB_writeback_running, &bdi->wb.state); -} -EXPORT_SYMBOL(writeback_in_progress); - static inline struct inode *wb_inode(struct list_head *head) { return list_entry(head, struct inode, i_wb_list); @@ -1532,7 +1519,7 @@ int try_to_writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr, enum wb_reason reason) { - if (writeback_in_progress(sb->s_bdi)) + if (writeback_in_progress(&sb->s_bdi->wb)) return 1; if (!down_read_trylock(&sb->s_umount)) diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 0ff40c228bee..f04956c900ec 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -156,7 +156,17 @@ int bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ratio); extern struct backing_dev_info noop_backing_dev_info; -int writeback_in_progress(struct backing_dev_info *bdi); +/** + * writeback_in_progress - determine whether there is writeback in progress + * @wb: bdi_writeback of interest + * + * Determine whether there is writeback waiting to be handled against a + * bdi_writeback. + */ +static inline bool writeback_in_progress(struct bdi_writeback *wb) +{ + return test_bit(WB_writeback_running, &wb->state); +} static inline struct backing_dev_info *inode_to_bdi(struct inode *inode) { diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 682e3a6a8b2e..e3b5c1dddf1d 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1455,7 +1455,7 @@ static void balance_dirty_pages(struct address_space *mapping, break; } - if (unlikely(!writeback_in_progress(bdi))) + if (unlikely(!writeback_in_progress(wb))) bdi_start_background_writeback(bdi); if (!strictlimit) @@ -1573,7 +1573,7 @@ pause: if (!dirty_exceeded && wb->dirty_exceeded) wb->dirty_exceeded = 0; - if (writeback_in_progress(bdi)) + if (writeback_in_progress(wb)) return; /* From 9ecf4866c018aeb304a7b49216c4d183665becb7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:54 -0400 Subject: [PATCH 40/85] writeback: make bdi_start_background_writeback() take bdi_writeback instead of backing_dev_info bdi_start_background_writeback() currently takes @bdi and kicks the root wb (bdi_writeback). In preparation for cgroup writeback support, make it take wb instead. This patch doesn't make any functional difference. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 12 ++++++------ include/linux/backing-dev.h | 2 +- mm/page-writeback.c | 4 ++-- 3 files changed, 9 insertions(+), 9 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 45baf6c89b99..92aaf641ee22 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -228,23 +228,23 @@ void wb_start_writeback(struct bdi_writeback *wb, long nr_pages, } /** - * bdi_start_background_writeback - start background writeback - * @bdi: the backing device to write from + * wb_start_background_writeback - start background writeback + * @wb: bdi_writback to write from * * Description: * This makes sure WB_SYNC_NONE background writeback happens. When - * this function returns, it is only guaranteed that for given BDI + * this function returns, it is only guaranteed that for given wb * some IO is happening if we are over background dirty threshold. * Caller need not hold sb s_umount semaphore. */ -void bdi_start_background_writeback(struct backing_dev_info *bdi) +void wb_start_background_writeback(struct bdi_writeback *wb) { /* * We just wake up the flusher thread. It will perform background * writeback as soon as there is no other work to do. */ - trace_writeback_wake_background(bdi); - wb_wakeup(&bdi->wb); + trace_writeback_wake_background(wb->bdi); + wb_wakeup(wb); } /* diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index f04956c900ec..9cc11e5b97ca 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -27,7 +27,7 @@ void bdi_unregister(struct backing_dev_info *bdi); int __must_check bdi_setup_and_register(struct backing_dev_info *, char *); void wb_start_writeback(struct bdi_writeback *wb, long nr_pages, bool range_cyclic, enum wb_reason reason); -void bdi_start_background_writeback(struct backing_dev_info *bdi); +void wb_start_background_writeback(struct bdi_writeback *wb); void wb_workfn(struct work_struct *work); void wb_wakeup_delayed(struct bdi_writeback *wb); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index e3b5c1dddf1d..70cf98dc3423 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1456,7 +1456,7 @@ static void balance_dirty_pages(struct address_space *mapping, } if (unlikely(!writeback_in_progress(wb))) - bdi_start_background_writeback(bdi); + wb_start_background_writeback(wb); if (!strictlimit) wb_dirty_limits(wb, dirty_thresh, background_thresh, @@ -1588,7 +1588,7 @@ pause: return; if (nr_reclaimable > background_thresh) - bdi_start_background_writeback(bdi); + wb_start_background_writeback(wb); } static DEFINE_PER_CPU(int, bdp_ratelimits); From f2b65121607631df0753eedd624262818707af32 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:55 -0400 Subject: [PATCH 41/85] writeback: make wakeup_flusher_threads() handle multiple bdi_writeback's wakeup_flusher_threads() currently only starts writeback on the root wb (bdi_writeback). For cgroup writeback support, update the function to wake up all wbs and distribute the number of pages to write according to the proportion of each wb's write bandwidth, which is implemented in wb_split_bdi_pages(). Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 48 +++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 46 insertions(+), 2 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 92aaf641ee22..508e10cc3fcd 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -198,6 +198,41 @@ int inode_congested(struct inode *inode, int cong_bits) } EXPORT_SYMBOL_GPL(inode_congested); +/** + * wb_split_bdi_pages - split nr_pages to write according to bandwidth + * @wb: target bdi_writeback to split @nr_pages to + * @nr_pages: number of pages to write for the whole bdi + * + * Split @wb's portion of @nr_pages according to @wb's write bandwidth in + * relation to the total write bandwidth of all wb's w/ dirty inodes on + * @wb->bdi. + */ +static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages) +{ + unsigned long this_bw = wb->avg_write_bandwidth; + unsigned long tot_bw = atomic_long_read(&wb->bdi->tot_write_bandwidth); + + if (nr_pages == LONG_MAX) + return LONG_MAX; + + /* + * This may be called on clean wb's and proportional distribution + * may not make sense, just use the original @nr_pages in those + * cases. In general, we wanna err on the side of writing more. + */ + if (!tot_bw || this_bw >= tot_bw) + return nr_pages; + else + return DIV_ROUND_UP_ULL((u64)nr_pages * this_bw, tot_bw); +} + +#else /* CONFIG_CGROUP_WRITEBACK */ + +static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages) +{ + return nr_pages; +} + #endif /* CONFIG_CGROUP_WRITEBACK */ void wb_start_writeback(struct bdi_writeback *wb, long nr_pages, @@ -1187,8 +1222,17 @@ void wakeup_flusher_threads(long nr_pages, enum wb_reason reason) nr_pages = get_nr_dirty_pages(); rcu_read_lock(); - list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) - wb_start_writeback(&bdi->wb, nr_pages, false, reason); + list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) { + struct bdi_writeback *wb; + struct wb_iter iter; + + if (!bdi_has_dirty_io(bdi)) + continue; + + bdi_for_each_wb(wb, bdi, &iter, 0) + wb_start_writeback(wb, wb_split_bdi_pages(wb, nr_pages), + false, reason); + } rcu_read_unlock(); } From 001fe6f617b1a352ad29cc3643384b3d8141dc07 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:56 -0400 Subject: [PATCH 42/85] writeback: make wakeup_dirtytime_writeback() handle multiple bdi_writeback's wakeup_dirtytime_writeback() currently only starts writeback on the root wb (bdi_writeback). For cgroup writeback support, update the function to check all wbs. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Theodore Ts'o Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 508e10cc3fcd..8ae212e90ec3 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1260,9 +1260,12 @@ static void wakeup_dirtytime_writeback(struct work_struct *w) rcu_read_lock(); list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) { - if (list_empty(&bdi->wb.b_dirty_time)) - continue; - wb_wakeup(&bdi->wb); + struct bdi_writeback *wb; + struct wb_iter iter; + + bdi_for_each_wb(wb, bdi, &iter, 0) + if (!list_empty(&bdi->wb.b_dirty_time)) + wb_wakeup(&bdi->wb); } rcu_read_unlock(); schedule_delayed_work(&dirtytime_work, dirtytime_expire_interval * HZ); From ac7b19a34f332fdf8b13aac8708b14188fd1fbd2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:57 -0400 Subject: [PATCH 43/85] writeback: add wb_writeback_work->auto_free Currently, a wb_writeback_work is freed automatically on completion if it doesn't have ->done set. Add wb_writeback_work->auto_free to make the switch explicit. This will help cgroup writeback support where waiting for completion and whether to free automatically don't necessarily move together. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 8ae212e90ec3..22f1def4e63a 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -47,6 +47,7 @@ struct wb_writeback_work { unsigned int range_cyclic:1; unsigned int for_background:1; unsigned int for_sync:1; /* sync(2) WB_SYNC_ALL writeback */ + unsigned int auto_free:1; /* free on completion */ enum wb_reason reason; /* why was writeback initiated? */ struct list_head list; /* pending work list */ @@ -258,6 +259,7 @@ void wb_start_writeback(struct bdi_writeback *wb, long nr_pages, work->nr_pages = nr_pages; work->range_cyclic = range_cyclic; work->reason = reason; + work->auto_free = 1; wb_queue_work(wb, work); } @@ -1141,19 +1143,16 @@ static long wb_do_writeback(struct bdi_writeback *wb) set_bit(WB_writeback_running, &wb->state); while ((work = get_next_work_item(wb)) != NULL) { + struct completion *done = work->done; trace_writeback_exec(wb->bdi, work); wrote += wb_writeback(wb, work); - /* - * Notify the caller of completion if this is a synchronous - * work item, otherwise just free it. - */ - if (work->done) - complete(work->done); - else + if (work->auto_free) kfree(work); + if (done) + complete(done); } /* From cc395d7f1f7b9c740ab6d367ef1f6eb248595dff Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:58 -0400 Subject: [PATCH 44/85] writeback: implement bdi_wait_for_completion() If the completion of a wb_writeback_work can be waited upon by setting its ->done to a struct completion and waiting on it; however, for cgroup writeback support, it's necessary to issue multiple work items to multiple bdi_writebacks and wait for the completion of all. This patch implements wb_completion which can wait for multiple work items and replaces the struct completion with it. It can be defined using DEFINE_WB_COMPLETION_ONSTACK(), used for multiple work items and waited for by wb_wait_for_completion(). Nobody currently issues multiple work items and this patch doesn't introduce any behavior changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 58 +++++++++++++++++++++++++------- include/linux/backing-dev-defs.h | 2 ++ mm/backing-dev.c | 1 + 3 files changed, 49 insertions(+), 12 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 22f1def4e63a..d7d4a1bcdd2f 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -34,6 +34,10 @@ */ #define MIN_WRITEBACK_PAGES (4096UL >> (PAGE_CACHE_SHIFT - 10)) +struct wb_completion { + atomic_t cnt; +}; + /* * Passed into wb_writeback(), essentially a subset of writeback_control */ @@ -51,9 +55,22 @@ struct wb_writeback_work { enum wb_reason reason; /* why was writeback initiated? */ struct list_head list; /* pending work list */ - struct completion *done; /* set if the caller waits */ + struct wb_completion *done; /* set if the caller waits */ }; +/* + * If one wants to wait for one or more wb_writeback_works, each work's + * ->done should be set to a wb_completion defined using the following + * macro. Once all work items are issued with wb_queue_work(), the caller + * can wait for the completion of all using wb_wait_for_completion(). Work + * items which are waited upon aren't freed automatically on completion. + */ +#define DEFINE_WB_COMPLETION_ONSTACK(cmpl) \ + struct wb_completion cmpl = { \ + .cnt = ATOMIC_INIT(1), \ + } + + /* * If an inode is constantly having its pages dirtied, but then the * updates stop dirtytime_expire_interval seconds in the past, it's @@ -161,17 +178,34 @@ static void wb_queue_work(struct bdi_writeback *wb, trace_writeback_queue(wb->bdi, work); spin_lock_bh(&wb->work_lock); - if (!test_bit(WB_registered, &wb->state)) { - if (work->done) - complete(work->done); + if (!test_bit(WB_registered, &wb->state)) goto out_unlock; - } + if (work->done) + atomic_inc(&work->done->cnt); list_add_tail(&work->list, &wb->work_list); mod_delayed_work(bdi_wq, &wb->dwork, 0); out_unlock: spin_unlock_bh(&wb->work_lock); } +/** + * wb_wait_for_completion - wait for completion of bdi_writeback_works + * @bdi: bdi work items were issued to + * @done: target wb_completion + * + * Wait for one or more work items issued to @bdi with their ->done field + * set to @done, which should have been defined with + * DEFINE_WB_COMPLETION_ONSTACK(). This function returns after all such + * work items are completed. Work items which are waited upon aren't freed + * automatically on completion. + */ +static void wb_wait_for_completion(struct backing_dev_info *bdi, + struct wb_completion *done) +{ + atomic_dec(&done->cnt); /* put down the initial count */ + wait_event(bdi->wb_waitq, !atomic_read(&done->cnt)); +} + #ifdef CONFIG_CGROUP_WRITEBACK /** @@ -1143,7 +1177,7 @@ static long wb_do_writeback(struct bdi_writeback *wb) set_bit(WB_writeback_running, &wb->state); while ((work = get_next_work_item(wb)) != NULL) { - struct completion *done = work->done; + struct wb_completion *done = work->done; trace_writeback_exec(wb->bdi, work); @@ -1151,8 +1185,8 @@ static long wb_do_writeback(struct bdi_writeback *wb) if (work->auto_free) kfree(work); - if (done) - complete(done); + if (done && atomic_dec_and_test(&done->cnt)) + wake_up_all(&wb->bdi->wb_waitq); } /* @@ -1518,7 +1552,7 @@ void writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr, enum wb_reason reason) { - DECLARE_COMPLETION_ONSTACK(done); + DEFINE_WB_COMPLETION_ONSTACK(done); struct wb_writeback_work work = { .sb = sb, .sync_mode = WB_SYNC_NONE, @@ -1533,7 +1567,7 @@ void writeback_inodes_sb_nr(struct super_block *sb, return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); wb_queue_work(&bdi->wb, &work); - wait_for_completion(&done); + wb_wait_for_completion(bdi, &done); } EXPORT_SYMBOL(writeback_inodes_sb_nr); @@ -1600,7 +1634,7 @@ EXPORT_SYMBOL(try_to_writeback_inodes_sb); */ void sync_inodes_sb(struct super_block *sb) { - DECLARE_COMPLETION_ONSTACK(done); + DEFINE_WB_COMPLETION_ONSTACK(done); struct wb_writeback_work work = { .sb = sb, .sync_mode = WB_SYNC_ALL, @@ -1618,7 +1652,7 @@ void sync_inodes_sb(struct super_block *sb) WARN_ON(!rwsem_is_locked(&sb->s_umount)); wb_queue_work(&bdi->wb, &work); - wait_for_completion(&done); + wb_wait_for_completion(bdi, &done); wait_sb_inodes(sb); } diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index 8c857d723023..97a92fa0cdb5 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -155,6 +155,8 @@ struct backing_dev_info { struct rb_root cgwb_congested_tree; /* their congested states */ atomic_t usage_cnt; /* counts both cgwbs and cgwb_contested's */ #endif + wait_queue_head_t wb_waitq; + struct device *dev; struct timer_list laptop_mode_wb_timer; diff --git a/mm/backing-dev.c b/mm/backing-dev.c index d2f16fc9069a..ad5608d01e8c 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -768,6 +768,7 @@ int bdi_init(struct backing_dev_info *bdi) bdi->max_ratio = 100; bdi->max_prop_frac = FPROP_FRAC_BASE; INIT_LIST_HEAD(&bdi->bdi_list); + init_waitqueue_head(&bdi->wb_waitq); err = wb_init(&bdi->wb, bdi, GFP_KERNEL); if (err) From 98754bf770c0de4be8242cecba53fae5eeedbebc Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:13:59 -0400 Subject: [PATCH 45/85] writeback: implement wb_wait_for_single_work() For cgroup writeback, multiple wb_writeback_work items may need to be issuedto accomplish a single task. The previous patch updated the waiting mechanism such that wb_wait_for_completion() can wait for multiple work items. Issuing mulitple work items involves memory allocation which may fail. As most writeback operations can't fail or blocked on memory allocation, in such cases, we'll fall back to sequential issuing of an on-stack work item, which would need to be waited upon sequentially. This patch implements wb_wait_for_single_work() which waits for a single work item independently from wb_completion waiting so that such fallback mechanism can be used without getting tangled with the usual issuing / completion operation. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 47 +++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 45 insertions(+), 2 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index d7d4a1bcdd2f..093b9594e846 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -52,6 +52,8 @@ struct wb_writeback_work { unsigned int for_background:1; unsigned int for_sync:1; /* sync(2) WB_SYNC_ALL writeback */ unsigned int auto_free:1; /* free on completion */ + unsigned int single_wait:1; + unsigned int single_done:1; enum wb_reason reason; /* why was writeback initiated? */ struct list_head list; /* pending work list */ @@ -178,8 +180,11 @@ static void wb_queue_work(struct bdi_writeback *wb, trace_writeback_queue(wb->bdi, work); spin_lock_bh(&wb->work_lock); - if (!test_bit(WB_registered, &wb->state)) + if (!test_bit(WB_registered, &wb->state)) { + if (work->single_wait) + work->single_done = 1; goto out_unlock; + } if (work->done) atomic_inc(&work->done->cnt); list_add_tail(&work->list, &wb->work_list); @@ -233,6 +238,32 @@ int inode_congested(struct inode *inode, int cong_bits) } EXPORT_SYMBOL_GPL(inode_congested); +/** + * wb_wait_for_single_work - wait for completion of a single bdi_writeback_work + * @bdi: bdi the work item was issued to + * @work: work item to wait for + * + * Wait for the completion of @work which was issued to one of @bdi's + * bdi_writeback's. The caller must have set @work->single_wait before + * issuing it. This wait operates independently fo + * wb_wait_for_completion() and also disables automatic freeing of @work. + */ +static void wb_wait_for_single_work(struct backing_dev_info *bdi, + struct wb_writeback_work *work) +{ + if (WARN_ON_ONCE(!work->single_wait)) + return; + + wait_event(bdi->wb_waitq, work->single_done); + + /* + * Paired with smp_wmb() in wb_do_writeback() and ensures that all + * modifications to @work prior to assertion of ->single_done is + * visible to the caller once this function returns. + */ + smp_rmb(); +} + /** * wb_split_bdi_pages - split nr_pages to write according to bandwidth * @wb: target bdi_writeback to split @nr_pages to @@ -1178,14 +1209,26 @@ static long wb_do_writeback(struct bdi_writeback *wb) set_bit(WB_writeback_running, &wb->state); while ((work = get_next_work_item(wb)) != NULL) { struct wb_completion *done = work->done; + bool need_wake_up = false; trace_writeback_exec(wb->bdi, work); wrote += wb_writeback(wb, work); - if (work->auto_free) + if (work->single_wait) { + WARN_ON_ONCE(work->auto_free); + /* paired w/ rmb in wb_wait_for_single_work() */ + smp_wmb(); + work->single_done = 1; + need_wake_up = true; + } else if (work->auto_free) { kfree(work); + } + if (done && atomic_dec_and_test(&done->cnt)) + need_wake_up = true; + + if (need_wake_up) wake_up_all(&wb->bdi->wb_waitq); } From f30a7d0cc8d9096d6728fadd0ab024e648010ec0 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:14:00 -0400 Subject: [PATCH 46/85] writeback: restructure try_writeback_inodes_sb[_nr]() try_writeback_inodes_sb_nr() wraps writeback_inodes_sb_nr() so that it handles s_umount locking and skips if writeback is already in progress. The in progress test is performed on the root wb (bdi_writeback) which isn't sufficient for cgroup writeback support. The test must be done per-wb. To prepare for the change, this patch factors out __writeback_inodes_sb_nr() from writeback_inodes_sb_nr() and adds @skip_if_busy and moves the in progress test right before queueing the wb_writeback_work. try_writeback_inodes_sb_nr() now just grabs s_umount and invokes __writeback_inodes_sb_nr() with asserted @skip_if_busy. This way, later addition of multiple wb handling can skip only the wb's which already have writeback in progress. This swaps the order between in progress test and s_umount test which can flip the return value when writeback is in progress and s_umount is being held by someone else but this shouldn't cause any meaningful difference. It's a fringe condition and the return value is an unsynchronized hint anyway. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 52 ++++++++++++++++++++++----------------- include/linux/writeback.h | 6 ++--- 2 files changed, 32 insertions(+), 26 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 093b9594e846..0039c5839cdd 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1581,19 +1581,8 @@ static void wait_sb_inodes(struct super_block *sb) iput(old_inode); } -/** - * writeback_inodes_sb_nr - writeback dirty inodes from given super_block - * @sb: the superblock - * @nr: the number of pages to write - * @reason: reason why some writeback work initiated - * - * Start writeback on some inodes on this super_block. No guarantees are made - * on how many (if any) will be written, and this function does not wait - * for IO completion of submitted IO. - */ -void writeback_inodes_sb_nr(struct super_block *sb, - unsigned long nr, - enum wb_reason reason) +static void __writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr, + enum wb_reason reason, bool skip_if_busy) { DEFINE_WB_COMPLETION_ONSTACK(done); struct wb_writeback_work work = { @@ -1609,9 +1598,30 @@ void writeback_inodes_sb_nr(struct super_block *sb, if (!bdi_has_dirty_io(bdi) || bdi == &noop_backing_dev_info) return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); + + if (skip_if_busy && writeback_in_progress(&bdi->wb)) + return; + wb_queue_work(&bdi->wb, &work); wb_wait_for_completion(bdi, &done); } + +/** + * writeback_inodes_sb_nr - writeback dirty inodes from given super_block + * @sb: the superblock + * @nr: the number of pages to write + * @reason: reason why some writeback work initiated + * + * Start writeback on some inodes on this super_block. No guarantees are made + * on how many (if any) will be written, and this function does not wait + * for IO completion of submitted IO. + */ +void writeback_inodes_sb_nr(struct super_block *sb, + unsigned long nr, + enum wb_reason reason) +{ + __writeback_inodes_sb_nr(sb, nr, reason, false); +} EXPORT_SYMBOL(writeback_inodes_sb_nr); /** @@ -1638,19 +1648,15 @@ EXPORT_SYMBOL(writeback_inodes_sb); * Invoke writeback_inodes_sb_nr if no writeback is currently underway. * Returns 1 if writeback was started, 0 if not. */ -int try_to_writeback_inodes_sb_nr(struct super_block *sb, - unsigned long nr, - enum wb_reason reason) +bool try_to_writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr, + enum wb_reason reason) { - if (writeback_in_progress(&sb->s_bdi->wb)) - return 1; - if (!down_read_trylock(&sb->s_umount)) - return 0; + return false; - writeback_inodes_sb_nr(sb, nr, reason); + __writeback_inodes_sb_nr(sb, nr, reason, true); up_read(&sb->s_umount); - return 1; + return true; } EXPORT_SYMBOL(try_to_writeback_inodes_sb_nr); @@ -1662,7 +1668,7 @@ EXPORT_SYMBOL(try_to_writeback_inodes_sb_nr); * Implement by try_to_writeback_inodes_sb_nr() * Returns 1 if writeback was started, 0 if not. */ -int try_to_writeback_inodes_sb(struct super_block *sb, enum wb_reason reason) +bool try_to_writeback_inodes_sb(struct super_block *sb, enum wb_reason reason) { return try_to_writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason); } diff --git a/include/linux/writeback.h b/include/linux/writeback.h index a6b9db7fcee8..23af355d5471 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -93,9 +93,9 @@ struct bdi_writeback; void writeback_inodes_sb(struct super_block *, enum wb_reason reason); void writeback_inodes_sb_nr(struct super_block *, unsigned long nr, enum wb_reason reason); -int try_to_writeback_inodes_sb(struct super_block *, enum wb_reason reason); -int try_to_writeback_inodes_sb_nr(struct super_block *, unsigned long nr, - enum wb_reason reason); +bool try_to_writeback_inodes_sb(struct super_block *, enum wb_reason reason); +bool try_to_writeback_inodes_sb_nr(struct super_block *, unsigned long nr, + enum wb_reason reason); void sync_inodes_sb(struct super_block *); void wakeup_flusher_threads(long nr_pages, enum wb_reason reason); void inode_wait_for_writeback(struct inode *inode); From db125360409fc320790e0d38218bd7a3e9cd2552 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:14:01 -0400 Subject: [PATCH 47/85] writeback: make writeback initiation functions handle multiple bdi_writeback's [try_]writeback_inodes_sb[_nr]() and sync_inodes_sb() currently only handle dirty inodes on the root wb (bdi_writeback) of the target bdi. This patch implements bdi_split_work_to_wbs() and use it to make these functions handle multiple wb's. bdi_split_work_to_wbs() takes a base wb_writeback_work and create clones of it and issue them to the wb's of the target bdi. The base work's nr_pages is distributed using wb_split_bdi_pages() - ie. according to each wb's write bandwidth's proportion in the bdi. Cloning a bdi involves memory allocation which may fail. In such cases, bdi_split_work_to_wbs() issues the base work directly and waits for its completion before proceeding to the next wb to guarantee forward progress and correctness under memory pressure. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 96 ++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 91 insertions(+), 5 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 0039c5839cdd..59d76f6c9d8d 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -292,6 +292,80 @@ static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages) return DIV_ROUND_UP_ULL((u64)nr_pages * this_bw, tot_bw); } +/** + * wb_clone_and_queue_work - clone a wb_writeback_work and issue it to a wb + * @wb: target bdi_writeback + * @base_work: source wb_writeback_work + * + * Try to make a clone of @base_work and issue it to @wb. If cloning + * succeeds, %true is returned; otherwise, @base_work is issued directly + * and %false is returned. In the latter case, the caller is required to + * wait for @base_work's completion using wb_wait_for_single_work(). + * + * A clone is auto-freed on completion. @base_work never is. + */ +static bool wb_clone_and_queue_work(struct bdi_writeback *wb, + struct wb_writeback_work *base_work) +{ + struct wb_writeback_work *work; + + work = kmalloc(sizeof(*work), GFP_ATOMIC); + if (work) { + *work = *base_work; + work->auto_free = 1; + work->single_wait = 0; + } else { + work = base_work; + work->auto_free = 0; + work->single_wait = 1; + } + work->single_done = 0; + wb_queue_work(wb, work); + return work != base_work; +} + +/** + * bdi_split_work_to_wbs - split a wb_writeback_work to all wb's of a bdi + * @bdi: target backing_dev_info + * @base_work: wb_writeback_work to issue + * @skip_if_busy: skip wb's which already have writeback in progress + * + * Split and issue @base_work to all wb's (bdi_writeback's) of @bdi which + * have dirty inodes. If @base_work->nr_page isn't %LONG_MAX, it's + * distributed to the busy wbs according to each wb's proportion in the + * total active write bandwidth of @bdi. + */ +static void bdi_split_work_to_wbs(struct backing_dev_info *bdi, + struct wb_writeback_work *base_work, + bool skip_if_busy) +{ + long nr_pages = base_work->nr_pages; + int next_blkcg_id = 0; + struct bdi_writeback *wb; + struct wb_iter iter; + + might_sleep(); + + if (!bdi_has_dirty_io(bdi)) + return; +restart: + rcu_read_lock(); + bdi_for_each_wb(wb, bdi, &iter, next_blkcg_id) { + if (!wb_has_dirty_io(wb) || + (skip_if_busy && writeback_in_progress(wb))) + continue; + + base_work->nr_pages = wb_split_bdi_pages(wb, nr_pages); + if (!wb_clone_and_queue_work(wb, base_work)) { + next_blkcg_id = wb->blkcg_css->id + 1; + rcu_read_unlock(); + wb_wait_for_single_work(bdi, base_work); + goto restart; + } + } + rcu_read_unlock(); +} + #else /* CONFIG_CGROUP_WRITEBACK */ static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages) @@ -299,6 +373,21 @@ static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages) return nr_pages; } +static void bdi_split_work_to_wbs(struct backing_dev_info *bdi, + struct wb_writeback_work *base_work, + bool skip_if_busy) +{ + might_sleep(); + + if (bdi_has_dirty_io(bdi) && + (!skip_if_busy || !writeback_in_progress(&bdi->wb))) { + base_work->auto_free = 0; + base_work->single_wait = 0; + base_work->single_done = 0; + wb_queue_work(&bdi->wb, base_work); + } +} + #endif /* CONFIG_CGROUP_WRITEBACK */ void wb_start_writeback(struct bdi_writeback *wb, long nr_pages, @@ -1599,10 +1688,7 @@ static void __writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr, return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); - if (skip_if_busy && writeback_in_progress(&bdi->wb)) - return; - - wb_queue_work(&bdi->wb, &work); + bdi_split_work_to_wbs(sb->s_bdi, &work, skip_if_busy); wb_wait_for_completion(bdi, &done); } @@ -1700,7 +1786,7 @@ void sync_inodes_sb(struct super_block *sb) return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); - wb_queue_work(&bdi->wb, &work); + bdi_split_work_to_wbs(bdi, &work, false); wb_wait_for_completion(bdi, &done); wait_sb_inodes(sb); From 0747259d13febfcc838980a63c414c9b920cea6f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:14:02 -0400 Subject: [PATCH 48/85] writeback: dirty inodes against their matching cgroup bdi_writeback's __mark_inode_dirty() always dirtied the inode against the root wb (bdi_writeback). The previous patches added all the infrastructure necessary to attribute an inode against the wb of the dirtying cgroup. This patch updates __mark_inode_dirty() so that it uses the wb associated with the inode instead of unconditionally using the root one. Currently, none of the filesystems has FS_CGROUP_WRITEBACK and all pages will keep being dirtied against the root wb. v2: Updated for per-inode wb association. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 59d76f6c9d8d..881ea5d97c00 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1504,7 +1504,6 @@ static noinline void block_dump___mark_inode_dirty(struct inode *inode) void __mark_inode_dirty(struct inode *inode, int flags) { struct super_block *sb = inode->i_sb; - struct backing_dev_info *bdi = NULL; int dirtytime; trace_writeback_mark_inode_dirty(inode, flags); @@ -1574,30 +1573,30 @@ void __mark_inode_dirty(struct inode *inode, int flags) * reposition it (that would break b_dirty time-ordering). */ if (!was_dirty) { + struct bdi_writeback *wb = inode_to_wb(inode); struct list_head *dirty_list; bool wakeup_bdi = false; - bdi = inode_to_bdi(inode); spin_unlock(&inode->i_lock); - spin_lock(&bdi->wb.list_lock); + spin_lock(&wb->list_lock); - WARN(bdi_cap_writeback_dirty(bdi) && - !test_bit(WB_registered, &bdi->wb.state), - "bdi-%s not registered\n", bdi->name); + WARN(bdi_cap_writeback_dirty(wb->bdi) && + !test_bit(WB_registered, &wb->state), + "bdi-%s not registered\n", wb->bdi->name); inode->dirtied_when = jiffies; if (dirtytime) inode->dirtied_time_when = jiffies; if (inode->i_state & (I_DIRTY_INODE | I_DIRTY_PAGES)) - dirty_list = &bdi->wb.b_dirty; + dirty_list = &wb->b_dirty; else - dirty_list = &bdi->wb.b_dirty_time; + dirty_list = &wb->b_dirty_time; - wakeup_bdi = inode_wb_list_move_locked(inode, &bdi->wb, + wakeup_bdi = inode_wb_list_move_locked(inode, wb, dirty_list); - spin_unlock(&bdi->wb.list_lock); + spin_unlock(&wb->list_lock); trace_writeback_dirty_inode_enqueue(inode); /* @@ -1606,8 +1605,8 @@ void __mark_inode_dirty(struct inode *inode, int flags) * to make sure background write-back happens * later. */ - if (bdi_cap_writeback_dirty(bdi) && wakeup_bdi) - wb_wakeup_delayed(&bdi->wb); + if (bdi_cap_writeback_dirty(wb->bdi) && wakeup_bdi) + wb_wakeup_delayed(wb); return; } } From bafc0dba1e20d84578d7098d32caf63441e5743d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 2 Jun 2015 08:37:23 -0600 Subject: [PATCH 49/85] buffer, writeback: make __block_write_full_page() honor cgroup writeback [__]block_write_full_page() is used to implement ->writepage in various filesystems. All writeback logic is now updated to handle cgroup writeback and the block cgroup to issue IOs for is encoded in writeback_control and can be retrieved from the inode; however, [__]block_write_full_page() currently ignores the blkcg indicated by inode and issues all bio's without explicit blkcg association. This patch adds submit_bh_blkcg() which associates the bio with the specified blkio cgroup before issuing and uses it in __block_write_full_page() so that the issued bio's are associated with inode_to_wb_blkcg_css(inode). v2: Updated for per-inode wb association. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Andrew Morton Signed-off-by: Jens Axboe --- fs/buffer.c | 26 ++++++++++++++++++++------ include/linux/backing-dev.h | 12 ++++++++++++ 2 files changed, 32 insertions(+), 6 deletions(-) diff --git a/fs/buffer.c b/fs/buffer.c index 23b640d5d6e9..b85e94134ea6 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include @@ -44,6 +45,9 @@ #include static int fsync_buffers_list(spinlock_t *lock, struct list_head *list); +static int submit_bh_blkcg(int rw, struct buffer_head *bh, + unsigned long bio_flags, + struct cgroup_subsys_state *blkcg_css); #define BH_ENTRY(list) list_entry((list), struct buffer_head, b_assoc_buffers) @@ -1704,8 +1708,8 @@ static int __block_write_full_page(struct inode *inode, struct page *page, struct buffer_head *bh, *head; unsigned int blocksize, bbits; int nr_underway = 0; - int write_op = (wbc->sync_mode == WB_SYNC_ALL ? - WRITE_SYNC : WRITE); + int write_op = (wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE); + struct cgroup_subsys_state *blkcg_css = inode_to_wb_blkcg_css(inode); head = create_page_buffers(page, inode, (1 << BH_Dirty)|(1 << BH_Uptodate)); @@ -1794,7 +1798,7 @@ static int __block_write_full_page(struct inode *inode, struct page *page, do { struct buffer_head *next = bh->b_this_page; if (buffer_async_write(bh)) { - submit_bh(write_op, bh); + submit_bh_blkcg(write_op, bh, 0, blkcg_css); nr_underway++; } bh = next; @@ -1848,7 +1852,7 @@ recover: struct buffer_head *next = bh->b_this_page; if (buffer_async_write(bh)) { clear_buffer_dirty(bh); - submit_bh(write_op, bh); + submit_bh_blkcg(write_op, bh, 0, blkcg_css); nr_underway++; } bh = next; @@ -3013,7 +3017,9 @@ void guard_bio_eod(int rw, struct bio *bio) } } -int _submit_bh(int rw, struct buffer_head *bh, unsigned long bio_flags) +static int submit_bh_blkcg(int rw, struct buffer_head *bh, + unsigned long bio_flags, + struct cgroup_subsys_state *blkcg_css) { struct bio *bio; @@ -3035,6 +3041,9 @@ int _submit_bh(int rw, struct buffer_head *bh, unsigned long bio_flags) */ bio = bio_alloc(GFP_NOIO, 1); + if (blkcg_css) + bio_associate_blkcg(bio, blkcg_css); + bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); bio->bi_bdev = bh->b_bdev; bio->bi_io_vec[0].bv_page = bh->b_page; @@ -3059,11 +3068,16 @@ int _submit_bh(int rw, struct buffer_head *bh, unsigned long bio_flags) submit_bio(rw, bio); return 0; } + +int _submit_bh(int rw, struct buffer_head *bh, unsigned long bio_flags) +{ + return submit_bh_blkcg(rw, bh, bio_flags, NULL); +} EXPORT_SYMBOL_GPL(_submit_bh); int submit_bh(int rw, struct buffer_head *bh) { - return _submit_bh(rw, bh, 0); + return submit_bh_blkcg(rw, bh, 0, NULL); } EXPORT_SYMBOL(submit_bh); diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 9cc11e5b97ca..e9d7373f5f93 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -393,6 +393,12 @@ static inline struct bdi_writeback *inode_to_wb(struct inode *inode) return inode->i_wb; } +static inline struct cgroup_subsys_state * +inode_to_wb_blkcg_css(struct inode *inode) +{ + return inode_to_wb(inode)->blkcg_css; +} + struct wb_iter { int start_blkcg_id; struct radix_tree_iter tree_iter; @@ -510,6 +516,12 @@ static inline void wb_blkcg_offline(struct blkcg *blkcg) { } +static inline struct cgroup_subsys_state * +inode_to_wb_blkcg_css(struct inode *inode) +{ + return blkcg_root_css; +} + struct wb_iter { int next_id; }; From 429b3fb027492b2b9f96e00bb1a9255fc56d4934 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:14:04 -0400 Subject: [PATCH 50/85] mpage: make __mpage_writepage() honor cgroup writeback __mpage_writepage() is used to implement mpage_writepages() which in turn is used for ->writepages() of various filesystems. All writeback logic is now updated to handle cgroup writeback and the block cgroup to issue IOs for is encoded in writeback_control and can be retrieved from the inode; however, __mpage_writepage() currently ignores the blkcg indicated by the inode and issues all bio's without explicit blkcg association. This patch updates __mpage_writepage() so that the issued bio's are associated with inode_to_writeback_blkcg_css(inode). v2: Updated for per-inode wb association. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Andrew Morton Cc: Alexander Viro Signed-off-by: Jens Axboe --- fs/mpage.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/mpage.c b/fs/mpage.c index 3e79220babac..a3ccb0bd465a 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -605,6 +605,8 @@ alloc_new: bio_get_nr_vecs(bdev), GFP_NOFS|__GFP_HIGH); if (bio == NULL) goto confused; + + bio_associate_blkcg(bio, inode_to_wb_blkcg_css(inode)); } /* From 108dad65be9786650c78ed9aea28c9b4a68810d9 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 17:14:05 -0400 Subject: [PATCH 51/85] ext2: enable cgroup writeback support Writeback now supports cgroup writeback and the generic writeback, buffer, libfs, and mpage helpers that ext2 uses are all updated to work with cgroup writeback. This patch enables cgroup writeback for ext2 by adding FS_CGROUP_WRITEBACK to its ->fs_flags. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: linux-ext4@vger.kernel.org Signed-off-by: Jens Axboe --- fs/ext2/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ext2/super.c b/fs/ext2/super.c index d0e746e96511..549219d44f1c 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -1543,7 +1543,7 @@ static struct file_system_type ext2_fs_type = { .name = "ext2", .mount = ext2_mount, .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, + .fs_flags = FS_REQUIRES_DEV | FS_CGROUP_WRITEBACK, }; MODULE_ALIAS_FS("ext2"); From 733a572e66d2a23c852fdce34dba5bbd40667817 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:18 -0400 Subject: [PATCH 52/85] memcg: make mem_cgroup_read_{stat|event}() iterate possible cpus instead of online cpu_possible_mask represents the CPUs which are actually possible during that boot instance. For systems which don't support CPU hotplug, this will match cpu_online_mask exactly in most cases. Even for systems which support CPU hotplug, the number of possible CPU slots is highly unlikely to diverge greatly from the number of online CPUs. The only cases where the difference between possible and online caused problems were when the boot code failed to initialize the possible mask and left it fully set at NR_CPUS - 1. As such, most per-cpu constructs allocate for all possible CPUs and often iterate over the possibles, which also has the benefit of avoiding the blocking CPU hotplug synchronization. memcg open codes per-cpu stat counting for mem_cgroup_read_stat() and mem_cgroup_read_events(), which iterates over online CPUs and handles CPU hotplug operations explicitly. This complexity doesn't actually buy anything. Switch to iterating over the possibles and drop the explicit CPU hotplug handling. Eventually, we want to convert memcg to use percpu_counter instead of its own custom implementation which also benefits from quick access w/o summing for cases where larger error margin is acceptable. This will allow mem_cgroup_read_stat() to be called from non-sleepable contexts which will be used by cgroup writeback. Signed-off-by: Tejun Heo Cc: Michal Hocko Acked-by: Johannes Weiner Signed-off-by: Jens Axboe --- mm/memcontrol.c | 51 ++----------------------------------------------- 1 file changed, 2 insertions(+), 49 deletions(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 49e5aa6abca6..701cbee9acba 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -324,11 +324,6 @@ struct mem_cgroup { * percpu counter. */ struct mem_cgroup_stat_cpu __percpu *stat; - /* - * used when a cpu is offlined or other synchronizations - * See mem_cgroup_read_stat(). - */ - struct mem_cgroup_stat_cpu nocpu_base; spinlock_t pcp_counter_lock; #if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_INET) @@ -834,15 +829,8 @@ static long mem_cgroup_read_stat(struct mem_cgroup *memcg, long val = 0; int cpu; - get_online_cpus(); - for_each_online_cpu(cpu) + for_each_possible_cpu(cpu) val += per_cpu(memcg->stat->count[idx], cpu); -#ifdef CONFIG_HOTPLUG_CPU - spin_lock(&memcg->pcp_counter_lock); - val += memcg->nocpu_base.count[idx]; - spin_unlock(&memcg->pcp_counter_lock); -#endif - put_online_cpus(); return val; } @@ -852,15 +840,8 @@ static unsigned long mem_cgroup_read_events(struct mem_cgroup *memcg, unsigned long val = 0; int cpu; - get_online_cpus(); - for_each_online_cpu(cpu) + for_each_possible_cpu(cpu) val += per_cpu(memcg->stat->events[idx], cpu); -#ifdef CONFIG_HOTPLUG_CPU - spin_lock(&memcg->pcp_counter_lock); - val += memcg->nocpu_base.events[idx]; - spin_unlock(&memcg->pcp_counter_lock); -#endif - put_online_cpus(); return val; } @@ -2210,37 +2191,12 @@ static void drain_all_stock(struct mem_cgroup *root_memcg) mutex_unlock(&percpu_charge_mutex); } -/* - * This function drains percpu counter value from DEAD cpu and - * move it to local cpu. Note that this function can be preempted. - */ -static void mem_cgroup_drain_pcp_counter(struct mem_cgroup *memcg, int cpu) -{ - int i; - - spin_lock(&memcg->pcp_counter_lock); - for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { - long x = per_cpu(memcg->stat->count[i], cpu); - - per_cpu(memcg->stat->count[i], cpu) = 0; - memcg->nocpu_base.count[i] += x; - } - for (i = 0; i < MEM_CGROUP_EVENTS_NSTATS; i++) { - unsigned long x = per_cpu(memcg->stat->events[i], cpu); - - per_cpu(memcg->stat->events[i], cpu) = 0; - memcg->nocpu_base.events[i] += x; - } - spin_unlock(&memcg->pcp_counter_lock); -} - static int memcg_cpu_hotplug_callback(struct notifier_block *nb, unsigned long action, void *hcpu) { int cpu = (unsigned long)hcpu; struct memcg_stock_pcp *stock; - struct mem_cgroup *iter; if (action == CPU_ONLINE) return NOTIFY_OK; @@ -2248,9 +2204,6 @@ static int memcg_cpu_hotplug_callback(struct notifier_block *nb, if (action != CPU_DEAD && action != CPU_DEAD_FROZEN) return NOTIFY_OK; - for_each_mem_cgroup(iter) - mem_cgroup_drain_pcp_counter(iter, cpu); - stock = &per_cpu(memcg_stock, cpu); drain_stock(stock); return NOTIFY_OK; From 0d960a383ae7aa791b2833e122ba7519d264cf92 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:19 -0400 Subject: [PATCH 53/85] writeback: clean up wb_dirty_limit() The function name wb_dirty_limit(), its argument @dirty and the local variable @wb_dirty are mortally confusing given that the function calculates per-wb threshold value not dirty pages, especially given that @dirty and @wb_dirty are used elsewhere for dirty pages. Let's rename the function to wb_calc_thresh() and wb_dirty to wb_thresh. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 2 +- include/linux/writeback.h | 2 +- mm/backing-dev.c | 6 +++--- mm/page-writeback.c | 30 +++++++++++++++--------------- 4 files changed, 20 insertions(+), 20 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 881ea5d97c00..b1b3b8184500 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1081,7 +1081,7 @@ static bool over_bground_thresh(struct bdi_writeback *wb) global_page_state(NR_UNSTABLE_NFS) > background_thresh) return true; - if (wb_stat(wb, WB_RECLAIMABLE) > wb_dirty_limit(wb, background_thresh)) + if (wb_stat(wb, WB_RECLAIMABLE) > wb_calc_thresh(wb, background_thresh)) return true; return false; diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 23af355d5471..0435c85d4cfa 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -155,7 +155,7 @@ int dirty_writeback_centisecs_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *); void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty); -unsigned long wb_dirty_limit(struct bdi_writeback *wb, unsigned long dirty); +unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh); void __wb_update_bandwidth(struct bdi_writeback *wb, unsigned long thresh, diff --git a/mm/backing-dev.c b/mm/backing-dev.c index ad5608d01e8c..9c8b7b5a0eee 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -49,7 +49,7 @@ static int bdi_debug_stats_show(struct seq_file *m, void *v) struct bdi_writeback *wb = &bdi->wb; unsigned long background_thresh; unsigned long dirty_thresh; - unsigned long bdi_thresh; + unsigned long wb_thresh; unsigned long nr_dirty, nr_io, nr_more_io, nr_dirty_time; struct inode *inode; @@ -67,7 +67,7 @@ static int bdi_debug_stats_show(struct seq_file *m, void *v) spin_unlock(&wb->list_lock); global_dirty_limits(&background_thresh, &dirty_thresh); - bdi_thresh = wb_dirty_limit(wb, dirty_thresh); + wb_thresh = wb_calc_thresh(wb, dirty_thresh); #define K(x) ((x) << (PAGE_SHIFT - 10)) seq_printf(m, @@ -87,7 +87,7 @@ static int bdi_debug_stats_show(struct seq_file *m, void *v) "state: %10lx\n", (unsigned long) K(wb_stat(wb, WB_WRITEBACK)), (unsigned long) K(wb_stat(wb, WB_RECLAIMABLE)), - K(bdi_thresh), + K(wb_thresh), K(dirty_thresh), K(background_thresh), (unsigned long) K(wb_stat(wb, WB_DIRTIED)), diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 70cf98dc3423..c7745a7fe11e 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -556,7 +556,7 @@ static unsigned long hard_dirty_limit(unsigned long thresh) } /** - * wb_dirty_limit - @wb's share of dirty throttling threshold + * wb_calc_thresh - @wb's share of dirty throttling threshold * @wb: bdi_writeback to query * @dirty: global dirty limit in pages * @@ -577,28 +577,28 @@ static unsigned long hard_dirty_limit(unsigned long thresh) * The wb's share of dirty limit will be adapting to its throughput and * bounded by the bdi->min_ratio and/or bdi->max_ratio parameters, if set. */ -unsigned long wb_dirty_limit(struct bdi_writeback *wb, unsigned long dirty) +unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh) { - u64 wb_dirty; + u64 wb_thresh; long numerator, denominator; unsigned long wb_min_ratio, wb_max_ratio; /* - * Calculate this BDI's share of the dirty ratio. + * Calculate this BDI's share of the thresh ratio. */ wb_writeout_fraction(wb, &numerator, &denominator); - wb_dirty = (dirty * (100 - bdi_min_ratio)) / 100; - wb_dirty *= numerator; - do_div(wb_dirty, denominator); + wb_thresh = (thresh * (100 - bdi_min_ratio)) / 100; + wb_thresh *= numerator; + do_div(wb_thresh, denominator); wb_min_max_ratio(wb, &wb_min_ratio, &wb_max_ratio); - wb_dirty += (dirty * wb_min_ratio) / 100; - if (wb_dirty > (dirty * wb_max_ratio) / 100) - wb_dirty = dirty * wb_max_ratio / 100; + wb_thresh += (thresh * wb_min_ratio) / 100; + if (wb_thresh > (thresh * wb_max_ratio) / 100) + wb_thresh = thresh * wb_max_ratio / 100; - return wb_dirty; + return wb_thresh; } /* @@ -750,7 +750,7 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, * total amount of RAM is 16GB, bdi->max_ratio is equal to 1%, global * limits are set by default to 10% and 20% (background and throttle). * Then wb_thresh is 1% of 20% of 16GB. This amounts to ~8K pages. - * wb_dirty_limit(wb, bg_thresh) is about ~4K pages. wb_setpoint is + * wb_calc_thresh(wb, bg_thresh) is about ~4K pages. wb_setpoint is * about ~6K pages (as the average of background and throttle wb * limits). The 3rd order polynomial will provide positive feedback if * wb_dirty is under wb_setpoint and vice versa. @@ -1115,7 +1115,7 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, * * We rampup dirty_ratelimit forcibly if wb_dirty is low because * it's possible that wb_thresh is close to zero due to inactivity - * of backing device (see the implementation of wb_dirty_limit()). + * of backing device (see the implementation of wb_calc_thresh()). */ if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { dirty = wb_dirty; @@ -1123,7 +1123,7 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, setpoint = wb_dirty + 1; else setpoint = (wb_thresh + - wb_dirty_limit(wb, bg_thresh)) / 2; + wb_calc_thresh(wb, bg_thresh)) / 2; } if (dirty < setpoint) { @@ -1352,7 +1352,7 @@ static inline void wb_dirty_limits(struct bdi_writeback *wb, * wb_position_ratio() will let the dirtier task progress * at some rate <= (write_bw / 2) for bringing down wb_dirty. */ - *wb_thresh = wb_dirty_limit(wb, dirty_thresh); + *wb_thresh = wb_calc_thresh(wb, dirty_thresh); if (wb_bg_thresh) *wb_bg_thresh = dirty_thresh ? div_u64((u64)*wb_thresh * From 8a73179956e649df0d4b3250db17734f272d8266 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:20 -0400 Subject: [PATCH 54/85] writeback: reorganize [__]wb_update_bandwidth() __wb_update_bandwidth() is called from two places - fs/fs-writeback.c::balance_dirty_pages() and mm/page-writeback.c::wb_writeback(). The latter updates only the write bandwidth while the former also deals with the dirty ratelimit. The two callsites are distinguished by whether @thresh parameter is zero or not, which is cryptic. In addition, the two files define their own different versions of wb_update_bandwidth() on top of __wb_update_bandwidth(), which is confusing to say the least. This patch cleans up [__]wb_update_bandwidth() in the following ways. * __wb_update_bandwidth() now takes explicit @update_ratelimit parameter to gate dirty ratelimit handling. * mm/page-writeback.c::wb_update_bandwidth() is flattened into its caller - balance_dirty_pages(). * fs/fs-writeback.c::wb_update_bandwidth() is moved to mm/page-writeback.c and __wb_update_bandwidth() is made static. * While at it, add a lockdep assertion to __wb_update_bandwidth(). Except for the lockdep addition, this is pure reorganization and doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 10 --------- include/linux/writeback.h | 9 +------- mm/page-writeback.c | 45 +++++++++++++++++++-------------------- 3 files changed, 23 insertions(+), 41 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index b1b3b8184500..cd89484486f6 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1087,16 +1087,6 @@ static bool over_bground_thresh(struct bdi_writeback *wb) return false; } -/* - * Called under wb->list_lock. If there are multiple wb per bdi, - * only the flusher working on the first wb should do it. - */ -static void wb_update_bandwidth(struct bdi_writeback *wb, - unsigned long start_time) -{ - __wb_update_bandwidth(wb, 0, 0, 0, 0, 0, start_time); -} - /* * Explicit flushing or periodic writeback of "old" data. * diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 0435c85d4cfa..80adf3d88d9d 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -157,14 +157,7 @@ int dirty_writeback_centisecs_handler(struct ctl_table *, int, void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty); unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh); -void __wb_update_bandwidth(struct bdi_writeback *wb, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long bdi_thresh, - unsigned long bdi_dirty, - unsigned long start_time); - +void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long start_time); void page_writeback_init(void); void balance_dirty_pages_ratelimited(struct address_space *mapping); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index c7745a7fe11e..bebdd41b8d8e 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1160,19 +1160,22 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, trace_bdi_dirty_ratelimit(wb->bdi, dirty_rate, task_ratelimit); } -void __wb_update_bandwidth(struct bdi_writeback *wb, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long wb_thresh, - unsigned long wb_dirty, - unsigned long start_time) +static void __wb_update_bandwidth(struct bdi_writeback *wb, + unsigned long thresh, + unsigned long bg_thresh, + unsigned long dirty, + unsigned long wb_thresh, + unsigned long wb_dirty, + unsigned long start_time, + bool update_ratelimit) { unsigned long now = jiffies; unsigned long elapsed = now - wb->bw_time_stamp; unsigned long dirtied; unsigned long written; + lockdep_assert_held(&wb->list_lock); + /* * rate-limit, only update once every 200ms. */ @@ -1189,7 +1192,7 @@ void __wb_update_bandwidth(struct bdi_writeback *wb, if (elapsed > HZ && time_before(wb->bw_time_stamp, start_time)) goto snapshot; - if (thresh) { + if (update_ratelimit) { global_update_bandwidth(thresh, dirty, now); wb_update_dirty_ratelimit(wb, thresh, bg_thresh, dirty, wb_thresh, wb_dirty, @@ -1203,20 +1206,9 @@ snapshot: wb->bw_time_stamp = now; } -static void wb_update_bandwidth(struct bdi_writeback *wb, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long wb_thresh, - unsigned long wb_dirty, - unsigned long start_time) +void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long start_time) { - if (time_is_after_eq_jiffies(wb->bw_time_stamp + BANDWIDTH_INTERVAL)) - return; - spin_lock(&wb->list_lock); - __wb_update_bandwidth(wb, thresh, bg_thresh, dirty, - wb_thresh, wb_dirty, start_time); - spin_unlock(&wb->list_lock); + __wb_update_bandwidth(wb, 0, 0, 0, 0, 0, start_time, false); } /* @@ -1467,8 +1459,15 @@ static void balance_dirty_pages(struct address_space *mapping, if (dirty_exceeded && !wb->dirty_exceeded) wb->dirty_exceeded = 1; - wb_update_bandwidth(wb, dirty_thresh, background_thresh, - nr_dirty, wb_thresh, wb_dirty, start_time); + if (time_is_before_jiffies(wb->bw_time_stamp + + BANDWIDTH_INTERVAL)) { + spin_lock(&wb->list_lock); + __wb_update_bandwidth(wb, dirty_thresh, + background_thresh, nr_dirty, + wb_thresh, wb_dirty, start_time, + true); + spin_unlock(&wb->list_lock); + } dirty_ratelimit = wb->dirty_ratelimit; pos_ratio = wb_position_ratio(wb, dirty_thresh, From 380c27ca33ebecc9da35aa90c8b3a9154f90aac2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:21 -0400 Subject: [PATCH 55/85] writeback: implement wb_domain Dirtyable memory is distributed to a wb (bdi_writeback) according to the relative bandwidth the wb is writing out in the whole system. This distribution is global - each wb is measured against all other wb's and gets the proportinately sized portion of the memory in the whole system. For cgroup writeback, the amount of dirtyable memory is scoped by memcg and thus each wb would need to be measured and controlled in its memcg. IOW, a wb will belong to two writeback domains - the global and memcg domains. Currently, what constitutes the global writeback domain are scattered across a number of global states. This patch starts collecting them into struct wb_domain. * fprop_global which serves as the basis for proportional bandwidth measurement and its period timer are moved into struct wb_domain. * global_wb_domain hosts the states for the global domain. * While at it, flatten wb_writeout_fraction() into its callers. This thin wrapper doesn't provide any actual benefits while getting in the way. This is pure reorganization and doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- include/linux/writeback.h | 32 +++++++++++++++++ mm/page-writeback.c | 72 +++++++++++++++------------------------ 2 files changed, 59 insertions(+), 45 deletions(-) diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 80adf3d88d9d..3148db1296a2 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -7,6 +7,7 @@ #include #include #include +#include DECLARE_PER_CPU(int, dirty_throttle_leaks); @@ -86,6 +87,36 @@ struct writeback_control { unsigned for_sync:1; /* sync(2) WB_SYNC_ALL writeback */ }; +/* + * A wb_domain represents a domain that wb's (bdi_writeback's) belong to + * and are measured against each other in. There always is one global + * domain, global_wb_domain, that every wb in the system is a member of. + * This allows measuring the relative bandwidth of each wb to distribute + * dirtyable memory accordingly. + */ +struct wb_domain { + /* + * Scale the writeback cache size proportional to the relative + * writeout speed. + * + * We do this by keeping a floating proportion between BDIs, based + * on page writeback completions [end_page_writeback()]. Those + * devices that write out pages fastest will get the larger share, + * while the slower will get a smaller share. + * + * We use page writeout completions because we are interested in + * getting rid of dirty pages. Having them written out is the + * primary goal. + * + * We introduce a concept of time, a period over which we measure + * these events, because demand can/will vary over time. The length + * of this period itself is measured in page writeback completions. + */ + struct fprop_global completions; + struct timer_list period_timer; /* timer for aging of completions */ + unsigned long period_time; +}; + /* * fs/fs-writeback.c */ @@ -120,6 +151,7 @@ static inline void laptop_sync_completion(void) { } #endif void throttle_vm_writeout(gfp_t gfp_mask); bool zone_dirty_ok(struct zone *zone); +int wb_domain_init(struct wb_domain *dom, gfp_t gfp); extern unsigned long global_dirty_limit; diff --git a/mm/page-writeback.c b/mm/page-writeback.c index bebdd41b8d8e..08e1737edb39 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -124,29 +124,7 @@ EXPORT_SYMBOL(laptop_mode); unsigned long global_dirty_limit; -/* - * Scale the writeback cache size proportional to the relative writeout speeds. - * - * We do this by keeping a floating proportion between BDIs, based on page - * writeback completions [end_page_writeback()]. Those devices that write out - * pages fastest will get the larger share, while the slower will get a smaller - * share. - * - * We use page writeout completions because we are interested in getting rid of - * dirty pages. Having them written out is the primary goal. - * - * We introduce a concept of time, a period over which we measure these events, - * because demand can/will vary over time. The length of this period itself is - * measured in page writeback completions. - * - */ -static struct fprop_global writeout_completions; - -static void writeout_period(unsigned long t); -/* Timer for aging of writeout_completions */ -static struct timer_list writeout_period_timer = - TIMER_DEFERRED_INITIALIZER(writeout_period, 0, 0); -static unsigned long writeout_period_time = 0; +static struct wb_domain global_wb_domain; /* * Length of period for aging writeout fractions of bdis. This is an @@ -433,24 +411,26 @@ static unsigned long wp_next_time(unsigned long cur_time) } /* - * Increment the BDI's writeout completion count and the global writeout + * Increment the wb's writeout completion count and the global writeout * completion count. Called from test_clear_page_writeback(). */ static inline void __wb_writeout_inc(struct bdi_writeback *wb) { + struct wb_domain *dom = &global_wb_domain; + __inc_wb_stat(wb, WB_WRITTEN); - __fprop_inc_percpu_max(&writeout_completions, &wb->completions, + __fprop_inc_percpu_max(&dom->completions, &wb->completions, wb->bdi->max_prop_frac); /* First event after period switching was turned off? */ - if (!unlikely(writeout_period_time)) { + if (!unlikely(dom->period_time)) { /* * We can race with other __bdi_writeout_inc calls here but * it does not cause any harm since the resulting time when * timer will fire and what is in writeout_period_time will be * roughly the same. */ - writeout_period_time = wp_next_time(jiffies); - mod_timer(&writeout_period_timer, writeout_period_time); + dom->period_time = wp_next_time(jiffies); + mod_timer(&dom->period_timer, dom->period_time); } } @@ -464,38 +444,38 @@ void wb_writeout_inc(struct bdi_writeback *wb) } EXPORT_SYMBOL_GPL(wb_writeout_inc); -/* - * Obtain an accurate fraction of the BDI's portion. - */ -static void wb_writeout_fraction(struct bdi_writeback *wb, - long *numerator, long *denominator) -{ - fprop_fraction_percpu(&writeout_completions, &wb->completions, - numerator, denominator); -} - /* * On idle system, we can be called long after we scheduled because we use * deferred timers so count with missed periods. */ static void writeout_period(unsigned long t) { - int miss_periods = (jiffies - writeout_period_time) / + struct wb_domain *dom = (void *)t; + int miss_periods = (jiffies - dom->period_time) / VM_COMPLETIONS_PERIOD_LEN; - if (fprop_new_period(&writeout_completions, miss_periods + 1)) { - writeout_period_time = wp_next_time(writeout_period_time + + if (fprop_new_period(&dom->completions, miss_periods + 1)) { + dom->period_time = wp_next_time(dom->period_time + miss_periods * VM_COMPLETIONS_PERIOD_LEN); - mod_timer(&writeout_period_timer, writeout_period_time); + mod_timer(&dom->period_timer, dom->period_time); } else { /* * Aging has zeroed all fractions. Stop wasting CPU on period * updates. */ - writeout_period_time = 0; + dom->period_time = 0; } } +int wb_domain_init(struct wb_domain *dom, gfp_t gfp) +{ + memset(dom, 0, sizeof(*dom)); + init_timer_deferrable(&dom->period_timer); + dom->period_timer.function = writeout_period; + dom->period_timer.data = (unsigned long)dom; + return fprop_global_init(&dom->completions, gfp); +} + /* * bdi_min_ratio keeps the sum of the minimum dirty shares of all * registered backing devices, which, for obvious reasons, can not @@ -579,6 +559,7 @@ static unsigned long hard_dirty_limit(unsigned long thresh) */ unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh) { + struct wb_domain *dom = &global_wb_domain; u64 wb_thresh; long numerator, denominator; unsigned long wb_min_ratio, wb_max_ratio; @@ -586,7 +567,8 @@ unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh) /* * Calculate this BDI's share of the thresh ratio. */ - wb_writeout_fraction(wb, &numerator, &denominator); + fprop_fraction_percpu(&dom->completions, &wb->completions, + &numerator, &denominator); wb_thresh = (thresh * (100 - bdi_min_ratio)) / 100; wb_thresh *= numerator; @@ -1831,7 +1813,7 @@ void __init page_writeback_init(void) writeback_set_ratelimit(); register_cpu_notifier(&ratelimit_nb); - fprop_global_init(&writeout_completions, GFP_KERNEL); + BUG_ON(wb_domain_init(&global_wb_domain, GFP_KERNEL)); } /** From dcc25ae76eb7b8ff883eaaab57e30e8f2f085be3 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:22 -0400 Subject: [PATCH 56/85] writeback: move global_dirty_limit into wb_domain This patch is a part of the series to define wb_domain which represents a domain that wb's (bdi_writeback's) belong to and are measured against each other in. This will enable IO backpressure propagation for cgroup writeback. global_dirty_limit exists to regulate the global dirty threshold which is a property of the wb_domain. This patch moves hard_dirty_limit, dirty_lock, and update_time into wb_domain. This is pure reorganization and doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 2 +- include/linux/writeback.h | 17 +++++++++++- include/trace/events/writeback.h | 7 ++--- mm/page-writeback.c | 46 ++++++++++++++++---------------- 4 files changed, 44 insertions(+), 28 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index cd89484486f6..51c8a5b14cdf 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -887,7 +887,7 @@ static long writeback_chunk_size(struct bdi_writeback *wb, pages = LONG_MAX; else { pages = min(wb->avg_write_bandwidth / 2, - global_dirty_limit / DIRTY_SCOPE); + global_wb_domain.dirty_limit / DIRTY_SCOPE); pages = min(pages, work->nr_pages); pages = round_down(pages + MIN_WRITEBACK_PAGES, MIN_WRITEBACK_PAGES); diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 3148db1296a2..5fdd4e1805e6 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -95,6 +95,8 @@ struct writeback_control { * dirtyable memory accordingly. */ struct wb_domain { + spinlock_t lock; + /* * Scale the writeback cache size proportional to the relative * writeout speed. @@ -115,6 +117,19 @@ struct wb_domain { struct fprop_global completions; struct timer_list period_timer; /* timer for aging of completions */ unsigned long period_time; + + /* + * The dirtyable memory and dirty threshold could be suddenly + * knocked down by a large amount (eg. on the startup of KVM in a + * swapless system). This may throw the system into deep dirty + * exceeded state and throttle heavy/light dirtiers alike. To + * retain good responsiveness, maintain global_dirty_limit for + * tracking slowly down to the knocked down dirty threshold. + * + * Both fields are protected by ->lock. + */ + unsigned long dirty_limit_tstamp; + unsigned long dirty_limit; }; /* @@ -153,7 +168,7 @@ void throttle_vm_writeout(gfp_t gfp_mask); bool zone_dirty_ok(struct zone *zone); int wb_domain_init(struct wb_domain *dom, gfp_t gfp); -extern unsigned long global_dirty_limit; +extern struct wb_domain global_wb_domain; /* These are exported to sysctl. */ extern int dirty_background_ratio; diff --git a/include/trace/events/writeback.h b/include/trace/events/writeback.h index 9b876f6cc81a..bec69995968f 100644 --- a/include/trace/events/writeback.h +++ b/include/trace/events/writeback.h @@ -361,7 +361,7 @@ TRACE_EVENT(global_dirty_state, __entry->nr_written = global_page_state(NR_WRITTEN); __entry->background_thresh = background_thresh; __entry->dirty_thresh = dirty_thresh; - __entry->dirty_limit = global_dirty_limit; + __entry->dirty_limit = global_wb_domain.dirty_limit; ), TP_printk("dirty=%lu writeback=%lu unstable=%lu " @@ -463,8 +463,9 @@ TRACE_EVENT(balance_dirty_pages, unsigned long freerun = (thresh + bg_thresh) / 2; strlcpy(__entry->bdi, dev_name(bdi->dev), 32); - __entry->limit = global_dirty_limit; - __entry->setpoint = (global_dirty_limit + freerun) / 2; + __entry->limit = global_wb_domain.dirty_limit; + __entry->setpoint = (global_wb_domain.dirty_limit + + freerun) / 2; __entry->dirty = dirty; __entry->bdi_setpoint = __entry->setpoint * bdi_thresh / (thresh + 1); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 08e1737edb39..27e60ba8e688 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -122,9 +122,7 @@ EXPORT_SYMBOL(laptop_mode); /* End of sysctl-exported parameters */ -unsigned long global_dirty_limit; - -static struct wb_domain global_wb_domain; +struct wb_domain global_wb_domain; /* * Length of period for aging writeout fractions of bdis. This is an @@ -470,9 +468,15 @@ static void writeout_period(unsigned long t) int wb_domain_init(struct wb_domain *dom, gfp_t gfp) { memset(dom, 0, sizeof(*dom)); + + spin_lock_init(&dom->lock); + init_timer_deferrable(&dom->period_timer); dom->period_timer.function = writeout_period; dom->period_timer.data = (unsigned long)dom; + + dom->dirty_limit_tstamp = jiffies; + return fprop_global_init(&dom->completions, gfp); } @@ -532,7 +536,9 @@ static unsigned long dirty_freerun_ceiling(unsigned long thresh, static unsigned long hard_dirty_limit(unsigned long thresh) { - return max(thresh, global_dirty_limit); + struct wb_domain *dom = &global_wb_domain; + + return max(thresh, dom->dirty_limit); } /** @@ -916,17 +922,10 @@ out: wb->avg_write_bandwidth = avg; } -/* - * The global dirtyable memory and dirty threshold could be suddenly knocked - * down by a large amount (eg. on the startup of KVM in a swapless system). - * This may throw the system into deep dirty exceeded state and throttle - * heavy/light dirtiers alike. To retain good responsiveness, maintain - * global_dirty_limit for tracking slowly down to the knocked down dirty - * threshold. - */ static void update_dirty_limit(unsigned long thresh, unsigned long dirty) { - unsigned long limit = global_dirty_limit; + struct wb_domain *dom = &global_wb_domain; + unsigned long limit = dom->dirty_limit; /* * Follow up in one step. @@ -939,7 +938,7 @@ static void update_dirty_limit(unsigned long thresh, unsigned long dirty) /* * Follow down slowly. Use the higher one as the target, because thresh * may drop below dirty. This is exactly the reason to introduce - * global_dirty_limit which is guaranteed to lie above the dirty pages. + * dom->dirty_limit which is guaranteed to lie above the dirty pages. */ thresh = max(thresh, dirty); if (limit > thresh) { @@ -948,28 +947,27 @@ static void update_dirty_limit(unsigned long thresh, unsigned long dirty) } return; update: - global_dirty_limit = limit; + dom->dirty_limit = limit; } static void global_update_bandwidth(unsigned long thresh, unsigned long dirty, unsigned long now) { - static DEFINE_SPINLOCK(dirty_lock); - static unsigned long update_time = INITIAL_JIFFIES; + struct wb_domain *dom = &global_wb_domain; /* * check locklessly first to optimize away locking for the most time */ - if (time_before(now, update_time + BANDWIDTH_INTERVAL)) + if (time_before(now, dom->dirty_limit_tstamp + BANDWIDTH_INTERVAL)) return; - spin_lock(&dirty_lock); - if (time_after_eq(now, update_time + BANDWIDTH_INTERVAL)) { + spin_lock(&dom->lock); + if (time_after_eq(now, dom->dirty_limit_tstamp + BANDWIDTH_INTERVAL)) { update_dirty_limit(thresh, dirty); - update_time = now; + dom->dirty_limit_tstamp = now; } - spin_unlock(&dirty_lock); + spin_unlock(&dom->lock); } /* @@ -1761,10 +1759,12 @@ void laptop_sync_completion(void) void writeback_set_ratelimit(void) { + struct wb_domain *dom = &global_wb_domain; unsigned long background_thresh; unsigned long dirty_thresh; + global_dirty_limits(&background_thresh, &dirty_thresh); - global_dirty_limit = dirty_thresh; + dom->dirty_limit = dirty_thresh; ratelimit_pages = dirty_thresh / (num_online_cpus() * 32); if (ratelimit_pages < 16) ratelimit_pages = 16; From 2bc00aef030f4f75550d5c88062ce1830e40097f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:23 -0400 Subject: [PATCH 57/85] writeback: consolidate dirty throttle parameters into dirty_throttle_control Dirty throttling implemented in balance_dirty_pages() and its subroutines makes use of a number of parameters which are passed around individually. This renders these functions somewhat unwieldy and makes it difficult to add or change the involved parameters. Also some functions use different or conflicting naming schemes for the same parameters making the code confusing to follow. This patch consolidates the main parameters into struct dirty_throttle_control so that they can be passed around easily and adding new paramters isn't painful. This also unifies how a given parameter is named and accessed. The drawback of using this type of control structure rather than explicit paramters is that it isn't immediately obvious which function accesses and modifies what; however, it's fairly clear that the benefits outweigh in this case. GDTC_INIT() macro is provided to ease initializing dirty_throttle_control for the global_wb_domain and balance_dirty_pages() uses a separate pointer to point to its global dirty_throttle_control. This is to make it uniform with memcg domain handling which will be added later. This patch doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 212 +++++++++++++++++++++----------------------- 1 file changed, 101 insertions(+), 111 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 27e60ba8e688..126e3c8bc1c6 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -124,6 +124,20 @@ EXPORT_SYMBOL(laptop_mode); struct wb_domain global_wb_domain; +/* consolidated parameters for balance_dirty_pages() and its subroutines */ +struct dirty_throttle_control { + struct bdi_writeback *wb; + + unsigned long dirty; /* file_dirty + write + nfs */ + unsigned long thresh; /* dirty threshold */ + unsigned long bg_thresh; /* dirty background threshold */ + + unsigned long wb_dirty; /* per-wb counterparts */ + unsigned long wb_thresh; +}; + +#define GDTC_INIT(__wb) .wb = (__wb) + /* * Length of period for aging writeout fractions of bdis. This is an * arbitrarily chosen number. The longer the period, the slower fractions will @@ -695,16 +709,13 @@ static long long pos_ratio_polynom(unsigned long setpoint, * card's wb_dirty may rush to many times higher than wb_setpoint. * - the wb dirty thresh drops quickly due to change of JBOD workload */ -static unsigned long wb_position_ratio(struct bdi_writeback *wb, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long wb_thresh, - unsigned long wb_dirty) +static unsigned long wb_position_ratio(struct dirty_throttle_control *dtc) { + struct bdi_writeback *wb = dtc->wb; unsigned long write_bw = wb->avg_write_bandwidth; - unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh); - unsigned long limit = hard_dirty_limit(thresh); + unsigned long freerun = dirty_freerun_ceiling(dtc->thresh, dtc->bg_thresh); + unsigned long limit = hard_dirty_limit(dtc->thresh); + unsigned long wb_thresh = dtc->wb_thresh; unsigned long x_intercept; unsigned long setpoint; /* dirty pages' target balance point */ unsigned long wb_setpoint; @@ -712,7 +723,7 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, long long pos_ratio; /* for scaling up/down the rate limit */ long x; - if (unlikely(dirty >= limit)) + if (unlikely(dtc->dirty >= limit)) return 0; /* @@ -721,7 +732,7 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, * See comment for pos_ratio_polynom(). */ setpoint = (freerun + limit) / 2; - pos_ratio = pos_ratio_polynom(setpoint, dirty, limit); + pos_ratio = pos_ratio_polynom(setpoint, dtc->dirty, limit); /* * The strictlimit feature is a tool preventing mistrusted filesystems @@ -752,20 +763,21 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, long long wb_pos_ratio; unsigned long wb_bg_thresh; - if (wb_dirty < 8) + if (dtc->wb_dirty < 8) return min_t(long long, pos_ratio * 2, 2 << RATELIMIT_CALC_SHIFT); - if (wb_dirty >= wb_thresh) + if (dtc->wb_dirty >= wb_thresh) return 0; - wb_bg_thresh = div_u64((u64)wb_thresh * bg_thresh, thresh); + wb_bg_thresh = div_u64((u64)wb_thresh * dtc->bg_thresh, + dtc->thresh); wb_setpoint = dirty_freerun_ceiling(wb_thresh, wb_bg_thresh); if (wb_setpoint == 0 || wb_setpoint == wb_thresh) return 0; - wb_pos_ratio = pos_ratio_polynom(wb_setpoint, wb_dirty, + wb_pos_ratio = pos_ratio_polynom(wb_setpoint, dtc->wb_dirty, wb_thresh); /* @@ -823,8 +835,8 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, * own size, so move the slope over accordingly and choose a slope that * yields 100% pos_ratio fluctuation on suddenly doubled wb_thresh. */ - if (unlikely(wb_thresh > thresh)) - wb_thresh = thresh; + if (unlikely(wb_thresh > dtc->thresh)) + wb_thresh = dtc->thresh; /* * It's very possible that wb_thresh is close to 0 not because the * device is slow, but that it has remained inactive for long time. @@ -832,12 +844,12 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, * threshold, so that the occasional writes won't be blocked and active * writes can rampup the threshold quickly. */ - wb_thresh = max(wb_thresh, (limit - dirty) / 8); + wb_thresh = max(wb_thresh, (limit - dtc->dirty) / 8); /* * scale global setpoint to wb's: * wb_setpoint = setpoint * wb_thresh / thresh */ - x = div_u64((u64)wb_thresh << 16, thresh + 1); + x = div_u64((u64)wb_thresh << 16, dtc->thresh + 1); wb_setpoint = setpoint * (u64)x >> 16; /* * Use span=(8*write_bw) in single wb case as indicated by @@ -847,12 +859,12 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, * span = --------- * (8 * write_bw) + ------------------ * wb_thresh * thresh thresh */ - span = (thresh - wb_thresh + 8 * write_bw) * (u64)x >> 16; + span = (dtc->thresh - wb_thresh + 8 * write_bw) * (u64)x >> 16; x_intercept = wb_setpoint + span; - if (wb_dirty < x_intercept - span / 4) { - pos_ratio = div64_u64(pos_ratio * (x_intercept - wb_dirty), - x_intercept - wb_setpoint + 1); + if (dtc->wb_dirty < x_intercept - span / 4) { + pos_ratio = div64_u64(pos_ratio * (x_intercept - dtc->wb_dirty), + x_intercept - wb_setpoint + 1); } else pos_ratio /= 4; @@ -862,9 +874,10 @@ static unsigned long wb_position_ratio(struct bdi_writeback *wb, * than setpoint. */ x_intercept = wb_thresh / 2; - if (wb_dirty < x_intercept) { - if (wb_dirty > x_intercept / 8) - pos_ratio = div_u64(pos_ratio * x_intercept, wb_dirty); + if (dtc->wb_dirty < x_intercept) { + if (dtc->wb_dirty > x_intercept / 8) + pos_ratio = div_u64(pos_ratio * x_intercept, + dtc->wb_dirty); else pos_ratio *= 8; } @@ -922,9 +935,10 @@ out: wb->avg_write_bandwidth = avg; } -static void update_dirty_limit(unsigned long thresh, unsigned long dirty) +static void update_dirty_limit(struct dirty_throttle_control *dtc) { struct wb_domain *dom = &global_wb_domain; + unsigned long thresh = dtc->thresh; unsigned long limit = dom->dirty_limit; /* @@ -940,7 +954,7 @@ static void update_dirty_limit(unsigned long thresh, unsigned long dirty) * may drop below dirty. This is exactly the reason to introduce * dom->dirty_limit which is guaranteed to lie above the dirty pages. */ - thresh = max(thresh, dirty); + thresh = max(thresh, dtc->dirty); if (limit > thresh) { limit -= (limit - thresh) >> 5; goto update; @@ -950,8 +964,7 @@ update: dom->dirty_limit = limit; } -static void global_update_bandwidth(unsigned long thresh, - unsigned long dirty, +static void global_update_bandwidth(struct dirty_throttle_control *dtc, unsigned long now) { struct wb_domain *dom = &global_wb_domain; @@ -964,7 +977,7 @@ static void global_update_bandwidth(unsigned long thresh, spin_lock(&dom->lock); if (time_after_eq(now, dom->dirty_limit_tstamp + BANDWIDTH_INTERVAL)) { - update_dirty_limit(thresh, dirty); + update_dirty_limit(dtc); dom->dirty_limit_tstamp = now; } spin_unlock(&dom->lock); @@ -976,17 +989,14 @@ static void global_update_bandwidth(unsigned long thresh, * Normal wb tasks will be curbed at or below it in long term. * Obviously it should be around (write_bw / N) when there are N dd tasks. */ -static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long wb_thresh, - unsigned long wb_dirty, +static void wb_update_dirty_ratelimit(struct dirty_throttle_control *dtc, unsigned long dirtied, unsigned long elapsed) { - unsigned long freerun = dirty_freerun_ceiling(thresh, bg_thresh); - unsigned long limit = hard_dirty_limit(thresh); + struct bdi_writeback *wb = dtc->wb; + unsigned long dirty = dtc->dirty; + unsigned long freerun = dirty_freerun_ceiling(dtc->thresh, dtc->bg_thresh); + unsigned long limit = hard_dirty_limit(dtc->thresh); unsigned long setpoint = (freerun + limit) / 2; unsigned long write_bw = wb->avg_write_bandwidth; unsigned long dirty_ratelimit = wb->dirty_ratelimit; @@ -1003,8 +1013,7 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, */ dirty_rate = (dirtied - wb->dirtied_stamp) * HZ / elapsed; - pos_ratio = wb_position_ratio(wb, thresh, bg_thresh, dirty, - wb_thresh, wb_dirty); + pos_ratio = wb_position_ratio(dtc); /* * task_ratelimit reflects each dd's dirty rate for the past 200ms. */ @@ -1098,12 +1107,12 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, * of backing device (see the implementation of wb_calc_thresh()). */ if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { - dirty = wb_dirty; - if (wb_dirty < 8) - setpoint = wb_dirty + 1; + dirty = dtc->wb_dirty; + if (dtc->wb_dirty < 8) + setpoint = dtc->wb_dirty + 1; else - setpoint = (wb_thresh + - wb_calc_thresh(wb, bg_thresh)) / 2; + setpoint = (dtc->wb_thresh + + wb_calc_thresh(wb, dtc->bg_thresh)) / 2; } if (dirty < setpoint) { @@ -1140,15 +1149,11 @@ static void wb_update_dirty_ratelimit(struct bdi_writeback *wb, trace_bdi_dirty_ratelimit(wb->bdi, dirty_rate, task_ratelimit); } -static void __wb_update_bandwidth(struct bdi_writeback *wb, - unsigned long thresh, - unsigned long bg_thresh, - unsigned long dirty, - unsigned long wb_thresh, - unsigned long wb_dirty, +static void __wb_update_bandwidth(struct dirty_throttle_control *dtc, unsigned long start_time, bool update_ratelimit) { + struct bdi_writeback *wb = dtc->wb; unsigned long now = jiffies; unsigned long elapsed = now - wb->bw_time_stamp; unsigned long dirtied; @@ -1173,10 +1178,8 @@ static void __wb_update_bandwidth(struct bdi_writeback *wb, goto snapshot; if (update_ratelimit) { - global_update_bandwidth(thresh, dirty, now); - wb_update_dirty_ratelimit(wb, thresh, bg_thresh, dirty, - wb_thresh, wb_dirty, - dirtied, elapsed); + global_update_bandwidth(dtc, now); + wb_update_dirty_ratelimit(dtc, dirtied, elapsed); } wb_update_write_bandwidth(wb, elapsed, written); @@ -1188,7 +1191,9 @@ snapshot: void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long start_time) { - __wb_update_bandwidth(wb, 0, 0, 0, 0, 0, start_time, false); + struct dirty_throttle_control gdtc = { GDTC_INIT(wb) }; + + __wb_update_bandwidth(&gdtc, start_time, false); } /* @@ -1302,13 +1307,10 @@ static long wb_min_pause(struct bdi_writeback *wb, return pages >= DIRTY_POLL_THRESH ? 1 + t / 2 : t; } -static inline void wb_dirty_limits(struct bdi_writeback *wb, - unsigned long dirty_thresh, - unsigned long background_thresh, - unsigned long *wb_dirty, - unsigned long *wb_thresh, +static inline void wb_dirty_limits(struct dirty_throttle_control *dtc, unsigned long *wb_bg_thresh) { + struct bdi_writeback *wb = dtc->wb; unsigned long wb_reclaimable; /* @@ -1324,12 +1326,12 @@ static inline void wb_dirty_limits(struct bdi_writeback *wb, * wb_position_ratio() will let the dirtier task progress * at some rate <= (write_bw / 2) for bringing down wb_dirty. */ - *wb_thresh = wb_calc_thresh(wb, dirty_thresh); + dtc->wb_thresh = wb_calc_thresh(dtc->wb, dtc->thresh); if (wb_bg_thresh) - *wb_bg_thresh = dirty_thresh ? div_u64((u64)*wb_thresh * - background_thresh, - dirty_thresh) : 0; + *wb_bg_thresh = dtc->thresh ? div_u64((u64)dtc->wb_thresh * + dtc->bg_thresh, + dtc->thresh) : 0; /* * In order to avoid the stacked BDI deadlock we need @@ -1341,12 +1343,12 @@ static inline void wb_dirty_limits(struct bdi_writeback *wb, * actually dirty; with m+n sitting in the percpu * deltas. */ - if (*wb_thresh < 2 * wb_stat_error(wb)) { + if (dtc->wb_thresh < 2 * wb_stat_error(wb)) { wb_reclaimable = wb_stat_sum(wb, WB_RECLAIMABLE); - *wb_dirty = wb_reclaimable + wb_stat_sum(wb, WB_WRITEBACK); + dtc->wb_dirty = wb_reclaimable + wb_stat_sum(wb, WB_WRITEBACK); } else { wb_reclaimable = wb_stat(wb, WB_RECLAIMABLE); - *wb_dirty = wb_reclaimable + wb_stat(wb, WB_WRITEBACK); + dtc->wb_dirty = wb_reclaimable + wb_stat(wb, WB_WRITEBACK); } } @@ -1361,10 +1363,9 @@ static void balance_dirty_pages(struct address_space *mapping, struct bdi_writeback *wb, unsigned long pages_dirtied) { + struct dirty_throttle_control gdtc_stor = { GDTC_INIT(wb) }; + struct dirty_throttle_control * const gdtc = &gdtc_stor; unsigned long nr_reclaimable; /* = file_dirty + unstable_nfs */ - unsigned long nr_dirty; /* = file_dirty + writeback + unstable_nfs */ - unsigned long background_thresh; - unsigned long dirty_thresh; long period; long pause; long max_pause; @@ -1380,11 +1381,7 @@ static void balance_dirty_pages(struct address_space *mapping, for (;;) { unsigned long now = jiffies; - unsigned long uninitialized_var(wb_thresh); - unsigned long thresh; - unsigned long uninitialized_var(wb_dirty); - unsigned long dirty; - unsigned long bg_thresh; + unsigned long dirty, thresh, bg_thresh; /* * Unstable writes are a feature of certain networked @@ -1394,20 +1391,19 @@ static void balance_dirty_pages(struct address_space *mapping, */ nr_reclaimable = global_page_state(NR_FILE_DIRTY) + global_page_state(NR_UNSTABLE_NFS); - nr_dirty = nr_reclaimable + global_page_state(NR_WRITEBACK); + gdtc->dirty = nr_reclaimable + global_page_state(NR_WRITEBACK); - global_dirty_limits(&background_thresh, &dirty_thresh); + global_dirty_limits(&gdtc->bg_thresh, &gdtc->thresh); if (unlikely(strictlimit)) { - wb_dirty_limits(wb, dirty_thresh, background_thresh, - &wb_dirty, &wb_thresh, &bg_thresh); + wb_dirty_limits(gdtc, &bg_thresh); - dirty = wb_dirty; - thresh = wb_thresh; + dirty = gdtc->wb_dirty; + thresh = gdtc->wb_thresh; } else { - dirty = nr_dirty; - thresh = dirty_thresh; - bg_thresh = background_thresh; + dirty = gdtc->dirty; + thresh = gdtc->thresh; + bg_thresh = gdtc->bg_thresh; } /* @@ -1431,31 +1427,25 @@ static void balance_dirty_pages(struct address_space *mapping, wb_start_background_writeback(wb); if (!strictlimit) - wb_dirty_limits(wb, dirty_thresh, background_thresh, - &wb_dirty, &wb_thresh, NULL); + wb_dirty_limits(gdtc, NULL); - dirty_exceeded = (wb_dirty > wb_thresh) && - ((nr_dirty > dirty_thresh) || strictlimit); + dirty_exceeded = (gdtc->wb_dirty > gdtc->wb_thresh) && + ((gdtc->dirty > gdtc->thresh) || strictlimit); if (dirty_exceeded && !wb->dirty_exceeded) wb->dirty_exceeded = 1; if (time_is_before_jiffies(wb->bw_time_stamp + BANDWIDTH_INTERVAL)) { spin_lock(&wb->list_lock); - __wb_update_bandwidth(wb, dirty_thresh, - background_thresh, nr_dirty, - wb_thresh, wb_dirty, start_time, - true); + __wb_update_bandwidth(gdtc, start_time, true); spin_unlock(&wb->list_lock); } dirty_ratelimit = wb->dirty_ratelimit; - pos_ratio = wb_position_ratio(wb, dirty_thresh, - background_thresh, nr_dirty, - wb_thresh, wb_dirty); + pos_ratio = wb_position_ratio(gdtc); task_ratelimit = ((u64)dirty_ratelimit * pos_ratio) >> RATELIMIT_CALC_SHIFT; - max_pause = wb_max_pause(wb, wb_dirty); + max_pause = wb_max_pause(wb, gdtc->wb_dirty); min_pause = wb_min_pause(wb, max_pause, task_ratelimit, dirty_ratelimit, &nr_dirtied_pause); @@ -1478,11 +1468,11 @@ static void balance_dirty_pages(struct address_space *mapping, */ if (pause < min_pause) { trace_balance_dirty_pages(bdi, - dirty_thresh, - background_thresh, - nr_dirty, - wb_thresh, - wb_dirty, + gdtc->thresh, + gdtc->bg_thresh, + gdtc->dirty, + gdtc->wb_thresh, + gdtc->wb_dirty, dirty_ratelimit, task_ratelimit, pages_dirtied, @@ -1507,11 +1497,11 @@ static void balance_dirty_pages(struct address_space *mapping, pause: trace_balance_dirty_pages(bdi, - dirty_thresh, - background_thresh, - nr_dirty, - wb_thresh, - wb_dirty, + gdtc->thresh, + gdtc->bg_thresh, + gdtc->dirty, + gdtc->wb_thresh, + gdtc->wb_dirty, dirty_ratelimit, task_ratelimit, pages_dirtied, @@ -1526,8 +1516,8 @@ pause: current->nr_dirtied_pause = nr_dirtied_pause; /* - * This is typically equal to (nr_dirty < dirty_thresh) and can - * also keep "1000+ dd on a slow USB stick" under control. + * This is typically equal to (dirty < thresh) and can also + * keep "1000+ dd on a slow USB stick" under control. */ if (task_ratelimit) break; @@ -1542,7 +1532,7 @@ pause: * more page. However wb_dirty has accounting errors. So use * the larger and more IO friendly wb_stat_error. */ - if (wb_dirty <= wb_stat_error(wb)) + if (gdtc->wb_dirty <= wb_stat_error(wb)) break; if (fatal_signal_pending(current)) @@ -1566,7 +1556,7 @@ pause: if (laptop_mode) return; - if (nr_reclaimable > background_thresh) + if (nr_reclaimable > gdtc->bg_thresh) wb_start_background_writeback(wb); } From 970fb01ad3a773b5612a9bba6b366abcefc18eaf Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:24 -0400 Subject: [PATCH 58/85] writeback: add dirty_throttle_control->wb_bg_thresh wb_bg_thresh is currently treated as a second-class citizen. It's only used when BDI_CAP_STRICTLIMIT is set and balance_dirty_pages() doesn't calculate it unless the cap is set. When the cap is set, the calculated value is not passed around but instead recalculated whenever it's used. wb_position_ratio() calculates it by scaling wb_thresh proportional to bg_thresh / thresh. wb_update_dirty_ratelimit() uses wb_dirty_limit() on bg_thresh, which should generally lead to a similar result as the proportional scaling but can also be way off in the presence of max/min_ratio settings. Avoiding wb_bg_thresh calculation saves us one u64 multiplication and divsion when BDI_CAP_STRICTLIMIT is not set. Given that balance_dirty_pages() is already ratelimited, this doesn't justify the incurred extra complexity. This patch adds wb_bg_thresh to dirty_throttle_control and makes wb_dirty_limits() always calculate it and updates the users to use the pre-calculated value. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 126e3c8bc1c6..3ec922309178 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -134,6 +134,7 @@ struct dirty_throttle_control { unsigned long wb_dirty; /* per-wb counterparts */ unsigned long wb_thresh; + unsigned long wb_bg_thresh; }; #define GDTC_INIT(__wb) .wb = (__wb) @@ -761,7 +762,6 @@ static unsigned long wb_position_ratio(struct dirty_throttle_control *dtc) */ if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { long long wb_pos_ratio; - unsigned long wb_bg_thresh; if (dtc->wb_dirty < 8) return min_t(long long, pos_ratio * 2, @@ -770,9 +770,8 @@ static unsigned long wb_position_ratio(struct dirty_throttle_control *dtc) if (dtc->wb_dirty >= wb_thresh) return 0; - wb_bg_thresh = div_u64((u64)wb_thresh * dtc->bg_thresh, - dtc->thresh); - wb_setpoint = dirty_freerun_ceiling(wb_thresh, wb_bg_thresh); + wb_setpoint = dirty_freerun_ceiling(wb_thresh, + dtc->wb_bg_thresh); if (wb_setpoint == 0 || wb_setpoint == wb_thresh) return 0; @@ -1104,15 +1103,14 @@ static void wb_update_dirty_ratelimit(struct dirty_throttle_control *dtc, * * We rampup dirty_ratelimit forcibly if wb_dirty is low because * it's possible that wb_thresh is close to zero due to inactivity - * of backing device (see the implementation of wb_calc_thresh()). + * of backing device. */ if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { dirty = dtc->wb_dirty; if (dtc->wb_dirty < 8) setpoint = dtc->wb_dirty + 1; else - setpoint = (dtc->wb_thresh + - wb_calc_thresh(wb, dtc->bg_thresh)) / 2; + setpoint = (dtc->wb_thresh + dtc->wb_bg_thresh) / 2; } if (dirty < setpoint) { @@ -1307,8 +1305,7 @@ static long wb_min_pause(struct bdi_writeback *wb, return pages >= DIRTY_POLL_THRESH ? 1 + t / 2 : t; } -static inline void wb_dirty_limits(struct dirty_throttle_control *dtc, - unsigned long *wb_bg_thresh) +static inline void wb_dirty_limits(struct dirty_throttle_control *dtc) { struct bdi_writeback *wb = dtc->wb; unsigned long wb_reclaimable; @@ -1327,11 +1324,8 @@ static inline void wb_dirty_limits(struct dirty_throttle_control *dtc, * at some rate <= (write_bw / 2) for bringing down wb_dirty. */ dtc->wb_thresh = wb_calc_thresh(dtc->wb, dtc->thresh); - - if (wb_bg_thresh) - *wb_bg_thresh = dtc->thresh ? div_u64((u64)dtc->wb_thresh * - dtc->bg_thresh, - dtc->thresh) : 0; + dtc->wb_bg_thresh = dtc->thresh ? + div_u64((u64)dtc->wb_thresh * dtc->bg_thresh, dtc->thresh) : 0; /* * In order to avoid the stacked BDI deadlock we need @@ -1396,10 +1390,11 @@ static void balance_dirty_pages(struct address_space *mapping, global_dirty_limits(&gdtc->bg_thresh, &gdtc->thresh); if (unlikely(strictlimit)) { - wb_dirty_limits(gdtc, &bg_thresh); + wb_dirty_limits(gdtc); dirty = gdtc->wb_dirty; thresh = gdtc->wb_thresh; + bg_thresh = gdtc->wb_bg_thresh; } else { dirty = gdtc->dirty; thresh = gdtc->thresh; @@ -1427,7 +1422,7 @@ static void balance_dirty_pages(struct address_space *mapping, wb_start_background_writeback(wb); if (!strictlimit) - wb_dirty_limits(gdtc, NULL); + wb_dirty_limits(gdtc); dirty_exceeded = (gdtc->wb_dirty > gdtc->wb_thresh) && ((gdtc->dirty > gdtc->thresh) || strictlimit); From b1cbc6d40c85639d405fee37c7bb688c3bf468d6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:25 -0400 Subject: [PATCH 59/85] writeback: make __wb_calc_thresh() take dirty_throttle_control wb_calc_thresh() calculates wb_thresh by scaling thresh according to the wb's portion in the system-wide write bandwidth. cgroup writeback support would need to calculate wb_thresh against memcg domain too. This patch renames wb_calc_thresh() to __wb_calc_thresh() and makes it take dirty_throttle_control so that the function can later be updated to calculate against different domains according to dirty_throttle_control. wb_calc_thresh() is now a thin wrapper around __wb_calc_thresh(). v2: The original version was incorrectly scaling dtc->dirty instead of dtc->thresh. This was due to the extremely confusing function and variable names. Added a rename patch and fixed this one. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 3ec922309178..2352c69bff12 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -557,9 +557,8 @@ static unsigned long hard_dirty_limit(unsigned long thresh) } /** - * wb_calc_thresh - @wb's share of dirty throttling threshold - * @wb: bdi_writeback to query - * @dirty: global dirty limit in pages + * __wb_calc_thresh - @wb's share of dirty throttling threshold + * @dtc: dirty_throttle_context of interest * * Returns @wb's dirty limit in pages. The term "dirty" in the context of * dirty balancing includes all PG_dirty, PG_writeback and NFS unstable pages. @@ -578,9 +577,10 @@ static unsigned long hard_dirty_limit(unsigned long thresh) * The wb's share of dirty limit will be adapting to its throughput and * bounded by the bdi->min_ratio and/or bdi->max_ratio parameters, if set. */ -unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh) +static unsigned long __wb_calc_thresh(struct dirty_throttle_control *dtc) { struct wb_domain *dom = &global_wb_domain; + unsigned long thresh = dtc->thresh; u64 wb_thresh; long numerator, denominator; unsigned long wb_min_ratio, wb_max_ratio; @@ -588,14 +588,14 @@ unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh) /* * Calculate this BDI's share of the thresh ratio. */ - fprop_fraction_percpu(&dom->completions, &wb->completions, + fprop_fraction_percpu(&dom->completions, &dtc->wb->completions, &numerator, &denominator); wb_thresh = (thresh * (100 - bdi_min_ratio)) / 100; wb_thresh *= numerator; do_div(wb_thresh, denominator); - wb_min_max_ratio(wb, &wb_min_ratio, &wb_max_ratio); + wb_min_max_ratio(dtc->wb, &wb_min_ratio, &wb_max_ratio); wb_thresh += (thresh * wb_min_ratio) / 100; if (wb_thresh > (thresh * wb_max_ratio) / 100) @@ -604,6 +604,13 @@ unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh) return wb_thresh; } +unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh) +{ + struct dirty_throttle_control gdtc = { GDTC_INIT(wb), + .thresh = thresh }; + return __wb_calc_thresh(&gdtc); +} + /* * setpoint - dirty 3 * f(dirty) := 1.0 + (----------------) @@ -1323,7 +1330,7 @@ static inline void wb_dirty_limits(struct dirty_throttle_control *dtc) * wb_position_ratio() will let the dirtier task progress * at some rate <= (write_bw / 2) for bringing down wb_dirty. */ - dtc->wb_thresh = wb_calc_thresh(dtc->wb, dtc->thresh); + dtc->wb_thresh = __wb_calc_thresh(dtc); dtc->wb_bg_thresh = dtc->thresh ? div_u64((u64)dtc->wb_thresh * dtc->bg_thresh, dtc->thresh) : 0; From daddfa3cb30ebfe322d50af146d830fd435ddb1f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:26 -0400 Subject: [PATCH 60/85] writeback: add dirty_throttle_control->pos_ratio wb_position_ratio() is used to calculate pos_ratio, which is used for two purposes. wb_update_dirty_ratelimit() uses it to adjust wb->[balanced_]dirty_ratelimit gradually and balance_dirty_pages() to immediately adjust dirty_ratelimit right before applying it to determine pause duration. While wb_update_dirty_ratelimit() is separately rate limited from balance_dirty_pages(), on the run where the ratelimit is updated, we end up calculating pos_ratio twice with the same parameters. This patch adds dirty_throttle_control->pos_ratio. balance_dirty_pages() calculates it once per run and wb_update_dirty_ratelimit() uses the value stored in dirty_throttle_control. This removes the duplicate calculation and also will help implementing memcg wb_domain. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 36 +++++++++++++++++++++--------------- 1 file changed, 21 insertions(+), 15 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 2352c69bff12..fcebae7db874 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -135,6 +135,8 @@ struct dirty_throttle_control { unsigned long wb_dirty; /* per-wb counterparts */ unsigned long wb_thresh; unsigned long wb_bg_thresh; + + unsigned long pos_ratio; }; #define GDTC_INIT(__wb) .wb = (__wb) @@ -717,7 +719,7 @@ static long long pos_ratio_polynom(unsigned long setpoint, * card's wb_dirty may rush to many times higher than wb_setpoint. * - the wb dirty thresh drops quickly due to change of JBOD workload */ -static unsigned long wb_position_ratio(struct dirty_throttle_control *dtc) +static void wb_position_ratio(struct dirty_throttle_control *dtc) { struct bdi_writeback *wb = dtc->wb; unsigned long write_bw = wb->avg_write_bandwidth; @@ -731,8 +733,10 @@ static unsigned long wb_position_ratio(struct dirty_throttle_control *dtc) long long pos_ratio; /* for scaling up/down the rate limit */ long x; + dtc->pos_ratio = 0; + if (unlikely(dtc->dirty >= limit)) - return 0; + return; /* * global setpoint @@ -770,18 +774,20 @@ static unsigned long wb_position_ratio(struct dirty_throttle_control *dtc) if (unlikely(wb->bdi->capabilities & BDI_CAP_STRICTLIMIT)) { long long wb_pos_ratio; - if (dtc->wb_dirty < 8) - return min_t(long long, pos_ratio * 2, - 2 << RATELIMIT_CALC_SHIFT); + if (dtc->wb_dirty < 8) { + dtc->pos_ratio = min_t(long long, pos_ratio * 2, + 2 << RATELIMIT_CALC_SHIFT); + return; + } if (dtc->wb_dirty >= wb_thresh) - return 0; + return; wb_setpoint = dirty_freerun_ceiling(wb_thresh, dtc->wb_bg_thresh); if (wb_setpoint == 0 || wb_setpoint == wb_thresh) - return 0; + return; wb_pos_ratio = pos_ratio_polynom(wb_setpoint, dtc->wb_dirty, wb_thresh); @@ -807,7 +813,8 @@ static unsigned long wb_position_ratio(struct dirty_throttle_control *dtc) * is 2. We might want to tweak this if we observe the control * system is too slow to adapt. */ - return min(pos_ratio, wb_pos_ratio); + dtc->pos_ratio = min(pos_ratio, wb_pos_ratio); + return; } /* @@ -888,7 +895,7 @@ static unsigned long wb_position_ratio(struct dirty_throttle_control *dtc) pos_ratio *= 8; } - return pos_ratio; + dtc->pos_ratio = pos_ratio; } static void wb_update_write_bandwidth(struct bdi_writeback *wb, @@ -1009,7 +1016,6 @@ static void wb_update_dirty_ratelimit(struct dirty_throttle_control *dtc, unsigned long dirty_rate; unsigned long task_ratelimit; unsigned long balanced_dirty_ratelimit; - unsigned long pos_ratio; unsigned long step; unsigned long x; @@ -1019,12 +1025,11 @@ static void wb_update_dirty_ratelimit(struct dirty_throttle_control *dtc, */ dirty_rate = (dirtied - wb->dirtied_stamp) * HZ / elapsed; - pos_ratio = wb_position_ratio(dtc); /* * task_ratelimit reflects each dd's dirty rate for the past 200ms. */ task_ratelimit = (u64)dirty_ratelimit * - pos_ratio >> RATELIMIT_CALC_SHIFT; + dtc->pos_ratio >> RATELIMIT_CALC_SHIFT; task_ratelimit++; /* it helps rampup dirty_ratelimit from tiny values */ /* @@ -1375,7 +1380,6 @@ static void balance_dirty_pages(struct address_space *mapping, bool dirty_exceeded = false; unsigned long task_ratelimit; unsigned long dirty_ratelimit; - unsigned long pos_ratio; struct backing_dev_info *bdi = wb->bdi; bool strictlimit = bdi->capabilities & BDI_CAP_STRICTLIMIT; unsigned long start_time = jiffies; @@ -1433,6 +1437,9 @@ static void balance_dirty_pages(struct address_space *mapping, dirty_exceeded = (gdtc->wb_dirty > gdtc->wb_thresh) && ((gdtc->dirty > gdtc->thresh) || strictlimit); + + wb_position_ratio(gdtc); + if (dirty_exceeded && !wb->dirty_exceeded) wb->dirty_exceeded = 1; @@ -1444,8 +1451,7 @@ static void balance_dirty_pages(struct address_space *mapping, } dirty_ratelimit = wb->dirty_ratelimit; - pos_ratio = wb_position_ratio(gdtc); - task_ratelimit = ((u64)dirty_ratelimit * pos_ratio) >> + task_ratelimit = ((u64)dirty_ratelimit * gdtc->pos_ratio) >> RATELIMIT_CALC_SHIFT; max_pause = wb_max_pause(wb, gdtc->wb_dirty); min_pause = wb_min_pause(wb, max_pause, From e9770b3487328b7e28803caf6c809292dd7adbf0 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:27 -0400 Subject: [PATCH 61/85] writeback: add dirty_throttle_control->wb_completions wb->completions measures the wb's proportional write bandwidth in global_wb_domain and thus naturally tied to the wb_domain. This patch adds dirty_throttle_control->wb_completions which is initialized to wb->completions by GDTC_INIT() and updates __wb_dirty_limits() to use it instead of dereferencing wb->completions directly. This will allow dirty_throttle_control to represent different wb_domains and the matching wb completions. This patch doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index fcebae7db874..5b439fc1e95e 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -127,6 +127,7 @@ struct wb_domain global_wb_domain; /* consolidated parameters for balance_dirty_pages() and its subroutines */ struct dirty_throttle_control { struct bdi_writeback *wb; + struct fprop_local_percpu *wb_completions; unsigned long dirty; /* file_dirty + write + nfs */ unsigned long thresh; /* dirty threshold */ @@ -139,7 +140,8 @@ struct dirty_throttle_control { unsigned long pos_ratio; }; -#define GDTC_INIT(__wb) .wb = (__wb) +#define GDTC_INIT(__wb) .wb = (__wb), \ + .wb_completions = &(__wb)->completions /* * Length of period for aging writeout fractions of bdis. This is an @@ -590,7 +592,7 @@ static unsigned long __wb_calc_thresh(struct dirty_throttle_control *dtc) /* * Calculate this BDI's share of the thresh ratio. */ - fprop_fraction_percpu(&dom->completions, &dtc->wb->completions, + fprop_fraction_percpu(&dom->completions, dtc->wb_completions, &numerator, &denominator); wb_thresh = (thresh * (100 - bdi_min_ratio)) / 100; From e9f07dfd7086a0b7e9ce98bb97b7422861aad40b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:28 -0400 Subject: [PATCH 62/85] writeback: add dirty_throttle_control->dom Currently all dirty throttle operations use global_wb_domain; however, cgroup writeback support requires considering per-memcg wb_domain too. This patch adds dirty_throttle_control->dom and updates functions which are directly using globabl_wb_domain to use it instead. As this makes global_update_bandwidth() a misnomer, the function is renamed to domain_update_bandwidth(). This patch doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 30 ++++++++++++++++++++++++------ 1 file changed, 24 insertions(+), 6 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 5b439fc1e95e..38d45d8c71f5 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -126,6 +126,9 @@ struct wb_domain global_wb_domain; /* consolidated parameters for balance_dirty_pages() and its subroutines */ struct dirty_throttle_control { +#ifdef CONFIG_CGROUP_WRITEBACK + struct wb_domain *dom; +#endif struct bdi_writeback *wb; struct fprop_local_percpu *wb_completions; @@ -140,7 +143,7 @@ struct dirty_throttle_control { unsigned long pos_ratio; }; -#define GDTC_INIT(__wb) .wb = (__wb), \ +#define DTC_INIT_COMMON(__wb) .wb = (__wb), \ .wb_completions = &(__wb)->completions /* @@ -152,6 +155,14 @@ struct dirty_throttle_control { #ifdef CONFIG_CGROUP_WRITEBACK +#define GDTC_INIT(__wb) .dom = &global_wb_domain, \ + DTC_INIT_COMMON(__wb) + +static struct wb_domain *dtc_dom(struct dirty_throttle_control *dtc) +{ + return dtc->dom; +} + static void wb_min_max_ratio(struct bdi_writeback *wb, unsigned long *minp, unsigned long *maxp) { @@ -181,6 +192,13 @@ static void wb_min_max_ratio(struct bdi_writeback *wb, #else /* CONFIG_CGROUP_WRITEBACK */ +#define GDTC_INIT(__wb) DTC_INIT_COMMON(__wb) + +static struct wb_domain *dtc_dom(struct dirty_throttle_control *dtc) +{ + return &global_wb_domain; +} + static void wb_min_max_ratio(struct bdi_writeback *wb, unsigned long *minp, unsigned long *maxp) { @@ -583,7 +601,7 @@ static unsigned long hard_dirty_limit(unsigned long thresh) */ static unsigned long __wb_calc_thresh(struct dirty_throttle_control *dtc) { - struct wb_domain *dom = &global_wb_domain; + struct wb_domain *dom = dtc_dom(dtc); unsigned long thresh = dtc->thresh; u64 wb_thresh; long numerator, denominator; @@ -952,7 +970,7 @@ out: static void update_dirty_limit(struct dirty_throttle_control *dtc) { - struct wb_domain *dom = &global_wb_domain; + struct wb_domain *dom = dtc_dom(dtc); unsigned long thresh = dtc->thresh; unsigned long limit = dom->dirty_limit; @@ -979,10 +997,10 @@ update: dom->dirty_limit = limit; } -static void global_update_bandwidth(struct dirty_throttle_control *dtc, +static void domain_update_bandwidth(struct dirty_throttle_control *dtc, unsigned long now) { - struct wb_domain *dom = &global_wb_domain; + struct wb_domain *dom = dtc_dom(dtc); /* * check locklessly first to optimize away locking for the most time @@ -1190,7 +1208,7 @@ static void __wb_update_bandwidth(struct dirty_throttle_control *dtc, goto snapshot; if (update_ratelimit) { - global_update_bandwidth(dtc, now); + domain_update_bandwidth(dtc, now); wb_update_dirty_ratelimit(dtc, dirtied, elapsed); } wb_update_write_bandwidth(wb, elapsed, written); From c7981433ef05e67b1b40740b2c40edbd4476b659 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:29 -0400 Subject: [PATCH 63/85] writeback: make __wb_writeout_inc() and hard_dirty_limit() take wb_domaas a parameter Currently __wb_writeout_inc() and hard_dirty_limit() assume global_wb_domain; however, cgroup writeback support requires considering per-memcg wb_domain too. This patch separates out domain-specific part of __wb_writeout_inc() into wb_domain_writeout_inc() which takes wb_domain as a parameter and adds the parameter to hard_dirty_limit(). This will allow these two functions to handle per-memcg wb_domains. This patch doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 37 +++++++++++++++++++++---------------- 1 file changed, 21 insertions(+), 16 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 38d45d8c71f5..a4d0ceefd097 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -445,17 +445,12 @@ static unsigned long wp_next_time(unsigned long cur_time) return cur_time; } -/* - * Increment the wb's writeout completion count and the global writeout - * completion count. Called from test_clear_page_writeback(). - */ -static inline void __wb_writeout_inc(struct bdi_writeback *wb) +static void wb_domain_writeout_inc(struct wb_domain *dom, + struct fprop_local_percpu *completions, + unsigned int max_prop_frac) { - struct wb_domain *dom = &global_wb_domain; - - __inc_wb_stat(wb, WB_WRITTEN); - __fprop_inc_percpu_max(&dom->completions, &wb->completions, - wb->bdi->max_prop_frac); + __fprop_inc_percpu_max(&dom->completions, completions, + max_prop_frac); /* First event after period switching was turned off? */ if (!unlikely(dom->period_time)) { /* @@ -469,6 +464,17 @@ static inline void __wb_writeout_inc(struct bdi_writeback *wb) } } +/* + * Increment @wb's writeout completion count and the global writeout + * completion count. Called from test_clear_page_writeback(). + */ +static inline void __wb_writeout_inc(struct bdi_writeback *wb) +{ + __inc_wb_stat(wb, WB_WRITTEN); + wb_domain_writeout_inc(&global_wb_domain, &wb->completions, + wb->bdi->max_prop_frac); +} + void wb_writeout_inc(struct bdi_writeback *wb) { unsigned long flags; @@ -571,10 +577,9 @@ static unsigned long dirty_freerun_ceiling(unsigned long thresh, return (thresh + bg_thresh) / 2; } -static unsigned long hard_dirty_limit(unsigned long thresh) +static unsigned long hard_dirty_limit(struct wb_domain *dom, + unsigned long thresh) { - struct wb_domain *dom = &global_wb_domain; - return max(thresh, dom->dirty_limit); } @@ -744,7 +749,7 @@ static void wb_position_ratio(struct dirty_throttle_control *dtc) struct bdi_writeback *wb = dtc->wb; unsigned long write_bw = wb->avg_write_bandwidth; unsigned long freerun = dirty_freerun_ceiling(dtc->thresh, dtc->bg_thresh); - unsigned long limit = hard_dirty_limit(dtc->thresh); + unsigned long limit = hard_dirty_limit(dtc_dom(dtc), dtc->thresh); unsigned long wb_thresh = dtc->wb_thresh; unsigned long x_intercept; unsigned long setpoint; /* dirty pages' target balance point */ @@ -1029,7 +1034,7 @@ static void wb_update_dirty_ratelimit(struct dirty_throttle_control *dtc, struct bdi_writeback *wb = dtc->wb; unsigned long dirty = dtc->dirty; unsigned long freerun = dirty_freerun_ceiling(dtc->thresh, dtc->bg_thresh); - unsigned long limit = hard_dirty_limit(dtc->thresh); + unsigned long limit = hard_dirty_limit(dtc_dom(dtc), dtc->thresh); unsigned long setpoint = (freerun + limit) / 2; unsigned long write_bw = wb->avg_write_bandwidth; unsigned long dirty_ratelimit = wb->dirty_ratelimit; @@ -1681,7 +1686,7 @@ void throttle_vm_writeout(gfp_t gfp_mask) for ( ; ; ) { global_dirty_limits(&background_thresh, &dirty_thresh); - dirty_thresh = hard_dirty_limit(dirty_thresh); + dirty_thresh = hard_dirty_limit(&global_wb_domain, dirty_thresh); /* * Boost the allowable dirty threshold a bit for page From 9fc3a43e1757ab6de0e8ce83b5d5a83083174e3b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:30 -0400 Subject: [PATCH 64/85] writeback: separate out domain_dirty_limits() global_dirty_limits() calculates thresh and bg_thresh (confusingly called *pdirty and *pbackground in the function) assuming global_wb_domain; however, cgroup writeback support requires considering per-memcg wb_domain too. This patch separates out domain_dirty_limits() which takes dirty_throttle_control out of global_dirty_limits(). As thresh and bg_thresh calculation needs the amount of dirtyable memory in the domain, dirty_throttle_control->avail is added. The new function calculates the two thresholds and store them directly in the dirty_throttle_control. Also, as memcg domains can't follow vm_dirty_bytes and dirty_background_bytes settings directly. If those are set and domain_dirty_limits() is invoked for a !global domain, the settings are translated to ratios by scaling them against globally available memory. dirty_throttle_control->gdtc is added to enable this when CONFIG_CGROUP_WRITEBACK. global_dirty_limits() is now a thin wrapper around domain_dirty_limits() and balance_dirty_pages() is updated to use the new function too. This patch doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 121 +++++++++++++++++++++++++++++++++----------- 1 file changed, 91 insertions(+), 30 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index a4d0ceefd097..c8ac8cea67dc 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -128,10 +128,12 @@ struct wb_domain global_wb_domain; struct dirty_throttle_control { #ifdef CONFIG_CGROUP_WRITEBACK struct wb_domain *dom; + struct dirty_throttle_control *gdtc; /* only set in memcg dtc's */ #endif struct bdi_writeback *wb; struct fprop_local_percpu *wb_completions; + unsigned long avail; /* dirtyable */ unsigned long dirty; /* file_dirty + write + nfs */ unsigned long thresh; /* dirty threshold */ unsigned long bg_thresh; /* dirty background threshold */ @@ -157,12 +159,18 @@ struct dirty_throttle_control { #define GDTC_INIT(__wb) .dom = &global_wb_domain, \ DTC_INIT_COMMON(__wb) +#define GDTC_INIT_NO_WB .dom = &global_wb_domain static struct wb_domain *dtc_dom(struct dirty_throttle_control *dtc) { return dtc->dom; } +static struct dirty_throttle_control *mdtc_gdtc(struct dirty_throttle_control *mdtc) +{ + return mdtc->gdtc; +} + static void wb_min_max_ratio(struct bdi_writeback *wb, unsigned long *minp, unsigned long *maxp) { @@ -193,12 +201,18 @@ static void wb_min_max_ratio(struct bdi_writeback *wb, #else /* CONFIG_CGROUP_WRITEBACK */ #define GDTC_INIT(__wb) DTC_INIT_COMMON(__wb) +#define GDTC_INIT_NO_WB static struct wb_domain *dtc_dom(struct dirty_throttle_control *dtc) { return &global_wb_domain; } +static struct dirty_throttle_control *mdtc_gdtc(struct dirty_throttle_control *mdtc) +{ + return NULL; +} + static void wb_min_max_ratio(struct bdi_writeback *wb, unsigned long *minp, unsigned long *maxp) { @@ -303,42 +317,88 @@ static unsigned long global_dirtyable_memory(void) return x + 1; /* Ensure that we never return 0 */ } -/* - * global_dirty_limits - background-writeback and dirty-throttling thresholds +/** + * domain_dirty_limits - calculate thresh and bg_thresh for a wb_domain + * @dtc: dirty_throttle_control of interest * - * Calculate the dirty thresholds based on sysctl parameters - * - vm.dirty_background_ratio or vm.dirty_background_bytes - * - vm.dirty_ratio or vm.dirty_bytes - * The dirty limits will be lifted by 1/4 for PF_LESS_THROTTLE (ie. nfsd) and + * Calculate @dtc->thresh and ->bg_thresh considering + * vm_dirty_{bytes|ratio} and dirty_background_{bytes|ratio}. The caller + * must ensure that @dtc->avail is set before calling this function. The + * dirty limits will be lifted by 1/4 for PF_LESS_THROTTLE (ie. nfsd) and * real-time tasks. */ +static void domain_dirty_limits(struct dirty_throttle_control *dtc) +{ + const unsigned long available_memory = dtc->avail; + struct dirty_throttle_control *gdtc = mdtc_gdtc(dtc); + unsigned long bytes = vm_dirty_bytes; + unsigned long bg_bytes = dirty_background_bytes; + unsigned long ratio = vm_dirty_ratio; + unsigned long bg_ratio = dirty_background_ratio; + unsigned long thresh; + unsigned long bg_thresh; + struct task_struct *tsk; + + /* gdtc is !NULL iff @dtc is for memcg domain */ + if (gdtc) { + unsigned long global_avail = gdtc->avail; + + /* + * The byte settings can't be applied directly to memcg + * domains. Convert them to ratios by scaling against + * globally available memory. + */ + if (bytes) + ratio = min(DIV_ROUND_UP(bytes, PAGE_SIZE) * 100 / + global_avail, 100UL); + if (bg_bytes) + bg_ratio = min(DIV_ROUND_UP(bg_bytes, PAGE_SIZE) * 100 / + global_avail, 100UL); + bytes = bg_bytes = 0; + } + + if (bytes) + thresh = DIV_ROUND_UP(bytes, PAGE_SIZE); + else + thresh = (ratio * available_memory) / 100; + + if (bg_bytes) + bg_thresh = DIV_ROUND_UP(bg_bytes, PAGE_SIZE); + else + bg_thresh = (bg_ratio * available_memory) / 100; + + if (bg_thresh >= thresh) + bg_thresh = thresh / 2; + tsk = current; + if (tsk->flags & PF_LESS_THROTTLE || rt_task(tsk)) { + bg_thresh += bg_thresh / 4; + thresh += thresh / 4; + } + dtc->thresh = thresh; + dtc->bg_thresh = bg_thresh; + + /* we should eventually report the domain in the TP */ + if (!gdtc) + trace_global_dirty_state(bg_thresh, thresh); +} + +/** + * global_dirty_limits - background-writeback and dirty-throttling thresholds + * @pbackground: out parameter for bg_thresh + * @pdirty: out parameter for thresh + * + * Calculate bg_thresh and thresh for global_wb_domain. See + * domain_dirty_limits() for details. + */ void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty) { - const unsigned long available_memory = global_dirtyable_memory(); - unsigned long background; - unsigned long dirty; - struct task_struct *tsk; + struct dirty_throttle_control gdtc = { GDTC_INIT_NO_WB }; - if (vm_dirty_bytes) - dirty = DIV_ROUND_UP(vm_dirty_bytes, PAGE_SIZE); - else - dirty = (vm_dirty_ratio * available_memory) / 100; + gdtc.avail = global_dirtyable_memory(); + domain_dirty_limits(&gdtc); - if (dirty_background_bytes) - background = DIV_ROUND_UP(dirty_background_bytes, PAGE_SIZE); - else - background = (dirty_background_ratio * available_memory) / 100; - - if (background >= dirty) - background = dirty / 2; - tsk = current; - if (tsk->flags & PF_LESS_THROTTLE || rt_task(tsk)) { - background += background / 4; - dirty += dirty / 4; - } - *pbackground = background; - *pdirty = dirty; - trace_global_dirty_state(background, dirty); + *pbackground = gdtc.bg_thresh; + *pdirty = gdtc.thresh; } /** @@ -1421,9 +1481,10 @@ static void balance_dirty_pages(struct address_space *mapping, */ nr_reclaimable = global_page_state(NR_FILE_DIRTY) + global_page_state(NR_UNSTABLE_NFS); + gdtc->avail = global_dirtyable_memory(); gdtc->dirty = nr_reclaimable + global_page_state(NR_WRITEBACK); - global_dirty_limits(&gdtc->bg_thresh, &gdtc->thresh); + domain_dirty_limits(gdtc); if (unlikely(strictlimit)) { wb_dirty_limits(gdtc); From aa661bbe1e61ce80ca4ae98804f673ede94b0827 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:31 -0400 Subject: [PATCH 65/85] writeback: move over_bground_thresh() to mm/page-writeback.c and rename it to wb_over_bg_thresh(). The function is closely tied to the dirty throttling mechanism implemented in page-writeback.c. This relocation will allow future updates necessary for cgroup writeback support. While at it, add function comment. This is pure reorganization and doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 20 ++------------------ include/linux/writeback.h | 1 + mm/page-writeback.c | 23 +++++++++++++++++++++++ 3 files changed, 26 insertions(+), 18 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 51c8a5b14cdf..da355879ba7c 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1071,22 +1071,6 @@ static long writeback_inodes_wb(struct bdi_writeback *wb, long nr_pages, return nr_pages - work.nr_pages; } -static bool over_bground_thresh(struct bdi_writeback *wb) -{ - unsigned long background_thresh, dirty_thresh; - - global_dirty_limits(&background_thresh, &dirty_thresh); - - if (global_page_state(NR_FILE_DIRTY) + - global_page_state(NR_UNSTABLE_NFS) > background_thresh) - return true; - - if (wb_stat(wb, WB_RECLAIMABLE) > wb_calc_thresh(wb, background_thresh)) - return true; - - return false; -} - /* * Explicit flushing or periodic writeback of "old" data. * @@ -1136,7 +1120,7 @@ static long wb_writeback(struct bdi_writeback *wb, * For background writeout, stop when we are below the * background dirty threshold */ - if (work->for_background && !over_bground_thresh(wb)) + if (work->for_background && !wb_over_bg_thresh(wb)) break; /* @@ -1227,7 +1211,7 @@ static unsigned long get_nr_dirty_pages(void) static long wb_check_background_flush(struct bdi_writeback *wb) { - if (over_bground_thresh(wb)) { + if (wb_over_bg_thresh(wb)) { struct wb_writeback_work work = { .nr_pages = LONG_MAX, diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 5fdd4e1805e6..b57c2786b5aa 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -207,6 +207,7 @@ unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh); void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long start_time); void page_writeback_init(void); void balance_dirty_pages_ratelimited(struct address_space *mapping); +bool wb_over_bg_thresh(struct bdi_writeback *wb); typedef int (*writepage_t)(struct page *page, struct writeback_control *wbc, void *data); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index c8ac8cea67dc..9d9a896fa7b5 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1740,6 +1740,29 @@ void balance_dirty_pages_ratelimited(struct address_space *mapping) } EXPORT_SYMBOL(balance_dirty_pages_ratelimited); +/** + * wb_over_bg_thresh - does @wb need to be written back? + * @wb: bdi_writeback of interest + * + * Determines whether background writeback should keep writing @wb or it's + * clean enough. Returns %true if writeback should continue. + */ +bool wb_over_bg_thresh(struct bdi_writeback *wb) +{ + unsigned long background_thresh, dirty_thresh; + + global_dirty_limits(&background_thresh, &dirty_thresh); + + if (global_page_state(NR_FILE_DIRTY) + + global_page_state(NR_UNSTABLE_NFS) > background_thresh) + return true; + + if (wb_stat(wb, WB_RECLAIMABLE) > wb_calc_thresh(wb, background_thresh)) + return true; + + return false; +} + void throttle_vm_writeout(gfp_t gfp_mask) { unsigned long background_thresh; From 947e9762a8ddefda38aa21e249e6a4fec215cd12 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:32 -0400 Subject: [PATCH 66/85] writeback: update wb_over_bg_thresh() to use wb_domain aware operations wb_over_bg_thresh() currently uses global_dirty_limits() and wb_dirty_limit() both of which are wrappers around operations which take dirty_throttle_control. For cgroup writeback support, the function will be updated to also consider memcg wb_domains which requires the context information carried in dirty_throttle_control. This patch updates wb_over_bg_thresh() so that it uses the underlying wb_domain aware operations directly and builds the global dirty_throttle_control in the process. This patch doesn't introduce any behavioral changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- mm/page-writeback.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 9d9a896fa7b5..a7ba5cee950b 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1749,15 +1749,22 @@ EXPORT_SYMBOL(balance_dirty_pages_ratelimited); */ bool wb_over_bg_thresh(struct bdi_writeback *wb) { - unsigned long background_thresh, dirty_thresh; + struct dirty_throttle_control gdtc_stor = { GDTC_INIT(wb) }; + struct dirty_throttle_control * const gdtc = &gdtc_stor; - global_dirty_limits(&background_thresh, &dirty_thresh); + /* + * Similar to balance_dirty_pages() but ignores pages being written + * as we're trying to decide whether to put more under writeback. + */ + gdtc->avail = global_dirtyable_memory(); + gdtc->dirty = global_page_state(NR_FILE_DIRTY) + + global_page_state(NR_UNSTABLE_NFS); + domain_dirty_limits(gdtc); - if (global_page_state(NR_FILE_DIRTY) + - global_page_state(NR_UNSTABLE_NFS) > background_thresh) + if (gdtc->dirty > gdtc->bg_thresh) return true; - if (wb_stat(wb, WB_RECLAIMABLE) > wb_calc_thresh(wb, background_thresh)) + if (wb_stat(wb, WB_RECLAIMABLE) > __wb_calc_thresh(gdtc)) return true; return false; From 841710aa6e4acd066ab9fe8c8cb6f4e4e6709d83 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:33 -0400 Subject: [PATCH 67/85] writeback: implement memcg wb_domain Dirtyable memory is distributed to a wb (bdi_writeback) according to the relative bandwidth the wb is writing out in the whole system. This distribution is global - each wb is measured against all other wb's and gets the proportinately sized portion of the memory in the whole system. For cgroup writeback, the amount of dirtyable memory is scoped by memcg and thus each wb would need to be measured and controlled in its memcg. IOW, a wb will belong to two writeback domains - the global and memcg domains. The previous patches laid the groundwork to support the two wb_domains and this patch implements memcg wb_domain. memcg->cgwb_domain is initialized on css online and destroyed on css release, wb->memcg_completions is added, and __wb_writeout_inc() is updated to increment completions against both global and memcg wb_domains. The following patches will update balance_dirty_pages() and its subroutines to actually consider memcg wb_domain for throttling. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- include/linux/backing-dev-defs.h | 1 + include/linux/memcontrol.h | 12 +++++++++- include/linux/writeback.h | 3 +++ mm/backing-dev.c | 9 +++++++- mm/memcontrol.c | 39 ++++++++++++++++++++++++++++++++ mm/page-writeback.c | 25 ++++++++++++++++++++ 6 files changed, 87 insertions(+), 2 deletions(-) diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index 97a92fa0cdb5..8d470b73824f 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -118,6 +118,7 @@ struct bdi_writeback { #ifdef CONFIG_CGROUP_WRITEBACK struct percpu_ref refcnt; /* used only for !root wb's */ + struct fprop_local_percpu memcg_completions; struct cgroup_subsys_state *memcg_css; /* the associated memcg */ struct cgroup_subsys_state *blkcg_css; /* and blkcg */ struct list_head memcg_node; /* anchored at memcg->cgwb_list */ diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 662a953ea8ad..e3177bed23ea 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -389,8 +389,18 @@ enum { }; #ifdef CONFIG_CGROUP_WRITEBACK + struct list_head *mem_cgroup_cgwb_list(struct mem_cgroup *memcg); -#endif +struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb); + +#else /* CONFIG_CGROUP_WRITEBACK */ + +static inline struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb) +{ + return NULL; +} + +#endif /* CONFIG_CGROUP_WRITEBACK */ struct sock; #if defined(CONFIG_INET) && defined(CONFIG_MEMCG_KMEM) diff --git a/include/linux/writeback.h b/include/linux/writeback.h index b57c2786b5aa..04a3786c456f 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -167,6 +167,9 @@ static inline void laptop_sync_completion(void) { } void throttle_vm_writeout(gfp_t gfp_mask); bool zone_dirty_ok(struct zone *zone); int wb_domain_init(struct wb_domain *dom, gfp_t gfp); +#ifdef CONFIG_CGROUP_WRITEBACK +void wb_domain_exit(struct wb_domain *dom); +#endif extern struct wb_domain global_wb_domain; diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 9c8b7b5a0eee..84ebf7c8d006 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -482,6 +482,7 @@ static void cgwb_release_workfn(struct work_struct *work) css_put(wb->blkcg_css); wb_congested_put(wb->congested); + fprop_local_destroy_percpu(&wb->memcg_completions); percpu_ref_exit(&wb->refcnt); wb_exit(wb); kfree_rcu(wb, rcu); @@ -548,9 +549,13 @@ static int cgwb_create(struct backing_dev_info *bdi, if (ret) goto err_wb_exit; + ret = fprop_local_init_percpu(&wb->memcg_completions, gfp); + if (ret) + goto err_ref_exit; + wb->congested = wb_congested_get_create(bdi, blkcg_css->id, gfp); if (!wb->congested) - goto err_ref_exit; + goto err_fprop_exit; wb->memcg_css = memcg_css; wb->blkcg_css = blkcg_css; @@ -587,6 +592,8 @@ static int cgwb_create(struct backing_dev_info *bdi, err_put_congested: wb_congested_put(wb->congested); +err_fprop_exit: + fprop_local_destroy_percpu(&wb->memcg_completions); err_ref_exit: percpu_ref_exit(&wb->refcnt); err_wb_exit: diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 701cbee9acba..ce113ddf2fb5 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -345,6 +345,7 @@ struct mem_cgroup { #ifdef CONFIG_CGROUP_WRITEBACK struct list_head cgwb_list; + struct wb_domain cgwb_domain; #endif /* List of events which userspace want to receive */ @@ -3994,6 +3995,37 @@ struct list_head *mem_cgroup_cgwb_list(struct mem_cgroup *memcg) return &memcg->cgwb_list; } +static int memcg_wb_domain_init(struct mem_cgroup *memcg, gfp_t gfp) +{ + return wb_domain_init(&memcg->cgwb_domain, gfp); +} + +static void memcg_wb_domain_exit(struct mem_cgroup *memcg) +{ + wb_domain_exit(&memcg->cgwb_domain); +} + +struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb) +{ + struct mem_cgroup *memcg = mem_cgroup_from_css(wb->memcg_css); + + if (!memcg->css.parent) + return NULL; + + return &memcg->cgwb_domain; +} + +#else /* CONFIG_CGROUP_WRITEBACK */ + +static int memcg_wb_domain_init(struct mem_cgroup *memcg, gfp_t gfp) +{ + return 0; +} + +static void memcg_wb_domain_exit(struct mem_cgroup *memcg) +{ +} + #endif /* CONFIG_CGROUP_WRITEBACK */ /* @@ -4380,9 +4412,15 @@ static struct mem_cgroup *mem_cgroup_alloc(void) memcg->stat = alloc_percpu(struct mem_cgroup_stat_cpu); if (!memcg->stat) goto out_free; + + if (memcg_wb_domain_init(memcg, GFP_KERNEL)) + goto out_free_stat; + spin_lock_init(&memcg->pcp_counter_lock); return memcg; +out_free_stat: + free_percpu(memcg->stat); out_free: kfree(memcg); return NULL; @@ -4409,6 +4447,7 @@ static void __mem_cgroup_free(struct mem_cgroup *memcg) free_mem_cgroup_per_zone_info(memcg, node); free_percpu(memcg->stat); + memcg_wb_domain_exit(memcg); kfree(memcg); } diff --git a/mm/page-writeback.c b/mm/page-writeback.c index a7ba5cee950b..a146e3389e78 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -171,6 +171,11 @@ static struct dirty_throttle_control *mdtc_gdtc(struct dirty_throttle_control *m return mdtc->gdtc; } +static struct fprop_local_percpu *wb_memcg_completions(struct bdi_writeback *wb) +{ + return &wb->memcg_completions; +} + static void wb_min_max_ratio(struct bdi_writeback *wb, unsigned long *minp, unsigned long *maxp) { @@ -213,6 +218,11 @@ static struct dirty_throttle_control *mdtc_gdtc(struct dirty_throttle_control *m return NULL; } +static struct fprop_local_percpu *wb_memcg_completions(struct bdi_writeback *wb) +{ + return NULL; +} + static void wb_min_max_ratio(struct bdi_writeback *wb, unsigned long *minp, unsigned long *maxp) { @@ -530,9 +540,16 @@ static void wb_domain_writeout_inc(struct wb_domain *dom, */ static inline void __wb_writeout_inc(struct bdi_writeback *wb) { + struct wb_domain *cgdom; + __inc_wb_stat(wb, WB_WRITTEN); wb_domain_writeout_inc(&global_wb_domain, &wb->completions, wb->bdi->max_prop_frac); + + cgdom = mem_cgroup_wb_domain(wb); + if (cgdom) + wb_domain_writeout_inc(cgdom, wb_memcg_completions(wb), + wb->bdi->max_prop_frac); } void wb_writeout_inc(struct bdi_writeback *wb) @@ -583,6 +600,14 @@ int wb_domain_init(struct wb_domain *dom, gfp_t gfp) return fprop_global_init(&dom->completions, gfp); } +#ifdef CONFIG_CGROUP_WRITEBACK +void wb_domain_exit(struct wb_domain *dom) +{ + del_timer_sync(&dom->period_timer); + fprop_global_destroy(&dom->completions); +} +#endif + /* * bdi_min_ratio keeps the sum of the minimum dirty shares of all * registered backing devices, which, for obvious reasons, can not From 2529bb3aadc40a93e642f5f3650f63379a964467 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:34 -0400 Subject: [PATCH 68/85] writeback: reset wb_domain->dirty_limit[_tstmp] when memcg domain size changes The amount of available memory to a memcg wb_domain can change as memcg configuration changes. A domain's ->dirty_limit exists to smooth out sudden drops in dirty threshold; however, when a domain's size actually drops significantly, it hinders the dirty throttling from adjusting to the new configuration leading to unexpected behaviors including unnecessary OOM kills. This patch resolves the issue by adding wb_domain_size_changed() which resets ->dirty_limit[_tstmp] and making memcg call it on configuration changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- include/linux/writeback.h | 20 ++++++++++++++++++++ mm/memcontrol.c | 12 ++++++++++++ 2 files changed, 32 insertions(+) diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 04a3786c456f..3b73e97ecfc7 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -132,6 +132,26 @@ struct wb_domain { unsigned long dirty_limit; }; +/** + * wb_domain_size_changed - memory available to a wb_domain has changed + * @dom: wb_domain of interest + * + * This function should be called when the amount of memory available to + * @dom has changed. It resets @dom's dirty limit parameters to prevent + * the past values which don't match the current configuration from skewing + * dirty throttling. Without this, when memory size of a wb_domain is + * greatly reduced, the dirty throttling logic may allow too many pages to + * be dirtied leading to consecutive unnecessary OOMs and may get stuck in + * that situation. + */ +static inline void wb_domain_size_changed(struct wb_domain *dom) +{ + spin_lock(&dom->lock); + dom->dirty_limit_tstamp = jiffies; + dom->dirty_limit = 0; + spin_unlock(&dom->lock); +} + /* * fs/fs-writeback.c */ diff --git a/mm/memcontrol.c b/mm/memcontrol.c index ce113ddf2fb5..c0b0406ae5ca 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4005,6 +4005,11 @@ static void memcg_wb_domain_exit(struct mem_cgroup *memcg) wb_domain_exit(&memcg->cgwb_domain); } +static void memcg_wb_domain_size_changed(struct mem_cgroup *memcg) +{ + wb_domain_size_changed(&memcg->cgwb_domain); +} + struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb) { struct mem_cgroup *memcg = mem_cgroup_from_css(wb->memcg_css); @@ -4026,6 +4031,10 @@ static void memcg_wb_domain_exit(struct mem_cgroup *memcg) { } +static void memcg_wb_domain_size_changed(struct mem_cgroup *memcg) +{ +} + #endif /* CONFIG_CGROUP_WRITEBACK */ /* @@ -4624,6 +4633,7 @@ static void mem_cgroup_css_reset(struct cgroup_subsys_state *css) memcg->low = 0; memcg->high = PAGE_COUNTER_MAX; memcg->soft_limit = PAGE_COUNTER_MAX; + memcg_wb_domain_size_changed(memcg); } #ifdef CONFIG_MMU @@ -5361,6 +5371,7 @@ static ssize_t memory_high_write(struct kernfs_open_file *of, memcg->high = high; + memcg_wb_domain_size_changed(memcg); return nbytes; } @@ -5393,6 +5404,7 @@ static ssize_t memory_max_write(struct kernfs_open_file *of, if (err) return err; + memcg_wb_domain_size_changed(memcg); return nbytes; } From c2aa723a6093633ae4ec15b08a4db276643cab3e Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:35 -0400 Subject: [PATCH 69/85] writeback: implement memcg writeback domain based throttling While cgroup writeback support now connects memcg and blkcg so that writeback IOs are properly attributed and controlled, the IO back pressure propagation mechanism implemented in balance_dirty_pages() and its subroutines wasn't aware of cgroup writeback. Processes belonging to a memcg may have access to only subset of total memory available in the system and not factoring this into dirty throttling rendered it completely ineffective for processes under memcg limits and memcg ended up building a separate ad-hoc degenerate mechanism directly into vmscan code to limit page dirtying. The previous patches updated balance_dirty_pages() and its subroutines so that they can deal with multiple wb_domain's (writeback domains) and defined per-memcg wb_domain. Processes belonging to a non-root memcg are bound to two wb_domains, global wb_domain and memcg wb_domain, and should be throttled according to IO pressures from both domains. This patch updates dirty throttling code so that it repeats similar calculations for the two domains - the differences between the two are few and minor - and applies the lower of the two sets of resulting constraints. wb_over_bg_thresh(), which controls when background writeback terminates, is also updated to consider both global and memcg wb_domains. It returns true if dirty is over bg_thresh for either domain. This makes the dirty throttling mechanism operational for memcg domains including writeback-bandwidth-proportional dirty page distribution inside them but the ad-hoc memcg throttling mechanism in vmscan is still in place. The next patch will rip it out. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- include/linux/memcontrol.h | 9 +++ mm/memcontrol.c | 43 ++++++++++ mm/page-writeback.c | 158 +++++++++++++++++++++++++++++++------ 3 files changed, 188 insertions(+), 22 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index e3177bed23ea..c3eb19e2bc1c 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -392,6 +392,8 @@ enum { struct list_head *mem_cgroup_cgwb_list(struct mem_cgroup *memcg); struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb); +void mem_cgroup_wb_stats(struct bdi_writeback *wb, unsigned long *pavail, + unsigned long *pdirty, unsigned long *pwriteback); #else /* CONFIG_CGROUP_WRITEBACK */ @@ -400,6 +402,13 @@ static inline struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb) return NULL; } +static inline void mem_cgroup_wb_stats(struct bdi_writeback *wb, + unsigned long *pavail, + unsigned long *pdirty, + unsigned long *pwriteback) +{ +} + #endif /* CONFIG_CGROUP_WRITEBACK */ struct sock; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index c0b0406ae5ca..f816d91c643b 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4020,6 +4020,49 @@ struct wb_domain *mem_cgroup_wb_domain(struct bdi_writeback *wb) return &memcg->cgwb_domain; } +/** + * mem_cgroup_wb_stats - retrieve writeback related stats from its memcg + * @wb: bdi_writeback in question + * @pavail: out parameter for number of available pages + * @pdirty: out parameter for number of dirty pages + * @pwriteback: out parameter for number of pages under writeback + * + * Determine the numbers of available, dirty, and writeback pages in @wb's + * memcg. Dirty and writeback are self-explanatory. Available is a bit + * more involved. + * + * A memcg's headroom is "min(max, high) - used". The available memory is + * calculated as the lowest headroom of itself and the ancestors plus the + * number of pages already being used for file pages. Note that this + * doesn't consider the actual amount of available memory in the system. + * The caller should further cap *@pavail accordingly. + */ +void mem_cgroup_wb_stats(struct bdi_writeback *wb, unsigned long *pavail, + unsigned long *pdirty, unsigned long *pwriteback) +{ + struct mem_cgroup *memcg = mem_cgroup_from_css(wb->memcg_css); + struct mem_cgroup *parent; + unsigned long head_room = PAGE_COUNTER_MAX; + unsigned long file_pages; + + *pdirty = mem_cgroup_read_stat(memcg, MEM_CGROUP_STAT_DIRTY); + + /* this should eventually include NR_UNSTABLE_NFS */ + *pwriteback = mem_cgroup_read_stat(memcg, MEM_CGROUP_STAT_WRITEBACK); + + file_pages = mem_cgroup_nr_lru_pages(memcg, (1 << LRU_INACTIVE_FILE) | + (1 << LRU_ACTIVE_FILE)); + while ((parent = parent_mem_cgroup(memcg))) { + unsigned long ceiling = min(memcg->memory.limit, memcg->high); + unsigned long used = page_counter_read(&memcg->memory); + + head_room = min(head_room, ceiling - min(ceiling, used)); + memcg = parent; + } + + *pavail = file_pages + head_room; +} + #else /* CONFIG_CGROUP_WRITEBACK */ static int memcg_wb_domain_init(struct mem_cgroup *memcg, gfp_t gfp) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index a146e3389e78..e8903356f423 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -160,6 +160,14 @@ struct dirty_throttle_control { #define GDTC_INIT(__wb) .dom = &global_wb_domain, \ DTC_INIT_COMMON(__wb) #define GDTC_INIT_NO_WB .dom = &global_wb_domain +#define MDTC_INIT(__wb, __gdtc) .dom = mem_cgroup_wb_domain(__wb), \ + .gdtc = __gdtc, \ + DTC_INIT_COMMON(__wb) + +static bool mdtc_valid(struct dirty_throttle_control *dtc) +{ + return dtc->dom; +} static struct wb_domain *dtc_dom(struct dirty_throttle_control *dtc) { @@ -207,6 +215,12 @@ static void wb_min_max_ratio(struct bdi_writeback *wb, #define GDTC_INIT(__wb) DTC_INIT_COMMON(__wb) #define GDTC_INIT_NO_WB +#define MDTC_INIT(__wb, __gdtc) + +static bool mdtc_valid(struct dirty_throttle_control *dtc) +{ + return false; +} static struct wb_domain *dtc_dom(struct dirty_throttle_control *dtc) { @@ -668,6 +682,15 @@ static unsigned long hard_dirty_limit(struct wb_domain *dom, return max(thresh, dom->dirty_limit); } +/* memory available to a memcg domain is capped by system-wide clean memory */ +static void mdtc_cap_avail(struct dirty_throttle_control *mdtc) +{ + struct dirty_throttle_control *gdtc = mdtc_gdtc(mdtc); + unsigned long clean = gdtc->avail - min(gdtc->avail, gdtc->dirty); + + mdtc->avail = min(mdtc->avail, clean); +} + /** * __wb_calc_thresh - @wb's share of dirty throttling threshold * @dtc: dirty_throttle_context of interest @@ -1269,11 +1292,12 @@ static void wb_update_dirty_ratelimit(struct dirty_throttle_control *dtc, trace_bdi_dirty_ratelimit(wb->bdi, dirty_rate, task_ratelimit); } -static void __wb_update_bandwidth(struct dirty_throttle_control *dtc, +static void __wb_update_bandwidth(struct dirty_throttle_control *gdtc, + struct dirty_throttle_control *mdtc, unsigned long start_time, bool update_ratelimit) { - struct bdi_writeback *wb = dtc->wb; + struct bdi_writeback *wb = gdtc->wb; unsigned long now = jiffies; unsigned long elapsed = now - wb->bw_time_stamp; unsigned long dirtied; @@ -1298,8 +1322,17 @@ static void __wb_update_bandwidth(struct dirty_throttle_control *dtc, goto snapshot; if (update_ratelimit) { - domain_update_bandwidth(dtc, now); - wb_update_dirty_ratelimit(dtc, dirtied, elapsed); + domain_update_bandwidth(gdtc, now); + wb_update_dirty_ratelimit(gdtc, dirtied, elapsed); + + /* + * @mdtc is always NULL if !CGROUP_WRITEBACK but the + * compiler has no way to figure that out. Help it. + */ + if (IS_ENABLED(CONFIG_CGROUP_WRITEBACK) && mdtc) { + domain_update_bandwidth(mdtc, now); + wb_update_dirty_ratelimit(mdtc, dirtied, elapsed); + } } wb_update_write_bandwidth(wb, elapsed, written); @@ -1313,7 +1346,7 @@ void wb_update_bandwidth(struct bdi_writeback *wb, unsigned long start_time) { struct dirty_throttle_control gdtc = { GDTC_INIT(wb) }; - __wb_update_bandwidth(&gdtc, start_time, false); + __wb_update_bandwidth(&gdtc, NULL, start_time, false); } /* @@ -1480,7 +1513,11 @@ static void balance_dirty_pages(struct address_space *mapping, unsigned long pages_dirtied) { struct dirty_throttle_control gdtc_stor = { GDTC_INIT(wb) }; + struct dirty_throttle_control mdtc_stor = { MDTC_INIT(wb, &gdtc_stor) }; struct dirty_throttle_control * const gdtc = &gdtc_stor; + struct dirty_throttle_control * const mdtc = mdtc_valid(&mdtc_stor) ? + &mdtc_stor : NULL; + struct dirty_throttle_control *sdtc; unsigned long nr_reclaimable; /* = file_dirty + unstable_nfs */ long period; long pause; @@ -1497,6 +1534,7 @@ static void balance_dirty_pages(struct address_space *mapping, for (;;) { unsigned long now = jiffies; unsigned long dirty, thresh, bg_thresh; + unsigned long m_dirty, m_thresh, m_bg_thresh; /* * Unstable writes are a feature of certain networked @@ -1523,6 +1561,32 @@ static void balance_dirty_pages(struct address_space *mapping, bg_thresh = gdtc->bg_thresh; } + if (mdtc) { + unsigned long writeback; + + /* + * If @wb belongs to !root memcg, repeat the same + * basic calculations for the memcg domain. + */ + mem_cgroup_wb_stats(wb, &mdtc->avail, &mdtc->dirty, + &writeback); + mdtc_cap_avail(mdtc); + mdtc->dirty += writeback; + + domain_dirty_limits(mdtc); + + if (unlikely(strictlimit)) { + wb_dirty_limits(mdtc); + m_dirty = mdtc->wb_dirty; + m_thresh = mdtc->wb_thresh; + m_bg_thresh = mdtc->wb_bg_thresh; + } else { + m_dirty = mdtc->dirty; + m_thresh = mdtc->thresh; + m_bg_thresh = mdtc->bg_thresh; + } + } + /* * Throttle it only when the background writeback cannot * catch-up. This avoids (excessively) small writeouts @@ -1531,18 +1595,31 @@ static void balance_dirty_pages(struct address_space *mapping, * In strictlimit case make decision based on the wb counters * and limits. Small writeouts when the wb limits are ramping * up are the price we consciously pay for strictlimit-ing. + * + * If memcg domain is in effect, @dirty should be under + * both global and memcg freerun ceilings. */ - if (dirty <= dirty_freerun_ceiling(thresh, bg_thresh)) { + if (dirty <= dirty_freerun_ceiling(thresh, bg_thresh) && + (!mdtc || + m_dirty <= dirty_freerun_ceiling(m_thresh, m_bg_thresh))) { + unsigned long intv = dirty_poll_interval(dirty, thresh); + unsigned long m_intv = ULONG_MAX; + current->dirty_paused_when = now; current->nr_dirtied = 0; - current->nr_dirtied_pause = - dirty_poll_interval(dirty, thresh); + if (mdtc) + m_intv = dirty_poll_interval(m_dirty, m_thresh); + current->nr_dirtied_pause = min(intv, m_intv); break; } if (unlikely(!writeback_in_progress(wb))) wb_start_background_writeback(wb); + /* + * Calculate global domain's pos_ratio and select the + * global dtc by default. + */ if (!strictlimit) wb_dirty_limits(gdtc); @@ -1550,6 +1627,25 @@ static void balance_dirty_pages(struct address_space *mapping, ((gdtc->dirty > gdtc->thresh) || strictlimit); wb_position_ratio(gdtc); + sdtc = gdtc; + + if (mdtc) { + /* + * If memcg domain is in effect, calculate its + * pos_ratio. @wb should satisfy constraints from + * both global and memcg domains. Choose the one + * w/ lower pos_ratio. + */ + if (!strictlimit) + wb_dirty_limits(mdtc); + + dirty_exceeded |= (mdtc->wb_dirty > mdtc->wb_thresh) && + ((mdtc->dirty > mdtc->thresh) || strictlimit); + + wb_position_ratio(mdtc); + if (mdtc->pos_ratio < gdtc->pos_ratio) + sdtc = mdtc; + } if (dirty_exceeded && !wb->dirty_exceeded) wb->dirty_exceeded = 1; @@ -1557,14 +1653,15 @@ static void balance_dirty_pages(struct address_space *mapping, if (time_is_before_jiffies(wb->bw_time_stamp + BANDWIDTH_INTERVAL)) { spin_lock(&wb->list_lock); - __wb_update_bandwidth(gdtc, start_time, true); + __wb_update_bandwidth(gdtc, mdtc, start_time, true); spin_unlock(&wb->list_lock); } + /* throttle according to the chosen dtc */ dirty_ratelimit = wb->dirty_ratelimit; - task_ratelimit = ((u64)dirty_ratelimit * gdtc->pos_ratio) >> + task_ratelimit = ((u64)dirty_ratelimit * sdtc->pos_ratio) >> RATELIMIT_CALC_SHIFT; - max_pause = wb_max_pause(wb, gdtc->wb_dirty); + max_pause = wb_max_pause(wb, sdtc->wb_dirty); min_pause = wb_min_pause(wb, max_pause, task_ratelimit, dirty_ratelimit, &nr_dirtied_pause); @@ -1587,11 +1684,11 @@ static void balance_dirty_pages(struct address_space *mapping, */ if (pause < min_pause) { trace_balance_dirty_pages(bdi, - gdtc->thresh, - gdtc->bg_thresh, - gdtc->dirty, - gdtc->wb_thresh, - gdtc->wb_dirty, + sdtc->thresh, + sdtc->bg_thresh, + sdtc->dirty, + sdtc->wb_thresh, + sdtc->wb_dirty, dirty_ratelimit, task_ratelimit, pages_dirtied, @@ -1616,11 +1713,11 @@ static void balance_dirty_pages(struct address_space *mapping, pause: trace_balance_dirty_pages(bdi, - gdtc->thresh, - gdtc->bg_thresh, - gdtc->dirty, - gdtc->wb_thresh, - gdtc->wb_dirty, + sdtc->thresh, + sdtc->bg_thresh, + sdtc->dirty, + sdtc->wb_thresh, + sdtc->wb_dirty, dirty_ratelimit, task_ratelimit, pages_dirtied, @@ -1651,7 +1748,7 @@ pause: * more page. However wb_dirty has accounting errors. So use * the larger and more IO friendly wb_stat_error. */ - if (gdtc->wb_dirty <= wb_stat_error(wb)) + if (sdtc->wb_dirty <= wb_stat_error(wb)) break; if (fatal_signal_pending(current)) @@ -1775,7 +1872,10 @@ EXPORT_SYMBOL(balance_dirty_pages_ratelimited); bool wb_over_bg_thresh(struct bdi_writeback *wb) { struct dirty_throttle_control gdtc_stor = { GDTC_INIT(wb) }; + struct dirty_throttle_control mdtc_stor = { MDTC_INIT(wb, &gdtc_stor) }; struct dirty_throttle_control * const gdtc = &gdtc_stor; + struct dirty_throttle_control * const mdtc = mdtc_valid(&mdtc_stor) ? + &mdtc_stor : NULL; /* * Similar to balance_dirty_pages() but ignores pages being written @@ -1792,6 +1892,20 @@ bool wb_over_bg_thresh(struct bdi_writeback *wb) if (wb_stat(wb, WB_RECLAIMABLE) > __wb_calc_thresh(gdtc)) return true; + if (mdtc) { + unsigned long writeback; + + mem_cgroup_wb_stats(wb, &mdtc->avail, &mdtc->dirty, &writeback); + mdtc_cap_avail(mdtc); + domain_dirty_limits(mdtc); /* ditto, ignore writeback */ + + if (mdtc->dirty > mdtc->bg_thresh) + return true; + + if (wb_stat(wb, WB_RECLAIMABLE) > __wb_calc_thresh(mdtc)) + return true; + } + return false; } From 97c9341f727105c29478da19f1687b0e0a917256 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 22 May 2015 18:23:36 -0400 Subject: [PATCH 70/85] mm: vmscan: disable memcg direct reclaim stalling if cgroup writeback support is in use Because writeback wasn't cgroup aware before, the usual dirty throttling mechanism in balance_dirty_pages() didn't work for processes under memcg limit. The writeback path didn't know how much memory is available or how fast the dirty pages are being written out for a given memcg and balance_dirty_pages() didn't have any measure of IO back pressure for the memcg. To work around the issue, memcg implemented an ad-hoc dirty throttling mechanism in the direct reclaim path by stalling on pages under writeback which are encountered during direct reclaim scan. This is rather ugly and crude - none of the configurability, fairness, or bandwidth-proportional distribution of the normal path. The previous patches implemented proper memcg aware dirty throttling when cgroup writeback is in use making the ad-hoc mechanism unnecessary. This patch disables direct reclaim stalling for such case. Note: I disabled the parts which seemed obvious and it behaves fine while testing but my understanding of this code path is rudimentary and it's quite possible that I got something wrong. Please let me know if I got some wrong or more global_reclaim() sites should be updated. v2: The original patch removed the direct stalling mechanism which breaks legacy hierarchies. Conditionalize instead of removing. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Cc: Vladimir Davydov Signed-off-by: Jens Axboe --- mm/vmscan.c | 51 +++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 41 insertions(+), 10 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index f46339870147..8cb16ebaf3ed 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -154,11 +154,42 @@ static bool global_reclaim(struct scan_control *sc) { return !sc->target_mem_cgroup; } + +/** + * sane_reclaim - is the usual dirty throttling mechanism operational? + * @sc: scan_control in question + * + * The normal page dirty throttling mechanism in balance_dirty_pages() is + * completely broken with the legacy memcg and direct stalling in + * shrink_page_list() is used for throttling instead, which lacks all the + * niceties such as fairness, adaptive pausing, bandwidth proportional + * allocation and configurability. + * + * This function tests whether the vmscan currently in progress can assume + * that the normal dirty throttling mechanism is operational. + */ +static bool sane_reclaim(struct scan_control *sc) +{ + struct mem_cgroup *memcg = sc->target_mem_cgroup; + + if (!memcg) + return true; +#ifdef CONFIG_CGROUP_WRITEBACK + if (cgroup_on_dfl(mem_cgroup_css(memcg)->cgroup)) + return true; +#endif + return false; +} #else static bool global_reclaim(struct scan_control *sc) { return true; } + +static bool sane_reclaim(struct scan_control *sc) +{ + return true; +} #endif static unsigned long zone_reclaimable_pages(struct zone *zone) @@ -941,10 +972,10 @@ static unsigned long shrink_page_list(struct list_head *page_list, * note that the LRU is being scanned too quickly and the * caller can stall after page list has been processed. * - * 2) Global reclaim encounters a page, memcg encounters a - * page that is not marked for immediate reclaim or - * the caller does not have __GFP_IO. In this case mark - * the page for immediate reclaim and continue scanning. + * 2) Global or new memcg reclaim encounters a page that is + * not marked for immediate reclaim or the caller does not + * have __GFP_IO. In this case mark the page for immediate + * reclaim and continue scanning. * * __GFP_IO is checked because a loop driver thread might * enter reclaim, and deadlock if it waits on a page for @@ -958,7 +989,7 @@ static unsigned long shrink_page_list(struct list_head *page_list, * grab_cache_page_write_begin(,,AOP_FLAG_NOFS), so testing * may_enter_fs here is liable to OOM on them. * - * 3) memcg encounters a page that is not already marked + * 3) Legacy memcg encounters a page that is not already marked * PageReclaim. memcg does not have any dirty pages * throttling so we could easily OOM just because too many * pages are in writeback and there is nothing else to @@ -973,7 +1004,7 @@ static unsigned long shrink_page_list(struct list_head *page_list, goto keep_locked; /* Case 2 above */ - } else if (global_reclaim(sc) || + } else if (sane_reclaim(sc) || !PageReclaim(page) || !(sc->gfp_mask & __GFP_IO)) { /* * This is slightly racy - end_page_writeback() @@ -1422,7 +1453,7 @@ static int too_many_isolated(struct zone *zone, int file, if (current_is_kswapd()) return 0; - if (!global_reclaim(sc)) + if (!sane_reclaim(sc)) return 0; if (file) { @@ -1614,10 +1645,10 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec, set_bit(ZONE_WRITEBACK, &zone->flags); /* - * memcg will stall in page writeback so only consider forcibly - * stalling for global reclaim + * Legacy memcg will stall in page writeback so avoid forcibly + * stalling here. */ - if (global_reclaim(sc)) { + if (sane_reclaim(sc)) { /* * Tag a zone as congested if all the dirty pages scanned were * backed by a congested BDI and wait_iff_congested will stall. From 21c6321fbb3a3787af07f1bc031d713a707fb69c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 28 May 2015 14:50:49 -0400 Subject: [PATCH 71/85] writeback: relocate wb[_try]_get(), wb_put(), inode_{attach|detach}_wb() Currently, majority of cgroup writeback support including all the above functions are implemented in include/linux/backing-dev.h and mm/backing-dev.c; however, the portion closely related to writeback logic implemented in include/linux/writeback.h and mm/page-writeback.c will expand to support foreign writeback detection and correction. This patch moves wb[_try]_get() and wb_put() to include/linux/backing-dev-defs.h so that they can be used from writeback.h and inode_{attach|detach}_wb() to writeback.h and page-writeback.c. This is pure reorganization and doesn't introduce any functional changes. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 31 ++++++++++++ include/linux/backing-dev-defs.h | 50 +++++++++++++++++++ include/linux/backing-dev.h | 82 -------------------------------- include/linux/writeback.h | 46 ++++++++++++++++++ mm/backing-dev.c | 30 ------------ 5 files changed, 127 insertions(+), 112 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index da355879ba7c..cf6ccfb01e03 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -27,6 +27,7 @@ #include #include #include +#include #include "internal.h" /* @@ -213,6 +214,36 @@ static void wb_wait_for_completion(struct backing_dev_info *bdi, #ifdef CONFIG_CGROUP_WRITEBACK +void __inode_attach_wb(struct inode *inode, struct page *page) +{ + struct backing_dev_info *bdi = inode_to_bdi(inode); + struct bdi_writeback *wb = NULL; + + if (inode_cgwb_enabled(inode)) { + struct cgroup_subsys_state *memcg_css; + + if (page) { + memcg_css = mem_cgroup_css_from_page(page); + wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); + } else { + /* must pin memcg_css, see wb_get_create() */ + memcg_css = task_get_css(current, memory_cgrp_id); + wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); + css_put(memcg_css); + } + } + + if (!wb) + wb = &bdi->wb; + + /* + * There may be multiple instances of this function racing to + * update the same inode. Use cmpxchg() to tell the winner. + */ + if (unlikely(cmpxchg(&inode->i_wb, NULL, wb))) + wb_put(wb); +} + /** * inode_congested - test whether an inode is congested * @inode: inode to test for congestion diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index 8d470b73824f..e047b496a0b9 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -186,4 +186,54 @@ static inline void set_bdi_congested(struct backing_dev_info *bdi, int sync) set_wb_congested(bdi->wb.congested, sync); } +#ifdef CONFIG_CGROUP_WRITEBACK + +/** + * wb_tryget - try to increment a wb's refcount + * @wb: bdi_writeback to get + */ +static inline bool wb_tryget(struct bdi_writeback *wb) +{ + if (wb != &wb->bdi->wb) + return percpu_ref_tryget(&wb->refcnt); + return true; +} + +/** + * wb_get - increment a wb's refcount + * @wb: bdi_writeback to get + */ +static inline void wb_get(struct bdi_writeback *wb) +{ + if (wb != &wb->bdi->wb) + percpu_ref_get(&wb->refcnt); +} + +/** + * wb_put - decrement a wb's refcount + * @wb: bdi_writeback to put + */ +static inline void wb_put(struct bdi_writeback *wb) +{ + if (wb != &wb->bdi->wb) + percpu_ref_put(&wb->refcnt); +} + +#else /* CONFIG_CGROUP_WRITEBACK */ + +static inline bool wb_tryget(struct bdi_writeback *wb) +{ + return true; +} + +static inline void wb_get(struct bdi_writeback *wb) +{ +} + +static inline void wb_put(struct bdi_writeback *wb) +{ +} + +#endif /* CONFIG_CGROUP_WRITEBACK */ + #endif /* __LINUX_BACKING_DEV_DEFS_H */ diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index e9d7373f5f93..5c978a924157 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -243,7 +243,6 @@ void wb_congested_put(struct bdi_writeback_congested *congested); struct bdi_writeback *wb_get_create(struct backing_dev_info *bdi, struct cgroup_subsys_state *memcg_css, gfp_t gfp); -void __inode_attach_wb(struct inode *inode, struct page *page); void wb_memcg_offline(struct mem_cgroup *memcg); void wb_blkcg_offline(struct blkcg *blkcg); int inode_congested(struct inode *inode, int cong_bits); @@ -264,37 +263,6 @@ static inline bool inode_cgwb_enabled(struct inode *inode) (inode->i_sb->s_type->fs_flags & FS_CGROUP_WRITEBACK); } -/** - * wb_tryget - try to increment a wb's refcount - * @wb: bdi_writeback to get - */ -static inline bool wb_tryget(struct bdi_writeback *wb) -{ - if (wb != &wb->bdi->wb) - return percpu_ref_tryget(&wb->refcnt); - return true; -} - -/** - * wb_get - increment a wb's refcount - * @wb: bdi_writeback to get - */ -static inline void wb_get(struct bdi_writeback *wb) -{ - if (wb != &wb->bdi->wb) - percpu_ref_get(&wb->refcnt); -} - -/** - * wb_put - decrement a wb's refcount - * @wb: bdi_writeback to put - */ -static inline void wb_put(struct bdi_writeback *wb) -{ - if (wb != &wb->bdi->wb) - percpu_ref_put(&wb->refcnt); -} - /** * wb_find_current - find wb for %current on a bdi * @bdi: bdi of interest @@ -353,35 +321,6 @@ wb_get_create_current(struct backing_dev_info *bdi, gfp_t gfp) return wb; } -/** - * inode_attach_wb - associate an inode with its wb - * @inode: inode of interest - * @page: page being dirtied (may be NULL) - * - * If @inode doesn't have its wb, associate it with the wb matching the - * memcg of @page or, if @page is NULL, %current. May be called w/ or w/o - * @inode->i_lock. - */ -static inline void inode_attach_wb(struct inode *inode, struct page *page) -{ - if (!inode->i_wb) - __inode_attach_wb(inode, page); -} - -/** - * inode_detach_wb - disassociate an inode from its wb - * @inode: inode of interest - * - * @inode is being freed. Detach from its wb. - */ -static inline void inode_detach_wb(struct inode *inode) -{ - if (inode->i_wb) { - wb_put(inode->i_wb); - inode->i_wb = NULL; - } -} - /** * inode_to_wb - determine the wb of an inode * @inode: inode of interest @@ -471,19 +410,6 @@ static inline void wb_congested_put(struct bdi_writeback_congested *congested) { } -static inline bool wb_tryget(struct bdi_writeback *wb) -{ - return true; -} - -static inline void wb_get(struct bdi_writeback *wb) -{ -} - -static inline void wb_put(struct bdi_writeback *wb) -{ -} - static inline struct bdi_writeback *wb_find_current(struct backing_dev_info *bdi) { return &bdi->wb; @@ -495,14 +421,6 @@ wb_get_create_current(struct backing_dev_info *bdi, gfp_t gfp) return &bdi->wb; } -static inline void inode_attach_wb(struct inode *inode, struct page *page) -{ -} - -static inline void inode_detach_wb(struct inode *inode) -{ -} - static inline struct bdi_writeback *inode_to_wb(struct inode *inode) { return &inode_to_bdi(inode)->wb; diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 3b73e97ecfc7..6726b7e56beb 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -8,6 +8,7 @@ #include #include #include +#include DECLARE_PER_CPU(int, dirty_throttle_leaks); @@ -173,6 +174,51 @@ static inline void wait_on_inode(struct inode *inode) wait_on_bit(&inode->i_state, __I_NEW, TASK_UNINTERRUPTIBLE); } +#ifdef CONFIG_CGROUP_WRITEBACK + +void __inode_attach_wb(struct inode *inode, struct page *page); + +/** + * inode_attach_wb - associate an inode with its wb + * @inode: inode of interest + * @page: page being dirtied (may be NULL) + * + * If @inode doesn't have its wb, associate it with the wb matching the + * memcg of @page or, if @page is NULL, %current. May be called w/ or w/o + * @inode->i_lock. + */ +static inline void inode_attach_wb(struct inode *inode, struct page *page) +{ + if (!inode->i_wb) + __inode_attach_wb(inode, page); +} + +/** + * inode_detach_wb - disassociate an inode from its wb + * @inode: inode of interest + * + * @inode is being freed. Detach from its wb. + */ +static inline void inode_detach_wb(struct inode *inode) +{ + if (inode->i_wb) { + wb_put(inode->i_wb); + inode->i_wb = NULL; + } +} + +#else /* CONFIG_CGROUP_WRITEBACK */ + +static inline void inode_attach_wb(struct inode *inode, struct page *page) +{ +} + +static inline void inode_detach_wb(struct inode *inode) +{ +} + +#endif /* CONFIG_CGROUP_WRITEBACK */ + /* * mm/page-writeback.c */ diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 84ebf7c8d006..887d72a85b5e 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -660,36 +660,6 @@ struct bdi_writeback *wb_get_create(struct backing_dev_info *bdi, return wb; } -void __inode_attach_wb(struct inode *inode, struct page *page) -{ - struct backing_dev_info *bdi = inode_to_bdi(inode); - struct bdi_writeback *wb = NULL; - - if (inode_cgwb_enabled(inode)) { - struct cgroup_subsys_state *memcg_css; - - if (page) { - memcg_css = mem_cgroup_css_from_page(page); - wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); - } else { - /* must pin memcg_css, see wb_get_create() */ - memcg_css = task_get_css(current, memory_cgrp_id); - wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); - css_put(memcg_css); - } - } - - if (!wb) - wb = &bdi->wb; - - /* - * There may be multiple instances of this function racing to - * update the same inode. Use cmpxchg() to tell the winner. - */ - if (unlikely(cmpxchg(&inode->i_wb, NULL, wb))) - wb_put(wb); -} - static void cgwb_bdi_init(struct backing_dev_info *bdi) { bdi->wb.memcg_css = mem_cgroup_root_css; From b16b1deb553adcd7b3b7ce3e6d6fd1b923f314da Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 2 Jun 2015 08:39:48 -0600 Subject: [PATCH 72/85] writeback: make writeback_control track the inode being written back Currently, for cgroup writeback, the IO submission paths directly associate the bio's with the blkcg from inode_to_wb_blkcg_css(); however, it'd be necessary to keep more writeback context to implement foreign inode writeback detection. wbc (writeback_control) is the natural fit for the extra context - it persists throughout the writeback of each inode and is passed all the way down to IO submission paths. This patch adds wbc_attach_and_unlock_inode(), wbc_detach_inode(), and wbc_attach_fdatawrite_inode() which are used to associate wbc with the inode being written back. IO submission paths now use wbc_init_bio() instead of directly associating bio's with blkcg themselves. This leaves inode_to_wb_blkcg_css() w/o any user. The function is removed. wbc currently only tracks the associated wb (bdi_writeback). Future patches will add more for foreign inode detection. The association is established under i_lock which will be depended upon when migrating foreign inodes to other wb's. As currently, once established, inode to wb association never changes, going through wbc when initializing bio's doesn't cause any behavior changes. v2: submit_blk_blkcg() now checks whether the wbc is associated with a wb before dereferencing it. This can happen when pageout() is writing pages directly without going through the usual writeback path. As pageout() path is single-threaded, we don't want it to be blocked behind a slow cgroup and ultimately want it to delegate actual writing to the usual writeback path. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/buffer.c | 25 +++++++------- fs/fs-writeback.c | 37 ++++++++++++++++++-- fs/mpage.c | 2 +- include/linux/backing-dev.h | 12 ------- include/linux/writeback.h | 68 +++++++++++++++++++++++++++++++++++++ mm/filemap.c | 2 ++ 6 files changed, 118 insertions(+), 28 deletions(-) diff --git a/fs/buffer.c b/fs/buffer.c index b85e94134ea6..d883c799fb45 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -45,9 +45,9 @@ #include static int fsync_buffers_list(spinlock_t *lock, struct list_head *list); -static int submit_bh_blkcg(int rw, struct buffer_head *bh, - unsigned long bio_flags, - struct cgroup_subsys_state *blkcg_css); +static int submit_bh_wbc(int rw, struct buffer_head *bh, + unsigned long bio_flags, + struct writeback_control *wbc); #define BH_ENTRY(list) list_entry((list), struct buffer_head, b_assoc_buffers) @@ -1709,7 +1709,6 @@ static int __block_write_full_page(struct inode *inode, struct page *page, unsigned int blocksize, bbits; int nr_underway = 0; int write_op = (wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE); - struct cgroup_subsys_state *blkcg_css = inode_to_wb_blkcg_css(inode); head = create_page_buffers(page, inode, (1 << BH_Dirty)|(1 << BH_Uptodate)); @@ -1798,7 +1797,7 @@ static int __block_write_full_page(struct inode *inode, struct page *page, do { struct buffer_head *next = bh->b_this_page; if (buffer_async_write(bh)) { - submit_bh_blkcg(write_op, bh, 0, blkcg_css); + submit_bh_wbc(write_op, bh, 0, wbc); nr_underway++; } bh = next; @@ -1852,7 +1851,7 @@ recover: struct buffer_head *next = bh->b_this_page; if (buffer_async_write(bh)) { clear_buffer_dirty(bh); - submit_bh_blkcg(write_op, bh, 0, blkcg_css); + submit_bh_wbc(write_op, bh, 0, wbc); nr_underway++; } bh = next; @@ -3017,11 +3016,11 @@ void guard_bio_eod(int rw, struct bio *bio) } } -static int submit_bh_blkcg(int rw, struct buffer_head *bh, - unsigned long bio_flags, - struct cgroup_subsys_state *blkcg_css) +static int submit_bh_wbc(int rw, struct buffer_head *bh, + unsigned long bio_flags, struct writeback_control *wbc) { struct bio *bio; + int ret = 0; BUG_ON(!buffer_locked(bh)); BUG_ON(!buffer_mapped(bh)); @@ -3041,8 +3040,8 @@ static int submit_bh_blkcg(int rw, struct buffer_head *bh, */ bio = bio_alloc(GFP_NOIO, 1); - if (blkcg_css) - bio_associate_blkcg(bio, blkcg_css); + if (wbc) + wbc_init_bio(wbc, bio); bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); bio->bi_bdev = bh->b_bdev; @@ -3071,13 +3070,13 @@ static int submit_bh_blkcg(int rw, struct buffer_head *bh, int _submit_bh(int rw, struct buffer_head *bh, unsigned long bio_flags) { - return submit_bh_blkcg(rw, bh, bio_flags, NULL); + return submit_bh_wbc(rw, bh, bio_flags, NULL); } EXPORT_SYMBOL_GPL(_submit_bh); int submit_bh(int rw, struct buffer_head *bh) { - return submit_bh_blkcg(rw, bh, 0, NULL); + return submit_bh_wbc(rw, bh, 0, NULL); } EXPORT_SYMBOL(submit_bh); diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index cf6ccfb01e03..755e8ef8d1f0 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -244,6 +244,37 @@ void __inode_attach_wb(struct inode *inode, struct page *page) wb_put(wb); } +/** + * wbc_attach_and_unlock_inode - associate wbc with target inode and unlock it + * @wbc: writeback_control of interest + * @inode: target inode + * + * @inode is locked and about to be written back under the control of @wbc. + * Record @inode's writeback context into @wbc and unlock the i_lock. On + * writeback completion, wbc_detach_inode() should be called. This is used + * to track the cgroup writeback context. + */ +void wbc_attach_and_unlock_inode(struct writeback_control *wbc, + struct inode *inode) +{ + wbc->wb = inode_to_wb(inode); + wb_get(wbc->wb); + spin_unlock(&inode->i_lock); +} + +/** + * wbc_detach_inode - disassociate wbc from its target inode + * @wbc: writeback_control of interest + * + * To be called after a writeback attempt of an inode finishes and undoes + * wbc_attach_and_unlock_inode(). Can be called under any context. + */ +void wbc_detach_inode(struct writeback_control *wbc) +{ + wb_put(wbc->wb); + wbc->wb = NULL; +} + /** * inode_congested - test whether an inode is congested * @inode: inode to test for congestion @@ -877,10 +908,11 @@ writeback_single_inode(struct inode *inode, struct bdi_writeback *wb, !mapping_tagged(inode->i_mapping, PAGECACHE_TAG_WRITEBACK))) goto out; inode->i_state |= I_SYNC; - spin_unlock(&inode->i_lock); + wbc_attach_and_unlock_inode(wbc, inode); ret = __writeback_single_inode(inode, wbc); + wbc_detach_inode(wbc); spin_lock(&wb->list_lock); spin_lock(&inode->i_lock); /* @@ -1013,7 +1045,7 @@ static long writeback_sb_inodes(struct super_block *sb, continue; } inode->i_state |= I_SYNC; - spin_unlock(&inode->i_lock); + wbc_attach_and_unlock_inode(&wbc, inode); write_chunk = writeback_chunk_size(wb, work); wbc.nr_to_write = write_chunk; @@ -1025,6 +1057,7 @@ static long writeback_sb_inodes(struct super_block *sb, */ __writeback_single_inode(inode, &wbc); + wbc_detach_inode(&wbc); work->nr_pages -= write_chunk - wbc.nr_to_write; wrote += write_chunk - wbc.nr_to_write; spin_lock(&wb->list_lock); diff --git a/fs/mpage.c b/fs/mpage.c index a3ccb0bd465a..388fde6ac255 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -606,7 +606,7 @@ alloc_new: if (bio == NULL) goto confused; - bio_associate_blkcg(bio, inode_to_wb_blkcg_css(inode)); + wbc_init_bio(wbc, bio); } /* diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 5c978a924157..b1d2489a6536 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -332,12 +332,6 @@ static inline struct bdi_writeback *inode_to_wb(struct inode *inode) return inode->i_wb; } -static inline struct cgroup_subsys_state * -inode_to_wb_blkcg_css(struct inode *inode) -{ - return inode_to_wb(inode)->blkcg_css; -} - struct wb_iter { int start_blkcg_id; struct radix_tree_iter tree_iter; @@ -434,12 +428,6 @@ static inline void wb_blkcg_offline(struct blkcg *blkcg) { } -static inline struct cgroup_subsys_state * -inode_to_wb_blkcg_css(struct inode *inode) -{ - return blkcg_root_css; -} - struct wb_iter { int next_id; }; diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 6726b7e56beb..8f964e558af5 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -86,6 +86,9 @@ struct writeback_control { unsigned for_reclaim:1; /* Invoked from the page allocator */ unsigned range_cyclic:1; /* range_start is cyclic */ unsigned for_sync:1; /* sync(2) WB_SYNC_ALL writeback */ +#ifdef CONFIG_CGROUP_WRITEBACK + struct bdi_writeback *wb; /* wb this writeback is issued under */ +#endif }; /* @@ -176,7 +179,14 @@ static inline void wait_on_inode(struct inode *inode) #ifdef CONFIG_CGROUP_WRITEBACK +#include +#include + void __inode_attach_wb(struct inode *inode, struct page *page); +void wbc_attach_and_unlock_inode(struct writeback_control *wbc, + struct inode *inode) + __releases(&inode->i_lock); +void wbc_detach_inode(struct writeback_control *wbc); /** * inode_attach_wb - associate an inode with its wb @@ -207,6 +217,44 @@ static inline void inode_detach_wb(struct inode *inode) } } +/** + * wbc_attach_fdatawrite_inode - associate wbc and inode for fdatawrite + * @wbc: writeback_control of interest + * @inode: target inode + * + * This function is to be used by __filemap_fdatawrite_range(), which is an + * alternative entry point into writeback code, and first ensures @inode is + * associated with a bdi_writeback and attaches it to @wbc. + */ +static inline void wbc_attach_fdatawrite_inode(struct writeback_control *wbc, + struct inode *inode) +{ + spin_lock(&inode->i_lock); + inode_attach_wb(inode, NULL); + wbc_attach_and_unlock_inode(wbc, inode); +} + +/** + * wbc_init_bio - writeback specific initializtion of bio + * @wbc: writeback_control for the writeback in progress + * @bio: bio to be initialized + * + * @bio is a part of the writeback in progress controlled by @wbc. Perform + * writeback specific initialization. This is used to apply the cgroup + * writeback context. + */ +static inline void wbc_init_bio(struct writeback_control *wbc, struct bio *bio) +{ + /* + * pageout() path doesn't attach @wbc to the inode being written + * out. This is intentional as we don't want the function to block + * behind a slow cgroup. Ultimately, we want pageout() to kick off + * regular writeback instead of writing things out itself. + */ + if (wbc->wb) + bio_associate_blkcg(bio, wbc->wb->blkcg_css); +} + #else /* CONFIG_CGROUP_WRITEBACK */ static inline void inode_attach_wb(struct inode *inode, struct page *page) @@ -217,6 +265,26 @@ static inline void inode_detach_wb(struct inode *inode) { } +static inline void wbc_attach_and_unlock_inode(struct writeback_control *wbc, + struct inode *inode) + __releases(&inode->i_lock) +{ + spin_unlock(&inode->i_lock); +} + +static inline void wbc_attach_fdatawrite_inode(struct writeback_control *wbc, + struct inode *inode) +{ +} + +static inline void wbc_detach_inode(struct writeback_control *wbc) +{ +} + +static inline void wbc_init_bio(struct writeback_control *wbc, struct bio *bio) +{ +} + #endif /* CONFIG_CGROUP_WRITEBACK */ /* diff --git a/mm/filemap.c b/mm/filemap.c index 7b1443dc3ad0..2f065b19b635 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -290,7 +290,9 @@ int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, if (!mapping_cap_writeback_dirty(mapping)) return 0; + wbc_attach_fdatawrite_inode(&wbc, mapping->host); ret = do_writepages(mapping, &wbc); + wbc_detach_inode(&wbc); return ret; } From 2a81490811d0296d390c571bb64eaa93e5ed7def Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 28 May 2015 14:50:51 -0400 Subject: [PATCH 73/85] writeback: implement foreign cgroup inode detection As concurrent write sharing of an inode is expected to be very rare and memcg only tracks page ownership on first-use basis severely confining the usefulness of such sharing, cgroup writeback tracks ownership per-inode. While the support for concurrent write sharing of an inode is deemed unnecessary, an inode being written to by different cgroups at different points in time is a lot more common, and, more importantly, charging only by first-use can too readily lead to grossly incorrect behaviors (single foreign page can lead to gigabytes of writeback to be incorrectly attributed). To resolve this issue, cgroup writeback detects the majority dirtier of an inode and will transfer the ownership to it. To avoid unnnecessary oscillation, the detection mechanism keeps track of history and gives out the switch verdict only if the foreign usage pattern is stable over a certain amount of time and/or writeback attempts. The detection mechanism has fairly low space and computation overhead. It adds 8 bytes to struct inode (one int and two u16's) and minimal amount of calculation per IO. The detection mechanism converges to the correct answer usually in several seconds of IO time when there's a clear majority dirtier. Even when there isn't, it can reach an acceptable answer fairly quickly under most circumstances. Please see wb_detach_inode() for more details. This patch only implements detection. Following patches will implement actual switching. v2: wbc_account_io() now checks whether the wbc is associated with a wb before dereferencing it. This can happen when pageout() is writing pages directly without going through the usual writeback path. As pageout() path is single-threaded, we don't want it to be blocked behind a slow cgroup and ultimately want it to delegate actual writing to the usual writeback path. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/buffer.c | 4 +- fs/fs-writeback.c | 177 +++++++++++++++++++++++++++++++++++++- fs/mpage.c | 1 + include/linux/fs.h | 5 ++ include/linux/writeback.h | 16 ++++ 5 files changed, 200 insertions(+), 3 deletions(-) diff --git a/fs/buffer.c b/fs/buffer.c index d883c799fb45..aca687f966d7 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -3040,8 +3040,10 @@ static int submit_bh_wbc(int rw, struct buffer_head *bh, */ bio = bio_alloc(GFP_NOIO, 1); - if (wbc) + if (wbc) { wbc_init_bio(wbc, bio); + wbc_account_io(wbc, bh->b_page, bh->b_size); + } bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); bio->bi_bdev = bh->b_bdev; diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 755e8ef8d1f0..f98d40333c85 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -214,6 +214,20 @@ static void wb_wait_for_completion(struct backing_dev_info *bdi, #ifdef CONFIG_CGROUP_WRITEBACK +/* parameters for foreign inode detection, see wb_detach_inode() */ +#define WB_FRN_TIME_SHIFT 13 /* 1s = 2^13, upto 8 secs w/ 16bit */ +#define WB_FRN_TIME_AVG_SHIFT 3 /* avg = avg * 7/8 + new * 1/8 */ +#define WB_FRN_TIME_CUT_DIV 2 /* ignore rounds < avg / 2 */ +#define WB_FRN_TIME_PERIOD (2 * (1 << WB_FRN_TIME_SHIFT)) /* 2s */ + +#define WB_FRN_HIST_SLOTS 16 /* inode->i_wb_frn_history is 16bit */ +#define WB_FRN_HIST_UNIT (WB_FRN_TIME_PERIOD / WB_FRN_HIST_SLOTS) + /* each slot's duration is 2s / 16 */ +#define WB_FRN_HIST_THR_SLOTS (WB_FRN_HIST_SLOTS / 2) + /* if foreign slots >= 8, switch */ +#define WB_FRN_HIST_MAX_SLOTS (WB_FRN_HIST_THR_SLOTS / 2 + 1) + /* one round can affect upto 5 slots */ + void __inode_attach_wb(struct inode *inode, struct page *page) { struct backing_dev_info *bdi = inode_to_bdi(inode); @@ -258,23 +272,182 @@ void wbc_attach_and_unlock_inode(struct writeback_control *wbc, struct inode *inode) { wbc->wb = inode_to_wb(inode); + wbc->inode = inode; + + wbc->wb_id = wbc->wb->memcg_css->id; + wbc->wb_lcand_id = inode->i_wb_frn_winner; + wbc->wb_tcand_id = 0; + wbc->wb_bytes = 0; + wbc->wb_lcand_bytes = 0; + wbc->wb_tcand_bytes = 0; + wb_get(wbc->wb); spin_unlock(&inode->i_lock); } /** - * wbc_detach_inode - disassociate wbc from its target inode - * @wbc: writeback_control of interest + * wbc_detach_inode - disassociate wbc from inode and perform foreign detection + * @wbc: writeback_control of the just finished writeback * * To be called after a writeback attempt of an inode finishes and undoes * wbc_attach_and_unlock_inode(). Can be called under any context. + * + * As concurrent write sharing of an inode is expected to be very rare and + * memcg only tracks page ownership on first-use basis severely confining + * the usefulness of such sharing, cgroup writeback tracks ownership + * per-inode. While the support for concurrent write sharing of an inode + * is deemed unnecessary, an inode being written to by different cgroups at + * different points in time is a lot more common, and, more importantly, + * charging only by first-use can too readily lead to grossly incorrect + * behaviors (single foreign page can lead to gigabytes of writeback to be + * incorrectly attributed). + * + * To resolve this issue, cgroup writeback detects the majority dirtier of + * an inode and transfers the ownership to it. To avoid unnnecessary + * oscillation, the detection mechanism keeps track of history and gives + * out the switch verdict only if the foreign usage pattern is stable over + * a certain amount of time and/or writeback attempts. + * + * On each writeback attempt, @wbc tries to detect the majority writer + * using Boyer-Moore majority vote algorithm. In addition to the byte + * count from the majority voting, it also counts the bytes written for the + * current wb and the last round's winner wb (max of last round's current + * wb, the winner from two rounds ago, and the last round's majority + * candidate). Keeping track of the historical winner helps the algorithm + * to semi-reliably detect the most active writer even when it's not the + * absolute majority. + * + * Once the winner of the round is determined, whether the winner is + * foreign or not and how much IO time the round consumed is recorded in + * inode->i_wb_frn_history. If the amount of recorded foreign IO time is + * over a certain threshold, the switch verdict is given. */ void wbc_detach_inode(struct writeback_control *wbc) { + struct bdi_writeback *wb = wbc->wb; + struct inode *inode = wbc->inode; + u16 history = inode->i_wb_frn_history; + unsigned long avg_time = inode->i_wb_frn_avg_time; + unsigned long max_bytes, max_time; + int max_id; + + /* pick the winner of this round */ + if (wbc->wb_bytes >= wbc->wb_lcand_bytes && + wbc->wb_bytes >= wbc->wb_tcand_bytes) { + max_id = wbc->wb_id; + max_bytes = wbc->wb_bytes; + } else if (wbc->wb_lcand_bytes >= wbc->wb_tcand_bytes) { + max_id = wbc->wb_lcand_id; + max_bytes = wbc->wb_lcand_bytes; + } else { + max_id = wbc->wb_tcand_id; + max_bytes = wbc->wb_tcand_bytes; + } + + /* + * Calculate the amount of IO time the winner consumed and fold it + * into the running average kept per inode. If the consumed IO + * time is lower than avag / WB_FRN_TIME_CUT_DIV, ignore it for + * deciding whether to switch or not. This is to prevent one-off + * small dirtiers from skewing the verdict. + */ + max_time = DIV_ROUND_UP((max_bytes >> PAGE_SHIFT) << WB_FRN_TIME_SHIFT, + wb->avg_write_bandwidth); + if (avg_time) + avg_time += (max_time >> WB_FRN_TIME_AVG_SHIFT) - + (avg_time >> WB_FRN_TIME_AVG_SHIFT); + else + avg_time = max_time; /* immediate catch up on first run */ + + if (max_time >= avg_time / WB_FRN_TIME_CUT_DIV) { + int slots; + + /* + * The switch verdict is reached if foreign wb's consume + * more than a certain proportion of IO time in a + * WB_FRN_TIME_PERIOD. This is loosely tracked by 16 slot + * history mask where each bit represents one sixteenth of + * the period. Determine the number of slots to shift into + * history from @max_time. + */ + slots = min(DIV_ROUND_UP(max_time, WB_FRN_HIST_UNIT), + (unsigned long)WB_FRN_HIST_MAX_SLOTS); + history <<= slots; + if (wbc->wb_id != max_id) + history |= (1U << slots) - 1; + + /* + * Switch if the current wb isn't the consistent winner. + * If there are multiple closely competing dirtiers, the + * inode may switch across them repeatedly over time, which + * is okay. The main goal is avoiding keeping an inode on + * the wrong wb for an extended period of time. + */ + if (hweight32(history) > WB_FRN_HIST_THR_SLOTS) { + /* switch */ + max_id = 0; + avg_time = 0; + history = 0; + } + } + + /* + * Multiple instances of this function may race to update the + * following fields but we don't mind occassional inaccuracies. + */ + inode->i_wb_frn_winner = max_id; + inode->i_wb_frn_avg_time = min(avg_time, (unsigned long)U16_MAX); + inode->i_wb_frn_history = history; + wb_put(wbc->wb); wbc->wb = NULL; } +/** + * wbc_account_io - account IO issued during writeback + * @wbc: writeback_control of the writeback in progress + * @page: page being written out + * @bytes: number of bytes being written out + * + * @bytes from @page are about to written out during the writeback + * controlled by @wbc. Keep the book for foreign inode detection. See + * wbc_detach_inode(). + */ +void wbc_account_io(struct writeback_control *wbc, struct page *page, + size_t bytes) +{ + int id; + + /* + * pageout() path doesn't attach @wbc to the inode being written + * out. This is intentional as we don't want the function to block + * behind a slow cgroup. Ultimately, we want pageout() to kick off + * regular writeback instead of writing things out itself. + */ + if (!wbc->wb) + return; + + rcu_read_lock(); + id = mem_cgroup_css_from_page(page)->id; + rcu_read_unlock(); + + if (id == wbc->wb_id) { + wbc->wb_bytes += bytes; + return; + } + + if (id == wbc->wb_lcand_id) + wbc->wb_lcand_bytes += bytes; + + /* Boyer-Moore majority vote algorithm */ + if (!wbc->wb_tcand_bytes) + wbc->wb_tcand_id = id; + if (id == wbc->wb_tcand_id) + wbc->wb_tcand_bytes += bytes; + else + wbc->wb_tcand_bytes -= min(bytes, wbc->wb_tcand_bytes); +} + /** * inode_congested - test whether an inode is congested * @inode: inode to test for congestion diff --git a/fs/mpage.c b/fs/mpage.c index 388fde6ac255..ca0244b69de8 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -614,6 +614,7 @@ alloc_new: * the confused fail path above (OOM) will be very confused when * it finds all bh marked clean (i.e. it will not write anything) */ + wbc_account_io(wbc, page, PAGE_SIZE); length = first_unmapped << blkbits; if (bio_add_page(bio, page, length, 0) < length) { bio = mpage_bio_submit(WRITE, bio); diff --git a/include/linux/fs.h b/include/linux/fs.h index 67a42ec95065..740126d7c44e 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -638,6 +638,11 @@ struct inode { struct list_head i_wb_list; /* backing dev IO list */ #ifdef CONFIG_CGROUP_WRITEBACK struct bdi_writeback *i_wb; /* the associated cgroup wb */ + + /* foreign inode detection, see wbc_detach_inode() */ + int i_wb_frn_winner; + u16 i_wb_frn_avg_time; + u16 i_wb_frn_history; #endif struct list_head i_lru; /* inode LRU list */ struct list_head i_sb_list; diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 8f964e558af5..b333c945e571 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -88,6 +88,15 @@ struct writeback_control { unsigned for_sync:1; /* sync(2) WB_SYNC_ALL writeback */ #ifdef CONFIG_CGROUP_WRITEBACK struct bdi_writeback *wb; /* wb this writeback is issued under */ + struct inode *inode; /* inode being written out */ + + /* foreign inode detection, see wbc_detach_inode() */ + int wb_id; /* current wb id */ + int wb_lcand_id; /* last foreign candidate wb id */ + int wb_tcand_id; /* this foreign candidate wb id */ + size_t wb_bytes; /* bytes written by current wb */ + size_t wb_lcand_bytes; /* bytes written by last candidate */ + size_t wb_tcand_bytes; /* bytes written by this candidate */ #endif }; @@ -187,6 +196,8 @@ void wbc_attach_and_unlock_inode(struct writeback_control *wbc, struct inode *inode) __releases(&inode->i_lock); void wbc_detach_inode(struct writeback_control *wbc); +void wbc_account_io(struct writeback_control *wbc, struct page *page, + size_t bytes); /** * inode_attach_wb - associate an inode with its wb @@ -285,6 +296,11 @@ static inline void wbc_init_bio(struct writeback_control *wbc, struct bio *bio) { } +static inline void wbc_account_io(struct writeback_control *wbc, + struct page *page, size_t bytes) +{ +} + #endif /* CONFIG_CGROUP_WRITEBACK */ /* From 87e1d789bf55b12fa7c1cdce024499aee3bc0af0 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 28 May 2015 14:50:52 -0400 Subject: [PATCH 74/85] writeback: implement [locked_]inode_to_wb_and_lock_list() cgroup writeback currently assumes that inode to wb association doesn't change; however, with the planned foreign inode wb switching mechanism, the association will change dynamically. When an inode needs to be put on one of the IO lists of its wb, the current code simply calls inode_to_wb() and locks the returned wb; however, with the planned wb switching, the association may change before locking the wb and may even get released. This patch implements [locked_]inode_to_wb_and_lock_list() which pins the associated wb while holding i_lock, releases it, acquires wb->list_lock and verifies that the association hasn't changed inbetween. As the association will be protected by both locks among other things, this guarantees that the wb is the inode's associated wb until the list_lock is released. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 80 ++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 75 insertions(+), 5 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index f98d40333c85..f67b956fb321 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -258,6 +258,56 @@ void __inode_attach_wb(struct inode *inode, struct page *page) wb_put(wb); } +/** + * locked_inode_to_wb_and_lock_list - determine a locked inode's wb and lock it + * @inode: inode of interest with i_lock held + * + * Returns @inode's wb with its list_lock held. @inode->i_lock must be + * held on entry and is released on return. The returned wb is guaranteed + * to stay @inode's associated wb until its list_lock is released. + */ +static struct bdi_writeback * +locked_inode_to_wb_and_lock_list(struct inode *inode) + __releases(&inode->i_lock) + __acquires(&wb->list_lock) +{ + while (true) { + struct bdi_writeback *wb = inode_to_wb(inode); + + /* + * inode_to_wb() association is protected by both + * @inode->i_lock and @wb->list_lock but list_lock nests + * outside i_lock. Drop i_lock and verify that the + * association hasn't changed after acquiring list_lock. + */ + wb_get(wb); + spin_unlock(&inode->i_lock); + spin_lock(&wb->list_lock); + wb_put(wb); /* not gonna deref it anymore */ + + if (likely(wb == inode_to_wb(inode))) + return wb; /* @inode already has ref */ + + spin_unlock(&wb->list_lock); + cpu_relax(); + spin_lock(&inode->i_lock); + } +} + +/** + * inode_to_wb_and_lock_list - determine an inode's wb and lock it + * @inode: inode of interest + * + * Same as locked_inode_to_wb_and_lock_list() but @inode->i_lock isn't held + * on entry. + */ +static struct bdi_writeback *inode_to_wb_and_lock_list(struct inode *inode) + __acquires(&wb->list_lock) +{ + spin_lock(&inode->i_lock); + return locked_inode_to_wb_and_lock_list(inode); +} + /** * wbc_attach_and_unlock_inode - associate wbc with target inode and unlock it * @wbc: writeback_control of interest @@ -603,6 +653,27 @@ restart: #else /* CONFIG_CGROUP_WRITEBACK */ +static struct bdi_writeback * +locked_inode_to_wb_and_lock_list(struct inode *inode) + __releases(&inode->i_lock) + __acquires(&wb->list_lock) +{ + struct bdi_writeback *wb = inode_to_wb(inode); + + spin_unlock(&inode->i_lock); + spin_lock(&wb->list_lock); + return wb; +} + +static struct bdi_writeback *inode_to_wb_and_lock_list(struct inode *inode) + __acquires(&wb->list_lock) +{ + struct bdi_writeback *wb = inode_to_wb(inode); + + spin_lock(&wb->list_lock); + return wb; +} + static long wb_split_bdi_pages(struct bdi_writeback *wb, long nr_pages) { return nr_pages; @@ -678,9 +749,9 @@ void wb_start_background_writeback(struct bdi_writeback *wb) */ void inode_wb_list_del(struct inode *inode) { - struct bdi_writeback *wb = inode_to_wb(inode); + struct bdi_writeback *wb; - spin_lock(&wb->list_lock); + wb = inode_to_wb_and_lock_list(inode); inode_wb_list_del_locked(inode, wb); spin_unlock(&wb->list_lock); } @@ -1784,12 +1855,11 @@ void __mark_inode_dirty(struct inode *inode, int flags) * reposition it (that would break b_dirty time-ordering). */ if (!was_dirty) { - struct bdi_writeback *wb = inode_to_wb(inode); + struct bdi_writeback *wb; struct list_head *dirty_list; bool wakeup_bdi = false; - spin_unlock(&inode->i_lock); - spin_lock(&wb->list_lock); + wb = locked_inode_to_wb_and_lock_list(inode); WARN(bdi_cap_writeback_dirty(wb->bdi) && !test_bit(WB_registered, &wb->state), From 682aa8e1a6a1504a4caaa62e6c2c9daae3757210 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 28 May 2015 14:50:53 -0400 Subject: [PATCH 75/85] writeback: implement unlocked_inode_to_wb transaction and use it for stat updates The mechanism for detecting whether an inode should switch its wb (bdi_writeback) association is now in place. This patch build the framework for the actual switching. This patch adds a new inode flag I_WB_SWITCHING, which has two functions. First, the easy one, it ensures that there's only one switching in progress for a give inode. Second, it's used as a mechanism to synchronize wb stat updates. The two stats, WB_RECLAIMABLE and WB_WRITEBACK, aren't event counters but track the current number of dirty pages and pages under writeback respectively. As such, when an inode is moved from one wb to another, the inode's portion of those stats have to be transferred together; unfortunately, this is a bit tricky as those stat updates are percpu operations which are performed without holding any lock in some places. This patch solves the problem in a similar way as memcg. Each such lockless stat updates are wrapped in transaction surrounded by unlocked_inode_to_wb_begin/end(). During normal operation, they map to rcu_read_lock/unlock(); however, if I_WB_SWITCHING is asserted, mapping->tree_lock is grabbed across the transaction. In turn, the switching path sets I_WB_SWITCHING and waits for a RCU grace period to pass before actually starting to switch, which guarantees that all stat update paths are synchronizing against mapping->tree_lock. This patch still doesn't implement the actual switching. v3: Updated on top of the recent cancel_dirty_page() updates. unlocked_inode_to_wb_begin() now nests inside mem_cgroup_begin_page_stat() to match the locking order. v2: The i_wb access transaction will be used for !stat accesses too. Function names and comments updated accordingly. s/inode_wb_stat_unlocked_{begin|end}/unlocked_inode_to_wb_{begin|end}/ s/switch_wb/switch_wbs/ Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 117 ++++++++++++++++++++++++++++++++++-- include/linux/backing-dev.h | 54 +++++++++++++++++ include/linux/fs.h | 6 ++ include/linux/mm.h | 3 +- mm/filemap.c | 3 +- mm/page-writeback.c | 27 +++++++-- 6 files changed, 196 insertions(+), 14 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index f67b956fb321..08f5496fcf1b 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -308,6 +308,115 @@ static struct bdi_writeback *inode_to_wb_and_lock_list(struct inode *inode) return locked_inode_to_wb_and_lock_list(inode); } +struct inode_switch_wbs_context { + struct inode *inode; + struct bdi_writeback *new_wb; + + struct rcu_head rcu_head; + struct work_struct work; +}; + +static void inode_switch_wbs_work_fn(struct work_struct *work) +{ + struct inode_switch_wbs_context *isw = + container_of(work, struct inode_switch_wbs_context, work); + struct inode *inode = isw->inode; + struct bdi_writeback *new_wb = isw->new_wb; + + /* + * By the time control reaches here, RCU grace period has passed + * since I_WB_SWITCH assertion and all wb stat update transactions + * between unlocked_inode_to_wb_begin/end() are guaranteed to be + * synchronizing against mapping->tree_lock. + */ + spin_lock(&inode->i_lock); + + inode->i_wb_frn_winner = 0; + inode->i_wb_frn_avg_time = 0; + inode->i_wb_frn_history = 0; + + /* + * Paired with load_acquire in unlocked_inode_to_wb_begin() and + * ensures that the new wb is visible if they see !I_WB_SWITCH. + */ + smp_store_release(&inode->i_state, inode->i_state & ~I_WB_SWITCH); + + spin_unlock(&inode->i_lock); + + iput(inode); + wb_put(new_wb); + kfree(isw); +} + +static void inode_switch_wbs_rcu_fn(struct rcu_head *rcu_head) +{ + struct inode_switch_wbs_context *isw = container_of(rcu_head, + struct inode_switch_wbs_context, rcu_head); + + /* needs to grab bh-unsafe locks, bounce to work item */ + INIT_WORK(&isw->work, inode_switch_wbs_work_fn); + schedule_work(&isw->work); +} + +/** + * inode_switch_wbs - change the wb association of an inode + * @inode: target inode + * @new_wb_id: ID of the new wb + * + * Switch @inode's wb association to the wb identified by @new_wb_id. The + * switching is performed asynchronously and may fail silently. + */ +static void inode_switch_wbs(struct inode *inode, int new_wb_id) +{ + struct backing_dev_info *bdi = inode_to_bdi(inode); + struct cgroup_subsys_state *memcg_css; + struct inode_switch_wbs_context *isw; + + /* noop if seems to be already in progress */ + if (inode->i_state & I_WB_SWITCH) + return; + + isw = kzalloc(sizeof(*isw), GFP_ATOMIC); + if (!isw) + return; + + /* find and pin the new wb */ + rcu_read_lock(); + memcg_css = css_from_id(new_wb_id, &memory_cgrp_subsys); + if (memcg_css) + isw->new_wb = wb_get_create(bdi, memcg_css, GFP_ATOMIC); + rcu_read_unlock(); + if (!isw->new_wb) + goto out_free; + + /* while holding I_WB_SWITCH, no one else can update the association */ + spin_lock(&inode->i_lock); + if (inode->i_state & (I_WB_SWITCH | I_FREEING) || + inode_to_wb(inode) == isw->new_wb) { + spin_unlock(&inode->i_lock); + goto out_free; + } + inode->i_state |= I_WB_SWITCH; + spin_unlock(&inode->i_lock); + + ihold(inode); + isw->inode = inode; + + /* + * In addition to synchronizing among switchers, I_WB_SWITCH tells + * the RCU protected stat update paths to grab the mapping's + * tree_lock so that stat transfer can synchronize against them. + * Let's continue after I_WB_SWITCH is guaranteed to be visible. + */ + call_rcu(&isw->rcu_head, inode_switch_wbs_rcu_fn); + return; + +out_free: + if (isw->new_wb) + wb_put(isw->new_wb); + kfree(isw); +} + /** * wbc_attach_and_unlock_inode - associate wbc with target inode and unlock it * @wbc: writeback_control of interest @@ -433,12 +542,8 @@ void wbc_detach_inode(struct writeback_control *wbc) * is okay. The main goal is avoiding keeping an inode on * the wrong wb for an extended period of time. */ - if (hweight32(history) > WB_FRN_HIST_THR_SLOTS) { - /* switch */ - max_id = 0; - avg_time = 0; - history = 0; - } + if (hweight32(history) > WB_FRN_HIST_THR_SLOTS) + inode_switch_wbs(inode, max_id); } /* diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index b1d2489a6536..73ffa32e58ee 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -332,6 +332,50 @@ static inline struct bdi_writeback *inode_to_wb(struct inode *inode) return inode->i_wb; } +/** + * unlocked_inode_to_wb_begin - begin unlocked inode wb access transaction + * @inode: target inode + * @lockedp: temp bool output param, to be passed to the end function + * + * The caller wants to access the wb associated with @inode but isn't + * holding inode->i_lock, mapping->tree_lock or wb->list_lock. This + * function determines the wb associated with @inode and ensures that the + * association doesn't change until the transaction is finished with + * unlocked_inode_to_wb_end(). + * + * The caller must call unlocked_inode_to_wb_end() with *@lockdep + * afterwards and can't sleep during transaction. IRQ may or may not be + * disabled on return. + */ +static inline struct bdi_writeback * +unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp) +{ + rcu_read_lock(); + + /* + * Paired with store_release in inode_switch_wb_work_fn() and + * ensures that we see the new wb if we see cleared I_WB_SWITCH. + */ + *lockedp = smp_load_acquire(&inode->i_state) & I_WB_SWITCH; + + if (unlikely(*lockedp)) + spin_lock_irq(&inode->i_mapping->tree_lock); + return inode_to_wb(inode); +} + +/** + * unlocked_inode_to_wb_end - end inode wb access transaction + * @inode: target inode + * @locked: *@lockedp from unlocked_inode_to_wb_begin() + */ +static inline void unlocked_inode_to_wb_end(struct inode *inode, bool locked) +{ + if (unlikely(locked)) + spin_unlock_irq(&inode->i_mapping->tree_lock); + + rcu_read_unlock(); +} + struct wb_iter { int start_blkcg_id; struct radix_tree_iter tree_iter; @@ -420,6 +464,16 @@ static inline struct bdi_writeback *inode_to_wb(struct inode *inode) return &inode_to_bdi(inode)->wb; } +static inline struct bdi_writeback * +unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp) +{ + return inode_to_wb(inode); +} + +static inline void unlocked_inode_to_wb_end(struct inode *inode, bool locked) +{ +} + static inline void wb_memcg_offline(struct mem_cgroup *memcg) { } diff --git a/include/linux/fs.h b/include/linux/fs.h index 740126d7c44e..b5e1dcfbc5e3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1815,6 +1815,11 @@ struct super_operations { * * I_DIO_WAKEUP Never set. Only used as a key for wait_on_bit(). * + * I_WB_SWITCH Cgroup bdi_writeback switching in progress. Used to + * synchronize competing switching instances and to tell + * wb stat updates to grab mapping->tree_lock. See + * inode_switch_wb_work_fn() for details. + * * Q: What is the difference between I_WILL_FREE and I_FREEING? */ #define I_DIRTY_SYNC (1 << 0) @@ -1834,6 +1839,7 @@ struct super_operations { #define I_DIRTY_TIME (1 << 11) #define __I_DIRTY_TIME_EXPIRED 12 #define I_DIRTY_TIME_EXPIRED (1 << __I_DIRTY_TIME_EXPIRED) +#define I_WB_SWITCH (1 << 13) #define I_DIRTY (I_DIRTY_SYNC | I_DIRTY_DATASYNC | I_DIRTY_PAGES) #define I_DIRTY_ALL (I_DIRTY | I_DIRTY_TIME) diff --git a/include/linux/mm.h b/include/linux/mm.h index f48d979ced4b..4024543b4203 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -27,6 +27,7 @@ struct anon_vma_chain; struct file_ra_state; struct user_struct; struct writeback_control; +struct bdi_writeback; #ifndef CONFIG_NEED_MULTIPLE_NODES /* Don't use mapnrs, do it properly */ extern unsigned long max_mapnr; @@ -1214,7 +1215,7 @@ int redirty_page_for_writepage(struct writeback_control *wbc, void account_page_dirtied(struct page *page, struct address_space *mapping, struct mem_cgroup *memcg); void account_page_cleaned(struct page *page, struct address_space *mapping, - struct mem_cgroup *memcg); + struct mem_cgroup *memcg, struct bdi_writeback *wb); int set_page_dirty(struct page *page); int set_page_dirty_lock(struct page *page); void cancel_dirty_page(struct page *page); diff --git a/mm/filemap.c b/mm/filemap.c index 2f065b19b635..bfc1ab053b12 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -213,7 +213,8 @@ void __delete_from_page_cache(struct page *page, void *shadow, * anyway will be cleared before returning page into buddy allocator. */ if (WARN_ON_ONCE(PageDirty(page))) - account_page_cleaned(page, mapping, memcg); + account_page_cleaned(page, mapping, memcg, + inode_to_wb(mapping->host)); } /** diff --git a/mm/page-writeback.c b/mm/page-writeback.c index e8903356f423..e1514d5b4e9b 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -2422,12 +2422,12 @@ EXPORT_SYMBOL(account_page_dirtied); * Caller must hold mem_cgroup_begin_page_stat(). */ void account_page_cleaned(struct page *page, struct address_space *mapping, - struct mem_cgroup *memcg) + struct mem_cgroup *memcg, struct bdi_writeback *wb) { if (mapping_cap_account_dirty(mapping)) { mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); dec_zone_page_state(page, NR_FILE_DIRTY); - dec_wb_stat(inode_to_wb(mapping->host), WB_RECLAIMABLE); + dec_wb_stat(wb, WB_RECLAIMABLE); task_io_account_cancelled_write(PAGE_CACHE_SIZE); } } @@ -2490,11 +2490,15 @@ void account_page_redirty(struct page *page) struct address_space *mapping = page->mapping; if (mapping && mapping_cap_account_dirty(mapping)) { - struct bdi_writeback *wb = inode_to_wb(mapping->host); + struct inode *inode = mapping->host; + struct bdi_writeback *wb; + bool locked; + wb = unlocked_inode_to_wb_begin(inode, &locked); current->nr_dirtied--; dec_zone_page_state(page, NR_DIRTIED); dec_wb_stat(wb, WB_DIRTIED); + unlocked_inode_to_wb_end(inode, locked); } } EXPORT_SYMBOL(account_page_redirty); @@ -2597,13 +2601,18 @@ void cancel_dirty_page(struct page *page) struct address_space *mapping = page_mapping(page); if (mapping_cap_account_dirty(mapping)) { + struct inode *inode = mapping->host; + struct bdi_writeback *wb; struct mem_cgroup *memcg; + bool locked; memcg = mem_cgroup_begin_page_stat(page); + wb = unlocked_inode_to_wb_begin(inode, &locked); if (TestClearPageDirty(page)) - account_page_cleaned(page, mapping, memcg); + account_page_cleaned(page, mapping, memcg, wb); + unlocked_inode_to_wb_end(inode, locked); mem_cgroup_end_page_stat(memcg); } else { ClearPageDirty(page); @@ -2628,12 +2637,16 @@ EXPORT_SYMBOL(cancel_dirty_page); int clear_page_dirty_for_io(struct page *page) { struct address_space *mapping = page_mapping(page); - struct mem_cgroup *memcg; int ret = 0; BUG_ON(!PageLocked(page)); if (mapping && mapping_cap_account_dirty(mapping)) { + struct inode *inode = mapping->host; + struct bdi_writeback *wb; + struct mem_cgroup *memcg; + bool locked; + /* * Yes, Virginia, this is indeed insane. * @@ -2670,12 +2683,14 @@ int clear_page_dirty_for_io(struct page *page) * exclusion. */ memcg = mem_cgroup_begin_page_stat(page); + wb = unlocked_inode_to_wb_begin(inode, &locked); if (TestClearPageDirty(page)) { mem_cgroup_dec_page_stat(memcg, MEM_CGROUP_STAT_DIRTY); dec_zone_page_state(page, NR_FILE_DIRTY); - dec_wb_stat(inode_to_wb(mapping->host), WB_RECLAIMABLE); + dec_wb_stat(wb, WB_RECLAIMABLE); ret = 1; } + unlocked_inode_to_wb_end(inode, locked); mem_cgroup_end_page_stat(memcg); return ret; } From 5cb8b8241e6146c487ca715551ac3a5e80358585 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 28 May 2015 14:50:54 -0400 Subject: [PATCH 76/85] writeback: use unlocked_inode_to_wb transaction in inode_congested() Similar to wb stat updates, inode_congested() accesses the associated wb of an inode locklessly, which will break with foreign inode wb switching. This path updates inode_congested() to use unlocked inode wb access transaction introduced by the previous patch. Combined with the previous two patches, this makes all wb list and access operations to be protected by either of inode->i_lock, wb->list_lock, or mapping->tree_lock while wb switching is in progress. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 08f5496fcf1b..25458fa4753d 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -618,10 +618,18 @@ void wbc_account_io(struct writeback_control *wbc, struct page *page, */ int inode_congested(struct inode *inode, int cong_bits) { - if (inode) { - struct bdi_writeback *wb = inode_to_wb(inode); - if (wb) - return wb_congested(wb, cong_bits); + /* + * Once set, ->i_wb never becomes NULL while the inode is alive. + * Start transaction iff ->i_wb is visible. + */ + if (inode && inode_to_wb(inode)) { + struct bdi_writeback *wb; + bool locked, congested; + + wb = unlocked_inode_to_wb_begin(inode, &locked); + congested = wb_congested(wb, cong_bits); + unlocked_inode_to_wb_end(inode, locked); + return congested; } return wb_congested(&inode_to_bdi(inode)->wb, cong_bits); From aaa2cacf8184e2a92accb8e443b1608d65f9a13f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 28 May 2015 14:50:55 -0400 Subject: [PATCH 77/85] writeback: add lockdep annotation to inode_to_wb() With the previous three patches, all operations which acquire wb from inode are either under one of inode->i_lock, mapping->tree_lock or wb->list_lock or protected by unlocked_inode_to_wb transaction. This will be depended upon by foreign inode wb switching. This patch adds lockdep assertion to inode_to_wb() so that usages outside the above list locks can be caught easily. There are three exceptions. * locked_inode_to_wb_and_lock_list() is holding wb->list_lock but the wb may not be the inode's. Ensuring that is the function's role after all. Updated to deref inode->i_wb directly. * inode_wb_stat_unlocked_begin() is usually protected by combination of !I_WB_SWITCH and rcu_read_lock(). Updated to deref inode->i_wb directly. * inode_congested() wants to test whether inode->i_wb is set before starting the transaction. Added inode_to_wb_is_valid() which tests inode->i_wb directly. v5: might_lock() removed. It annotates that the lock is grabbed w/ irq enabled which isn't the case and triggering lockdep warning spuriously. v4: might_lock() added to unlocked_inode_to_wb_begin(). v3: inode_congested() conversion added. v2: locked_inode_to_wb_and_lock_list() was missing in the first version. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 5 +++-- include/linux/backing-dev.h | 34 ++++++++++++++++++++++++++++++++-- 2 files changed, 35 insertions(+), 4 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 25458fa4753d..6b99deeb9de0 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -285,7 +285,8 @@ locked_inode_to_wb_and_lock_list(struct inode *inode) spin_lock(&wb->list_lock); wb_put(wb); /* not gonna deref it anymore */ - if (likely(wb == inode_to_wb(inode))) + /* i_wb may have changed inbetween, can't use inode_to_wb() */ + if (likely(wb == inode->i_wb)) return wb; /* @inode already has ref */ spin_unlock(&wb->list_lock); @@ -622,7 +623,7 @@ int inode_congested(struct inode *inode, int cong_bits) * Once set, ->i_wb never becomes NULL while the inode is alive. * Start transaction iff ->i_wb is visible. */ - if (inode && inode_to_wb(inode)) { + if (inode && inode_to_wb_is_valid(inode)) { struct bdi_writeback *wb; bool locked, congested; diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 73ffa32e58ee..dfce80869145 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -321,14 +321,34 @@ wb_get_create_current(struct backing_dev_info *bdi, gfp_t gfp) return wb; } +/** + * inode_to_wb_is_valid - test whether an inode has a wb associated + * @inode: inode of interest + * + * Returns %true if @inode has a wb associated. May be called without any + * locking. + */ +static inline bool inode_to_wb_is_valid(struct inode *inode) +{ + return inode->i_wb; +} + /** * inode_to_wb - determine the wb of an inode * @inode: inode of interest * - * Returns the wb @inode is currently associated with. + * Returns the wb @inode is currently associated with. The caller must be + * holding either @inode->i_lock, @inode->i_mapping->tree_lock, or the + * associated wb's list_lock. */ static inline struct bdi_writeback *inode_to_wb(struct inode *inode) { +#ifdef CONFIG_LOCKDEP + WARN_ON_ONCE(debug_locks && + (!lockdep_is_held(&inode->i_lock) && + !lockdep_is_held(&inode->i_mapping->tree_lock) && + !lockdep_is_held(&inode->i_wb->list_lock))); +#endif return inode->i_wb; } @@ -360,7 +380,12 @@ unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp) if (unlikely(*lockedp)) spin_lock_irq(&inode->i_mapping->tree_lock); - return inode_to_wb(inode); + + /* + * Protected by either !I_WB_SWITCH + rcu_read_lock() or tree_lock. + * inode_to_wb() will bark. Deref directly. + */ + return inode->i_wb; } /** @@ -459,6 +484,11 @@ wb_get_create_current(struct backing_dev_info *bdi, gfp_t gfp) return &bdi->wb; } +static inline bool inode_to_wb_is_valid(struct inode *inode) +{ + return true; +} + static inline struct bdi_writeback *inode_to_wb(struct inode *inode) { return &inode_to_bdi(inode)->wb; From d10c809552659d8a0089062b9d73da6d47e84cbf Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 28 May 2015 14:50:56 -0400 Subject: [PATCH 78/85] writeback: implement foreign cgroup inode bdi_writeback switching As concurrent write sharing of an inode is expected to be very rare and memcg only tracks page ownership on first-use basis severely confining the usefulness of such sharing, cgroup writeback tracks ownership per-inode. While the support for concurrent write sharing of an inode is deemed unnecessary, an inode being written to by different cgroups at different points in time is a lot more common, and, more importantly, charging only by first-use can too readily lead to grossly incorrect behaviors (single foreign page can lead to gigabytes of writeback to be incorrectly attributed). To resolve this issue, cgroup writeback detects the majority dirtier of an inode and transfers the ownership to it. The previous patches implemented the foreign condition detection mechanism and laid the groundwork. This patch implements the actual switching. With the previously implemented [unlocked_]inode_to_wb_and_list_lock() and wb stat transaction, grabbing wb->list_lock, inode->i_lock and mapping->tree_lock gives us full exclusion against all wb operations on the target inode. inode_switch_wb_work_fn() grabs all the locks and transfers the inode atomically along with its RECLAIMABLE and WRITEBACK stats. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 86 +++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 84 insertions(+), 2 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 6b99deeb9de0..5eeb24a8082f 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -322,30 +322,112 @@ static void inode_switch_wbs_work_fn(struct work_struct *work) struct inode_switch_wbs_context *isw = container_of(work, struct inode_switch_wbs_context, work); struct inode *inode = isw->inode; + struct address_space *mapping = inode->i_mapping; + struct bdi_writeback *old_wb = inode->i_wb; struct bdi_writeback *new_wb = isw->new_wb; + struct radix_tree_iter iter; + bool switched = false; + void **slot; /* * By the time control reaches here, RCU grace period has passed * since I_WB_SWITCH assertion and all wb stat update transactions * between unlocked_inode_to_wb_begin/end() are guaranteed to be * synchronizing against mapping->tree_lock. + * + * Grabbing old_wb->list_lock, inode->i_lock and mapping->tree_lock + * gives us exclusion against all wb related operations on @inode + * including IO list manipulations and stat updates. */ + if (old_wb < new_wb) { + spin_lock(&old_wb->list_lock); + spin_lock_nested(&new_wb->list_lock, SINGLE_DEPTH_NESTING); + } else { + spin_lock(&new_wb->list_lock); + spin_lock_nested(&old_wb->list_lock, SINGLE_DEPTH_NESTING); + } spin_lock(&inode->i_lock); + spin_lock_irq(&mapping->tree_lock); + /* + * Once I_FREEING is visible under i_lock, the eviction path owns + * the inode and we shouldn't modify ->i_wb_list. + */ + if (unlikely(inode->i_state & I_FREEING)) + goto skip_switch; + + /* + * Count and transfer stats. Note that PAGECACHE_TAG_DIRTY points + * to possibly dirty pages while PAGECACHE_TAG_WRITEBACK points to + * pages actually under underwriteback. + */ + radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, 0, + PAGECACHE_TAG_DIRTY) { + struct page *page = radix_tree_deref_slot_protected(slot, + &mapping->tree_lock); + if (likely(page) && PageDirty(page)) { + __dec_wb_stat(old_wb, WB_RECLAIMABLE); + __inc_wb_stat(new_wb, WB_RECLAIMABLE); + } + } + + radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, 0, + PAGECACHE_TAG_WRITEBACK) { + struct page *page = radix_tree_deref_slot_protected(slot, + &mapping->tree_lock); + if (likely(page)) { + WARN_ON_ONCE(!PageWriteback(page)); + __dec_wb_stat(old_wb, WB_WRITEBACK); + __inc_wb_stat(new_wb, WB_WRITEBACK); + } + } + + wb_get(new_wb); + + /* + * Transfer to @new_wb's IO list if necessary. The specific list + * @inode was on is ignored and the inode is put on ->b_dirty which + * is always correct including from ->b_dirty_time. The transfer + * preserves @inode->dirtied_when ordering. + */ + if (!list_empty(&inode->i_wb_list)) { + struct inode *pos; + + inode_wb_list_del_locked(inode, old_wb); + inode->i_wb = new_wb; + list_for_each_entry(pos, &new_wb->b_dirty, i_wb_list) + if (time_after_eq(inode->dirtied_when, + pos->dirtied_when)) + break; + inode_wb_list_move_locked(inode, new_wb, pos->i_wb_list.prev); + } else { + inode->i_wb = new_wb; + } + + /* ->i_wb_frn updates may race wbc_detach_inode() but doesn't matter */ inode->i_wb_frn_winner = 0; inode->i_wb_frn_avg_time = 0; inode->i_wb_frn_history = 0; - + switched = true; +skip_switch: /* * Paired with load_acquire in unlocked_inode_to_wb_begin() and * ensures that the new wb is visible if they see !I_WB_SWITCH. */ smp_store_release(&inode->i_state, inode->i_state & ~I_WB_SWITCH); + spin_unlock_irq(&mapping->tree_lock); spin_unlock(&inode->i_lock); + spin_unlock(&new_wb->list_lock); + spin_unlock(&old_wb->list_lock); + + if (switched) { + wb_wakeup(new_wb); + wb_put(old_wb); + } + wb_put(new_wb); iput(inode); - wb_put(new_wb); kfree(isw); } From e8a7abf5a5bd302a1e06a3c21a629eaa4cba57d6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 28 May 2015 14:50:57 -0400 Subject: [PATCH 79/85] writeback: disassociate inodes from dying bdi_writebacks For the purpose of foreign inode detection, wb's (bdi_writeback's) are identified by the associated memcg ID. As we create a separate wb for each memcg, this is enough to identify the active wb's; however, when blkcg is enabled or disabled higher up in the hierarchy, the mapping between memcg and blkcg changes which in turn creates a new wb to service the new mapping. The old wb is unlinked from index and released after all references are drained. The foreign inode detection logic can't detect this condition because both the old and new wb's point to the same memcg and thus never decides to move inodes attached to the old wb to the new one. This patch adds logic to initiate switching immediately in wbc_attach_and_unlock_inode() if the associated wb is dying. We can make the usual foreign detection logic to distinguish the different wb's mapped to the memcg but the dying wb is never gonna be in active service again and there's no point in tracking the usage history and reaching the switch verdict after enough data points are collected. It's already known that the wb has to be switched. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Jan Kara Cc: Wu Fengguang Cc: Greg Thelen Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 7 +++++++ include/linux/backing-dev-defs.h | 16 ++++++++++++++++ 2 files changed, 23 insertions(+) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 5eeb24a8082f..f60de54d2042 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -525,6 +525,13 @@ void wbc_attach_and_unlock_inode(struct writeback_control *wbc, wb_get(wbc->wb); spin_unlock(&inode->i_lock); + + /* + * A dying wb indicates that the memcg-blkcg mapping has changed + * and a new wb is already serving the memcg. Switch immediately. + */ + if (unlikely(wb_dying(wbc->wb))) + inode_switch_wbs(inode, wbc->wb_id); } /** diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index e047b496a0b9..a48d90e3bcbb 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -219,6 +219,17 @@ static inline void wb_put(struct bdi_writeback *wb) percpu_ref_put(&wb->refcnt); } +/** + * wb_dying - is a wb dying? + * @wb: bdi_writeback of interest + * + * Returns whether @wb is unlinked and being drained. + */ +static inline bool wb_dying(struct bdi_writeback *wb) +{ + return percpu_ref_is_dying(&wb->refcnt); +} + #else /* CONFIG_CGROUP_WRITEBACK */ static inline bool wb_tryget(struct bdi_writeback *wb) @@ -234,6 +245,11 @@ static inline void wb_put(struct bdi_writeback *wb) { } +static inline bool wb_dying(struct bdi_writeback *wb) +{ + return false; +} + #endif /* CONFIG_CGROUP_WRITEBACK */ #endif /* __LINUX_BACKING_DEV_DEFS_H */ From d2e73fcceb18570e82cf3f4725b9e4dbd6cb3c31 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 2 Jun 2015 09:22:34 -0600 Subject: [PATCH 80/85] buffer: remove unusued 'ret' variable Merge hickup on my part, due to a clash between the writeback changes and the EOPNOTSUPP removal in _submit_bh(). Signed-off-by: Jens Axboe --- fs/buffer.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/buffer.c b/fs/buffer.c index aca687f966d7..1cf7a53a0277 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -3020,7 +3020,6 @@ static int submit_bh_wbc(int rw, struct buffer_head *bh, unsigned long bio_flags, struct writeback_control *wbc) { struct bio *bio; - int ret = 0; BUG_ON(!buffer_locked(bh)); BUG_ON(!buffer_mapped(bh)); From 5857cd637bc0d60dc7e37af396b01324f199d89b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 5 Jun 2015 06:20:51 +0900 Subject: [PATCH 81/85] bdi: fix wrong error return value in cgwb_create() On wb_congested_get_create() failure, cgwb_create() forgot to set @ret to -ENOMEM ending up returning 0. Fix it so that it returns -ENOMEM. Signed-off-by: Tejun Heo Reported-by: Dan Carpenter Signed-off-by: Jens Axboe --- mm/backing-dev.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 887d72a85b5e..436bb53dd383 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -554,8 +554,10 @@ static int cgwb_create(struct backing_dev_info *bdi, goto err_ref_exit; wb->congested = wb_congested_get_create(bdi, blkcg_css->id, gfp); - if (!wb->congested) + if (!wb->congested) { + ret = -ENOMEM; goto err_fprop_exit; + } wb->memcg_css = memcg_css; wb->blkcg_css = blkcg_css; From 412a19b64ad17f7650ff778fd2cb9032938cf71f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 8 Jun 2015 14:57:31 +0900 Subject: [PATCH 82/85] v9fs: fix error handling in v9fs_session_init() On failure, v9fs_session_init() returns with the v9fs_session_info struct partially initialized and expects the caller to invoke v9fs_session_close() to clean it up; however, it doesn't track whether the bdi is initialized or not and curiously invokes bdi_destroy() in both vfs_session_init() failure path too. A. If v9fs_session_init() fails before the bdi is initialized, the follow-up v9fs_session_close() will invoke bdi_destroy() on an uninitialized bdi. B. If v9fs_session_init() fails after the bdi is initialized, bdi_destroy() will be called twice on the same bdi - once in the failure path of v9fs_session_init() and then by v9fs_session_close(). A is broken no matter what. B used to be okay because bdi_destroy() allowed being invoked multiple times on the same bdi, which BTW was broken in its own way - if bdi_destroy() was invoked on an initialiezd but !registered bdi, it'd fail to free percpu counters. Since f0054bb1e1f3 ("writeback: move backing_dev_info->wb_lock and ->worklist into bdi_writeback"), this no longer work - bdi_destroy() on an initialized but not registered bdi works correctly but multiple invocations of bdi_destroy() is no longer allowed. The obvious culprit here is v9fs_session_init()'s odd and broken error behavior. It should simply clean up after itself on failures. This patch makes the following updates to v9fs_session_init(). * @rc -> @retval error return propagation removed. It didn't serve any purpose. Just use @rc. * Move addition to v9fs_sessionlist to the end of the function so that incomplete sessions are not put on the list or iterated and error path doesn't have to worry about it. * Update error handling so that it cleans up after itself. Signed-off-by: Tejun Heo Reported-by: Sasha Levin Signed-off-by: Jens Axboe --- fs/9p/v9fs.c | 50 +++++++++++++++++++++-------------------------- fs/9p/vfs_super.c | 8 ++------ 2 files changed, 24 insertions(+), 34 deletions(-) diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 620d93489539..8aa56bb6e861 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -320,31 +320,21 @@ fail_option_alloc: struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, const char *dev_name, char *data) { - int retval = -EINVAL; struct p9_fid *fid; - int rc; + int rc = -ENOMEM; v9ses->uname = kstrdup(V9FS_DEFUSER, GFP_KERNEL); if (!v9ses->uname) - return ERR_PTR(-ENOMEM); + goto err_names; v9ses->aname = kstrdup(V9FS_DEFANAME, GFP_KERNEL); - if (!v9ses->aname) { - kfree(v9ses->uname); - return ERR_PTR(-ENOMEM); - } + if (!v9ses->aname) + goto err_names; init_rwsem(&v9ses->rename_sem); rc = bdi_setup_and_register(&v9ses->bdi, "9p"); - if (rc) { - kfree(v9ses->aname); - kfree(v9ses->uname); - return ERR_PTR(rc); - } - - spin_lock(&v9fs_sessionlist_lock); - list_add(&v9ses->slist, &v9fs_sessionlist); - spin_unlock(&v9fs_sessionlist_lock); + if (rc) + goto err_names; v9ses->uid = INVALID_UID; v9ses->dfltuid = V9FS_DEFUID; @@ -352,10 +342,9 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, v9ses->clnt = p9_client_create(dev_name, data); if (IS_ERR(v9ses->clnt)) { - retval = PTR_ERR(v9ses->clnt); - v9ses->clnt = NULL; + rc = PTR_ERR(v9ses->clnt); p9_debug(P9_DEBUG_ERROR, "problem initializing 9p client\n"); - goto error; + goto err_bdi; } v9ses->flags = V9FS_ACCESS_USER; @@ -368,10 +357,8 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, } rc = v9fs_parse_options(v9ses, data); - if (rc < 0) { - retval = rc; - goto error; - } + if (rc < 0) + goto err_clnt; v9ses->maxdata = v9ses->clnt->msize - P9_IOHDRSZ; @@ -405,10 +392,9 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, fid = p9_client_attach(v9ses->clnt, NULL, v9ses->uname, INVALID_UID, v9ses->aname); if (IS_ERR(fid)) { - retval = PTR_ERR(fid); - fid = NULL; + rc = PTR_ERR(fid); p9_debug(P9_DEBUG_ERROR, "cannot attach\n"); - goto error; + goto err_clnt; } if ((v9ses->flags & V9FS_ACCESS_MASK) == V9FS_ACCESS_SINGLE) @@ -420,12 +406,20 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, /* register the session for caching */ v9fs_cache_session_get_cookie(v9ses); #endif + spin_lock(&v9fs_sessionlist_lock); + list_add(&v9ses->slist, &v9fs_sessionlist); + spin_unlock(&v9fs_sessionlist_lock); return fid; -error: +err_clnt: + p9_client_destroy(v9ses->clnt); +err_bdi: bdi_destroy(&v9ses->bdi); - return ERR_PTR(retval); +err_names: + kfree(v9ses->uname); + kfree(v9ses->aname); + return ERR_PTR(rc); } /** diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index e99a338a4638..bf495cedec26 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -130,11 +130,7 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, fid = v9fs_session_init(v9ses, dev_name, data); if (IS_ERR(fid)) { retval = PTR_ERR(fid); - /* - * we need to call session_close to tear down some - * of the data structure setup by session_init - */ - goto close_session; + goto free_session; } sb = sget(fs_type, NULL, v9fs_set_super, flags, v9ses); @@ -195,8 +191,8 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, clunk_fid: p9_client_clunk(fid); -close_session: v9fs_session_close(v9ses); +free_session: kfree(v9ses); return ERR_PTR(retval); From dd73e4b7df9589fa603fc6d128b4f6cabb250e8e Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 16 Jun 2015 18:48:30 -0400 Subject: [PATCH 83/85] writeback: do foreign inode detection iff cgroup writeback is enabled Currently, even when a filesystem doesn't set the FS_CGROUP_WRITEBACK flag, if the filesystem uses wbc_init_bio() and wbc_account_io(), the foreign inode detection and migration logic still ends up activating cgroup writeback which is unexpected. This patch ensures that the foreign inode detection logic stays disabled when inode_cgwb_enabled() is false by not associating writeback_control's with bdi_writeback's. This also avoids unnecessary operations in wbc_init_bio(), wbc_account_io() and wbc_detach_inode() for filesystems which don't support cgroup writeback. Signed-off-by: Tejun Heo Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index f60de54d2042..f0520bcf2094 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -513,6 +513,11 @@ out_free: void wbc_attach_and_unlock_inode(struct writeback_control *wbc, struct inode *inode) { + if (!inode_cgwb_enabled(inode)) { + spin_unlock(&inode->i_lock); + return; + } + wbc->wb = inode_to_wb(inode); wbc->inode = inode; @@ -575,11 +580,16 @@ void wbc_detach_inode(struct writeback_control *wbc) { struct bdi_writeback *wb = wbc->wb; struct inode *inode = wbc->inode; - u16 history = inode->i_wb_frn_history; - unsigned long avg_time = inode->i_wb_frn_avg_time; - unsigned long max_bytes, max_time; + unsigned long avg_time, max_bytes, max_time; + u16 history; int max_id; + if (!wb) + return; + + history = inode->i_wb_frn_history; + avg_time = inode->i_wb_frn_avg_time; + /* pick the winner of this round */ if (wbc->wb_bytes >= wbc->wb_lcand_bytes && wbc->wb_bytes >= wbc->wb_tcand_bytes) { From 46b15caa7cb19b0f6e3bc8ebaee5bc1bb2e35110 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 16 Jun 2015 18:48:31 -0400 Subject: [PATCH 84/85] vfs, writeback: replace FS_CGROUP_WRITEBACK with SB_I_CGROUPWB FS_CGROUP_WRITEBACK indicates whether a file_system_type supports cgroup writeback; however, different super_blocks of the same file_system_type may or may not support cgroup writeback depending on filesystem options. This patch replaces FS_CGROUP_WRITEBACK with a per-super_block flag. super_block->s_flags carries some internal flags in the high bits but it's exposd to userland through uapi header and running out of space anyway. This patch adds a new field super_block->s_iflags to carry kernel-internal flags. It is currently only used by the new SB_I_CGROUPWB flag whose concatenated and abbreviated name is for consistency with other super_block flags. ext2_fill_super() is updated to set SB_I_CGROUPWB. v2: Added super_block->s_iflags instead of stealing another high bit from sb->s_flags as suggested by Christoph and Jan. Signed-off-by: Tejun Heo Cc: Alexander Viro Cc: linux-fsdevel@vger.kernel.org Cc: Christoph Hellwig Cc: Jan Kara Cc: linux-ext4@vger.kernel.org Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe --- fs/ext2/super.c | 3 ++- include/linux/backing-dev.h | 2 +- include/linux/fs.h | 4 +++- 3 files changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 549219d44f1c..900e19cf9ef6 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -882,6 +882,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | ((EXT2_SB(sb)->s_mount_opt & EXT2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); + sb->s_iflags |= SB_I_CGROUPWB; if (le32_to_cpu(es->s_rev_level) == EXT2_GOOD_OLD_REV && (EXT2_HAS_COMPAT_FEATURE(sb, ~0U) || @@ -1543,7 +1544,7 @@ static struct file_system_type ext2_fs_type = { .name = "ext2", .mount = ext2_mount, .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV | FS_CGROUP_WRITEBACK, + .fs_flags = FS_REQUIRES_DEV, }; MODULE_ALIAS_FS("ext2"); diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index dfce80869145..a13181a42b9a 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -260,7 +260,7 @@ static inline bool inode_cgwb_enabled(struct inode *inode) return bdi_cap_account_dirty(bdi) && (bdi->capabilities & BDI_CAP_CGROUP_WRITEBACK) && - (inode->i_sb->s_type->fs_flags & FS_CGROUP_WRITEBACK); + (inode->i_sb->s_iflags & SB_I_CGROUPWB); } /** diff --git a/include/linux/fs.h b/include/linux/fs.h index b5e1dcfbc5e3..2c5e33a5b2af 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1241,6 +1241,8 @@ struct mm_struct; #define UMOUNT_NOFOLLOW 0x00000008 /* Don't follow symlink on umount */ #define UMOUNT_UNUSED 0x80000000 /* Flag guaranteed to be unused */ +/* sb->s_iflags */ +#define SB_I_CGROUPWB 0x00000001 /* cgroup-aware writeback enabled */ /* Possible states of 'frozen' field */ enum { @@ -1279,6 +1281,7 @@ struct super_block { const struct quotactl_ops *s_qcop; const struct export_operations *s_export_op; unsigned long s_flags; + unsigned long s_iflags; /* internal SB_I_* flags */ unsigned long s_magic; struct dentry *s_root; struct rw_semaphore s_umount; @@ -1912,7 +1915,6 @@ struct file_system_type { #define FS_HAS_SUBTYPE 4 #define FS_USERNS_MOUNT 8 /* Can be mounted by userns root */ #define FS_USERNS_DEV_MOUNT 16 /* A userns mount does not imply MNT_NODEV */ -#define FS_CGROUP_WRITEBACK 32 /* Supports cgroup-aware writeback */ #define FS_RENAME_DOES_D_MOVE 32768 /* FS will handle d_move() during rename() internally. */ struct dentry *(*mount) (struct file_system_type *, int, const char *, void *); From 3e1534cf4a2a8278e811e7c84a79da1a02347b8b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 16 Jun 2015 18:48:32 -0400 Subject: [PATCH 85/85] writeback, blkio: add documentation for cgroup writeback support Update Documentation/cgroups/blkio-controller.txt to reflect the recently added cgroup writeback support. Signed-off-by: Tejun Heo Cc: Li Zefan Cc: Vivek Goyal Cc: cgroups@vger.kernel.org Cc: linux-fsdevel@vger.kernel.org Signed-off-by: Jens Axboe --- Documentation/cgroups/blkio-controller.txt | 83 ++++++++++++++++++++-- 1 file changed, 78 insertions(+), 5 deletions(-) diff --git a/Documentation/cgroups/blkio-controller.txt b/Documentation/cgroups/blkio-controller.txt index cd556b914786..68b6a6a470b0 100644 --- a/Documentation/cgroups/blkio-controller.txt +++ b/Documentation/cgroups/blkio-controller.txt @@ -387,8 +387,81 @@ groups and put applications in that group which are not driving enough IO to keep disk busy. In that case set group_idle=0, and CFQ will not idle on individual groups and throughput should improve. -What works -========== -- Currently only sync IO queues are support. All the buffered writes are - still system wide and not per group. Hence we will not see service - differentiation between buffered writes between groups. +Writeback +========= + +Page cache is dirtied through buffered writes and shared mmaps and +written asynchronously to the backing filesystem by the writeback +mechanism. Writeback sits between the memory and IO domains and +regulates the proportion of dirty memory by balancing dirtying and +write IOs. + +On traditional cgroup hierarchies, relationships between different +controllers cannot be established making it impossible for writeback +to operate accounting for cgroup resource restrictions and all +writeback IOs are attributed to the root cgroup. + +If both the blkio and memory controllers are used on the v2 hierarchy +and the filesystem supports cgroup writeback, writeback operations +correctly follow the resource restrictions imposed by both memory and +blkio controllers. + +Writeback examines both system-wide and per-cgroup dirty memory status +and enforces the more restrictive of the two. Also, writeback control +parameters which are absolute values - vm.dirty_bytes and +vm.dirty_background_bytes - are distributed across cgroups according +to their current writeback bandwidth. + +There's a peculiarity stemming from the discrepancy in ownership +granularity between memory controller and writeback. While memory +controller tracks ownership per page, writeback operates on inode +basis. cgroup writeback bridges the gap by tracking ownership by +inode but migrating ownership if too many foreign pages, pages which +don't match the current inode ownership, have been encountered while +writing back the inode. + +This is a conscious design choice as writeback operations are +inherently tied to inodes making strictly following page ownership +complicated and inefficient. The only use case which suffers from +this compromise is multiple cgroups concurrently dirtying disjoint +regions of the same inode, which is an unlikely use case and decided +to be unsupported. Note that as memory controller assigns page +ownership on the first use and doesn't update it until the page is +released, even if cgroup writeback strictly follows page ownership, +multiple cgroups dirtying overlapping areas wouldn't work as expected. +In general, write-sharing an inode across multiple cgroups is not well +supported. + +Filesystem support for cgroup writeback +--------------------------------------- + +A filesystem can make writeback IOs cgroup-aware by updating +address_space_operations->writepage[s]() to annotate bio's using the +following two functions. + +* wbc_init_bio(@wbc, @bio) + + Should be called for each bio carrying writeback data and associates + the bio with the inode's owner cgroup. Can be called anytime + between bio allocation and submission. + +* wbc_account_io(@wbc, @page, @bytes) + + Should be called for each data segment being written out. While + this function doesn't care exactly when it's called during the + writeback session, it's the easiest and most natural to call it as + data segments are added to a bio. + +With writeback bio's annotated, cgroup support can be enabled per +super_block by setting MS_CGROUPWB in ->s_flags. This allows for +selective disabling of cgroup writeback support which is helpful when +certain filesystem features, e.g. journaled data mode, are +incompatible. + +wbc_init_bio() binds the specified bio to its cgroup. Depending on +the configuration, the bio may be executed at a lower priority and if +the writeback session is holding shared resources, e.g. a journal +entry, may lead to priority inversion. There is no one easy solution +for the problem. Filesystems can try to work around specific problem +cases by skipping wbc_init_bio() or using bio_associate_blkcg() +directly.