Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs

Pull btrfs update from Chris Mason:
 "A big set of fixes and features.

  In terms of line count, most of the code comes from Stefan, who added
  the ability to replace a single drive in place.  This is different
  from how btrfs normally replaces drives, and is much much much faster.

  Josef is plowing through our synchronous write performance.  This pull
  request does not include the DIO_OWN_WAITING patch that was discussed
  on the list, but it has a number of other improvements to cut down our
  latencies and CPU time during fsync/O_DIRECT writes.

  Miao Xie has a big series of fixes and is spreading out ordered
  operations over more CPUs.  This improves performance and reduces
  contention.

  I've put in fixes for error handling around hash collisions.  These
  are going back to individual stable kernels as I test against them.

  Otherwise we have a lot of fixes and cleanups, thanks everyone!
  raid5/6 is being rebased against the device replacement code.  I'll
  have it posted this Friday along with a nice series of benchmarks."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (115 commits)
  Btrfs: fix a bug of per-file nocow
  Btrfs: fix hash overflow handling
  Btrfs: don't take inode delalloc mutex if we're a free space inode
  Btrfs: fix autodefrag and umount lockup
  Btrfs: fix permissions of empty files not affected by umask
  Btrfs: put raid properties into global table
  Btrfs: fix BUG() in scrub when first superblock reading gives EIO
  Btrfs: do not call file_update_time in aio_write
  Btrfs: only unlock and relock if we have to
  Btrfs: use tokens where we can in the tree log
  Btrfs: optimize leaf_space_used
  Btrfs: don't memset new tokens
  Btrfs: only clear dirty on the buffer if it is marked as dirty
  Btrfs: move checks in set_page_dirty under DEBUG
  Btrfs: log changed inodes based on the extent map tree
  Btrfs: add path->really_keep_locks
  Btrfs: do not mark ems as prealloc if we are writing to them
  Btrfs: keep track of the extents original block length
  Btrfs: inline csums if we're fsyncing
  Btrfs: don't bother copying if we're only logging the inode
  ...

1 files changed, 92 insertions, 50 deletions

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 22a0439e5a86..a8f652dc940b 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -45,6 +45,7 @@
 #include "inode-map.h"
 #include "check-integrity.h"
 #include "rcu-string.h"
+#include "dev-replace.h"
 
 #ifdef CONFIG_X86
 #include <asm/cpufeature.h>
@@ -387,7 +388,7 @@ static int btree_read_extent_buffer_pages(struct btrfs_root *root,
                 if (test_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags))
                         break;
 
-                num_copies = btrfs_num_copies(&root->fs_info->mapping_tree,
+                num_copies = btrfs_num_copies(root->fs_info,
                                               eb->start, eb->len);
                 if (num_copies == 1)
                         break;
@@ -852,11 +853,16 @@ static int __btree_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
                                  int mirror_num, unsigned long bio_flags,
                                  u64 bio_offset)
 {
+        int ret;
+
         /*
          * when we're called for a write, we're already in the async
          * submission context.  Just jump into btrfs_map_bio
          */
-        return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 1);
+        ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 1);
+        if (ret)
+                bio_endio(bio, ret);
+        return ret;
 }
 
 static int check_async_write(struct inode *inode, unsigned long bio_flags)
@@ -878,7 +884,6 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
         int ret;
 
         if (!(rw & REQ_WRITE)) {
-
                 /*
                  * called for a read, do the setup so that checksum validation
                  * can happen in the async kernel threads
@@ -886,26 +891,32 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
                 ret = btrfs_bio_wq_end_io(BTRFS_I(inode)->root->fs_info,
                                           bio, 1);
                 if (ret)
-                        return ret;
-                return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio,
-                                     mirror_num, 0);
+                        goto out_w_error;
+                ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio,
+                                    mirror_num, 0);
         } else if (!async) {
                 ret = btree_csum_one_bio(bio);
                 if (ret)
-                        return ret;
-                return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio,
-                                     mirror_num, 0);
+                        goto out_w_error;
+                ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio,
+                                    mirror_num, 0);
+        } else {
+                /*
+                 * kthread helpers are used to submit writes so that
+                 * checksumming can happen in parallel across all CPUs
+                 */
+                ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
+                                          inode, rw, bio, mirror_num, 0,
+                                          bio_offset,
+                                          __btree_submit_bio_start,
+                                          __btree_submit_bio_done);
         }
 
-        /*
-         * kthread helpers are used to submit writes so that checksumming
-         * can happen in parallel across all CPUs
-         */
-        return btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
-                                   inode, rw, bio, mirror_num, 0,
-                                   bio_offset,
-                                   __btree_submit_bio_start,
-                                   __btree_submit_bio_done);
+        if (ret) {
+out_w_error:
+                bio_endio(bio, ret);
+        }
+        return ret;
 }
 
 #ifdef CONFIG_MIGRATION
@@ -990,6 +1001,7 @@ static void btree_invalidatepage(struct page *page, unsigned long offset)
 
 static int btree_set_page_dirty(struct page *page)
 {
+#ifdef DEBUG
         struct extent_buffer *eb;
 
         BUG_ON(!PagePrivate(page));
@@ -998,6 +1010,7 @@ static int btree_set_page_dirty(struct page *page)
         BUG_ON(!test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags));
         BUG_ON(!atomic_read(&eb->refs));
         btrfs_assert_tree_locked(eb);
+#endif
         return __set_page_dirty_nobuffers(page);
 }
 
@@ -1129,11 +1142,11 @@ void clean_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root,
                                           root->fs_info->dirty_metadata_bytes);
                         }
                         spin_unlock(&root->fs_info->delalloc_lock);
-                }
 
-                /* ugh, clear_extent_buffer_dirty needs to lock the page */
-                btrfs_set_lock_blocking(buf);
-                clear_extent_buffer_dirty(buf);
+                        /* ugh, clear_extent_buffer_dirty needs to lock the page */
+                        btrfs_set_lock_blocking(buf);
+                        clear_extent_buffer_dirty(buf);
+                }
         }
 }
 
@@ -1193,7 +1206,7 @@ static void __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
         root->root_key.objectid = objectid;
         root->anon_dev = 0;
 
-        spin_lock_init(&root->root_times_lock);
+        spin_lock_init(&root->root_item_lock);
 }
 
 static int __must_check find_and_setup_root(struct btrfs_root *tree_root,
@@ -2131,6 +2144,11 @@ int open_ctree(struct super_block *sb,
         init_rwsem(&fs_info->extent_commit_sem);
         init_rwsem(&fs_info->cleanup_work_sem);
         init_rwsem(&fs_info->subvol_sem);
+        fs_info->dev_replace.lock_owner = 0;
+        atomic_set(&fs_info->dev_replace.nesting_level, 0);
+        mutex_init(&fs_info->dev_replace.lock_finishing_cancel_unmount);
+        mutex_init(&fs_info->dev_replace.lock_management_lock);
+        mutex_init(&fs_info->dev_replace.lock);
 
         spin_lock_init(&fs_info->qgroup_lock);
         fs_info->qgroup_tree = RB_ROOT;
@@ -2279,6 +2297,10 @@ int open_ctree(struct super_block *sb,
                            fs_info->thread_pool_size,
                            &fs_info->generic_worker);
 
+        btrfs_init_workers(&fs_info->flush_workers, "flush_delalloc",
+                           fs_info->thread_pool_size,
+                           &fs_info->generic_worker);
+
         btrfs_init_workers(&fs_info->submit_workers, "submit",
                            min_t(u64, fs_devices->num_devices,
                            fs_info->thread_pool_size),
@@ -2350,6 +2372,7 @@ int open_ctree(struct super_block *sb,
         ret |= btrfs_start_workers(&fs_info->delayed_workers);
         ret |= btrfs_start_workers(&fs_info->caching_workers);
         ret |= btrfs_start_workers(&fs_info->readahead_workers);
+        ret |= btrfs_start_workers(&fs_info->flush_workers);
         if (ret) {
                 err = -ENOMEM;
                 goto fail_sb_buffer;
@@ -2418,7 +2441,11 @@ int open_ctree(struct super_block *sb,
                 goto fail_tree_roots;
         }
 
-        btrfs_close_extra_devices(fs_devices);
+        /*
+         * keep the device that is marked to be the target device for the
+         * dev_replace procedure
+         */
+        btrfs_close_extra_devices(fs_info, fs_devices, 0);
 
         if (!fs_devices->latest_bdev) {
                 printk(KERN_CRIT "btrfs: failed to read devices on %s\n",
@@ -2490,6 +2517,14 @@ retry_root_backup:
                 goto fail_block_groups;
         }
 
+        ret = btrfs_init_dev_replace(fs_info);
+        if (ret) {
+                pr_err("btrfs: failed to init dev_replace: %d\n", ret);
+                goto fail_block_groups;
+        }
+
+        btrfs_close_extra_devices(fs_info, fs_devices, 1);
+
         ret = btrfs_init_space_info(fs_info);
         if (ret) {
                 printk(KERN_ERR "Failed to initial space info: %d\n", ret);
@@ -2503,6 +2538,13 @@ retry_root_backup:
         }
         fs_info->num_tolerated_disk_barrier_failures =
                 btrfs_calc_num_tolerated_disk_barrier_failures(fs_info);
+        if (fs_info->fs_devices->missing_devices >
+             fs_info->num_tolerated_disk_barrier_failures &&
+            !(sb->s_flags & MS_RDONLY)) {
+                printk(KERN_WARNING
+                       "Btrfs: too many missing devices, writeable mount is not allowed\n");
+                goto fail_block_groups;
+        }
 
         fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root,
                                                "btrfs-cleaner");
@@ -2631,6 +2673,13 @@ retry_root_backup:
                 return ret;
         }
 
+        ret = btrfs_resume_dev_replace_async(fs_info);
+        if (ret) {
+                pr_warn("btrfs: failed to resume dev_replace\n");
+                close_ctree(tree_root);
+                return ret;
+        }
+
         return 0;
 
 fail_qgroup:
@@ -2667,6 +2716,7 @@ fail_sb_buffer:
         btrfs_stop_workers(&fs_info->submit_workers);
         btrfs_stop_workers(&fs_info->delayed_workers);
         btrfs_stop_workers(&fs_info->caching_workers);
+        btrfs_stop_workers(&fs_info->flush_workers);
 fail_alloc:
 fail_iput:
         btrfs_mapping_tree_free(&fs_info->mapping_tree);
@@ -3270,16 +3320,18 @@ int close_ctree(struct btrfs_root *root)
         smp_mb();
 
         /* pause restriper - we want to resume on mount */
-        btrfs_pause_balance(root->fs_info);
+        btrfs_pause_balance(fs_info);
 
-        btrfs_scrub_cancel(root);
+        btrfs_dev_replace_suspend_for_unmount(fs_info);
+
+        btrfs_scrub_cancel(fs_info);
 
         /* wait for any defraggers to finish */
         wait_event(fs_info->transaction_wait,
                    (atomic_read(&fs_info->defrag_running) == 0));
 
         /* clear out the rbtree of defraggable inodes */
-        btrfs_run_defrag_inodes(fs_info);
+        btrfs_cleanup_defrag_inodes(fs_info);
 
         if (!(fs_info->sb->s_flags & MS_RDONLY)) {
                 ret = btrfs_commit_super(root);
@@ -3339,6 +3391,7 @@ int close_ctree(struct btrfs_root *root)
         btrfs_stop_workers(&fs_info->delayed_workers);
         btrfs_stop_workers(&fs_info->caching_workers);
         btrfs_stop_workers(&fs_info->readahead_workers);
+        btrfs_stop_workers(&fs_info->flush_workers);
 
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
         if (btrfs_test_opt(root, CHECK_INTEGRITY))
@@ -3383,14 +3436,12 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
         int was_dirty;
 
         btrfs_assert_tree_locked(buf);
-        if (transid != root->fs_info->generation) {
-                printk(KERN_CRIT "btrfs transid mismatch buffer %llu, "
+        if (transid != root->fs_info->generation)
+                WARN(1, KERN_CRIT "btrfs transid mismatch buffer %llu, "
                        "found %llu running %llu\n",
                         (unsigned long long)buf->start,
                         (unsigned long long)transid,
                         (unsigned long long)root->fs_info->generation);
-                WARN_ON(1);
-        }
         was_dirty = set_extent_buffer_dirty(buf);
         if (!was_dirty) {
                 spin_lock(&root->fs_info->delalloc_lock);
@@ -3399,7 +3450,8 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
         }
 }
 
-void btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
+static void __btrfs_btree_balance_dirty(struct btrfs_root *root,
+                                        int flush_delayed)
 {
         /*
          * looks as though older kernels can get into trouble with
@@ -3411,7 +3463,8 @@ void btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
         if (current->flags & PF_MEMALLOC)
                 return;
 
-        btrfs_balance_delayed_items(root);
+        if (flush_delayed)
+                btrfs_balance_delayed_items(root);
 
         num_dirty = root->fs_info->dirty_metadata_bytes;
 
@@ -3422,25 +3475,14 @@ void btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
         return;
 }
 
-void __btrfs_btree_balance_dirty(struct btrfs_root *root, unsigned long nr)
+void btrfs_btree_balance_dirty(struct btrfs_root *root)
 {
-        /*
-         * looks as though older kernels can get into trouble with
-         * this code, they end up stuck in balance_dirty_pages forever
-         */
-        u64 num_dirty;
-        unsigned long thresh = 32 * 1024 * 1024;
-
-        if (current->flags & PF_MEMALLOC)
-                return;
-
-        num_dirty = root->fs_info->dirty_metadata_bytes;
+        __btrfs_btree_balance_dirty(root, 1);
+}
 
-        if (num_dirty > thresh) {
-                balance_dirty_pages_ratelimited(
-                                   root->fs_info->btree_inode->i_mapping);
-        }
-        return;
+void btrfs_btree_balance_dirty_nodelay(struct btrfs_root *root)
+{
+        __btrfs_btree_balance_dirty(root, 0);
 }
 
 int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid)