Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 updates from Theodore Ts'o:
 "The one new feature added in this patch series is the ability to use
  the "punch hole" functionality for inodes that are not using extent
  maps.

  In the bug fix category, we fixed some races in the AIO and fstrim
  code, and some potential NULL pointer dereferences and memory leaks in
  error handling code paths.

  In the optimization category, we fixed a performance regression in the
  jbd2 layer introduced by commit d9b01934d56a ("jbd: fix fsync() tid
  wraparound bug", introduced in v3.0) which shows up in the AIM7
  benchmark.  We also further optimized jbd2 by minimize the amount of
  time that transaction handles are held active.

  This patch series also features some additional enhancement of the
  extent status tree, which is now used to cache extent information in a
  more efficient/compact form than what we use on-disk."

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (65 commits)
  ext4: fix free clusters calculation in bigalloc filesystem
  ext4: no need to remove extent if len is 0 in ext4_es_remove_extent()
  ext4: fix xattr block allocation/release with bigalloc
  ext4: reclaim extents from extent status tree
  ext4: adjust some functions for reclaiming extents from extent status tree
  ext4: remove single extent cache
  ext4: lookup block mapping in extent status tree
  ext4: track all extent status in extent status tree
  ext4: let ext4_ext_map_blocks return EXT4_MAP_UNWRITTEN flag
  ext4: rename and improbe ext4_es_find_extent()
  ext4: add physical block and status member into extent status tree
  ext4: refine extent status tree
  ext4: use ERR_PTR() abstraction for ext4_append()
  ext4: refactor code to read directory blocks into ext4_read_dirblock()
  ext4: add debugging context for warning in ext4_da_update_reserve_space()
  ext4: use KERN_WARNING for warning messages
  jbd2: use module parameters instead of debugfs for jbd_debug
  ext4: use module parameters instead of debugfs for mballoc_debug
  ext4: start handle at the last possible moment when creating inodes
  ext4: fix the number of credits needed for acl ops with inline data
  ...

1 files changed, 22 insertions, 47 deletions

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 1bf6fe785c4..6540ebe058e 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -23,11 +23,18 @@
 
 #include "ext4_jbd2.h"
 #include "mballoc.h"
-#include <linux/debugfs.h>
 #include <linux/log2.h>
+#include <linux/module.h>
 #include <linux/slab.h>
 #include <trace/events/ext4.h>
 
+#ifdef CONFIG_EXT4_DEBUG
+ushort ext4_mballoc_debug __read_mostly;
+
+module_param_named(mballoc_debug, ext4_mballoc_debug, ushort, 0644);
+MODULE_PARM_DESC(mballoc_debug, "Debugging level for ext4's mballoc");
+#endif
+
 /*
  * MUSTDO:
  *   - test ext4_ext_search_left() and ext4_ext_search_right()
@@ -1884,15 +1891,19 @@ static int ext4_mb_good_group(struct ext4_allocation_context *ac,
 	case 0:
 		BUG_ON(ac->ac_2order == 0);
 
-		if (grp->bb_largest_free_order < ac->ac_2order)
-			return 0;
-
 		/* Avoid using the first bg of a flexgroup for data files */
 		if ((ac->ac_flags & EXT4_MB_HINT_DATA) &&
 		    (flex_size >= EXT4_FLEX_SIZE_DIR_ALLOC_SCHEME) &&
 		    ((group % flex_size) == 0))
 			return 0;
 
+		if ((ac->ac_2order > ac->ac_sb->s_blocksize_bits+1) ||
+		    (free / fragments) >= ac->ac_g_ex.fe_len)
+			return 1;
+
+		if (grp->bb_largest_free_order < ac->ac_2order)
+			return 0;
+
 		return 1;
 	case 1:
 		if ((free / fragments) >= ac->ac_g_ex.fe_len)
@@ -2007,7 +2018,7 @@ repeat:
 			}
 
 			ac->ac_groups_scanned++;
-			if (cr == 0)
+			if (cr == 0 && ac->ac_2order < sb->s_blocksize_bits+2)
 				ext4_mb_simple_scan_group(ac, &e4b);
 			else if (cr == 1 && sbi->s_stripe &&
 					!(ac->ac_g_ex.fe_len % sbi->s_stripe))
@@ -2656,40 +2667,6 @@ static void ext4_free_data_callback(struct super_block *sb,
 	mb_debug(1, "freed %u blocks in %u structures\n", count, count2);
 }
 
-#ifdef CONFIG_EXT4_DEBUG
-u8 mb_enable_debug __read_mostly;
-
-static struct dentry *debugfs_dir;
-static struct dentry *debugfs_debug;
-
-static void __init ext4_create_debugfs_entry(void)
-{
-	debugfs_dir = debugfs_create_dir("ext4", NULL);
-	if (debugfs_dir)
-		debugfs_debug = debugfs_create_u8("mballoc-debug",
-						  S_IRUGO | S_IWUSR,
-						  debugfs_dir,
-						  &mb_enable_debug);
-}
-
-static void ext4_remove_debugfs_entry(void)
-{
-	debugfs_remove(debugfs_debug);
-	debugfs_remove(debugfs_dir);
-}
-
-#else
-
-static void __init ext4_create_debugfs_entry(void)
-{
-}
-
-static void ext4_remove_debugfs_entry(void)
-{
-}
-
-#endif
-
 int __init ext4_init_mballoc(void)
 {
 	ext4_pspace_cachep = KMEM_CACHE(ext4_prealloc_space,
@@ -2711,7 +2688,6 @@ int __init ext4_init_mballoc(void)
 		kmem_cache_destroy(ext4_ac_cachep);
 		return -ENOMEM;
 	}
-	ext4_create_debugfs_entry();
 	return 0;
 }
 
@@ -2726,7 +2702,6 @@ void ext4_exit_mballoc(void)
 	kmem_cache_destroy(ext4_ac_cachep);
 	kmem_cache_destroy(ext4_free_data_cachep);
 	ext4_groupinfo_destroy_slabs();
-	ext4_remove_debugfs_entry();
 }
 
 
@@ -3872,7 +3847,7 @@ static void ext4_mb_show_ac(struct ext4_allocation_context *ac)
 	struct super_block *sb = ac->ac_sb;
 	ext4_group_t ngroups, i;
 
-	if (!mb_enable_debug ||
+	if (!ext4_mballoc_debug ||
 	    (EXT4_SB(sb)->s_mount_flags & EXT4_MF_FS_ABORTED))
 		return;
 
@@ -4005,8 +3980,8 @@ ext4_mb_initialize_context(struct ext4_allocation_context *ac,
 	len = ar->len;
 
 	/* just a dirty hack to filter too big requests  */
-	if (len >= EXT4_CLUSTERS_PER_GROUP(sb) - 10)
-		len = EXT4_CLUSTERS_PER_GROUP(sb) - 10;
+	if (len >= EXT4_CLUSTERS_PER_GROUP(sb))
+		len = EXT4_CLUSTERS_PER_GROUP(sb);
 
 	/* start searching from the goal */
 	goal = ar->goal;
@@ -4136,7 +4111,7 @@ static void ext4_mb_add_n_trim(struct ext4_allocation_context *ac)
 		/* The max size of hash table is PREALLOC_TB_SIZE */
 		order = PREALLOC_TB_SIZE - 1;
 	/* Add the prealloc space to lg */
-	rcu_read_lock();
+	spin_lock(&lg->lg_prealloc_lock);
 	list_for_each_entry_rcu(tmp_pa, &lg->lg_prealloc_list[order],
 						pa_inode_list) {
 		spin_lock(&tmp_pa->pa_lock);
@@ -4160,12 +4135,12 @@ static void ext4_mb_add_n_trim(struct ext4_allocation_context *ac)
 	if (!added)
 		list_add_tail_rcu(&pa->pa_inode_list,
 					&lg->lg_prealloc_list[order]);
-	rcu_read_unlock();
+	spin_unlock(&lg->lg_prealloc_lock);
 
 	/* Now trim the list to be not more than 8 elements */
 	if (lg_prealloc_count > 8) {
 		ext4_mb_discard_lg_preallocations(sb, lg,
-						order, lg_prealloc_count);
+						  order, lg_prealloc_count);
 		return;
 	}
 	return ;