Merge git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw: (25 commits)
  GFS2: Merge gfs2_get_sb into gfs2_get_sb_meta
  GFS2: Fix cache coherency between truncate and O_DIRECT read
  GFS2: Fix locking issue mounting gfs2meta fs
  GFS2: Remove unused variable
  GFS2: smbd proccess hangs with flock() call.
  GFS2: Remove args subdir from gfs2 sysfs files
  GFS2: Remove lockstruct subdir from gfs2 sysfs files
  GFS2: Move gfs2_unlink_ok into ops_inode.c
  GFS2: Move gfs2_readlinki into ops_inode.c
  GFS2: Move gfs2_rmdiri into ops_inode.c
  GFS2: Merge mount.c and ops_super.c into super.c
  GFS2: Clean up some file names
  GFS2: Be more aggressive in reclaiming unlinked inodes
  GFS2: Add a rgrp bitmap full flag
  GFS2: Improve resource group error handling
  GFS2: Don't warn when delete inode fails on ro filesystem
  GFS2: Update docs
  GFS2: Umount recovery race fix
  GFS2: Remove a couple of unused sysfs entries
  GFS2: Add commit= mount option
  ...

Documentation/filesystems/gfs2-glocks.txt

@@ -60,7 +60,7 @@ go_lock          | Called for the first local holder of a lock
 go_unlock        | Called on the final local unlock of a lock
 go_dump          | Called to print content of object for debugfs file, or on
                  | error to dump glock to the log.
-go_type;         | The type of the glock, LM_TYPE_.....
+go_type          | The type of the glock, LM_TYPE_.....
 go_min_hold_time | The minimum hold time
 
 The minimum hold time for each lock is the time after a remote lock

Documentation/filesystems/gfs2.txt

@@ -11,18 +11,15 @@ their I/O so file system consistency is maintained.  One of the nifty
 features of GFS is perfect consistency -- changes made to the file system
 on one machine show up immediately on all other machines in the cluster.
 
-GFS uses interchangable inter-node locking mechanisms.  Different lock
-modules can plug into GFS and each file system selects the appropriate
-lock module at mount time.  Lock modules include:
+GFS uses interchangable inter-node locking mechanisms, the currently
+supported mechanisms are:
 
   lock_nolock -- allows gfs to be used as a local file system
 
   lock_dlm -- uses a distributed lock manager (dlm) for inter-node locking
   The dlm is found at linux/fs/dlm/
 
-In addition to interfacing with an external locking manager, a gfs lock
-module is responsible for interacting with external cluster management
-systems.  Lock_dlm depends on user space cluster management systems found
+Lock_dlm depends on user space cluster management systems found
 at the URL above.
 
 To use gfs as a local file system, no external clustering systems are
@@ -31,13 +28,19 @@ needed, simply:
   $ mkfs -t gfs2 -p lock_nolock -j 1 /dev/block_device
   $ mount -t gfs2 /dev/block_device /dir
 
-GFS2 is not on-disk compatible with previous versions of GFS.
+If you are using Fedora, you need to install the gfs2-utils package
+and, for lock_dlm, you will also need to install the cman package
+and write a cluster.conf as per the documentation.
+
+GFS2 is not on-disk compatible with previous versions of GFS, but it
+is pretty close.
 
 The following man pages can be found at the URL above:
-  gfs2_fsck	to repair a filesystem
+  fsck.gfs2	to repair a filesystem
   gfs2_grow	to expand a filesystem online
   gfs2_jadd	to add journals to a filesystem online
   gfs2_tool	to manipulate, examine and tune a filesystem
   gfs2_quota	to examine and change quota values in a filesystem
+  gfs2_convert	to convert a gfs filesystem to gfs2 in-place
   mount.gfs2	to help mount(8) mount a filesystem
   mkfs.gfs2	to make a filesystem

fs/gfs2/Kconfig

@@ -7,6 +7,7 @@ config GFS2_FS
 	select IP_SCTP if DLM_SCTP
 	select FS_POSIX_ACL
 	select CRC32
+	select SLOW_WORK
 	help
 	  A cluster filesystem.
 

fs/gfs2/Makefile

 obj-$(CONFIG_GFS2_FS) += gfs2.o
 gfs2-y := acl.o bmap.o dir.o eaops.o eattr.o glock.o \
 	glops.o inode.o log.o lops.o main.o meta_io.o \
-	mount.o ops_address.o ops_dentry.o ops_export.o ops_file.o \
-	ops_fstype.o ops_inode.o ops_super.o quota.o \
+	aops.o dentry.o export.o file.o \
+	ops_fstype.o ops_inode.o quota.o \
 	recovery.o rgrp.o super.o sys.o trans.o util.o
 
 gfs2-$(CONFIG_GFS2_FS_LOCKING_DLM) += lock_dlm.o

fs/gfs2/ops_address.c → fs/gfs2/aops.c

@@ -28,7 +28,6 @@
 #include "inode.h"
 #include "log.h"
 #include "meta_io.h"
-#include "ops_address.h"
 #include "quota.h"
 #include "trans.h"
 #include "rgrp.h"
@@ -781,10 +780,12 @@ static int gfs2_stuffed_write_end(struct inode *inode, struct buffer_head *dibh,
 	unlock_page(page);
 	page_cache_release(page);
 
-	if (inode->i_size < to) {
-		i_size_write(inode, to);
-		ip->i_disksize = inode->i_size;
-		di->di_size = cpu_to_be64(inode->i_size);
+	if (copied) {
+		if (inode->i_size < to) {
+			i_size_write(inode, to);
+			ip->i_disksize = inode->i_size;
+		}
+		gfs2_dinode_out(ip, di);
 		mark_inode_dirty(inode);
 	}
 
@@ -824,7 +825,6 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping,
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct buffer_head *dibh;
 	struct gfs2_alloc *al = ip->i_alloc;
-	struct gfs2_dinode *di;
 	unsigned int from = pos & (PAGE_CACHE_SIZE - 1);
 	unsigned int to = from + len;
 	int ret;
@@ -847,11 +847,10 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping,
 		gfs2_page_add_databufs(ip, page, from, to);
 
 	ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata);
-
-	if (likely(ret >= 0) && (inode->i_size > ip->i_disksize)) {
-		di = (struct gfs2_dinode *)dibh->b_data;
-		ip->i_disksize = inode->i_size;
-		di->di_size = cpu_to_be64(inode->i_size);
+	if (ret > 0) {
+		if (inode->i_size > ip->i_disksize)
+			ip->i_disksize = inode->i_size;
+		gfs2_dinode_out(ip, dibh->b_data);
 		mark_inode_dirty(inode);
 	}
 

fs/gfs2/bmap.c

@@ -25,7 +25,6 @@
 #include "trans.h"
 #include "dir.h"
 #include "util.h"
-#include "ops_address.h"
 
 /* This doesn't need to be that large as max 64 bit pointers in a 4k
  * block is 512, so __u16 is fine for that. It saves stack space to
@@ -136,7 +135,9 @@ int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page)
 		   and write it out to disk */
 
 		unsigned int n = 1;
-		block = gfs2_alloc_block(ip, &n);
+		error = gfs2_alloc_block(ip, &block, &n);
+		if (error)
+			goto out_brelse;
 		if (isdir) {
 			gfs2_trans_add_unrevoke(GFS2_SB(&ip->i_inode), block, 1);
 			error = gfs2_dir_get_new_buffer(ip, block, &bh);
@@ -476,8 +477,11 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock,
 	blks = dblks + iblks;
 	i = sheight;
 	do {
+		int error;
 		n = blks - alloced;
-		bn = gfs2_alloc_block(ip, &n);
+		error = gfs2_alloc_block(ip, &bn, &n);
+		if (error)
+			return error;
 		alloced += n;
 		if (state != ALLOC_DATA || gfs2_is_jdata(ip))
 			gfs2_trans_add_unrevoke(sdp, bn, n);
@@ -1008,7 +1012,7 @@ static int gfs2_block_truncate_page(struct address_space *mapping)
 		gfs2_trans_add_bh(ip->i_gl, bh, 0);
 
 	zero_user(page, offset, length);
-
+	mark_buffer_dirty(bh);
 unlock:
 	unlock_page(page);
 	page_cache_release(page);

fs/gfs2/dir.c

@@ -803,13 +803,20 @@ static struct gfs2_leaf *new_leaf(struct inode *inode, struct buffer_head **pbh,
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
 	unsigned int n = 1;
-	u64 bn = gfs2_alloc_block(ip, &n);
-	struct buffer_head *bh = gfs2_meta_new(ip->i_gl, bn);
+	u64 bn;
+	int error;
+	struct buffer_head *bh;
 	struct gfs2_leaf *leaf;
 	struct gfs2_dirent *dent;
 	struct qstr name = { .name = "", .len = 0, .hash = 0 };
+
+	error = gfs2_alloc_block(ip, &bn, &n);
+	if (error)
+		return NULL;
+	bh = gfs2_meta_new(ip->i_gl, bn);
 	if (!bh)
 		return NULL;
+
 	gfs2_trans_add_unrevoke(GFS2_SB(inode), bn, 1);
 	gfs2_trans_add_bh(ip->i_gl, bh, 1);
 	gfs2_metatype_set(bh, GFS2_METATYPE_LF, GFS2_FORMAT_LF);

fs/gfs2/eattr.c

@@ -582,8 +582,11 @@ static int ea_alloc_blk(struct gfs2_inode *ip, struct buffer_head **bhp)
 	struct gfs2_ea_header *ea;
 	unsigned int n = 1;
 	u64 block;
+	int error;
 
-	block = gfs2_alloc_block(ip, &n);
+	error = gfs2_alloc_block(ip, &block, &n);
+	if (error)
+		return error;
 	gfs2_trans_add_unrevoke(sdp, block, 1);
 	*bhp = gfs2_meta_new(ip->i_gl, block);
 	gfs2_trans_add_bh(ip->i_gl, *bhp, 1);
@@ -617,6 +620,7 @@ static int ea_write(struct gfs2_inode *ip, struct gfs2_ea_header *ea,
 		    struct gfs2_ea_request *er)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+	int error;
 
 	ea->ea_data_len = cpu_to_be32(er->er_data_len);
 	ea->ea_name_len = er->er_name_len;
@@ -642,7 +646,9 @@ static int ea_write(struct gfs2_inode *ip, struct gfs2_ea_header *ea,
 			int mh_size = sizeof(struct gfs2_meta_header);
 			unsigned int n = 1;
 
-			block = gfs2_alloc_block(ip, &n);
+			error = gfs2_alloc_block(ip, &block, &n);
+			if (error)
+				return error;
 			gfs2_trans_add_unrevoke(sdp, block, 1);
 			bh = gfs2_meta_new(ip->i_gl, block);
 			gfs2_trans_add_bh(ip->i_gl, bh, 1);
@@ -963,7 +969,9 @@ static int ea_set_block(struct gfs2_inode *ip, struct gfs2_ea_request *er,
 	} else {
 		u64 blk;
 		unsigned int n = 1;
-		blk = gfs2_alloc_block(ip, &n);
+		error = gfs2_alloc_block(ip, &blk, &n);
+		if (error)
+			return error;
 		gfs2_trans_add_unrevoke(sdp, blk, 1);
 		indbh = gfs2_meta_new(ip->i_gl, blk);
 		gfs2_trans_add_bh(ip->i_gl, indbh, 1);

fs/gfs2/ops_file.c → fs/gfs2/file.c

@@ -39,7 +39,6 @@
 #include "trans.h"
 #include "util.h"
 #include "eaops.h"
-#include "ops_address.h"
 
 /**
  * gfs2_llseek - seek to a location in a file
@@ -425,33 +424,36 @@ static struct vm_operations_struct gfs2_vm_ops = {
 	.page_mkwrite = gfs2_page_mkwrite,
 };
 
-
 /**
  * gfs2_mmap -
  * @file: The file to map
  * @vma: The VMA which described the mapping
  *
- * Returns: 0 or error code
+ * There is no need to get a lock here unless we should be updating
+ * atime. We ignore any locking errors since the only consequence is
+ * a missed atime update (which will just be deferred until later).
+ *
+ * Returns: 0
  */
 
 static int gfs2_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	struct gfs2_inode *ip = GFS2_I(file->f_mapping->host);
-	struct gfs2_holder i_gh;
-	int error;
 
-	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh);
-	error = gfs2_glock_nq(&i_gh);
-	if (error) {
-		gfs2_holder_uninit(&i_gh);
-		return error;
-	}
+	if (!(file->f_flags & O_NOATIME)) {
+		struct gfs2_holder i_gh;
+		int error;
 
+		gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh);
+		error = gfs2_glock_nq(&i_gh);
+		file_accessed(file);
+		if (error == 0)
+			gfs2_glock_dq_uninit(&i_gh);
+	}
 	vma->vm_ops = &gfs2_vm_ops;
+	vma->vm_flags |= VM_CAN_NONLINEAR;
 
-	gfs2_glock_dq_uninit(&i_gh);
-
-	return error;
+	return 0;
 }
 
 /**
@@ -692,12 +694,10 @@ static void do_unflock(struct file *file, struct file_lock *fl)
 
 static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl)
 {
-	struct gfs2_inode *ip = GFS2_I(file->f_mapping->host);
-
 	if (!(fl->fl_flags & FL_FLOCK))
 		return -ENOLCK;
-	if (__mandatory_lock(&ip->i_inode))
-		return -ENOLCK;
+	if (fl->fl_type & LOCK_MAND)
+		return -EOPNOTSUPP;
 
 	if (fl->fl_type == F_UNLCK) {
 		do_unflock(file, fl);

fs/gfs2/glock.c

@@ -796,22 +796,37 @@ void gfs2_holder_uninit(struct gfs2_holder *gh)
 	gh->gh_ip = 0;
 }
 
-static int just_schedule(void *word)
+/**
+ * gfs2_glock_holder_wait
+ * @word: unused
+ *
+ * This function and gfs2_glock_demote_wait both show up in the WCHAN
+ * field. Thus I've separated these otherwise identical functions in
+ * order to be more informative to the user.
+ */
+
+static int gfs2_glock_holder_wait(void *word)
 {
         schedule();
         return 0;
 }
 
+static int gfs2_glock_demote_wait(void *word)
+{
+	schedule();
+	return 0;
+}
+
 static void wait_on_holder(struct gfs2_holder *gh)
 {
 	might_sleep();
-	wait_on_bit(&gh->gh_iflags, HIF_WAIT, just_schedule, TASK_UNINTERRUPTIBLE);
+	wait_on_bit(&gh->gh_iflags, HIF_WAIT, gfs2_glock_holder_wait, TASK_UNINTERRUPTIBLE);
 }
 
 static void wait_on_demote(struct gfs2_glock *gl)
 {
 	might_sleep();
-	wait_on_bit(&gl->gl_flags, GLF_DEMOTE, just_schedule, TASK_UNINTERRUPTIBLE);
+	wait_on_bit(&gl->gl_flags, GLF_DEMOTE, gfs2_glock_demote_wait, TASK_UNINTERRUPTIBLE);
 }
 
 /**

fs/gfs2/glops.c

@@ -309,24 +309,6 @@ static void rgrp_go_unlock(struct gfs2_holder *gh)
 	gfs2_rgrp_bh_put(gh->gh_gl->gl_object);
 }
 
-/**
- * rgrp_go_dump - print out an rgrp
- * @seq: The iterator
- * @gl: The glock in question
- *
- */
-
-static int rgrp_go_dump(struct seq_file *seq, const struct gfs2_glock *gl)
-{
-	const struct gfs2_rgrpd *rgd = gl->gl_object;
-	if (rgd == NULL)
-		return 0;
-	gfs2_print_dbg(seq, " R: n:%llu f:%02x b:%u/%u i:%u\n",
-		       (unsigned long long)rgd->rd_addr, rgd->rd_flags,
-		       rgd->rd_free, rgd->rd_free_clone, rgd->rd_dinodes);
-	return 0;
-}
-
 /**
  * trans_go_sync - promote/demote the transaction glock
  * @gl: the glock
@@ -410,7 +392,7 @@ const struct gfs2_glock_operations gfs2_rgrp_glops = {
 	.go_demote_ok = rgrp_go_demote_ok,
 	.go_lock = rgrp_go_lock,
 	.go_unlock = rgrp_go_unlock,
-	.go_dump = rgrp_go_dump,
+	.go_dump = gfs2_rgrp_dump,
 	.go_type = LM_TYPE_RGRP,
 	.go_min_hold_time = HZ / 5,
 };

fs/gfs2/incore.h

@@ -12,6 +12,7 @@
 
 #include <linux/fs.h>
 #include <linux/workqueue.h>
+#include <linux/slow-work.h>
 #include <linux/dlm.h>
 #include <linux/buffer_head.h>
 
@@ -63,9 +64,12 @@ struct gfs2_log_element {
 	const struct gfs2_log_operations *le_ops;
 };
 
+#define GBF_FULL 1
+
 struct gfs2_bitmap {
 	struct buffer_head *bi_bh;
 	char *bi_clone;
+	unsigned long bi_flags;
 	u32 bi_offset;
 	u32 bi_start;
 	u32 bi_len;
@@ -90,10 +94,11 @@ struct gfs2_rgrpd {
 	struct gfs2_sbd *rd_sbd;
 	unsigned int rd_bh_count;
 	u32 rd_last_alloc;
-	unsigned char rd_flags;
-#define GFS2_RDF_CHECK        0x01      /* Need to check for unlinked inodes */
-#define GFS2_RDF_NOALLOC      0x02      /* rg prohibits allocation */
-#define GFS2_RDF_UPTODATE     0x04      /* rg is up to date */
+	u32 rd_flags;
+#define GFS2_RDF_CHECK		0x10000000 /* check for unlinked inodes */
+#define GFS2_RDF_UPTODATE	0x20000000 /* rg is up to date */
+#define GFS2_RDF_ERROR		0x40000000 /* error in rg */
+#define GFS2_RDF_MASK		0xf0000000 /* mask for internal flags */
 };
 
 enum gfs2_state_bits {
@@ -376,11 +381,11 @@ struct gfs2_journal_extent {
 struct gfs2_jdesc {
 	struct list_head jd_list;
 	struct list_head extent_list;
-
+	struct slow_work jd_work;
 	struct inode *jd_inode;
+	unsigned long jd_flags;
+#define JDF_RECOVERY 1
 	unsigned int jd_jid;
-	int jd_dirty;
-
 	unsigned int jd_blocks;
 };
 
@@ -390,9 +395,6 @@ struct gfs2_statfs_change_host {
 	s64 sc_dinodes;
 };
 
-#define GFS2_GLOCKD_DEFAULT	1
-#define GFS2_GLOCKD_MAX		16
-
 #define GFS2_QUOTA_DEFAULT	GFS2_QUOTA_OFF
 #define GFS2_QUOTA_OFF		0
 #define GFS2_QUOTA_ACCOUNT	1
@@ -418,6 +420,7 @@ struct gfs2_args {
 	unsigned int ar_data:2;			/* ordered/writeback */
 	unsigned int ar_meta:1;			/* mount metafs */
 	unsigned int ar_discard:1;		/* discard requests */
+	int ar_commit;				/* Commit interval */
 };
 
 struct gfs2_tune {
@@ -426,7 +429,6 @@ struct gfs2_tune {
 	unsigned int gt_incore_log_blocks;
 	unsigned int gt_log_flush_secs;
 
-	unsigned int gt_recoverd_secs;
 	unsigned int gt_logd_secs;
 
 	unsigned int gt_quota_simul_sync; /* Max quotavals to sync at once */
@@ -447,6 +449,7 @@ enum {
 	SDF_JOURNAL_LIVE	= 1,
 	SDF_SHUTDOWN		= 2,
 	SDF_NOBARRIERS		= 3,
+	SDF_NORECOVERY		= 4,
 };
 
 #define GFS2_FSNAME_LEN		256
@@ -493,7 +496,6 @@ struct lm_lockstruct {
 	unsigned long ls_flags;
 	dlm_lockspace_t *ls_dlm;
 
-	int ls_recover_jid;
 	int ls_recover_jid_done;
 	int ls_recover_jid_status;
 };
@@ -582,7 +584,6 @@ struct gfs2_sbd {
 
 	/* Daemon stuff */
 
-	struct task_struct *sd_recoverd_process;
 	struct task_struct *sd_logd_process;
 	struct task_struct *sd_quotad_process;
 

fs/gfs2/inode.c

@@ -30,7 +30,6 @@
 #include "inode.h"
 #include "log.h"
 #include "meta_io.h"
-#include "ops_address.h"
 #include "quota.h"
 #include "rgrp.h"
 #include "trans.h"
@@ -1047,154 +1046,7 @@ fail:
 	return ERR_PTR(error);
 }
 
-/**
- * gfs2_rmdiri - Remove a directory
- * @dip: The parent directory of the directory to be removed
- * @name: The name of the directory to be removed
- * @ip: The GFS2 inode of the directory to be removed
- *
- * Assumes Glocks on dip and ip are held
- *
- * Returns: errno
- */
-
-int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name,
-		struct gfs2_inode *ip)
-{
-	struct qstr dotname;
-	int error;
-
-	if (ip->i_entries != 2) {
-		if (gfs2_consist_inode(ip))
-			gfs2_dinode_print(ip);
-		return -EIO;
-	}
-
-	error = gfs2_dir_del(dip, name);
-	if (error)
-		return error;
-
-	error = gfs2_change_nlink(dip, -1);
-	if (error)
-		return error;
-
-	gfs2_str2qstr(&dotname, ".");
-	error = gfs2_dir_del(ip, &dotname);
-	if (error)
-		return error;
-
-	gfs2_str2qstr(&dotname, "..");
-	error = gfs2_dir_del(ip, &dotname);
-	if (error)
-		return error;
-
-	/* It looks odd, but it really should be done twice */
-	error = gfs2_change_nlink(ip, -1);
-	if (error)
-		return error;
-
-	error = gfs2_change_nlink(ip, -1);
-	if (error)
-		return error;
-
-	return error;
-}
-
-/*
- * gfs2_unlink_ok - check to see that a inode is still in a directory
- * @dip: the directory
- * @name: the name of the file
- * @ip: the inode
- *
- * Assumes that the lock on (at least) @dip is held.
- *
- * Returns: 0 if the parent/child relationship is correct, errno if it isn't
- */
-
-int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name,
-		   const struct gfs2_inode *ip)
-{
-	int error;
-
-	if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode))
-		return -EPERM;
-
-	if ((dip->i_inode.i_mode & S_ISVTX) &&
-	    dip->i_inode.i_uid != current_fsuid() &&
-	    ip->i_inode.i_uid != current_fsuid() && !capable(CAP_FOWNER))
-		return -EPERM;
-
-	if (IS_APPEND(&dip->i_inode))
-		return -EPERM;
-
-	error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC);
-	if (error)
-		return error;
-
-	error = gfs2_dir_check(&dip->i_inode, name, ip);
-	if (error)
-		return error;
-
-	return 0;
-}
-
-/**
- * gfs2_readlinki - return the contents of a symlink
- * @ip: the symlink's inode
- * @buf: a pointer to the buffer to be filled
- * @len: a pointer to the length of @buf
- *
- * If @buf is too small, a piece of memory is kmalloc()ed and needs
- * to be freed by the caller.
- *
- * Returns: errno
- */
-
-int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len)
-{
-	struct gfs2_holder i_gh;
-	struct buffer_head *dibh;
-	unsigned int x;
-	int error;
-
-	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh);
-	error = gfs2_glock_nq(&i_gh);
-	if (error) {
-		gfs2_holder_uninit(&i_gh);
-		return error;
-	}
-
-	if (!ip->i_disksize) {
-		gfs2_consist_inode(ip);
-		error = -EIO;
-		goto out;
-	}
-
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-	if (error)
-		goto out;
-
-	x = ip->i_disksize + 1;
-	if (x > *len) {
-		*buf = kmalloc(x, GFP_NOFS);
-		if (!*buf) {
-			error = -ENOMEM;
-			goto out_brelse;
-		}
-	}
-
-	memcpy(*buf, dibh->b_data + sizeof(struct gfs2_dinode), x);
-	*len = x;
-
-out_brelse:
-	brelse(dibh);
-out:
-	gfs2_glock_dq_uninit(&i_gh);
-	return error;
-}
-
-static int
-__gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr)
+static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr)
 {
 	struct buffer_head *dibh;
 	int error;

fs/gfs2/inode.h

@@ -11,8 +11,16 @@
 #define __INODE_DOT_H__
 
 #include <linux/fs.h>
+#include <linux/buffer_head.h>
+#include <linux/mm.h>
 #include "util.h"
 
+extern int gfs2_releasepage(struct page *page, gfp_t gfp_mask);
+extern int gfs2_internal_read(struct gfs2_inode *ip,
+			      struct file_ra_state *ra_state,
+			      char *buf, loff_t *pos, unsigned size);
+extern void gfs2_set_aops(struct inode *inode);
+
 static inline int gfs2_is_stuffed(const struct gfs2_inode *ip)
 {
 	return !ip->i_height;
@@ -73,30 +81,26 @@ static inline void gfs2_inum_out(const struct gfs2_inode *ip,
 }
 
 
-void gfs2_set_iop(struct inode *inode);
-struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, 
-				u64 no_addr, u64 no_formal_ino,
-				int skip_freeing);
-struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr);
-
-int gfs2_inode_refresh(struct gfs2_inode *ip);
-
-int gfs2_dinode_dealloc(struct gfs2_inode *inode);
-int gfs2_change_nlink(struct gfs2_inode *ip, int diff);
-struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name,
-			   int is_root);
-struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
-			   unsigned int mode, dev_t dev);
-int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name,
-		struct gfs2_inode *ip);
-int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name,
-		   const struct gfs2_inode *ip);
-int gfs2_permission(struct inode *inode, int mask);
-int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len);
-int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr);
-struct inode *gfs2_lookup_simple(struct inode *dip, const char *name);
-void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf);
-void gfs2_dinode_print(const struct gfs2_inode *ip);
+extern void gfs2_set_iop(struct inode *inode);
+extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, 
+				       u64 no_addr, u64 no_formal_ino,
+				       int skip_freeing);
+extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr);
+
+extern int gfs2_inode_refresh(struct gfs2_inode *ip);
+
+extern int gfs2_dinode_dealloc(struct gfs2_inode *inode);
+extern int gfs2_change_nlink(struct gfs2_inode *ip, int diff);
+extern struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name,
+				  int is_root);
+extern struct inode *gfs2_createi(struct gfs2_holder *ghs,
+				  const struct qstr *name,
+				  unsigned int mode, dev_t dev);
+extern int gfs2_permission(struct inode *inode, int mask);
+extern int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr);
+extern struct inode *gfs2_lookup_simple(struct inode *dip, const char *name);
+extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf);
+extern void gfs2_dinode_print(const struct gfs2_inode *ip);
 
 extern const struct inode_operations gfs2_file_iops;
 extern const struct inode_operations gfs2_dir_iops;

fs/gfs2/log.c

@@ -120,7 +120,7 @@ __acquires(&sdp->sd_log_lock)
 			lock_buffer(bh);
 			if (test_clear_buffer_dirty(bh)) {
 				bh->b_end_io = end_buffer_write_sync;
-				submit_bh(WRITE, bh);
+				submit_bh(WRITE_SYNC_PLUG, bh);
 			} else {
 				unlock_buffer(bh);
 				brelse(bh);
@@ -604,7 +604,7 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull)
 	if (test_bit(SDF_NOBARRIERS, &sdp->sd_flags))
 		goto skip_barrier;
 	get_bh(bh);
-	submit_bh(WRITE_BARRIER | (1 << BIO_RW_META), bh);
+	submit_bh(WRITE_SYNC | (1 << BIO_RW_BARRIER) | (1 << BIO_RW_META), bh);
 	wait_on_buffer(bh);
 	if (buffer_eopnotsupp(bh)) {
 		clear_buffer_eopnotsupp(bh);
@@ -664,7 +664,7 @@ static void gfs2_ordered_write(struct gfs2_sbd *sdp)
 		lock_buffer(bh);
 		if (buffer_mapped(bh) && test_clear_buffer_dirty(bh)) {
 			bh->b_end_io = end_buffer_write_sync;
-			submit_bh(WRITE, bh);
+			submit_bh(WRITE_SYNC_PLUG, bh);
 		} else {
 			unlock_buffer(bh);
 			brelse(bh);

fs/gfs2/lops.c

@@ -13,6 +13,8 @@
 #include <linux/completion.h>
 #include <linux/buffer_head.h>
 #include <linux/gfs2_ondisk.h>
+#include <linux/bio.h>
+#include <linux/fs.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -189,7 +191,7 @@ static void buf_lo_before_commit(struct gfs2_sbd *sdp)
 		}
 
 		gfs2_log_unlock(sdp);
-		submit_bh(WRITE, bh);
+		submit_bh(WRITE_SYNC_PLUG, bh);
 		gfs2_log_lock(sdp);
 
 		n = 0;
@@ -199,7 +201,7 @@ static void buf_lo_before_commit(struct gfs2_sbd *sdp)
 			gfs2_log_unlock(sdp);
 			lock_buffer(bd2->bd_bh);
 			bh = gfs2_log_fake_buf(sdp, bd2->bd_bh);
-			submit_bh(WRITE, bh);
+			submit_bh(WRITE_SYNC_PLUG, bh);
 			gfs2_log_lock(sdp);
 			if (++n >= num)
 				break;
@@ -341,7 +343,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp)
 		sdp->sd_log_num_revoke--;
 
 		if (offset + sizeof(u64) > sdp->sd_sb.sb_bsize) {
-			submit_bh(WRITE, bh);
+			submit_bh(WRITE_SYNC_PLUG, bh);
 
 			bh = gfs2_log_get_buf(sdp);
 			mh = (struct gfs2_meta_header *)bh->b_data;
@@ -358,7 +360,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp)
 	}
 	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke);
 
-	submit_bh(WRITE, bh);
+	submit_bh(WRITE_SYNC_PLUG, bh);
 }
 
 static void revoke_lo_before_scan(struct gfs2_jdesc *jd,
@@ -560,7 +562,7 @@ static void gfs2_write_blocks(struct gfs2_sbd *sdp, struct buffer_head *bh,
 	ptr = bh_log_ptr(bh);
 	
 	get_bh(bh);
-	submit_bh(WRITE, bh);
+	submit_bh(WRITE_SYNC_PLUG, bh);
 	gfs2_log_lock(sdp);
 	while(!list_empty(list)) {
 		bd = list_entry(list->next, struct gfs2_bufdata, bd_le.le_list);
@@ -586,7 +588,7 @@ static void gfs2_write_blocks(struct gfs2_sbd *sdp, struct buffer_head *bh,
 		} else {
 			bh1 = gfs2_log_fake_buf(sdp, bd->bd_bh);
 		}
-		submit_bh(WRITE, bh1);
+		submit_bh(WRITE_SYNC_PLUG, bh1);
 		gfs2_log_lock(sdp);
 		ptr += 2;
 	}

fs/gfs2/main.c

@@ -15,6 +15,7 @@
 #include <linux/init.h>
 #include <linux/gfs2_ondisk.h>
 #include <asm/atomic.h>
+#include <linux/slow-work.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -113,12 +114,18 @@ static int __init init_gfs2_fs(void)
 	if (error)
 		goto fail_unregister;
 
+	error = slow_work_register_user();
+	if (error)
+		goto fail_slow;
+
 	gfs2_register_debugfs();
 
 	printk("GFS2 (built %s %s) installed\n", __DATE__, __TIME__);
 
 	return 0;
 
+fail_slow:
+	unregister_filesystem(&gfs2meta_fs_type);
 fail_unregister:
 	unregister_filesystem(&gfs2_fs_type);
 fail:
@@ -156,6 +163,7 @@ static void __exit exit_gfs2_fs(void)
 	gfs2_unregister_debugfs();
 	unregister_filesystem(&gfs2_fs_type);
 	unregister_filesystem(&gfs2meta_fs_type);
+	slow_work_unregister_user();
 
 	kmem_cache_destroy(gfs2_quotad_cachep);
 	kmem_cache_destroy(gfs2_rgrpd_cachep);

fs/gfs2/meta_io.c

@@ -31,19 +31,66 @@
 #include "rgrp.h"
 #include "trans.h"
 #include "util.h"
-#include "ops_address.h"
 
-static int aspace_get_block(struct inode *inode, sector_t lblock,
-			    struct buffer_head *bh_result, int create)
+static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wbc)
 {
-	gfs2_assert_warn(inode->i_sb->s_fs_info, 0);
-	return -EOPNOTSUPP;
-}
+	int err;
+	struct buffer_head *bh, *head;
+	int nr_underway = 0;
+	int write_op = (1 << BIO_RW_META) | ((wbc->sync_mode == WB_SYNC_ALL ?
+			WRITE_SYNC_PLUG : WRITE));
+
+	BUG_ON(!PageLocked(page));
+	BUG_ON(!page_has_buffers(page));
+
+	head = page_buffers(page);
+	bh = head;
+
+	do {
+		if (!buffer_mapped(bh))
+			continue;
+		/*
+		 * If it's a fully non-blocking write attempt and we cannot
+		 * lock the buffer then redirty the page.  Note that this can
+		 * potentially cause a busy-wait loop from pdflush and kswapd
+		 * activity, but those code paths have their own higher-level
+		 * throttling.
+		 */
+		if (wbc->sync_mode != WB_SYNC_NONE || !wbc->nonblocking) {
+			lock_buffer(bh);
+		} else if (!trylock_buffer(bh)) {
+			redirty_page_for_writepage(wbc, page);
+			continue;
+		}
+		if (test_clear_buffer_dirty(bh)) {
+			mark_buffer_async_write(bh);
+		} else {
+			unlock_buffer(bh);
+		}
+	} while ((bh = bh->b_this_page) != head);
+
+	/*
+	 * The page and its buffers are protected by PageWriteback(), so we can
+	 * drop the bh refcounts early.
+	 */
+	BUG_ON(PageWriteback(page));
+	set_page_writeback(page);
+
+	do {
+		struct buffer_head *next = bh->b_this_page;
+		if (buffer_async_write(bh)) {
+			submit_bh(write_op, bh);
+			nr_underway++;
+		}
+		bh = next;
+	} while (bh != head);
+	unlock_page(page);
 
-static int gfs2_aspace_writepage(struct page *page,
-				 struct writeback_control *wbc)
-{
-	return block_write_full_page(page, aspace_get_block, wbc);
+	err = 0;
+	if (nr_underway == 0)
+		end_page_writeback(page);
+
+	return err;
 }
 
 static const struct address_space_operations aspace_aops = {
@@ -201,16 +248,32 @@ struct buffer_head *gfs2_meta_new(struct gfs2_glock *gl, u64 blkno)
 int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags,
 		   struct buffer_head **bhp)
 {
-	*bhp = gfs2_getbuf(gl, blkno, CREATE);
-	if (!buffer_uptodate(*bhp)) {
-		ll_rw_block(READ_META, 1, bhp);
-		if (flags & DIO_WAIT) {
-			int error = gfs2_meta_wait(gl->gl_sbd, *bhp);
-			if (error) {
-				brelse(*bhp);
-				return error;
-			}
-		}
+	struct gfs2_sbd *sdp = gl->gl_sbd;
+	struct buffer_head *bh;
+
+	if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
+		return -EIO;
+
+	*bhp = bh = gfs2_getbuf(gl, blkno, CREATE);
+
+	lock_buffer(bh);
+	if (buffer_uptodate(bh)) {
+		unlock_buffer(bh);
+		return 0;
+	}
+	bh->b_end_io = end_buffer_read_sync;
+	get_bh(bh);
+	submit_bh(READ_SYNC | (1 << BIO_RW_META), bh);
+	if (!(flags & DIO_WAIT))
+		return 0;
+
+	wait_on_buffer(bh);
+	if (unlikely(!buffer_uptodate(bh))) {
+		struct gfs2_trans *tr = current->journal_info;
+		if (tr && tr->tr_touched)
+			gfs2_io_error_bh(sdp, bh);
+		brelse(bh);
+		return -EIO;
 	}
 
 	return 0;
@@ -404,7 +467,7 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen)
 	if (buffer_uptodate(first_bh))
 		goto out;
 	if (!buffer_locked(first_bh))
-		ll_rw_block(READ_META, 1, &first_bh);
+		ll_rw_block(READ_SYNC | (1 << BIO_RW_META), 1, &first_bh);
 
 	dblock++;
 	extlen--;

fs/gfs2/mount.c

-/*
- * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
- * Copyright (C) 2004-2006 Red Hat, Inc.  All rights reserved.
- *
- * This copyrighted material is made available to anyone wishing to use,
- * modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License version 2.
- */
-
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/completion.h>
-#include <linux/buffer_head.h>
-#include <linux/gfs2_ondisk.h>
-#include <linux/parser.h>
-
-#include "gfs2.h"
-#include "incore.h"
-#include "super.h"
-#include "sys.h"
-#include "util.h"
-
-enum {
-	Opt_lockproto,
-	Opt_locktable,
-	Opt_hostdata,
-	Opt_spectator,
-	Opt_ignore_local_fs,
-	Opt_localflocks,
-	Opt_localcaching,
-	Opt_debug,
-	Opt_nodebug,
-	Opt_upgrade,
-	Opt_acl,
-	Opt_noacl,
-	Opt_quota_off,
-	Opt_quota_account,
-	Opt_quota_on,
-	Opt_quota,
-	Opt_noquota,
-	Opt_suiddir,
-	Opt_nosuiddir,
-	Opt_data_writeback,
-	Opt_data_ordered,
-	Opt_meta,
-	Opt_discard,
-	Opt_nodiscard,
-	Opt_err,
-};
-
-static const match_table_t tokens = {
-	{Opt_lockproto, "lockproto=%s"},
-	{Opt_locktable, "locktable=%s"},
-	{Opt_hostdata, "hostdata=%s"},
-	{Opt_spectator, "spectator"},
-	{Opt_ignore_local_fs, "ignore_local_fs"},
-	{Opt_localflocks, "localflocks"},
-	{Opt_localcaching, "localcaching"},
-	{Opt_debug, "debug"},
-	{Opt_nodebug, "nodebug"},
-	{Opt_upgrade, "upgrade"},
-	{Opt_acl, "acl"},
-	{Opt_noacl, "noacl"},
-	{Opt_quota_off, "quota=off"},
-	{Opt_quota_account, "quota=account"},
-	{Opt_quota_on, "quota=on"},
-	{Opt_quota, "quota"},
-	{Opt_noquota, "noquota"},
-	{Opt_suiddir, "suiddir"},
-	{Opt_nosuiddir, "nosuiddir"},
-	{Opt_data_writeback, "data=writeback"},
-	{Opt_data_ordered, "data=ordered"},
-	{Opt_meta, "meta"},
-	{Opt_discard, "discard"},
-	{Opt_nodiscard, "nodiscard"},
-	{Opt_err, NULL}
-};
-
-/**
- * gfs2_mount_args - Parse mount options
- * @sdp:
- * @data:
- *
- * Return: errno
- */
-
-int gfs2_mount_args(struct gfs2_sbd *sdp, struct gfs2_args *args, char *options)
-{
-	char *o;
-	int token;
-	substring_t tmp[MAX_OPT_ARGS];
-
-	/* Split the options into tokens with the "," character and
-	   process them */
-
-	while (1) {
-		o = strsep(&options, ",");
-		if (o == NULL)
-			break;
-		if (*o == '\0')
-			continue;
-
-		token = match_token(o, tokens, tmp);
-		switch (token) {
-		case Opt_lockproto:
-			match_strlcpy(args->ar_lockproto, &tmp[0],
-				      GFS2_LOCKNAME_LEN);
-			break;
-		case Opt_locktable:
-			match_strlcpy(args->ar_locktable, &tmp[0],
-				      GFS2_LOCKNAME_LEN);
-			break;
-		case Opt_hostdata:
-			match_strlcpy(args->ar_hostdata, &tmp[0],
-				      GFS2_LOCKNAME_LEN);
-			break;
-		case Opt_spectator:
-			args->ar_spectator = 1;
-			break;
-		case Opt_ignore_local_fs:
-			args->ar_ignore_local_fs = 1;
-			break;
-		case Opt_localflocks:
-			args->ar_localflocks = 1;
-			break;
-		case Opt_localcaching:
-			args->ar_localcaching = 1;
-			break;
-		case Opt_debug:
-			args->ar_debug = 1;
-			break;
-		case Opt_nodebug:
-			args->ar_debug = 0;
-			break;
-		case Opt_upgrade:
-			args->ar_upgrade = 1;
-			break;
-		case Opt_acl:
-			args->ar_posix_acl = 1;
-			break;
-		case Opt_noacl:
-			args->ar_posix_acl = 0;
-			break;
-		case Opt_quota_off:
-		case Opt_noquota:
-			args->ar_quota = GFS2_QUOTA_OFF;
-			break;
-		case Opt_quota_account:
-			args->ar_quota = GFS2_QUOTA_ACCOUNT;
-			break;
-		case Opt_quota_on:
-		case Opt_quota:
-			args->ar_quota = GFS2_QUOTA_ON;
-			break;
-		case Opt_suiddir:
-			args->ar_suiddir = 1;
-			break;
-		case Opt_nosuiddir:
-			args->ar_suiddir = 0;
-			break;
-		case Opt_data_writeback:
-			args->ar_data = GFS2_DATA_WRITEBACK;
-			break;
-		case Opt_data_ordered:
-			args->ar_data = GFS2_DATA_ORDERED;
-			break;
-		case Opt_meta:
-			args->ar_meta = 1;
-			break;
-		case Opt_discard:
-			args->ar_discard = 1;
-			break;
-		case Opt_nodiscard:
-			args->ar_discard = 0;
-			break;
-		case Opt_err:
-		default:
-			fs_info(sdp, "invalid mount option: %s\n", o);
-			return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-

fs/gfs2/ops_address.h

-/*
- * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
- * Copyright (C) 2004-2007 Red Hat, Inc.  All rights reserved.
- *
- * This copyrighted material is made available to anyone wishing to use,
- * modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License version 2.
- */
-
-#ifndef __OPS_ADDRESS_DOT_H__
-#define __OPS_ADDRESS_DOT_H__
-
-#include <linux/fs.h>
-#include <linux/buffer_head.h>
-#include <linux/mm.h>
-
-extern int gfs2_releasepage(struct page *page, gfp_t gfp_mask);
-extern int gfs2_internal_read(struct gfs2_inode *ip,
-			      struct file_ra_state *ra_state,
-			      char *buf, loff_t *pos, unsigned size);
-extern void gfs2_set_aops(struct inode *inode);
-
-#endif /* __OPS_ADDRESS_DOT_H__ */

fs/gfs2/ops_fstype.c

@@ -17,6 +17,7 @@
 #include <linux/namei.h>
 #include <linux/mount.h>
 #include <linux/gfs2_ondisk.h>
+#include <linux/slow-work.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -55,8 +56,6 @@ static void gfs2_tune_init(struct gfs2_tune *gt)
 	spin_lock_init(&gt->gt_spin);
 
 	gt->gt_incore_log_blocks = 1024;
-	gt->gt_log_flush_secs = 60;
-	gt->gt_recoverd_secs = 60;
 	gt->gt_logd_secs = 1;
 	gt->gt_quota_simul_sync = 64;
 	gt->gt_quota_warn_period = 10;
@@ -676,6 +675,7 @@ static int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh)
 			break;
 
 		INIT_LIST_HEAD(&jd->extent_list);
+		slow_work_init(&jd->jd_work, &gfs2_recover_ops);
 		jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1);
 		if (!jd->jd_inode || IS_ERR(jd->jd_inode)) {
 			if (!jd->jd_inode)
@@ -701,14 +701,13 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 {
 	struct inode *master = sdp->sd_master_dir->d_inode;
 	struct gfs2_holder ji_gh;
-	struct task_struct *p;
 	struct gfs2_inode *ip;
 	int jindex = 1;
 	int error = 0;
 
 	if (undo) {
 		jindex = 0;
-		goto fail_recoverd;
+		goto fail_jinode_gh;
 	}
 
 	sdp->sd_jindex = gfs2_lookup_simple(master, "jindex");
@@ -801,18 +800,8 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 	gfs2_glock_dq_uninit(&ji_gh);
 	jindex = 0;
 
-	p = kthread_run(gfs2_recoverd, sdp, "gfs2_recoverd");
-	error = IS_ERR(p);
-	if (error) {
-		fs_err(sdp, "can't start recoverd thread: %d\n", error);
-		goto fail_jinode_gh;
-	}
-	sdp->sd_recoverd_process = p;
-
 	return 0;
 
-fail_recoverd:
-	kthread_stop(sdp->sd_recoverd_process);
 fail_jinode_gh:
 	if (!sdp->sd_args.ar_spectator)
 		gfs2_glock_dq_uninit(&sdp->sd_jinode_gh);
@@ -1165,6 +1154,7 @@ static int fill_super(struct super_block *sb, void *data, int silent)
 
 	sdp->sd_args.ar_quota = GFS2_QUOTA_DEFAULT;
 	sdp->sd_args.ar_data = GFS2_DATA_DEFAULT;
+	sdp->sd_args.ar_commit = 60;
 
 	error = gfs2_mount_args(sdp, &sdp->sd_args, data);
 	if (error) {
@@ -1172,8 +1162,10 @@ static int fill_super(struct super_block *sb, void *data, int silent)
 		goto fail;
 	}
 
-	if (sdp->sd_args.ar_spectator)
+	if (sdp->sd_args.ar_spectator) {
                 sb->s_flags |= MS_RDONLY;
+		set_bit(SDF_NORECOVERY, &sdp->sd_flags);
+	}
 	if (sdp->sd_args.ar_posix_acl)
 		sb->s_flags |= MS_POSIXACL;
 
@@ -1191,6 +1183,8 @@ static int fill_super(struct super_block *sb, void *data, int silent)
                                GFS2_BASIC_BLOCK_SHIFT;
 	sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift;
 
+	sdp->sd_tune.gt_log_flush_secs = sdp->sd_args.ar_commit;
+
 	error = init_names(sdp, silent);
 	if (error)
 		goto fail;
@@ -1279,9 +1273,22 @@ static int gfs2_get_sb(struct file_system_type *fs_type, int flags,
 	return get_sb_bdev(fs_type, flags, dev_name, data, fill_super, mnt);
 }
 
-static struct super_block *get_gfs2_sb(const char *dev_name)
+static int test_meta_super(struct super_block *s, void *ptr)
+{
+	struct block_device *bdev = ptr;
+	return (bdev == s->s_bdev);
+}
+
+static int set_meta_super(struct super_block *s, void *ptr)
 {
-	struct super_block *sb;
+	return -EINVAL;
+}
+
+static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags,
+			    const char *dev_name, void *data, struct vfsmount *mnt)
+{
+	struct super_block *s;
+	struct gfs2_sbd *sdp;
 	struct path path;
 	int error;
 
@@ -1289,30 +1296,17 @@ static struct super_block *get_gfs2_sb(const char *dev_name)
 	if (error) {
 		printk(KERN_WARNING "GFS2: path_lookup on %s returned error %d\n",
 		       dev_name, error);
-		return NULL;
+		return error;
 	}
-	sb = path.dentry->d_inode->i_sb;
-	if (sb && (sb->s_type == &gfs2_fs_type))
-		atomic_inc(&sb->s_active);
-	else
-		sb = NULL;
+	s = sget(&gfs2_fs_type, test_meta_super, set_meta_super,
+		 path.dentry->d_inode->i_sb->s_bdev);
 	path_put(&path);
-	return sb;
-}
-
-static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags,
-			    const char *dev_name, void *data, struct vfsmount *mnt)
-{
-	struct super_block *sb = NULL;
-	struct gfs2_sbd *sdp;
-
-	sb = get_gfs2_sb(dev_name);
-	if (!sb) {
+	if (IS_ERR(s)) {
 		printk(KERN_WARNING "GFS2: gfs2 mount does not exist\n");
-		return -ENOENT;
+		return PTR_ERR(s);
 	}
-	sdp = sb->s_fs_info;
-	mnt->mnt_sb = sb;
+	sdp = s->s_fs_info;
+	mnt->mnt_sb = s;
 	mnt->mnt_root = dget(sdp->sd_master_dir);
 	return 0;
 }

fs/gfs2/ops_inode.c

@@ -262,6 +262,44 @@ out_parent:
 	return error;
 }
 
+/*
+ * gfs2_unlink_ok - check to see that a inode is still in a directory
+ * @dip: the directory
+ * @name: the name of the file
+ * @ip: the inode
+ *
+ * Assumes that the lock on (at least) @dip is held.
+ *
+ * Returns: 0 if the parent/child relationship is correct, errno if it isn't
+ */
+
+static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name,
+			  const struct gfs2_inode *ip)
+{
+	int error;
+
+	if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode))
+		return -EPERM;
+
+	if ((dip->i_inode.i_mode & S_ISVTX) &&
+	    dip->i_inode.i_uid != current_fsuid() &&
+	    ip->i_inode.i_uid != current_fsuid() && !capable(CAP_FOWNER))
+		return -EPERM;
+
+	if (IS_APPEND(&dip->i_inode))
+		return -EPERM;
+
+	error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC);
+	if (error)
+		return error;
+
+	error = gfs2_dir_check(&dip->i_inode, name, ip);
+	if (error)
+		return error;
+
+	return 0;
+}
+
 /**
  * gfs2_unlink - Unlink a file
  * @dir: The inode of the directory containing the file to unlink
@@ -472,6 +510,59 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	return 0;
 }
 
+/**
+ * gfs2_rmdiri - Remove a directory
+ * @dip: The parent directory of the directory to be removed
+ * @name: The name of the directory to be removed
+ * @ip: The GFS2 inode of the directory to be removed
+ *
+ * Assumes Glocks on dip and ip are held
+ *
+ * Returns: errno
+ */
+
+static int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name,
+		       struct gfs2_inode *ip)
+{
+	struct qstr dotname;
+	int error;
+
+	if (ip->i_entries != 2) {
+		if (gfs2_consist_inode(ip))
+			gfs2_dinode_print(ip);
+		return -EIO;
+	}
+
+	error = gfs2_dir_del(dip, name);
+	if (error)
+		return error;
+
+	error = gfs2_change_nlink(dip, -1);
+	if (error)
+		return error;
+
+	gfs2_str2qstr(&dotname, ".");
+	error = gfs2_dir_del(ip, &dotname);
+	if (error)
+		return error;
+
+	gfs2_str2qstr(&dotname, "..");
+	error = gfs2_dir_del(ip, &dotname);
+	if (error)
+		return error;
+
+	/* It looks odd, but it really should be done twice */
+	error = gfs2_change_nlink(ip, -1);
+	if (error)
+		return error;
+
+	error = gfs2_change_nlink(ip, -1);
+	if (error)
+		return error;
+
+	return error;
+}
+
 /**
  * gfs2_rmdir - Remove a directory
  * @dir: The parent directory of the directory to be removed
@@ -884,6 +975,61 @@ out:
 	return error;
 }
 
+/**
+ * gfs2_readlinki - return the contents of a symlink
+ * @ip: the symlink's inode
+ * @buf: a pointer to the buffer to be filled
+ * @len: a pointer to the length of @buf
+ *
+ * If @buf is too small, a piece of memory is kmalloc()ed and needs
+ * to be freed by the caller.
+ *
+ * Returns: errno
+ */
+
+static int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len)
+{
+	struct gfs2_holder i_gh;
+	struct buffer_head *dibh;
+	unsigned int x;
+	int error;
+
+	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh);
+	error = gfs2_glock_nq(&i_gh);
+	if (error) {
+		gfs2_holder_uninit(&i_gh);
+		return error;
+	}
+
+	if (!ip->i_disksize) {
+		gfs2_consist_inode(ip);
+		error = -EIO;
+		goto out;
+	}
+
+	error = gfs2_meta_inode_buffer(ip, &dibh);
+	if (error)
+		goto out;
+
+	x = ip->i_disksize + 1;
+	if (x > *len) {
+		*buf = kmalloc(x, GFP_NOFS);
+		if (!*buf) {
+			error = -ENOMEM;
+			goto out_brelse;
+		}
+	}
+
+	memcpy(*buf, dibh->b_data + sizeof(struct gfs2_dinode), x);
+	*len = x;
+
+out_brelse:
+	brelse(dibh);
+out:
+	gfs2_glock_dq_uninit(&i_gh);
+	return error;
+}
+
 /**
  * gfs2_readlink - Read the value of a symlink
  * @dentry: the symlink

fs/gfs2/quota.c

@@ -60,7 +60,6 @@
 #include "super.h"
 #include "trans.h"
 #include "inode.h"
-#include "ops_address.h"
 #include "util.h"
 
 #define QUOTA_USER 1

fs/gfs2/recovery.c

@@ -13,8 +13,7 @@
 #include <linux/buffer_head.h>
 #include <linux/gfs2_ondisk.h>
 #include <linux/crc32.h>
-#include <linux/kthread.h>
-#include <linux/freezer.h>
+#include <linux/slow-work.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -441,18 +440,25 @@ static void gfs2_recovery_done(struct gfs2_sbd *sdp, unsigned int jid,
         kobject_uevent_env(&sdp->sd_kobj, KOBJ_CHANGE, envp);
 }
 
-/**
- * gfs2_recover_journal - recover a given journal
- * @jd: the struct gfs2_jdesc describing the journal
- *
- * Acquire the journal's lock, check to see if the journal is clean, and
- * do recovery if necessary.
- *
- * Returns: errno
- */
+static int gfs2_recover_get_ref(struct slow_work *work)
+{
+	struct gfs2_jdesc *jd = container_of(work, struct gfs2_jdesc, jd_work);
+	if (test_and_set_bit(JDF_RECOVERY, &jd->jd_flags))
+		return -EBUSY;
+	return 0;
+}
 
-int gfs2_recover_journal(struct gfs2_jdesc *jd)
+static void gfs2_recover_put_ref(struct slow_work *work)
+{
+	struct gfs2_jdesc *jd = container_of(work, struct gfs2_jdesc, jd_work);
+	clear_bit(JDF_RECOVERY, &jd->jd_flags);
+	smp_mb__after_clear_bit();
+	wake_up_bit(&jd->jd_flags, JDF_RECOVERY);
+}
+
+static void gfs2_recover_work(struct slow_work *work)
 {
+	struct gfs2_jdesc *jd = container_of(work, struct gfs2_jdesc, jd_work);
 	struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
 	struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
 	struct gfs2_log_header_host head;
@@ -569,7 +575,7 @@ int gfs2_recover_journal(struct gfs2_jdesc *jd)
 		gfs2_glock_dq_uninit(&j_gh);
 
 	fs_info(sdp, "jid=%u: Done\n", jd->jd_jid);
-	return 0;
+	return;
 
 fail_gunlock_tr:
 	gfs2_glock_dq_uninit(&t_gh);
@@ -584,70 +590,28 @@ fail_gunlock_j:
 
 fail:
 	gfs2_recovery_done(sdp, jd->jd_jid, LM_RD_GAVEUP);
-	return error;
 }
 
-static struct gfs2_jdesc *gfs2_jdesc_find_dirty(struct gfs2_sbd *sdp)
-{
-	struct gfs2_jdesc *jd;
-	int found = 0;
-
-	spin_lock(&sdp->sd_jindex_spin);
+struct slow_work_ops gfs2_recover_ops = {
+	.get_ref = gfs2_recover_get_ref,
+	.put_ref = gfs2_recover_put_ref,
+	.execute = gfs2_recover_work,
+};
 
-	list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
-		if (jd->jd_dirty) {
-			jd->jd_dirty = 0;
-			found = 1;
-			break;
-		}
-	}
-	spin_unlock(&sdp->sd_jindex_spin);
-
-	if (!found)
-		jd = NULL;
 
-	return jd;
-}
-
-/**
- * gfs2_check_journals - Recover any dirty journals
- * @sdp: the filesystem
- *
- */
-
-static void gfs2_check_journals(struct gfs2_sbd *sdp)
+static int gfs2_recovery_wait(void *word)
 {
-	struct gfs2_jdesc *jd;
-
-	for (;;) {
-		jd = gfs2_jdesc_find_dirty(sdp);
-		if (!jd)
-			break;
-
-		if (jd != sdp->sd_jdesc)
-			gfs2_recover_journal(jd);
-	}
+	schedule();
+	return 0;
 }
 
-/**
- * gfs2_recoverd - Recover dead machine's journals
- * @sdp: Pointer to GFS2 superblock
- *
- */
-
-int gfs2_recoverd(void *data)
+int gfs2_recover_journal(struct gfs2_jdesc *jd)
 {
-	struct gfs2_sbd *sdp = data;
-	unsigned long t;
-
-	while (!kthread_should_stop()) {
-		gfs2_check_journals(sdp);
-		t = gfs2_tune_get(sdp,  gt_recoverd_secs) * HZ;
-		if (freezing(current))
-			refrigerator();
-		schedule_timeout_interruptible(t);
-	}
-
+	int rv;
+	rv = slow_work_enqueue(&jd->jd_work);
+	if (rv)
+		return rv;
+	wait_on_bit(&jd->jd_flags, JDF_RECOVERY, gfs2_recovery_wait, TASK_UNINTERRUPTIBLE);
 	return 0;
 }
 

fs/gfs2/recovery.h

@@ -28,7 +28,7 @@ extern void gfs2_revoke_clean(struct gfs2_sbd *sdp);
 extern int gfs2_find_jhead(struct gfs2_jdesc *jd,
 		    struct gfs2_log_header_host *head);
 extern int gfs2_recover_journal(struct gfs2_jdesc *gfs2_jd);
-extern int gfs2_recoverd(void *data);
+extern struct slow_work_ops gfs2_recover_ops;
 
 #endif /* __RECOVERY_DOT_H__ */
 

fs/gfs2/rgrp.c

@@ -29,7 +29,6 @@
 #include "util.h"
 #include "log.h"
 #include "inode.h"
-#include "ops_address.h"
 
 #define BFITNOENT ((u32)~0)
 #define NO_BLOCK ((u64)~0)
@@ -442,6 +441,7 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd)
 	for (x = 0; x < length; x++) {
 		bi = rgd->rd_bits + x;
 
+		bi->bi_flags = 0;
 		/* small rgrp; bitmap stored completely in header block */
 		if (length == 1) {
 			bytes = bytes_left;
@@ -580,7 +580,6 @@ static int read_rindex_entry(struct gfs2_inode *ip,
 
 	rgd->rd_gl->gl_object = rgd;
 	rgd->rd_flags &= ~GFS2_RDF_UPTODATE;
-	rgd->rd_flags |= GFS2_RDF_CHECK;
 	return error;
 }
 
@@ -701,10 +700,9 @@ static void gfs2_rgrp_in(struct gfs2_rgrpd *rgd, const void *buf)
 	u32 rg_flags;
 
 	rg_flags = be32_to_cpu(str->rg_flags);
-	if (rg_flags & GFS2_RGF_NOALLOC)
-		rgd->rd_flags |= GFS2_RDF_NOALLOC;
-	else
-		rgd->rd_flags &= ~GFS2_RDF_NOALLOC;
+	rg_flags &= ~GFS2_RDF_MASK;
+	rgd->rd_flags &= GFS2_RDF_MASK;
+	rgd->rd_flags |= rg_flags;
 	rgd->rd_free = be32_to_cpu(str->rg_free);
 	rgd->rd_dinodes = be32_to_cpu(str->rg_dinodes);
 	rgd->rd_igeneration = be64_to_cpu(str->rg_igeneration);
@@ -713,11 +711,8 @@ static void gfs2_rgrp_in(struct gfs2_rgrpd *rgd, const void *buf)
 static void gfs2_rgrp_out(struct gfs2_rgrpd *rgd, void *buf)
 {
 	struct gfs2_rgrp *str = buf;
-	u32 rg_flags = 0;
 
-	if (rgd->rd_flags & GFS2_RDF_NOALLOC)
-		rg_flags |= GFS2_RGF_NOALLOC;
-	str->rg_flags = cpu_to_be32(rg_flags);
+	str->rg_flags = cpu_to_be32(rgd->rd_flags & ~GFS2_RDF_MASK);
 	str->rg_free = cpu_to_be32(rgd->rd_free);
 	str->rg_dinodes = cpu_to_be32(rgd->rd_dinodes);
 	str->__pad = cpu_to_be32(0);
@@ -775,8 +770,10 @@ int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd)
 	}
 
 	if (!(rgd->rd_flags & GFS2_RDF_UPTODATE)) {
+		for (x = 0; x < length; x++)
+			clear_bit(GBF_FULL, &rgd->rd_bits[x].bi_flags);
 		gfs2_rgrp_in(rgd, (rgd->rd_bits[0].bi_bh)->b_data);
-		rgd->rd_flags |= GFS2_RDF_UPTODATE;
+		rgd->rd_flags |= (GFS2_RDF_UPTODATE | GFS2_RDF_CHECK);
 	}
 
 	spin_lock(&sdp->sd_rindex_spin);
@@ -903,6 +900,7 @@ void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd)
 			continue;
 		if (sdp->sd_args.ar_discard)
 			gfs2_rgrp_send_discards(sdp, rgd->rd_data0, bi);
+		clear_bit(GBF_FULL, &bi->bi_flags);
 		memcpy(bi->bi_clone + bi->bi_offset,
 		       bi->bi_bh->b_data + bi->bi_offset, bi->bi_len);
 	}
@@ -942,7 +940,7 @@ static int try_rgrp_fit(struct gfs2_rgrpd *rgd, struct gfs2_alloc *al)
 	struct gfs2_sbd *sdp = rgd->rd_sbd;
 	int ret = 0;
 
-	if (rgd->rd_flags & GFS2_RDF_NOALLOC)
+	if (rgd->rd_flags & (GFS2_RGF_NOALLOC | GFS2_RDF_ERROR))
 		return 0;
 
 	spin_lock(&sdp->sd_rindex_spin);
@@ -1315,30 +1313,37 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal,
 {
 	struct gfs2_bitmap *bi = NULL;
 	const u32 length = rgd->rd_length;
-	u32 blk = 0;
+	u32 blk = BFITNOENT;
 	unsigned int buf, x;
 	const unsigned int elen = *n;
-	const u8 *buffer;
+	const u8 *buffer = NULL;
 
 	*n = 0;
 	/* Find bitmap block that contains bits for goal block */
 	for (buf = 0; buf < length; buf++) {
 		bi = rgd->rd_bits + buf;
-		if (goal < (bi->bi_start + bi->bi_len) * GFS2_NBBY)
-			break;
+		/* Convert scope of "goal" from rgrp-wide to within found bit block */
+		if (goal < (bi->bi_start + bi->bi_len) * GFS2_NBBY) {
+			goal -= bi->bi_start * GFS2_NBBY;
+			goto do_search;
+		}
 	}
+	buf = 0;
+	goal = 0;
 
-	gfs2_assert(rgd->rd_sbd, buf < length);
-
-	/* Convert scope of "goal" from rgrp-wide to within found bit block */
-	goal -= bi->bi_start * GFS2_NBBY;
-
+do_search:
 	/* Search (up to entire) bitmap in this rgrp for allocatable block.
 	   "x <= length", instead of "x < length", because we typically start
 	   the search in the middle of a bit block, but if we can't find an
 	   allocatable block anywhere else, we want to be able wrap around and
 	   search in the first part of our first-searched bit block.  */
 	for (x = 0; x <= length; x++) {
+		bi = rgd->rd_bits + buf;
+
+		if (test_bit(GBF_FULL, &bi->bi_flags) &&
+		    (old_state == GFS2_BLKST_FREE))
+			goto skip;
+
 		/* The GFS2_BLKST_UNLINKED state doesn't apply to the clone
 		   bitmaps, so we must search the originals for that. */
 		buffer = bi->bi_bh->b_data + bi->bi_offset;
@@ -1349,33 +1354,39 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal,
 		if (blk != BFITNOENT)
 			break;
 
+		if ((goal == 0) && (old_state == GFS2_BLKST_FREE))
+			set_bit(GBF_FULL, &bi->bi_flags);
+
 		/* Try next bitmap block (wrap back to rgrp header if at end) */
-		buf = (buf + 1) % length;
-		bi = rgd->rd_bits + buf;
+skip:
+		buf++;
+		buf %= length;
 		goal = 0;
 	}
 
-	if (blk != BFITNOENT && old_state != new_state) {
-		*n = 1;
-		gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1);
+	if (blk == BFITNOENT)
+		return blk;
+	*n = 1;
+	if (old_state == new_state)
+		goto out;
+
+	gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1);
+	gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset,
+		    bi->bi_len, blk, new_state);
+	goal = blk;
+	while (*n < elen) {
+		goal++;
+		if (goal >= (bi->bi_len * GFS2_NBBY))
+			break;
+		if (gfs2_testbit(rgd, buffer, bi->bi_len, goal) !=
+		    GFS2_BLKST_FREE)
+			break;
 		gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset,
-			    bi->bi_len, blk, new_state);
-		goal = blk;
-		while (*n < elen) {
-			goal++;
-			if (goal >= (bi->bi_len * GFS2_NBBY))
-				break;
-			if (gfs2_testbit(rgd, buffer, bi->bi_len, goal) !=
-			    GFS2_BLKST_FREE)
-				break;
-			gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone,
-				    bi->bi_offset, bi->bi_len, goal,
-				    new_state);
-			(*n)++;
-		}
+			    bi->bi_len, goal, new_state);
+		(*n)++;
 	}
-
-	return (blk == BFITNOENT) ? blk : (bi->bi_start * GFS2_NBBY) + blk;
+out:
+	return (bi->bi_start * GFS2_NBBY) + blk;
 }
 
 /**
@@ -1435,13 +1446,33 @@ static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart,
 }
 
 /**
- * gfs2_alloc_block - Allocate a block
+ * gfs2_rgrp_dump - print out an rgrp
+ * @seq: The iterator
+ * @gl: The glock in question
+ *
+ */
+
+int gfs2_rgrp_dump(struct seq_file *seq, const struct gfs2_glock *gl)
+{
+	const struct gfs2_rgrpd *rgd = gl->gl_object;
+	if (rgd == NULL)
+		return 0;
+	gfs2_print_dbg(seq, " R: n:%llu f:%02x b:%u/%u i:%u\n",
+		       (unsigned long long)rgd->rd_addr, rgd->rd_flags,
+		       rgd->rd_free, rgd->rd_free_clone, rgd->rd_dinodes);
+	return 0;
+}
+
+/**
+ * gfs2_alloc_block - Allocate one or more blocks
  * @ip: the inode to allocate the block for
+ * @bn: Used to return the starting block number
+ * @n: requested number of blocks/extent length (value/result)
  *
- * Returns: the allocated block
+ * Returns: 0 or error
  */
 
-u64 gfs2_alloc_block(struct gfs2_inode *ip, unsigned int *n)
+int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	struct buffer_head *dibh;
@@ -1457,7 +1488,10 @@ u64 gfs2_alloc_block(struct gfs2_inode *ip, unsigned int *n)
 		goal = rgd->rd_last_alloc;
 
 	blk = rgblk_search(rgd, goal, GFS2_BLKST_FREE, GFS2_BLKST_USED, n);
-	BUG_ON(blk == BFITNOENT);
+
+	/* Since all blocks are reserved in advance, this shouldn't happen */
+	if (blk == BFITNOENT)
+		goto rgrp_error;
 
 	rgd->rd_last_alloc = blk;
 	block = rgd->rd_data0 + blk;
@@ -1469,7 +1503,9 @@ u64 gfs2_alloc_block(struct gfs2_inode *ip, unsigned int *n)
 		di->di_goal_meta = di->di_goal_data = cpu_to_be64(ip->i_goal);
 		brelse(dibh);
 	}
-	gfs2_assert_withdraw(sdp, rgd->rd_free >= *n);
+	if (rgd->rd_free < *n)
+		goto rgrp_error;
+
 	rgd->rd_free -= *n;
 
 	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
@@ -1484,7 +1520,16 @@ u64 gfs2_alloc_block(struct gfs2_inode *ip, unsigned int *n)
 	rgd->rd_free_clone -= *n;
 	spin_unlock(&sdp->sd_rindex_spin);
 
-	return block;
+	*bn = block;
+	return 0;
+
+rgrp_error:
+	fs_warn(sdp, "rgrp %llu has an error, marking it readonly until umount\n",
+	        (unsigned long long)rgd->rd_addr);
+	fs_warn(sdp, "umount on all nodes and run fsck.gfs2 to fix the error\n");
+	gfs2_rgrp_dump(NULL, rgd->rd_gl);
+	rgd->rd_flags |= GFS2_RDF_ERROR;
+	return -EIO;
 }
 
 /**

fs/gfs2/rgrp.h

@@ -14,22 +14,22 @@ struct gfs2_rgrpd;
 struct gfs2_sbd;
 struct gfs2_holder;
 
-void gfs2_rgrp_verify(struct gfs2_rgrpd *rgd);
+extern void gfs2_rgrp_verify(struct gfs2_rgrpd *rgd);
 
 struct gfs2_rgrpd *gfs2_blk2rgrpd(struct gfs2_sbd *sdp, u64 blk);
 struct gfs2_rgrpd *gfs2_rgrpd_get_first(struct gfs2_sbd *sdp);
 struct gfs2_rgrpd *gfs2_rgrpd_get_next(struct gfs2_rgrpd *rgd);
 
-void gfs2_clear_rgrpd(struct gfs2_sbd *sdp);
-int gfs2_rindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ri_gh);
+extern void gfs2_clear_rgrpd(struct gfs2_sbd *sdp);
+extern int gfs2_rindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ri_gh);
 
-int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd);
-void gfs2_rgrp_bh_hold(struct gfs2_rgrpd *rgd);
-void gfs2_rgrp_bh_put(struct gfs2_rgrpd *rgd);
+extern int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd);
+extern void gfs2_rgrp_bh_hold(struct gfs2_rgrpd *rgd);
+extern void gfs2_rgrp_bh_put(struct gfs2_rgrpd *rgd);
 
-void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd);
+extern void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd);
 
-struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip);
+extern struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip);
 static inline void gfs2_alloc_put(struct gfs2_inode *ip)
 {
 	BUG_ON(ip->i_alloc == NULL);
@@ -37,22 +37,22 @@ static inline void gfs2_alloc_put(struct gfs2_inode *ip)
 	ip->i_alloc = NULL;
 }
 
-int gfs2_inplace_reserve_i(struct gfs2_inode *ip,
-			 char *file, unsigned int line);
+extern int gfs2_inplace_reserve_i(struct gfs2_inode *ip, char *file,
+				  unsigned int line);
 #define gfs2_inplace_reserve(ip) \
 gfs2_inplace_reserve_i((ip), __FILE__, __LINE__)
 
-void gfs2_inplace_release(struct gfs2_inode *ip);
+extern void gfs2_inplace_release(struct gfs2_inode *ip);
 
-unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block);
+extern unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block);
 
-u64 gfs2_alloc_block(struct gfs2_inode *ip, unsigned int *n);
-u64 gfs2_alloc_di(struct gfs2_inode *ip, u64 *generation);
+extern int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n);
+extern u64 gfs2_alloc_di(struct gfs2_inode *ip, u64 *generation);
 
-void gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen);
-void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen);
-void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip);
-void gfs2_unlink_di(struct inode *inode);
+extern void gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen);
+extern void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen);
+extern void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip);
+extern void gfs2_unlink_di(struct inode *inode);
 
 struct gfs2_rgrp_list {
 	unsigned int rl_rgrps;
@@ -61,10 +61,11 @@ struct gfs2_rgrp_list {
 	struct gfs2_holder *rl_ghs;
 };
 
-void gfs2_rlist_add(struct gfs2_sbd *sdp, struct gfs2_rgrp_list *rlist,
-		    u64 block);
-void gfs2_rlist_alloc(struct gfs2_rgrp_list *rlist, unsigned int state);
-void gfs2_rlist_free(struct gfs2_rgrp_list *rlist);
-u64 gfs2_ri_total(struct gfs2_sbd *sdp);
+extern void gfs2_rlist_add(struct gfs2_sbd *sdp, struct gfs2_rgrp_list *rlist,
+			   u64 block);
+extern void gfs2_rlist_alloc(struct gfs2_rgrp_list *rlist, unsigned int state);
+extern void gfs2_rlist_free(struct gfs2_rgrp_list *rlist);
+extern u64 gfs2_ri_total(struct gfs2_sbd *sdp);
+extern int gfs2_rgrp_dump(struct seq_file *seq, const struct gfs2_glock *gl);
 
 #endif /* __RGRP_DOT_H__ */

fs/gfs2/trans.c

@@ -33,6 +33,9 @@ int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks,
 	BUG_ON(current->journal_info);
 	BUG_ON(blocks == 0 && revokes == 0);
 
+	if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))
+		return -EROFS;
+
 	tr = kzalloc(sizeof(struct gfs2_trans), GFP_NOFS);
 	if (!tr)
 		return -ENOMEM;
@@ -54,12 +57,6 @@ int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks,
 	if (error)
 		goto fail_holder_uninit;
 
-	if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) {
-		tr->tr_t_gh.gh_flags |= GL_NOCACHE;
-		error = -EROFS;
-		goto fail_gunlock;
-	}
-
 	error = gfs2_log_reserve(sdp, tr->tr_reserved);
 	if (error)
 		goto fail_gunlock;