Merge tag 'gfs2-for-5.12' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2

Pull gfs2 updates from Andreas Gruenbacher:

 - Log space and revoke accounting rework to fix some failed asserts.

 - Local resource group glock sharing for better local performance.

 - Add support for version 1802 filesystems: trusted xattr support and
   '-o rgrplvb' mounts by default.

 - Actually synchronize on the inode glock's FREEING bit during withdraw
   ("gfs2: fix glock confusion in function signal_our_withdraw").

 - Fix parallel recovery of multiple journals ("gfs2: keep bios separate
   for each journal").

 - Various other bug fixes.

* tag 'gfs2-for-5.12' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2: (49 commits)
  gfs2: Don't get stuck with I/O plugged in gfs2_ail1_flush
  gfs2: Per-revoke accounting in transactions
  gfs2: Rework the log space allocation logic
  gfs2: Minor calc_reserved cleanup
  gfs2: Use resource group glock sharing
  gfs2: Allow node-wide exclusive glock sharing
  gfs2: Add local resource group locking
  gfs2: Add per-reservation reserved block accounting
  gfs2: Rename rs_{free -> requested} and rd_{reserved -> requested}
  gfs2: Check for active reservation in gfs2_release
  gfs2: Don't search for unreserved space twice
  gfs2: Only pass reservation down to gfs2_rbm_find
  gfs2: Also reflect single-block allocations in rgd->rd_extfail_pt
  gfs2: Recursive gfs2_quota_hold in gfs2_iomap_end
  gfs2: Add trusted xattr support
  gfs2: Enable rgrplvb for sb_fs_format 1802
  gfs2: Don't skip dlm unlock if glock has an lvb
  gfs2: Lock imbalance on error path in gfs2_recover_one
  gfs2: Move function gfs2_ail_empty_tr
  gfs2: Get rid of current_tail()
  ...

fs/gfs2/bmap.c

@@ -1230,6 +1230,9 @@ static int gfs2_iomap_end(struct inode *inode, loff_t pos, loff_t length,
 
 	gfs2_inplace_release(ip);
 
+	if (ip->i_qadata && ip->i_qadata->qa_qd_num)
+		gfs2_quota_unlock(ip);
+
 	if (length != written && (iomap->flags & IOMAP_F_NEW)) {
 		/* Deallocate blocks that were just allocated. */
 		loff_t blockmask = i_blocksize(inode) - 1;
@@ -1242,9 +1245,6 @@ static int gfs2_iomap_end(struct inode *inode, loff_t pos, loff_t length,
 		}
 	}
 
-	if (ip->i_qadata && ip->i_qadata->qa_qd_num)
-		gfs2_quota_unlock(ip);
-
 	if (unlikely(!written))
 		goto out_unlock;
 
@@ -1538,13 +1538,13 @@ static int sweep_bh_for_rgrps(struct gfs2_inode *ip, struct gfs2_holder *rd_gh,
 				goto out;
 			}
 			ret = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE,
-						 0, rd_gh);
+						 LM_FLAG_NODE_SCOPE, rd_gh);
 			if (ret)
 				goto out;
 
 			/* Must be done with the rgrp glock held: */
 			if (gfs2_rs_active(&ip->i_res) &&
-			    rgd == ip->i_res.rs_rbm.rgd)
+			    rgd == ip->i_res.rs_rgd)
 				gfs2_rs_deltree(&ip->i_res);
 		}
 

fs/gfs2/file.c

@@ -716,10 +716,10 @@ static int gfs2_release(struct inode *inode, struct file *file)
 	kfree(file->private_data);
 	file->private_data = NULL;
 
-	if (file->f_mode & FMODE_WRITE) {
+	if (gfs2_rs_active(&ip->i_res))
 		gfs2_rs_delete(ip, &inode->i_writecount);
+	if (file->f_mode & FMODE_WRITE)
 		gfs2_qa_put(ip);
-	}
 	return 0;
 }
 
@@ -1112,8 +1112,8 @@ static long __gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t
 			goto out_qunlock;
 
 		/* check if the selected rgrp limits our max_blks further */
-		if (ap.allowed && ap.allowed < max_blks)
-			max_blks = ap.allowed;
+		if (ip->i_res.rs_reserved < max_blks)
+			max_blks = ip->i_res.rs_reserved;
 
 		/* Almost done. Calculate bytes that can be written using
 		 * max_blks. We also recompute max_bytes, data_blocks and

fs/gfs2/glock.c

@@ -313,9 +313,23 @@ void gfs2_glock_put(struct gfs2_glock *gl)
 static inline int may_grant(const struct gfs2_glock *gl, const struct gfs2_holder *gh)
 {
 	const struct gfs2_holder *gh_head = list_first_entry(&gl->gl_holders, const struct gfs2_holder, gh_list);
-	if ((gh->gh_state == LM_ST_EXCLUSIVE ||
-	     gh_head->gh_state == LM_ST_EXCLUSIVE) && gh != gh_head)
-		return 0;
+
+	if (gh != gh_head) {
+		/**
+		 * Here we make a special exception to grant holders who agree
+		 * to share the EX lock with other holders who also have the
+		 * bit set. If the original holder has the LM_FLAG_NODE_SCOPE bit
+		 * is set, we grant more holders with the bit set.
+		 */
+		if (gh_head->gh_state == LM_ST_EXCLUSIVE &&
+		    (gh_head->gh_flags & LM_FLAG_NODE_SCOPE) &&
+		    gh->gh_state == LM_ST_EXCLUSIVE &&
+		    (gh->gh_flags & LM_FLAG_NODE_SCOPE))
+			return 1;
+		if ((gh->gh_state == LM_ST_EXCLUSIVE ||
+		     gh_head->gh_state == LM_ST_EXCLUSIVE))
+			return 0;
+	}
 	if (gl->gl_state == gh->gh_state)
 		return 1;
 	if (gh->gh_flags & GL_EXACT)
@@ -2030,6 +2044,8 @@ static const char *hflags2str(char *buf, u16 flags, unsigned long iflags)
 		*p++ = 'A';
 	if (flags & LM_FLAG_PRIORITY)
 		*p++ = 'p';
+	if (flags & LM_FLAG_NODE_SCOPE)
+		*p++ = 'n';
 	if (flags & GL_ASYNC)
 		*p++ = 'a';
 	if (flags & GL_EXACT)

fs/gfs2/glock.h

@@ -75,6 +75,11 @@ enum {
  * request and directly join the other shared lock.  A shared lock request
  * without the priority flag might be forced to wait until the deferred
  * requested had acquired and released the lock.
+ *
+ * LM_FLAG_NODE_SCOPE
+ * This holder agrees to share the lock within this node. In other words,
+ * the glock is held in EX mode according to DLM, but local holders on the
+ * same node can share it.
  */
 
 #define LM_FLAG_TRY		0x0001
@@ -82,6 +87,7 @@ enum {
 #define LM_FLAG_NOEXP		0x0004
 #define LM_FLAG_ANY		0x0008
 #define LM_FLAG_PRIORITY	0x0010
+#define LM_FLAG_NODE_SCOPE	0x0020
 #define GL_ASYNC		0x0040
 #define GL_EXACT		0x0080
 #define GL_SKIP			0x0100

fs/gfs2/glops.c

@@ -86,16 +86,12 @@ static int gfs2_ail_empty_gl(struct gfs2_glock *gl)
 {
 	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
 	struct gfs2_trans tr;
+	unsigned int revokes;
 	int ret;
 
-	memset(&tr, 0, sizeof(tr));
-	INIT_LIST_HEAD(&tr.tr_buf);
-	INIT_LIST_HEAD(&tr.tr_databuf);
-	INIT_LIST_HEAD(&tr.tr_ail1_list);
-	INIT_LIST_HEAD(&tr.tr_ail2_list);
-	tr.tr_revokes = atomic_read(&gl->gl_ail_count);
+	revokes = atomic_read(&gl->gl_ail_count);
 
-	if (!tr.tr_revokes) {
+	if (!revokes) {
 		bool have_revokes;
 		bool log_in_flight;
 
@@ -122,20 +118,14 @@ static int gfs2_ail_empty_gl(struct gfs2_glock *gl)
 		return 0;
 	}
 
-	/* A shortened, inline version of gfs2_trans_begin()
-         * tr->alloced is not set since the transaction structure is
-         * on the stack */
-	tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes);
-	tr.tr_ip = _RET_IP_;
-	ret = gfs2_log_reserve(sdp, tr.tr_reserved);
-	if (ret < 0)
-		return ret;
-	WARN_ON_ONCE(current->journal_info);
-	current->journal_info = &tr;
-
-	__gfs2_ail_flush(gl, 0, tr.tr_revokes);
-
+	memset(&tr, 0, sizeof(tr));
+	set_bit(TR_ONSTACK, &tr.tr_flags);
+	ret = __gfs2_trans_begin(&tr, sdp, 0, revokes, _RET_IP_);
+	if (ret)
+		goto flush;
+	__gfs2_ail_flush(gl, 0, revokes);
 	gfs2_trans_end(sdp);
+
 flush:
 	gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL |
 		       GFS2_LFC_AIL_EMPTY_GL);
@@ -146,19 +136,15 @@ void gfs2_ail_flush(struct gfs2_glock *gl, bool fsync)
 {
 	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
 	unsigned int revokes = atomic_read(&gl->gl_ail_count);
-	unsigned int max_revokes = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / sizeof(u64);
 	int ret;
 
 	if (!revokes)
 		return;
 
-	while (revokes > max_revokes)
-		max_revokes += (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header)) / sizeof(u64);
-
-	ret = gfs2_trans_begin(sdp, 0, max_revokes);
+	ret = gfs2_trans_begin(sdp, 0, revokes);
 	if (ret)
 		return;
-	__gfs2_ail_flush(gl, fsync, max_revokes);
+	__gfs2_ail_flush(gl, fsync, revokes);
 	gfs2_trans_end(sdp);
 	gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL |
 		       GFS2_LFC_AIL_FLUSH);

fs/gfs2/incore.h

@@ -20,6 +20,7 @@
 #include <linux/percpu.h>
 #include <linux/lockref.h>
 #include <linux/rhashtable.h>
+#include <linux/mutex.h>
 
 #define DIO_WAIT	0x00000010
 #define DIO_METADATA	0x00000020
@@ -106,7 +107,8 @@ struct gfs2_rgrpd {
 	u32 rd_data;			/* num of data blocks in rgrp */
 	u32 rd_bitbytes;		/* number of bytes in data bitmaps */
 	u32 rd_free;
-	u32 rd_reserved;                /* number of blocks reserved */
+	u32 rd_requested;		/* number of blocks in rd_rstree */
+	u32 rd_reserved;		/* number of reserved blocks */
 	u32 rd_free_clone;
 	u32 rd_dinodes;
 	u64 rd_igeneration;
@@ -122,34 +124,10 @@ struct gfs2_rgrpd {
 #define GFS2_RDF_PREFERRED	0x80000000 /* This rgrp is preferred */
 #define GFS2_RDF_MASK		0xf0000000 /* mask for internal flags */
 	spinlock_t rd_rsspin;           /* protects reservation related vars */
+	struct mutex rd_mutex;
 	struct rb_root rd_rstree;       /* multi-block reservation tree */
 };
 
-struct gfs2_rbm {
-	struct gfs2_rgrpd *rgd;
-	u32 offset;		/* The offset is bitmap relative */
-	int bii;		/* Bitmap index */
-};
-
-static inline struct gfs2_bitmap *rbm_bi(const struct gfs2_rbm *rbm)
-{
-	return rbm->rgd->rd_bits + rbm->bii;
-}
-
-static inline u64 gfs2_rbm_to_block(const struct gfs2_rbm *rbm)
-{
-	BUG_ON(rbm->offset >= rbm->rgd->rd_data);
-	return rbm->rgd->rd_data0 + (rbm_bi(rbm)->bi_start * GFS2_NBBY) +
-		rbm->offset;
-}
-
-static inline bool gfs2_rbm_eq(const struct gfs2_rbm *rbm1,
-			       const struct gfs2_rbm *rbm2)
-{
-	return (rbm1->rgd == rbm2->rgd) && (rbm1->bii == rbm2->bii) &&
-	       (rbm1->offset == rbm2->offset);
-}
-
 enum gfs2_state_bits {
 	BH_Pinned = BH_PrivateStart,
 	BH_Escaped = BH_PrivateStart + 1,
@@ -313,9 +291,11 @@ struct gfs2_qadata { /* quota allocation data */
 */
 
 struct gfs2_blkreserv {
-	struct rb_node rs_node;       /* link to other block reservations */
-	struct gfs2_rbm rs_rbm;       /* Start of reservation */
-	u32 rs_free;                  /* how many blocks are still free */
+	struct rb_node rs_node;       /* node within rd_rstree */
+	struct gfs2_rgrpd *rs_rgd;
+	u64 rs_start;
+	u32 rs_requested;
+	u32 rs_reserved;              /* number of reserved blocks */
 };
 
 /*
@@ -490,7 +470,7 @@ struct gfs2_quota_data {
 enum {
 	TR_TOUCHED = 1,
 	TR_ATTACHED = 2,
-	TR_ALLOCED = 3,
+	TR_ONSTACK = 3,
 };
 
 struct gfs2_trans {
@@ -506,7 +486,6 @@ struct gfs2_trans {
 	unsigned int tr_num_buf_rm;
 	unsigned int tr_num_databuf_rm;
 	unsigned int tr_num_revoke;
-	unsigned int tr_num_revoke_rm;
 
 	struct list_head tr_list;
 	struct list_head tr_databuf;
@@ -531,6 +510,7 @@ struct gfs2_jdesc {
 	unsigned int nr_extents;
 	struct work_struct jd_work;
 	struct inode *jd_inode;
+	struct bio *jd_log_bio;
 	unsigned long jd_flags;
 #define JDF_RECOVERY 1
 	unsigned int jd_jid;
@@ -585,6 +565,7 @@ struct gfs2_args {
 	unsigned int ar_errors:2;               /* errors=withdraw | panic */
 	unsigned int ar_nobarrier:1;            /* do not send barriers */
 	unsigned int ar_rgrplvb:1;		/* use lvbs for rgrp info */
+	unsigned int ar_got_rgrplvb:1;		/* Was the rgrplvb opt given? */
 	unsigned int ar_loccookie:1;		/* use location based readdir
 						   cookies */
 	s32 ar_commit;				/* Commit interval */
@@ -821,7 +802,6 @@ struct gfs2_sbd {
 
 	struct gfs2_trans *sd_log_tr;
 	unsigned int sd_log_blks_reserved;
-	int sd_log_committed_revoke;
 
 	atomic_t sd_log_pinned;
 	unsigned int sd_log_num_revoke;
@@ -834,24 +814,22 @@ struct gfs2_sbd {
 	atomic_t sd_log_thresh2;
 	atomic_t sd_log_blks_free;
 	atomic_t sd_log_blks_needed;
+	atomic_t sd_log_revokes_available;
 	wait_queue_head_t sd_log_waitq;
 	wait_queue_head_t sd_logd_waitq;
 
 	u64 sd_log_sequence;
-	unsigned int sd_log_head;
-	unsigned int sd_log_tail;
 	int sd_log_idle;
 
 	struct rw_semaphore sd_log_flush_lock;
 	atomic_t sd_log_in_flight;
-	struct bio *sd_log_bio;
 	wait_queue_head_t sd_log_flush_wait;
 	int sd_log_error; /* First log error */
 	wait_queue_head_t sd_withdraw_wait;
 
-	atomic_t sd_reserving_log;
-	wait_queue_head_t sd_reserving_log_wait;
-
+	unsigned int sd_log_tail;
+	unsigned int sd_log_flush_tail;
+	unsigned int sd_log_head;
 	unsigned int sd_log_flush_head;
 
 	spinlock_t sd_ail_lock;

fs/gfs2/inode.c

@@ -1147,7 +1147,7 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry)
 	if (!rgd)
 		goto out_inodes;
 
-	gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2);
+	gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, LM_FLAG_NODE_SCOPE, ghs + 2);
 
 
 	error = gfs2_glock_nq(ghs); /* parent */
@@ -1453,8 +1453,8 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
 			error = -ENOENT;
 			goto out_gunlock;
 		}
-		error = gfs2_glock_nq_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0,
-					   &rd_gh);
+		error = gfs2_glock_nq_init(nrgd->rd_gl, LM_ST_EXCLUSIVE,
+					   LM_FLAG_NODE_SCOPE, &rd_gh);
 		if (error)
 			goto out_gunlock;
 	}

fs/gfs2/lock_dlm.c

@@ -284,7 +284,6 @@ static void gdlm_put_lock(struct gfs2_glock *gl)
 {
 	struct gfs2_sbd *sdp = gl->gl_name.ln_sbd;
 	struct lm_lockstruct *ls = &sdp->sd_lockstruct;
-	int lvb_needs_unlock = 0;
 	int error;
 
 	if (gl->gl_lksb.sb_lkid == 0) {
@@ -297,13 +296,10 @@ static void gdlm_put_lock(struct gfs2_glock *gl)
 	gfs2_sbstats_inc(gl, GFS2_LKS_DCOUNT);
 	gfs2_update_request_times(gl);
 
-	/* don't want to skip dlm_unlock writing the lvb when lock is ex */
-
-	if (gl->gl_lksb.sb_lvbptr && (gl->gl_state == LM_ST_EXCLUSIVE))
-		lvb_needs_unlock = 1;
+	/* don't want to skip dlm_unlock writing the lvb when lock has one */
 
 	if (test_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags) &&
-	    !lvb_needs_unlock) {
+	    !gl->gl_lksb.sb_lvbptr) {
 		gfs2_glock_free(gl);
 		return;
 	}

fs/gfs2/log.h

@@ -13,6 +13,13 @@
 #include "incore.h"
 #include "inode.h"
 
+/*
+ * The minimum amount of log space required for a log flush is one block for
+ * revokes and one block for the log header.  Log flushes other than
+ * GFS2_LOG_HEAD_FLUSH_NORMAL may write one or two more log headers.
+ */
+#define GFS2_LOG_FLUSH_MIN_BLOCKS 4
+
 /**
  * gfs2_log_lock - acquire the right to mess with the log manager
  * @sdp: the filesystem
@@ -43,7 +50,9 @@ static inline void gfs2_log_pointers_init(struct gfs2_sbd *sdp,
 	if (++value == sdp->sd_jdesc->jd_blocks) {
 		value = 0;
 	}
-	sdp->sd_log_head = sdp->sd_log_tail = value;
+	sdp->sd_log_tail = value;
+	sdp->sd_log_flush_tail = value;
+	sdp->sd_log_head = value;
 }
 
 static inline void gfs2_ordered_add_inode(struct gfs2_inode *ip)
@@ -64,8 +73,13 @@ static inline void gfs2_ordered_add_inode(struct gfs2_inode *ip)
 extern void gfs2_ordered_del_inode(struct gfs2_inode *ip);
 extern unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct);
 extern void gfs2_remove_from_ail(struct gfs2_bufdata *bd);
+extern bool gfs2_log_is_empty(struct gfs2_sbd *sdp);
+extern void gfs2_log_release_revokes(struct gfs2_sbd *sdp, unsigned int revokes);
 extern void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks);
-extern int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks);
+extern bool gfs2_log_try_reserve(struct gfs2_sbd *sdp, struct gfs2_trans *tr,
+				 unsigned int *extra_revokes);
+extern void gfs2_log_reserve(struct gfs2_sbd *sdp, struct gfs2_trans *tr,
+			     unsigned int *extra_revokes);
 extern void gfs2_write_log_header(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
 				  u64 seq, u32 tail, u32 lblock, u32 flags,
 				  int op_flags);
@@ -78,6 +92,6 @@ extern void log_flush_wait(struct gfs2_sbd *sdp);
 extern int gfs2_logd(void *data);
 extern void gfs2_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd);
 extern void gfs2_glock_remove_revoke(struct gfs2_glock *gl);
-extern void gfs2_write_revokes(struct gfs2_sbd *sdp);
+extern void gfs2_flush_revokes(struct gfs2_sbd *sdp);
 
 #endif /* __LOG_DOT_H__ */

fs/gfs2/lops.c

@@ -76,15 +76,20 @@ static void maybe_release_space(struct gfs2_bufdata *bd)
 	unsigned int index = bd->bd_bh->b_blocknr - gl->gl_name.ln_number;
 	struct gfs2_bitmap *bi = rgd->rd_bits + index;
 
+	rgrp_lock_local(rgd);
 	if (bi->bi_clone == NULL)
-		return;
+		goto out;
 	if (sdp->sd_args.ar_discard)
 		gfs2_rgrp_send_discards(sdp, rgd->rd_data0, bd->bd_bh, bi, 1, NULL);
 	memcpy(bi->bi_clone + bi->bi_offset,
 	       bd->bd_bh->b_data + bi->bi_offset, bi->bi_bytes);
 	clear_bit(GBF_FULL, &bi->bi_flags);
 	rgd->rd_free_clone = rgd->rd_free;
+	BUG_ON(rgd->rd_free_clone < rgd->rd_reserved);
 	rgd->rd_extfail_pt = rgd->rd_free;
+
+out:
+	rgrp_unlock_local(rgd);
 }
 
 /**
@@ -322,17 +327,18 @@ static struct bio *gfs2_log_get_bio(struct gfs2_sbd *sdp, u64 blkno,
  * then add the page segment to that.
  */
 
-void gfs2_log_write(struct gfs2_sbd *sdp, struct page *page,
-		    unsigned size, unsigned offset, u64 blkno)
+void gfs2_log_write(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
+		    struct page *page, unsigned size, unsigned offset,
+		    u64 blkno)
 {
 	struct bio *bio;
 	int ret;
 
-	bio = gfs2_log_get_bio(sdp, blkno, &sdp->sd_log_bio, REQ_OP_WRITE,
+	bio = gfs2_log_get_bio(sdp, blkno, &jd->jd_log_bio, REQ_OP_WRITE,
 			       gfs2_end_log_write, false);
 	ret = bio_add_page(bio, page, size, offset);
 	if (ret == 0) {
-		bio = gfs2_log_get_bio(sdp, blkno, &sdp->sd_log_bio,
+		bio = gfs2_log_get_bio(sdp, blkno, &jd->jd_log_bio,
 				       REQ_OP_WRITE, gfs2_end_log_write, true);
 		ret = bio_add_page(bio, page, size, offset);
 		WARN_ON(ret == 0);
@@ -355,7 +361,8 @@ static void gfs2_log_write_bh(struct gfs2_sbd *sdp, struct buffer_head *bh)
 
 	dblock = gfs2_log_bmap(sdp->sd_jdesc, sdp->sd_log_flush_head);
 	gfs2_log_incr_head(sdp);
-	gfs2_log_write(sdp, bh->b_page, bh->b_size, bh_offset(bh), dblock);
+	gfs2_log_write(sdp, sdp->sd_jdesc, bh->b_page, bh->b_size,
+		       bh_offset(bh), dblock);
 }
 
 /**
@@ -369,14 +376,14 @@ static void gfs2_log_write_bh(struct gfs2_sbd *sdp, struct buffer_head *bh)
  * the page may be freed at any time.
  */
 
-void gfs2_log_write_page(struct gfs2_sbd *sdp, struct page *page)
+static void gfs2_log_write_page(struct gfs2_sbd *sdp, struct page *page)
 {
 	struct super_block *sb = sdp->sd_vfs;
 	u64 dblock;
 
 	dblock = gfs2_log_bmap(sdp->sd_jdesc, sdp->sd_log_flush_head);
 	gfs2_log_incr_head(sdp);
-	gfs2_log_write(sdp, page, sb->s_blocksize, 0, dblock);
+	gfs2_log_write(sdp, sdp->sd_jdesc, page, sb->s_blocksize, 0, dblock);
 }
 
 /**
@@ -845,7 +852,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 	struct page *page;
 	unsigned int length;
 
-	gfs2_write_revokes(sdp);
+	gfs2_flush_revokes(sdp);
 	if (!sdp->sd_log_num_revoke)
 		return;
 
@@ -857,7 +864,6 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 		sdp->sd_log_num_revoke--;
 
 		if (offset + sizeof(u64) > sdp->sd_sb.sb_bsize) {
-
 			gfs2_log_write_page(sdp, page);
 			page = mempool_alloc(gfs2_page_pool, GFP_NOIO);
 			mh = page_address(page);

fs/gfs2/lops.h

@@ -10,37 +10,24 @@
 #include <linux/list.h>
 #include "incore.h"
 
-#define BUF_OFFSET \
-	((sizeof(struct gfs2_log_descriptor) + sizeof(__be64) - 1) & \
-	 ~(sizeof(__be64) - 1))
-#define DATABUF_OFFSET \
-	((sizeof(struct gfs2_log_descriptor) + (2 * sizeof(__be64) - 1)) & \
-	 ~(2 * sizeof(__be64) - 1))
-
 extern const struct gfs2_log_operations *gfs2_log_ops[];
 extern void gfs2_log_incr_head(struct gfs2_sbd *sdp);
 extern u64 gfs2_log_bmap(struct gfs2_jdesc *jd, unsigned int lbn);
-extern void gfs2_log_write(struct gfs2_sbd *sdp, struct page *page,
-			   unsigned size, unsigned offset, u64 blkno);
-extern void gfs2_log_write_page(struct gfs2_sbd *sdp, struct page *page);
+extern void gfs2_log_write(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
+			   struct page *page, unsigned size, unsigned offset,
+			   u64 blkno);
 extern void gfs2_log_submit_bio(struct bio **biop, int opf);
 extern void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh);
 extern int gfs2_find_jhead(struct gfs2_jdesc *jd,
 			   struct gfs2_log_header_host *head, bool keep_cache);
 static inline unsigned int buf_limit(struct gfs2_sbd *sdp)
 {
-	unsigned int limit;
-
-	limit = (sdp->sd_sb.sb_bsize - BUF_OFFSET) / sizeof(__be64);
-	return limit;
+	return sdp->sd_ldptrs;
 }
 
 static inline unsigned int databuf_limit(struct gfs2_sbd *sdp)
 {
-	unsigned int limit;
-
-	limit = (sdp->sd_sb.sb_bsize - DATABUF_OFFSET) / (2 * sizeof(__be64));
-	return limit;
+	return sdp->sd_ldptrs / 2;
 }
 
 static inline void lops_before_commit(struct gfs2_sbd *sdp,

fs/gfs2/main.c

@@ -98,7 +98,7 @@ static int __init init_gfs2_fs(void)
 	error = -ENOMEM;
 	gfs2_glock_cachep = kmem_cache_create("gfs2_glock",
 					      sizeof(struct gfs2_glock),
-					      0, 0,
+					      0, SLAB_RECLAIM_ACCOUNT,
 					      gfs2_init_glock_once);
 	if (!gfs2_glock_cachep)
 		goto fail_cachep1;
@@ -134,7 +134,7 @@ static int __init init_gfs2_fs(void)
 
 	gfs2_quotad_cachep = kmem_cache_create("gfs2_quotad",
 					       sizeof(struct gfs2_quota_data),
-					       0, 0, NULL);
+					       0, SLAB_RECLAIM_ACCOUNT, NULL);
 	if (!gfs2_quotad_cachep)
 		goto fail_cachep6;
 

fs/gfs2/ops_fstype.c

@@ -136,8 +136,6 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb)
 
 	init_rwsem(&sdp->sd_log_flush_lock);
 	atomic_set(&sdp->sd_log_in_flight, 0);
-	atomic_set(&sdp->sd_reserving_log, 0);
-	init_waitqueue_head(&sdp->sd_reserving_log_wait);
 	init_waitqueue_head(&sdp->sd_log_flush_wait);
 	atomic_set(&sdp->sd_freeze_state, SFS_UNFROZEN);
 	mutex_init(&sdp->sd_freeze_mutex);
@@ -171,7 +169,8 @@ static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent)
 		return -EINVAL;
 	}
 
-	if (sb->sb_fs_format != GFS2_FORMAT_FS ||
+	if (sb->sb_fs_format < GFS2_FS_FORMAT_MIN ||
+	    sb->sb_fs_format > GFS2_FS_FORMAT_MAX ||
 	    sb->sb_multihost_format != GFS2_FORMAT_MULTI) {
 		fs_warn(sdp, "Unknown on-disk format, unable to mount\n");
 		return -EINVAL;
@@ -179,7 +178,7 @@ static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent)
 
 	if (sb->sb_bsize < 512 || sb->sb_bsize > PAGE_SIZE ||
 	    (sb->sb_bsize & (sb->sb_bsize - 1))) {
-		pr_warn("Invalid superblock size\n");
+		pr_warn("Invalid block size\n");
 		return -EINVAL;
 	}
 
@@ -317,6 +316,13 @@ static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent)
 				     sizeof(struct gfs2_meta_header))
 		* GFS2_NBBY; /* not the rgrp bitmap, subsequent bitmaps only */
 
+	/*
+	 * We always keep at least one block reserved for revokes in
+	 * transactions.  This greatly simplifies allocating additional
+	 * revoke blocks.
+	 */
+	atomic_set(&sdp->sd_log_revokes_available, sdp->sd_ldptrs);
+
 	/* Compute maximum reservation required to add a entry to a directory */
 
 	hash_blocks = DIV_ROUND_UP(sizeof(u64) * BIT(GFS2_DIR_MAX_DEPTH),
@@ -488,6 +494,19 @@ static int init_sb(struct gfs2_sbd *sdp, int silent)
 		goto out;
 	}
 
+	switch(sdp->sd_sb.sb_fs_format) {
+	case GFS2_FS_FORMAT_MAX:
+		sb->s_xattr = gfs2_xattr_handlers_max;
+		break;
+
+	case GFS2_FS_FORMAT_MIN:
+		sb->s_xattr = gfs2_xattr_handlers_min;
+		break;
+
+	default:
+		BUG();
+	}
+
 	/* Set up the buffer cache and SB for real */
 	if (sdp->sd_sb.sb_bsize < bdev_logical_block_size(sb->s_bdev)) {
 		ret = -EINVAL;
@@ -1032,13 +1051,14 @@ static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent)
 	}
 
 	if (lm->lm_mount == NULL) {
-		fs_info(sdp, "Now mounting FS...\n");
+		fs_info(sdp, "Now mounting FS (format %u)...\n", sdp->sd_sb.sb_fs_format);
 		complete_all(&sdp->sd_locking_init);
 		return 0;
 	}
 	ret = lm->lm_mount(sdp, table);
 	if (ret == 0)
-		fs_info(sdp, "Joined cluster. Now mounting FS...\n");
+		fs_info(sdp, "Joined cluster. Now mounting FS (format %u)...\n",
+		        sdp->sd_sb.sb_fs_format);
 	complete_all(&sdp->sd_locking_init);
 	return ret;
 }
@@ -1084,6 +1104,7 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc)
 	int silent = fc->sb_flags & SB_SILENT;
 	struct gfs2_sbd *sdp;
 	struct gfs2_holder mount_gh;
+	struct gfs2_holder freeze_gh;
 	int error;
 
 	sdp = init_sbd(sb);
@@ -1107,7 +1128,6 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc)
 	sb->s_op = &gfs2_super_ops;
 	sb->s_d_op = &gfs2_dops;
 	sb->s_export_op = &gfs2_export_ops;
-	sb->s_xattr = gfs2_xattr_handlers;
 	sb->s_qcop = &gfs2_quotactl_ops;
 	sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP;
 	sb_dqopt(sb)->flags |= DQUOT_QUOTA_SYS_FILE;
@@ -1156,6 +1176,10 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc)
 	if (error)
 		goto fail_locking;
 
+	/* Turn rgrplvb on by default if fs format is recent enough */
+	if (!sdp->sd_args.ar_got_rgrplvb && sdp->sd_sb.sb_fs_format > 1801)
+		sdp->sd_args.ar_rgrplvb = 1;
+
 	error = wait_on_journal(sdp);
 	if (error)
 		goto fail_sb;
@@ -1195,25 +1219,18 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc)
 		goto fail_per_node;
 	}
 
-	if (sb_rdonly(sb)) {
-		struct gfs2_holder freeze_gh;
+	error = gfs2_freeze_lock(sdp, &freeze_gh, 0);
+	if (error)
+		goto fail_per_node;
 
-		error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED,
-					   LM_FLAG_NOEXP | GL_EXACT,
-					   &freeze_gh);
-		if (error) {
-			fs_err(sdp, "can't make FS RO: %d\n", error);
-			goto fail_per_node;
-		}
-		gfs2_glock_dq_uninit(&freeze_gh);
-	} else {
+	if (!sb_rdonly(sb))
 		error = gfs2_make_fs_rw(sdp);
-		if (error) {
-			fs_err(sdp, "can't make FS RW: %d\n", error);
-			goto fail_per_node;
-		}
-	}
 
+	gfs2_freeze_unlock(&freeze_gh);
+	if (error) {
+		fs_err(sdp, "can't make FS RW: %d\n", error);
+		goto fail_per_node;
+	}
 	gfs2_glock_dq_uninit(&mount_gh);
 	gfs2_online_uevent(sdp);
 	return 0;
@@ -1456,6 +1473,7 @@ static int gfs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
 		break;
 	case Opt_rgrplvb:
 		args->ar_rgrplvb = result.boolean;
+		args->ar_got_rgrplvb = 1;
 		break;
 	case Opt_loccookie:
 		args->ar_loccookie = result.boolean;
@@ -1514,6 +1532,12 @@ static int gfs2_reconfigure(struct fs_context *fc)
 		fc->sb_flags |= SB_RDONLY;
 
 	if ((sb->s_flags ^ fc->sb_flags) & SB_RDONLY) {
+		struct gfs2_holder freeze_gh;
+
+		error = gfs2_freeze_lock(sdp, &freeze_gh, 0);
+		if (error)
+			return -EINVAL;
+
 		if (fc->sb_flags & SB_RDONLY) {
 			error = gfs2_make_fs_ro(sdp);
 			if (error)
@@ -1523,6 +1547,7 @@ static int gfs2_reconfigure(struct fs_context *fc)
 			if (error)
 				errorfc(fc, "unable to remount read-write");
 		}
+		gfs2_freeze_unlock(&freeze_gh);
 	}
 	sdp->sd_args = *newargs;
 

fs/gfs2/recovery.c

@@ -470,9 +470,7 @@ void gfs2_recover_func(struct work_struct *work)
 
 		/* Acquire a shared hold on the freeze lock */
 
-		error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED,
-					   LM_FLAG_NOEXP | LM_FLAG_PRIORITY |
-					   GL_EXACT, &thaw_gh);
+		error = gfs2_freeze_lock(sdp, &thaw_gh, LM_FLAG_PRIORITY);
 		if (error)
 			goto fail_gunlock_ji;
 
@@ -507,22 +505,24 @@ void gfs2_recover_func(struct work_struct *work)
 
 		/* We take the sd_log_flush_lock here primarily to prevent log
 		 * flushes and simultaneous journal replays from stomping on
-		 * each other wrt sd_log_bio. */
+		 * each other wrt jd_log_bio. */
 		down_read(&sdp->sd_log_flush_lock);
 		for (pass = 0; pass < 2; pass++) {
 			lops_before_scan(jd, &head, pass);
 			error = foreach_descriptor(jd, head.lh_tail,
 						   head.lh_blkno, pass);
 			lops_after_scan(jd, error, pass);
-			if (error)
+			if (error) {
+				up_read(&sdp->sd_log_flush_lock);
 				goto fail_gunlock_thaw;
+			}
 		}
 
 		recover_local_statfs(jd, &head);
 		clean_journal(jd, &head);
 		up_read(&sdp->sd_log_flush_lock);
 
-		gfs2_glock_dq_uninit(&thaw_gh);
+		gfs2_freeze_unlock(&thaw_gh);
 		t_rep = ktime_get();
 		fs_info(sdp, "jid=%u: Journal replayed in %lldms [jlck:%lldms, "
 			"jhead:%lldms, tlck:%lldms, replay:%lldms]\n",
@@ -544,7 +544,7 @@ void gfs2_recover_func(struct work_struct *work)
 	goto done;
 
 fail_gunlock_thaw:
-	gfs2_glock_dq_uninit(&thaw_gh);
+	gfs2_freeze_unlock(&thaw_gh);
 fail_gunlock_ji:
 	if (jlocked) {
 		gfs2_glock_dq_uninit(&ji_gh);

fs/gfs2/rgrp.h

@@ -77,7 +77,7 @@ extern int gfs2_fitrim(struct file *filp, void __user *argp);
 /* This is how to tell if a reservation is in the rgrp tree: */
 static inline bool gfs2_rs_active(const struct gfs2_blkreserv *rs)
 {
-	return rs && !RB_EMPTY_NODE(&rs->rs_node);
+	return !RB_EMPTY_NODE(&rs->rs_node);
 }
 
 static inline int rgrp_contains_block(struct gfs2_rgrpd *rgd, u64 block)
@@ -88,4 +88,8 @@ static inline int rgrp_contains_block(struct gfs2_rgrpd *rgd, u64 block)
 }
 
 extern void check_and_update_goal(struct gfs2_inode *ip);
+
+extern void rgrp_lock_local(struct gfs2_rgrpd *rgd);
+extern void rgrp_unlock_local(struct gfs2_rgrpd *rgd);
+
 #endif /* __RGRP_DOT_H__ */

fs/gfs2/super.c

@@ -81,19 +81,12 @@ void gfs2_jindex_free(struct gfs2_sbd *sdp)
 static struct gfs2_jdesc *jdesc_find_i(struct list_head *head, unsigned int jid)
 {
 	struct gfs2_jdesc *jd;
-	int found = 0;
 
 	list_for_each_entry(jd, head, jd_list) {
-		if (jd->jd_jid == jid) {
-			found = 1;
-			break;
-		}
+		if (jd->jd_jid == jid)
+			return jd;
 	}
-
-	if (!found)
-		jd = NULL;
-
-	return jd;
+	return NULL;
 }
 
 struct gfs2_jdesc *gfs2_jdesc_find(struct gfs2_sbd *sdp, unsigned int jid)
@@ -165,7 +158,6 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
 {
 	struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode);
 	struct gfs2_glock *j_gl = ip->i_gl;
-	struct gfs2_holder freeze_gh;
 	struct gfs2_log_header_host head;
 	int error;
 
@@ -173,12 +165,6 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
 	if (error)
 		return error;
 
-	error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED,
-				   LM_FLAG_NOEXP | GL_EXACT,
-				   &freeze_gh);
-	if (error)
-		goto fail_threads;
-
 	j_gl->gl_ops->go_inval(j_gl, DIO_METADATA);
 	if (gfs2_withdrawn(sdp)) {
 		error = -EIO;
@@ -205,13 +191,9 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
 
 	set_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
 
-	gfs2_glock_dq_uninit(&freeze_gh);
-
 	return 0;
 
 fail:
-	gfs2_glock_dq_uninit(&freeze_gh);
-fail_threads:
 	if (sdp->sd_quotad_process)
 		kthread_stop(sdp->sd_quotad_process);
 	sdp->sd_quotad_process = NULL;
@@ -452,7 +434,7 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp)
 	}
 
 	if (error)
-		gfs2_glock_dq_uninit(&sdp->sd_freeze_gh);
+		gfs2_freeze_unlock(&sdp->sd_freeze_gh);
 
 out:
 	while (!list_empty(&list)) {
@@ -607,30 +589,9 @@ static void gfs2_dirty_inode(struct inode *inode, int flags)
 
 int gfs2_make_fs_ro(struct gfs2_sbd *sdp)
 {
-	struct gfs2_holder freeze_gh;
 	int error = 0;
 	int log_write_allowed = test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
 
-	gfs2_holder_mark_uninitialized(&freeze_gh);
-	if (sdp->sd_freeze_gl &&
-	    !gfs2_glock_is_locked_by_me(sdp->sd_freeze_gl)) {
-		if (!log_write_allowed) {
-			error = gfs2_glock_nq_init(sdp->sd_freeze_gl,
-						   LM_ST_SHARED, LM_FLAG_TRY |
-						   LM_FLAG_NOEXP | GL_EXACT,
-						   &freeze_gh);
-			if (error == GLR_TRYFAILED)
-				error = 0;
-		} else {
-			error = gfs2_glock_nq_init(sdp->sd_freeze_gl,
-						   LM_ST_SHARED,
-						   LM_FLAG_NOEXP | GL_EXACT,
-						   &freeze_gh);
-			if (error && !gfs2_withdrawn(sdp))
-				return error;
-		}
-	}
-
 	gfs2_flush_delete_work(sdp);
 	if (!log_write_allowed && current == sdp->sd_quotad_process)
 		fs_warn(sdp, "The quotad daemon is withdrawing.\n");
@@ -650,18 +611,15 @@ int gfs2_make_fs_ro(struct gfs2_sbd *sdp)
 
 		gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_SHUTDOWN |
 			       GFS2_LFC_MAKE_FS_RO);
-		wait_event(sdp->sd_reserving_log_wait,
-			   atomic_read(&sdp->sd_reserving_log) == 0);
-		gfs2_assert_warn(sdp, atomic_read(&sdp->sd_log_blks_free) ==
-				 sdp->sd_jdesc->jd_blocks);
+		wait_event_timeout(sdp->sd_log_waitq,
+				   gfs2_log_is_empty(sdp),
+				   HZ * 5);
+		gfs2_assert_warn(sdp, gfs2_log_is_empty(sdp));
 	} else {
-		wait_event_timeout(sdp->sd_reserving_log_wait,
-				   atomic_read(&sdp->sd_reserving_log) == 0,
+		wait_event_timeout(sdp->sd_log_waitq,
+				   gfs2_log_is_empty(sdp),
 				   HZ * 5);
 	}
-	if (gfs2_holder_initialized(&freeze_gh))
-		gfs2_glock_dq_uninit(&freeze_gh);
-
 	gfs2_quota_cleanup(sdp);
 
 	if (!log_write_allowed)
@@ -770,10 +728,8 @@ void gfs2_freeze_func(struct work_struct *work)
 	struct super_block *sb = sdp->sd_vfs;
 
 	atomic_inc(&sb->s_active);
-	error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED,
-				   LM_FLAG_NOEXP | GL_EXACT, &freeze_gh);
+	error = gfs2_freeze_lock(sdp, &freeze_gh, 0);
 	if (error) {
-		fs_info(sdp, "GFS2: couldn't get freeze lock : %d\n", error);
 		gfs2_assert_withdraw(sdp, 0);
 	} else {
 		atomic_set(&sdp->sd_freeze_state, SFS_UNFROZEN);
@@ -783,7 +739,7 @@ void gfs2_freeze_func(struct work_struct *work)
 				error);
 			gfs2_assert_withdraw(sdp, 0);
 		}
-		gfs2_glock_dq_uninit(&freeze_gh);
+		gfs2_freeze_unlock(&freeze_gh);
 	}
 	deactivate_super(sb);
 	clear_bit_unlock(SDF_FS_FROZEN, &sdp->sd_flags);
@@ -851,7 +807,7 @@ static int gfs2_unfreeze(struct super_block *sb)
                 return 0;
 	}
 
-	gfs2_glock_dq_uninit(&sdp->sd_freeze_gh);
+	gfs2_freeze_unlock(&sdp->sd_freeze_gh);
 	mutex_unlock(&sdp->sd_freeze_mutex);
 	return wait_on_bit(&sdp->sd_flags, SDF_FS_FROZEN, TASK_INTERRUPTIBLE);
 }
@@ -1227,7 +1183,8 @@ static int gfs2_dinode_dealloc(struct gfs2_inode *ip)
 		goto out_qs;
 	}
 
-	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh);
+	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE,
+				   LM_FLAG_NODE_SCOPE, &gh);
 	if (error)
 		goto out_qs;
 

fs/gfs2/super.h

@@ -11,6 +11,10 @@
 #include <linux/dcache.h>
 #include "incore.h"
 
+/* Supported fs format version range */
+#define GFS2_FS_FORMAT_MIN (1801)
+#define GFS2_FS_FORMAT_MAX (1802)
+
 extern void gfs2_lm_unmount(struct gfs2_sbd *sdp);
 
 static inline unsigned int gfs2_jindex_size(struct gfs2_sbd *sdp)
@@ -54,7 +58,9 @@ extern struct file_system_type gfs2meta_fs_type;
 extern const struct export_operations gfs2_export_ops;
 extern const struct super_operations gfs2_super_ops;
 extern const struct dentry_operations gfs2_dops;
-extern const struct xattr_handler *gfs2_xattr_handlers[];
+
+extern const struct xattr_handler *gfs2_xattr_handlers_max[];
+extern const struct xattr_handler **gfs2_xattr_handlers_min;
 
 #endif /* __SUPER_DOT_H__ */
 

fs/gfs2/trace_gfs2.h

@@ -560,6 +560,7 @@ TRACE_EVENT(gfs2_block_alloc,
 		__field(	u8,	block_state		)
 		__field(        u64,	rd_addr			)
 		__field(        u32,	rd_free_clone		)
+		__field(	u32,	rd_requested		)
 		__field(	u32,	rd_reserved		)
 	),
 
@@ -571,17 +572,20 @@ TRACE_EVENT(gfs2_block_alloc,
 		__entry->block_state	= block_state;
 		__entry->rd_addr	= rgd->rd_addr;
 		__entry->rd_free_clone	= rgd->rd_free_clone;
+		__entry->rd_requested	= rgd->rd_requested;
 		__entry->rd_reserved	= rgd->rd_reserved;
 	),
 
-	TP_printk("%u,%u bmap %llu alloc %llu/%lu %s rg:%llu rf:%u rr:%lu",
+	TP_printk("%u,%u bmap %llu alloc %llu/%lu %s rg:%llu rf:%u rq:%u rr:%u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  (unsigned long long)__entry->inum,
 		  (unsigned long long)__entry->start,
 		  (unsigned long)__entry->len,
 		  block_state_name(__entry->block_state),
 		  (unsigned long long)__entry->rd_addr,
-		  __entry->rd_free_clone, (unsigned long)__entry->rd_reserved)
+		  __entry->rd_free_clone,
+		  __entry->rd_requested,
+		  __entry->rd_reserved)
 );
 
 /* Keep track of multi-block reservations as they are allocated/freed */
@@ -595,33 +599,40 @@ TRACE_EVENT(gfs2_rs,
 		__field(        dev_t,  dev                     )
 		__field(	u64,	rd_addr			)
 		__field(	u32,	rd_free_clone		)
+		__field(	u32,	rd_requested		)
 		__field(	u32,	rd_reserved		)
 		__field(	u64,	inum			)
 		__field(	u64,	start			)
-		__field(	u32,	free			)
+		__field(	u32,	requested		)
+		__field(	u32,	reserved		)
 		__field(	u8,	func			)
 	),
 
 	TP_fast_assign(
-		__entry->dev		= rs->rs_rbm.rgd->rd_sbd->sd_vfs->s_dev;
-		__entry->rd_addr	= rs->rs_rbm.rgd->rd_addr;
-		__entry->rd_free_clone	= rs->rs_rbm.rgd->rd_free_clone;
-		__entry->rd_reserved	= rs->rs_rbm.rgd->rd_reserved;
+		__entry->dev		= rs->rs_rgd->rd_sbd->sd_vfs->s_dev;
+		__entry->rd_addr	= rs->rs_rgd->rd_addr;
+		__entry->rd_free_clone	= rs->rs_rgd->rd_free_clone;
+		__entry->rd_requested	= rs->rs_rgd->rd_requested;
+		__entry->rd_reserved	= rs->rs_rgd->rd_reserved;
 		__entry->inum		= container_of(rs, struct gfs2_inode,
 						       i_res)->i_no_addr;
-		__entry->start		= gfs2_rbm_to_block(&rs->rs_rbm);
-		__entry->free		= rs->rs_free;
+		__entry->start		= rs->rs_start;
+		__entry->requested	= rs->rs_requested;
+		__entry->reserved	= rs->rs_reserved;
 		__entry->func		= func;
 	),
 
-	TP_printk("%u,%u bmap %llu resrv %llu rg:%llu rf:%lu rr:%lu %s f:%lu",
+	TP_printk("%u,%u bmap %llu resrv %llu rg:%llu rf:%u rq:%u rr:%u %s q:%u r:%u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  (unsigned long long)__entry->inum,
 		  (unsigned long long)__entry->start,
 		  (unsigned long long)__entry->rd_addr,
-		  (unsigned long)__entry->rd_free_clone,
-		  (unsigned long)__entry->rd_reserved,
-		  rs_func_name(__entry->func), (unsigned long)__entry->free)
+		  __entry->rd_free_clone,
+		  __entry->rd_requested,
+		  __entry->rd_reserved,
+		  rs_func_name(__entry->func),
+		  __entry->requested,
+		  __entry->reserved)
 );
 
 #endif /* _TRACE_GFS2_H */

fs/gfs2/trans.c

@@ -31,17 +31,17 @@ static void gfs2_print_trans(struct gfs2_sbd *sdp, const struct gfs2_trans *tr)
 	fs_warn(sdp, "blocks=%u revokes=%u reserved=%u touched=%u\n",
 		tr->tr_blocks, tr->tr_revokes, tr->tr_reserved,
 		test_bit(TR_TOUCHED, &tr->tr_flags));
-	fs_warn(sdp, "Buf %u/%u Databuf %u/%u Revoke %u/%u\n",
+	fs_warn(sdp, "Buf %u/%u Databuf %u/%u Revoke %u\n",
 		tr->tr_num_buf_new, tr->tr_num_buf_rm,
 		tr->tr_num_databuf_new, tr->tr_num_databuf_rm,
-		tr->tr_num_revoke, tr->tr_num_revoke_rm);
+		tr->tr_num_revoke);
 }
 
-int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks,
-		     unsigned int revokes)
+int __gfs2_trans_begin(struct gfs2_trans *tr, struct gfs2_sbd *sdp,
+		       unsigned int blocks, unsigned int revokes,
+		       unsigned long ip)
 {
-	struct gfs2_trans *tr;
-	int error;
+	unsigned int extra_revokes;
 
 	if (current->journal_info) {
 		gfs2_print_trans(sdp, current->journal_info);
@@ -52,39 +52,72 @@ int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks,
 	if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))
 		return -EROFS;
 
-	tr = kmem_cache_zalloc(gfs2_trans_cachep, GFP_NOFS);
-	if (!tr)
-		return -ENOMEM;
-
-	tr->tr_ip = _RET_IP_;
+	tr->tr_ip = ip;
 	tr->tr_blocks = blocks;
 	tr->tr_revokes = revokes;
-	tr->tr_reserved = 1;
-	set_bit(TR_ALLOCED, &tr->tr_flags);
-	if (blocks)
-		tr->tr_reserved += 6 + blocks;
-	if (revokes)
-		tr->tr_reserved += gfs2_struct2blk(sdp, revokes);
+	tr->tr_reserved = GFS2_LOG_FLUSH_MIN_BLOCKS;
+	if (blocks) {
+		/*
+		 * The reserved blocks are either used for data or metadata.
+		 * We can have mixed data and metadata, each with its own log
+		 * descriptor block; see calc_reserved().
+		 */
+		tr->tr_reserved += blocks + 1 + DIV_ROUND_UP(blocks - 1, databuf_limit(sdp));
+	}
 	INIT_LIST_HEAD(&tr->tr_databuf);
 	INIT_LIST_HEAD(&tr->tr_buf);
 	INIT_LIST_HEAD(&tr->tr_list);
 	INIT_LIST_HEAD(&tr->tr_ail1_list);
 	INIT_LIST_HEAD(&tr->tr_ail2_list);
 
+	if (gfs2_assert_warn(sdp, tr->tr_reserved <= sdp->sd_jdesc->jd_blocks))
+		return -EINVAL;
+
 	sb_start_intwrite(sdp->sd_vfs);
 
-	error = gfs2_log_reserve(sdp, tr->tr_reserved);
-	if (error)
-		goto fail;
+	/*
+	 * Try the reservations under sd_log_flush_lock to prevent log flushes
+	 * from creating inconsistencies between the number of allocated and
+	 * reserved revokes.  If that fails, do a full-block allocation outside
+	 * of the lock to avoid stalling log flushes.  Then, allot the
+	 * appropriate number of blocks to revokes, use as many revokes locally
+	 * as needed, and "release" the surplus into the revokes pool.
+	 */
+
+	down_read(&sdp->sd_log_flush_lock);
+	if (gfs2_log_try_reserve(sdp, tr, &extra_revokes))
+		goto reserved;
+	up_read(&sdp->sd_log_flush_lock);
+	gfs2_log_reserve(sdp, tr, &extra_revokes);
+	down_read(&sdp->sd_log_flush_lock);
+
+reserved:
+	gfs2_log_release_revokes(sdp, extra_revokes);
+	if (unlikely(!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))) {
+		gfs2_log_release_revokes(sdp, tr->tr_revokes);
+		up_read(&sdp->sd_log_flush_lock);
+		gfs2_log_release(sdp, tr->tr_reserved);
+		sb_end_intwrite(sdp->sd_vfs);
+		return -EROFS;
+	}
 
 	current->journal_info = tr;
 
 	return 0;
+}
 
-fail:
-	sb_end_intwrite(sdp->sd_vfs);
-	kmem_cache_free(gfs2_trans_cachep, tr);
+int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks,
+		     unsigned int revokes)
+{
+	struct gfs2_trans *tr;
+	int error;
 
+	tr = kmem_cache_zalloc(gfs2_trans_cachep, GFP_NOFS);
+	if (!tr)
+		return -ENOMEM;
+	error = __gfs2_trans_begin(tr, sdp, blocks, revokes, _RET_IP_);
+	if (error)
+		kmem_cache_free(gfs2_trans_cachep, tr);
 	return error;
 }
 
@@ -92,37 +125,39 @@ void gfs2_trans_end(struct gfs2_sbd *sdp)
 {
 	struct gfs2_trans *tr = current->journal_info;
 	s64 nbuf;
-	int alloced = test_bit(TR_ALLOCED, &tr->tr_flags);
 
 	current->journal_info = NULL;
 
 	if (!test_bit(TR_TOUCHED, &tr->tr_flags)) {
+		gfs2_log_release_revokes(sdp, tr->tr_revokes);
+		up_read(&sdp->sd_log_flush_lock);
 		gfs2_log_release(sdp, tr->tr_reserved);
-		if (alloced) {
+		if (!test_bit(TR_ONSTACK, &tr->tr_flags))
 			gfs2_trans_free(sdp, tr);
-			sb_end_intwrite(sdp->sd_vfs);
-		}
+		sb_end_intwrite(sdp->sd_vfs);
 		return;
 	}
 
+	gfs2_log_release_revokes(sdp, tr->tr_revokes - tr->tr_num_revoke);
+
 	nbuf = tr->tr_num_buf_new + tr->tr_num_databuf_new;
 	nbuf -= tr->tr_num_buf_rm;
 	nbuf -= tr->tr_num_databuf_rm;
 
-	if (gfs2_assert_withdraw(sdp, (nbuf <= tr->tr_blocks) &&
-				       (tr->tr_num_revoke <= tr->tr_revokes)))
+	if (gfs2_assert_withdraw(sdp, nbuf <= tr->tr_blocks) ||
+	    gfs2_assert_withdraw(sdp, tr->tr_num_revoke <= tr->tr_revokes))
 		gfs2_print_trans(sdp, tr);
 
 	gfs2_log_commit(sdp, tr);
-	if (alloced && !test_bit(TR_ATTACHED, &tr->tr_flags))
+	if (!test_bit(TR_ONSTACK, &tr->tr_flags) &&
+	    !test_bit(TR_ATTACHED, &tr->tr_flags))
 		gfs2_trans_free(sdp, tr);
 	up_read(&sdp->sd_log_flush_lock);
 
 	if (sdp->sd_vfs->s_flags & SB_SYNCHRONOUS)
 		gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL |
 			       GFS2_LFC_TRANS_END);
-	if (alloced)
-		sb_end_intwrite(sdp->sd_vfs);
+	sb_end_intwrite(sdp->sd_vfs);
 }
 
 static struct gfs2_bufdata *gfs2_alloc_bufdata(struct gfs2_glock *gl,
@@ -262,7 +297,6 @@ void gfs2_trans_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd)
 void gfs2_trans_remove_revoke(struct gfs2_sbd *sdp, u64 blkno, unsigned int len)
 {
 	struct gfs2_bufdata *bd, *tmp;
-	struct gfs2_trans *tr = current->journal_info;
 	unsigned int n = len;
 
 	gfs2_log_lock(sdp);
@@ -274,7 +308,7 @@ void gfs2_trans_remove_revoke(struct gfs2_sbd *sdp, u64 blkno, unsigned int len)
 			if (bd->bd_gl)
 				gfs2_glock_remove_revoke(bd->bd_gl);
 			kmem_cache_free(gfs2_bufdata_cachep, bd);
-			tr->tr_num_revoke_rm++;
+			gfs2_log_release_revokes(sdp, 1);
 			if (--n == 0)
 				break;
 		}

fs/gfs2/trans.h

@@ -27,13 +27,16 @@ struct gfs2_glock;
  * block, or all of the blocks in the rg, whichever is smaller */
 static inline unsigned int gfs2_rg_blocks(const struct gfs2_inode *ip, unsigned requested)
 {
-	struct gfs2_rgrpd *rgd = ip->i_res.rs_rbm.rgd;
+	struct gfs2_rgrpd *rgd = ip->i_res.rs_rgd;
 
 	if (requested < rgd->rd_length)
 		return requested + 1;
 	return rgd->rd_length;
 }
 
+extern int __gfs2_trans_begin(struct gfs2_trans *tr, struct gfs2_sbd *sdp,
+			      unsigned int blocks, unsigned int revokes,
+			      unsigned long ip);
 extern int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks,
 			    unsigned int revokes);
 

fs/gfs2/util.c

@@ -91,12 +91,39 @@ int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
 	return error;
 }
 
+/**
+ * gfs2_freeze_lock - hold the freeze glock
+ * @sdp: the superblock
+ * @freeze_gh: pointer to the requested holder
+ * @caller_flags: any additional flags needed by the caller
+ */
+int gfs2_freeze_lock(struct gfs2_sbd *sdp, struct gfs2_holder *freeze_gh,
+		     int caller_flags)
+{
+	int flags = LM_FLAG_NOEXP | GL_EXACT | caller_flags;
+	int error;
+
+	error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, flags,
+				   freeze_gh);
+	if (error && error != GLR_TRYFAILED)
+		fs_err(sdp, "can't lock the freeze lock: %d\n", error);
+	return error;
+}
+
+void gfs2_freeze_unlock(struct gfs2_holder *freeze_gh)
+{
+	if (gfs2_holder_initialized(freeze_gh))
+		gfs2_glock_dq_uninit(freeze_gh);
+}
+
 static void signal_our_withdraw(struct gfs2_sbd *sdp)
 {
-	struct gfs2_glock *gl = sdp->sd_live_gh.gh_gl;
+	struct gfs2_glock *live_gl = sdp->sd_live_gh.gh_gl;
 	struct inode *inode = sdp->sd_jdesc->jd_inode;
 	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_glock *i_gl = ip->i_gl;
 	u64 no_formal_ino = ip->i_no_formal_ino;
+	int log_write_allowed = test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
 	int ret = 0;
 	int tries;
 
@@ -117,8 +144,21 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp)
 	 * therefore we need to clear SDF_JOURNAL_LIVE manually.
 	 */
 	clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
-	if (!sb_rdonly(sdp->sd_vfs))
-		ret = gfs2_make_fs_ro(sdp);
+	if (!sb_rdonly(sdp->sd_vfs)) {
+		struct gfs2_holder freeze_gh;
+
+		gfs2_holder_mark_uninitialized(&freeze_gh);
+		if (sdp->sd_freeze_gl &&
+		    !gfs2_glock_is_locked_by_me(sdp->sd_freeze_gl)) {
+			ret = gfs2_freeze_lock(sdp, &freeze_gh,
+				       log_write_allowed ? 0 : LM_FLAG_TRY);
+			if (ret == GLR_TRYFAILED)
+				ret = 0;
+		}
+		if (!ret)
+			ret = gfs2_make_fs_ro(sdp);
+		gfs2_freeze_unlock(&freeze_gh);
+	}
 
 	if (sdp->sd_lockstruct.ls_ops->lm_lock == NULL) { /* lock_nolock */
 		if (!ret)
@@ -141,7 +181,8 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp)
 		atomic_set(&sdp->sd_freeze_state, SFS_FROZEN);
 		thaw_super(sdp->sd_vfs);
 	} else {
-		wait_on_bit(&gl->gl_flags, GLF_DEMOTE, TASK_UNINTERRUPTIBLE);
+		wait_on_bit(&i_gl->gl_flags, GLF_DEMOTE,
+			    TASK_UNINTERRUPTIBLE);
 	}
 
 	/*
@@ -161,15 +202,15 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp)
 	 * on other nodes to be successful, otherwise we remain the owner of
 	 * the glock as far as dlm is concerned.
 	 */
-	if (gl->gl_ops->go_free) {
-		set_bit(GLF_FREEING, &gl->gl_flags);
-		wait_on_bit(&gl->gl_flags, GLF_FREEING, TASK_UNINTERRUPTIBLE);
+	if (i_gl->gl_ops->go_free) {
+		set_bit(GLF_FREEING, &i_gl->gl_flags);
+		wait_on_bit(&i_gl->gl_flags, GLF_FREEING, TASK_UNINTERRUPTIBLE);
 	}
 
 	/*
 	 * Dequeue the "live" glock, but keep a reference so it's never freed.
 	 */
-	gfs2_glock_hold(gl);
+	gfs2_glock_hold(live_gl);
 	gfs2_glock_dq_wait(&sdp->sd_live_gh);
 	/*
 	 * We enqueue the "live" glock in EX so that all other nodes
@@ -208,7 +249,7 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp)
 		gfs2_glock_nq(&sdp->sd_live_gh);
 	}
 
-	gfs2_glock_queue_put(gl); /* drop the extra reference we acquired */
+	gfs2_glock_queue_put(live_gl); /* drop extra reference we acquired */
 	clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags);
 
 	/*

fs/gfs2/util.h

@@ -149,6 +149,9 @@ int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function,
 
 extern int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
 			       bool verbose);
+extern int gfs2_freeze_lock(struct gfs2_sbd *sdp,
+			    struct gfs2_holder *freeze_gh, int caller_flags);
+extern void gfs2_freeze_unlock(struct gfs2_holder *freeze_gh);
 
 #define gfs2_io_error(sdp) \
 gfs2_io_error_i((sdp), __func__, __FILE__, __LINE__)

fs/gfs2/xattr.c

@@ -70,6 +70,20 @@ static int ea_check_size(struct gfs2_sbd *sdp, unsigned int nsize, size_t dsize)
 	return 0;
 }
 
+static bool gfs2_eatype_valid(struct gfs2_sbd *sdp, u8 type)
+{
+	switch(sdp->sd_sb.sb_fs_format) {
+	case GFS2_FS_FORMAT_MAX:
+		return true;
+
+	case GFS2_FS_FORMAT_MIN:
+		return type <= GFS2_EATYPE_SECURITY;
+
+	default:
+		return false;
+	}
+}
+
 typedef int (*ea_call_t) (struct gfs2_inode *ip, struct buffer_head *bh,
 			  struct gfs2_ea_header *ea,
 			  struct gfs2_ea_header *prev, void *private);
@@ -77,6 +91,7 @@ typedef int (*ea_call_t) (struct gfs2_inode *ip, struct buffer_head *bh,
 static int ea_foreach_i(struct gfs2_inode *ip, struct buffer_head *bh,
 			ea_call_t ea_call, void *data)
 {
+	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	struct gfs2_ea_header *ea, *prev = NULL;
 	int error = 0;
 
@@ -89,9 +104,8 @@ static int ea_foreach_i(struct gfs2_inode *ip, struct buffer_head *bh,
 		if (!(bh->b_data <= (char *)ea && (char *)GFS2_EA2NEXT(ea) <=
 						  bh->b_data + bh->b_size))
 			goto fail;
-		if (!GFS2_EATYPE_VALID(ea->ea_type))
+		if (!gfs2_eatype_valid(sdp, ea->ea_type))
 			goto fail;
-
 		error = ea_call(ip, bh, ea, prev, data);
 		if (error)
 			return error;
@@ -259,7 +273,8 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh,
 		return -EIO;
 	}
 
-	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &rg_gh);
+	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE,
+				   LM_FLAG_NODE_SCOPE, &rg_gh);
 	if (error)
 		return error;
 
@@ -344,6 +359,7 @@ static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh,
 		     struct gfs2_ea_header *ea, struct gfs2_ea_header *prev,
 		     void *private)
 {
+	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	struct ea_list *ei = private;
 	struct gfs2_ea_request *er = ei->ei_er;
 	unsigned int ea_size;
@@ -353,6 +369,8 @@ static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh,
 	if (ea->ea_type == GFS2_EATYPE_UNUSED)
 		return 0;
 
+	BUG_ON(ea->ea_type > GFS2_EATYPE_SECURITY &&
+	       sdp->sd_sb.sb_fs_format == GFS2_FS_FORMAT_MIN);
 	switch (ea->ea_type) {
 	case GFS2_EATYPE_USR:
 		prefix = "user.";
@@ -366,8 +384,12 @@ static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh,
 		prefix = "security.";
 		l = 9;
 		break;
+	case GFS2_EATYPE_TRUSTED:
+		prefix = "trusted.";
+		l = 8;
+		break;
 	default:
-		BUG();
+		return 0;
 	}
 
 	ea_size = l + ea->ea_name_len + 1;
@@ -1386,7 +1408,8 @@ static int ea_dealloc_block(struct gfs2_inode *ip)
 		return -EIO;
 	}
 
-	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh);
+	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE,
+				   LM_FLAG_NODE_SCOPE, &gh);
 	if (error)
 		return error;
 
@@ -1464,7 +1487,25 @@ static const struct xattr_handler gfs2_xattr_security_handler = {
 	.set    = gfs2_xattr_set,
 };
 
-const struct xattr_handler *gfs2_xattr_handlers[] = {
+static bool
+gfs2_xattr_trusted_list(struct dentry *dentry)
+{
+	return capable(CAP_SYS_ADMIN);
+}
+
+static const struct xattr_handler gfs2_xattr_trusted_handler = {
+	.prefix = XATTR_TRUSTED_PREFIX,
+	.flags  = GFS2_EATYPE_TRUSTED,
+	.list	= gfs2_xattr_trusted_list,
+	.get    = gfs2_xattr_get,
+	.set    = gfs2_xattr_set,
+};
+
+const struct xattr_handler *gfs2_xattr_handlers_max[] = {
+	/* GFS2_FS_FORMAT_MAX */
+	&gfs2_xattr_trusted_handler,
+
+	/* GFS2_FS_FORMAT_MIN */
 	&gfs2_xattr_user_handler,
 	&gfs2_xattr_security_handler,
 	&posix_acl_access_xattr_handler,
@@ -1472,3 +1513,4 @@ const struct xattr_handler *gfs2_xattr_handlers[] = {
 	NULL,
 };
 
+const struct xattr_handler **gfs2_xattr_handlers_min = gfs2_xattr_handlers_max + 1;

include/uapi/linux/gfs2_ondisk.h

@@ -47,7 +47,7 @@
 #define GFS2_FORMAT_DE		1200
 #define GFS2_FORMAT_QU		1500
 /* These are part of the superblock */
-#define GFS2_FORMAT_FS		1801
+#define GFS2_FORMAT_FS		1802
 #define GFS2_FORMAT_MULTI	1900
 
 /*
@@ -389,8 +389,9 @@ struct gfs2_leaf {
 #define GFS2_EATYPE_USR		1
 #define GFS2_EATYPE_SYS		2
 #define GFS2_EATYPE_SECURITY	3
+#define GFS2_EATYPE_TRUSTED	4
 
-#define GFS2_EATYPE_LAST	3
+#define GFS2_EATYPE_LAST	4
 #define GFS2_EATYPE_VALID(x)	((x) <= GFS2_EATYPE_LAST)
 
 #define GFS2_EAFLAG_LAST	0x01	/* last ea in block */