Merge git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-nmw

Pull GFS2 updates from Steven Whitehouse:
 "The major feature this time is the "rbm" conversion in the resource
  group code.  The new struct gfs2_rbm specifies the location of an
  allocatable block in (resource group, bitmap, offset) form.  There are
  a number of added helper functions, and later patches then rewrite
  some of the resource group code in terms of this new structure.  Not
  only does this give us a nice code clean up, but it also removes some
  of the previous restrictions where extents could not cross bitmap
  boundaries, for example.

  In addition to that, there are a few bug fixes and clean ups, but the
  rbm work is by far the majority of this patch set in terms of number
  of changed lines."

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-nmw: (27 commits)
  GFS2: Write out dirty inode metadata in delayed deletes
  GFS2: fix s_writers.counter imbalance in gfs2_ail_empty_gl
  GFS2: Fix infinite loop in rbm_find
  GFS2: Consolidate free block searching functions
  GFS2: Get rid of I_MUTEX_QUOTA usage
  GFS2: Stop block extents at the end of bitmaps
  GFS2: Fix unclaimed_blocks() wrapping bug and clean up
  GFS2: Improve block reservation tracing
  GFS2: Fall back to ignoring reservations, if there are no other blocks left
  GFS2: Fix ->show_options() for statfs slow
  GFS2: Use rbm for gfs2_setbit()
  GFS2: Use rbm for gfs2_testbit()
  GFS2: Eliminate unnecessary check for state > 3 in bitfit
  GFS2: Eliminate redundant calls to may_grant
  GFS2: Combine functions gfs2_glock_dq_wait and wait_on_demote
  GFS2: Combine functions gfs2_glock_wait and wait_on_holder
  GFS2: inline __gfs2_glock_schedule_for_reclaim
  GFS2: change function gfs2_direct_IO to use a normal gfs2_glock_dq
  GFS2: rbm code cleanup
  GFS2: Fix case where reservation finished at end of rgrp
  ...

fs/gfs2/aops.c

@@ -612,6 +612,7 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping,
 	struct gfs2_sbd *sdp = GFS2_SB(mapping->host);
 	struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
 	unsigned int data_blocks = 0, ind_blocks = 0, rblocks;
+	unsigned requested = 0;
 	int alloc_required;
 	int error = 0;
 	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
@@ -641,7 +642,8 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping,
 		if (error)
 			goto out_unlock;
 
-		error = gfs2_inplace_reserve(ip, data_blocks + ind_blocks);
+		requested = data_blocks + ind_blocks;
+		error = gfs2_inplace_reserve(ip, requested);
 		if (error)
 			goto out_qunlock;
 	}
@@ -654,7 +656,7 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping,
 	if (&ip->i_inode == sdp->sd_rindex)
 		rblocks += 2 * RES_STATFS;
 	if (alloc_required)
-		rblocks += gfs2_rg_blocks(ip);
+		rblocks += gfs2_rg_blocks(ip, requested);
 
 	error = gfs2_trans_begin(sdp, rblocks,
 				 PAGE_CACHE_SIZE/sdp->sd_sb.sb_bsize);
@@ -868,8 +870,7 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping,
 	brelse(dibh);
 failed:
 	gfs2_trans_end(sdp);
-	if (gfs2_mb_reserved(ip))
-		gfs2_inplace_release(ip);
+	gfs2_inplace_release(ip);
 	if (ip->i_res->rs_qa_qd_num)
 		gfs2_quota_unlock(ip);
 	if (inode == sdp->sd_rindex) {
@@ -1023,7 +1024,7 @@ static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb,
 				  offset, nr_segs, gfs2_get_block_direct,
 				  NULL, NULL, 0);
 out:
-	gfs2_glock_dq_m(1, &gh);
+	gfs2_glock_dq(&gh);
 	gfs2_holder_uninit(&gh);
 	return rv;
 }

fs/gfs2/bmap.c

@@ -786,7 +786,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh,
 		goto out_rlist;
 
 	if (gfs2_rs_active(ip->i_res)) /* needs to be done with the rgrp glock held */
-		gfs2_rs_deltree(ip->i_res);
+		gfs2_rs_deltree(ip, ip->i_res);
 
 	error = gfs2_trans_begin(sdp, rg_blocks + RES_DINODE +
 				 RES_INDIRECT + RES_STATFS + RES_QUOTA,

fs/gfs2/file.c

@@ -441,7 +441,7 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 		rblocks += data_blocks ? data_blocks : 1;
 	if (ind_blocks || data_blocks) {
 		rblocks += RES_STATFS + RES_QUOTA;
-		rblocks += gfs2_rg_blocks(ip);
+		rblocks += gfs2_rg_blocks(ip, data_blocks + ind_blocks);
 	}
 	ret = gfs2_trans_begin(sdp, rblocks, 0);
 	if (ret)
@@ -845,7 +845,7 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset,
 				&max_bytes, &data_blocks, &ind_blocks);
 
 		rblocks = RES_DINODE + ind_blocks + RES_STATFS + RES_QUOTA +
-			  RES_RG_HDR + gfs2_rg_blocks(ip);
+			  RES_RG_HDR + gfs2_rg_blocks(ip, data_blocks + ind_blocks);
 		if (gfs2_is_jdata(ip))
 			rblocks += data_blocks ? data_blocks : 1;
 

fs/gfs2/glock.c

@@ -185,20 +185,6 @@ static void gfs2_glock_remove_from_lru(struct gfs2_glock *gl)
 	spin_unlock(&lru_lock);
 }
 
-/**
- * __gfs2_glock_schedule_for_reclaim - Add a glock to the reclaim list
- * @gl: the glock
- *
- * If the glock is demotable, then we add it (or move it) to the end
- * of the glock LRU list.
- */
-
-static void __gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
-{
-	if (demote_ok(gl))
-		gfs2_glock_add_to_lru(gl);
-}
-
 /**
  * gfs2_glock_put_nolock() - Decrement reference count on glock
  * @gl: The glock to put
@@ -883,7 +869,14 @@ static int gfs2_glock_demote_wait(void *word)
 	return 0;
 }
 
-static void wait_on_holder(struct gfs2_holder *gh)
+/**
+ * gfs2_glock_wait - wait on a glock acquisition
+ * @gh: the glock holder
+ *
+ * Returns: 0 on success
+ */
+
+int gfs2_glock_wait(struct gfs2_holder *gh)
 {
 	unsigned long time1 = jiffies;
 
@@ -894,12 +887,7 @@ static void wait_on_holder(struct gfs2_holder *gh)
 		gh->gh_gl->gl_hold_time = min(gh->gh_gl->gl_hold_time +
 					      GL_GLOCK_HOLD_INCR,
 					      GL_GLOCK_MAX_HOLD);
-}
-
-static void wait_on_demote(struct gfs2_glock *gl)
-{
-	might_sleep();
-	wait_on_bit(&gl->gl_flags, GLF_DEMOTE, gfs2_glock_demote_wait, TASK_UNINTERRUPTIBLE);
+	return gh->gh_error;
 }
 
 /**
@@ -929,19 +917,6 @@ static void handle_callback(struct gfs2_glock *gl, unsigned int state,
 	trace_gfs2_demote_rq(gl);
 }
 
-/**
- * gfs2_glock_wait - wait on a glock acquisition
- * @gh: the glock holder
- *
- * Returns: 0 on success
- */
-
-int gfs2_glock_wait(struct gfs2_holder *gh)
-{
-	wait_on_holder(gh);
-	return gh->gh_error;
-}
-
 void gfs2_print_dbg(struct seq_file *seq, const char *fmt, ...)
 {
 	struct va_format vaf;
@@ -979,15 +954,15 @@ __acquires(&gl->gl_spin)
 	struct gfs2_sbd *sdp = gl->gl_sbd;
 	struct list_head *insert_pt = NULL;
 	struct gfs2_holder *gh2;
-	int try_lock = 0;
+	int try_futile = 0;
 
 	BUG_ON(gh->gh_owner_pid == NULL);
 	if (test_and_set_bit(HIF_WAIT, &gh->gh_iflags))
 		BUG();
 
 	if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)) {
 		if (test_bit(GLF_LOCK, &gl->gl_flags))
-			try_lock = 1;
+			try_futile = !may_grant(gl, gh);
 		if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags))
 			goto fail;
 	}
@@ -996,9 +971,8 @@ __acquires(&gl->gl_spin)
 		if (unlikely(gh2->gh_owner_pid == gh->gh_owner_pid &&
 		    (gh->gh_gl->gl_ops->go_type != LM_TYPE_FLOCK)))
 			goto trap_recursive;
-		if (try_lock &&
-		    !(gh2->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)) &&
-		    !may_grant(gl, gh)) {
+		if (try_futile &&
+		    !(gh2->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) {
 fail:
 			gh->gh_error = GLR_TRYFAILED;
 			gfs2_holder_wake(gh);
@@ -1121,8 +1095,9 @@ void gfs2_glock_dq(struct gfs2_holder *gh)
 		    !test_bit(GLF_DEMOTE, &gl->gl_flags))
 			fast_path = 1;
 	}
-	if (!test_bit(GLF_LFLUSH, &gl->gl_flags))
-		__gfs2_glock_schedule_for_reclaim(gl);
+	if (!test_bit(GLF_LFLUSH, &gl->gl_flags) && demote_ok(gl))
+		gfs2_glock_add_to_lru(gl);
+
 	trace_gfs2_glock_queue(gh, 0);
 	spin_unlock(&gl->gl_spin);
 	if (likely(fast_path))
@@ -1141,7 +1116,8 @@ void gfs2_glock_dq_wait(struct gfs2_holder *gh)
 {
 	struct gfs2_glock *gl = gh->gh_gl;
 	gfs2_glock_dq(gh);
-	wait_on_demote(gl);
+	might_sleep();
+	wait_on_bit(&gl->gl_flags, GLF_DEMOTE, gfs2_glock_demote_wait, TASK_UNINTERRUPTIBLE);
 }
 
 /**

fs/gfs2/glops.c

@@ -94,6 +94,7 @@ static void gfs2_ail_empty_gl(struct gfs2_glock *gl)
 	/* A shortened, inline version of gfs2_trans_begin() */
 	tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64));
 	tr.tr_ip = (unsigned long)__builtin_return_address(0);
+	sb_start_intwrite(sdp->sd_vfs);
 	gfs2_log_reserve(sdp, tr.tr_reserved);
 	BUG_ON(current->journal_info);
 	current->journal_info = &tr;

fs/gfs2/incore.h

@@ -99,9 +99,26 @@ struct gfs2_rgrpd {
 #define GFS2_RDF_MASK		0xf0000000 /* mask for internal flags */
 	spinlock_t rd_rsspin;           /* protects reservation related vars */
 	struct rb_root rd_rstree;       /* multi-block reservation tree */
-	u32 rd_rs_cnt;                  /* count of current reservations */
 };
 
+struct gfs2_rbm {
+	struct gfs2_rgrpd *rgd;
+	struct gfs2_bitmap *bi;	/* Bitmap must belong to the rgd */
+	u32 offset;		/* The offset is bitmap relative */
+};
+
+static inline u64 gfs2_rbm_to_block(const struct gfs2_rbm *rbm)
+{
+	return rbm->rgd->rd_data0 + (rbm->bi->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->bi == rbm2->bi) && 
+	       (rbm1->offset == rbm2->offset);
+}
+
 enum gfs2_state_bits {
 	BH_Pinned = BH_PrivateStart,
 	BH_Escaped = BH_PrivateStart + 1,
@@ -250,18 +267,11 @@ struct gfs2_blkreserv {
 	/* components used during write (step 1): */
 	atomic_t rs_sizehint;         /* hint of the write size */
 
-	/* components used during inplace_reserve (step 2): */
-	u32 rs_requested; /* Filled in by caller of gfs2_inplace_reserve() */
-
-	/* components used during get_local_rgrp (step 3): */
-	struct gfs2_rgrpd *rs_rgd;    /* pointer to the gfs2_rgrpd */
 	struct gfs2_holder rs_rgd_gh; /* Filled in by get_local_rgrp */
 	struct rb_node rs_node;       /* link to other block reservations */
-
-	/* components used during block searches and assignments (step 4): */
-	struct gfs2_bitmap *rs_bi;    /* bitmap for the current allocation */
-	u32 rs_biblk;                 /* start block relative to the bi */
+	struct gfs2_rbm rs_rbm;       /* Start of reservation */
 	u32 rs_free;                  /* how many blocks are still free */
+	u64 rs_inum;                  /* Inode number for reservation */
 
 	/* ancillary quota stuff */
 	struct gfs2_quota_data *rs_qa_qd[2 * MAXQUOTAS];

fs/gfs2/inode.c

@@ -712,14 +712,9 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 	if (error)
 		goto fail_gunlock2;
 
-	/* The newly created inode needs a reservation so it can allocate
-	   xattrs. At the same time, we want new blocks allocated to the new
-	   dinode to be as contiguous as possible. Since we allocated the
-	   dinode block under the directory's reservation, we transfer
-	   ownership of that reservation to the new inode. The directory
-	   doesn't need a reservation unless it needs a new allocation. */
-	ip->i_res = dip->i_res;
-	dip->i_res = NULL;
+	error = gfs2_rs_alloc(ip);
+	if (error)
+		goto fail_gunlock2;
 
 	error = gfs2_acl_create(dip, inode);
 	if (error)
@@ -737,10 +732,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 		brelse(bh);
 
 	gfs2_trans_end(sdp);
-	/* Check if we reserved space in the rgrp. Function link_dinode may
-	   not, depending on whether alloc is required. */
-	if (gfs2_mb_reserved(dip))
-		gfs2_inplace_release(dip);
+	gfs2_inplace_release(dip);
 	gfs2_quota_unlock(dip);
 	mark_inode_dirty(inode);
 	gfs2_glock_dq_uninit_m(2, ghs);
@@ -897,7 +889,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir,
 			goto out_gunlock_q;
 
 		error = gfs2_trans_begin(sdp, sdp->sd_max_dirres +
-					 gfs2_rg_blocks(dip) +
+					 gfs2_rg_blocks(dip, sdp->sd_max_dirres) +
 					 2 * RES_DINODE + RES_STATFS +
 					 RES_QUOTA, 0);
 		if (error)
@@ -1378,7 +1370,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
 			goto out_gunlock_q;
 
 		error = gfs2_trans_begin(sdp, sdp->sd_max_dirres +
-					 gfs2_rg_blocks(ndip) +
+					 gfs2_rg_blocks(ndip, sdp->sd_max_dirres) +
 					 4 * RES_DINODE + 4 * RES_LEAF +
 					 RES_STATFS + RES_QUOTA + 4, 0);
 		if (error)

fs/gfs2/ops_fstype.c

@@ -19,6 +19,7 @@
 #include <linux/mount.h>
 #include <linux/gfs2_ondisk.h>
 #include <linux/quotaops.h>
+#include <linux/lockdep.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -766,6 +767,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 	return error;
 }
 
+static struct lock_class_key gfs2_quota_imutex_key;
 
 static int init_inodes(struct gfs2_sbd *sdp, int undo)
 {
@@ -803,6 +805,12 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo)
 		fs_err(sdp, "can't get quota file inode: %d\n", error);
 		goto fail_rindex;
 	}
+	/*
+	 * i_mutex on quota files is special. Since this inode is hidden system
+	 * file, we are safe to define locking ourselves.
+	 */
+	lockdep_set_class(&sdp->sd_quota_inode->i_mutex,
+			  &gfs2_quota_imutex_key);
 
 	error = gfs2_rindex_update(sdp);
 	if (error)

fs/gfs2/quota.c

@@ -765,6 +765,7 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda)
 	struct gfs2_holder *ghs, i_gh;
 	unsigned int qx, x;
 	struct gfs2_quota_data *qd;
+	unsigned reserved;
 	loff_t offset;
 	unsigned int nalloc = 0, blocks;
 	int error;
@@ -781,7 +782,7 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda)
 		return -ENOMEM;
 
 	sort(qda, num_qd, sizeof(struct gfs2_quota_data *), sort_qd, NULL);
-	mutex_lock_nested(&ip->i_inode.i_mutex, I_MUTEX_QUOTA);
+	mutex_lock(&ip->i_inode.i_mutex);
 	for (qx = 0; qx < num_qd; qx++) {
 		error = gfs2_glock_nq_init(qda[qx]->qd_gl, LM_ST_EXCLUSIVE,
 					   GL_NOCACHE, &ghs[qx]);
@@ -811,13 +812,13 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda)
 	 * two blocks need to be updated instead of 1 */
 	blocks = num_qd * data_blocks + RES_DINODE + num_qd + 3;
 
-	error = gfs2_inplace_reserve(ip, 1 +
-				     (nalloc * (data_blocks + ind_blocks)));
+	reserved = 1 + (nalloc * (data_blocks + ind_blocks));
+	error = gfs2_inplace_reserve(ip, reserved);
 	if (error)
 		goto out_alloc;
 
 	if (nalloc)
-		blocks += gfs2_rg_blocks(ip) + nalloc * ind_blocks + RES_STATFS;
+		blocks += gfs2_rg_blocks(ip, reserved) + nalloc * ind_blocks + RES_STATFS;
 
 	error = gfs2_trans_begin(sdp, blocks, 0);
 	if (error)
@@ -1598,7 +1599,7 @@ static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id,
 		error = gfs2_inplace_reserve(ip, blocks);
 		if (error)
 			goto out_i;
-		blocks += gfs2_rg_blocks(ip);
+		blocks += gfs2_rg_blocks(ip, blocks);
 	}
 
 	/* Some quotas span block boundaries and can update two blocks,