Merge tag 'xfs-5.9-merge-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull xfs updates from Darrick Wong:
 "There are quite a few changes in this release, the most notable of
  which is that we've made inode flushing fully asynchronous, and we no
  longer block memory reclaim on this.

  Furthermore, we have fixed a long-standing bug in the quota code where
  soft limit warnings and inode limits were never tracked properly.

  Moving further down the line, the reflink control loops have been
  redesigned to behave more efficiently; and numerous small bugs have
  been fixed (see below). The xattr and quota code have been extensively
  refactored in preparation for more new features coming down the line.

  Finally, the behavior of DAX between ext4 and xfs has been stabilized,
  which gets us a step closer to removing the experimental tag from that
  feature.

  We have a few new contributors this time around. Welcome, all!

  I anticipate a second pull request next week for a few small bugfixes
  that have been trickling in, but this is it for big changes.

  Summary:

   - Fix some btree block pingponging problems when swapping extents

   - Redesign the reflink copy loop so that we only run one remapping
     operation per transaction. This helps us avoid running out of block
     reservation on highly deduped filesystems.

   - Take the MMAPLOCK around filemap_map_pages.

   - Make inode reclaim fully async so that we avoid stalling processes
     on flushing inodes to disk.

   - Reduce inode cluster buffer RMW cycles by attaching the buffer to
     dirty inodes so we won't let go of the cluster buffer when we know
     we're going to need it soon.

   - Add some more checks to the realtime bitmap file scrubber.

   - Don't trip false lockdep warnings in fs freeze.

   - Remove various redundant lines of code.

   - Remove unnecessary calls to xfs_perag_{get,put}.

   - Preserve I_VERSION state across remounts.

   - Fix an unmount hang due to AIL going to sleep with a non-empty
     delwri buffer list.

   - Fix an error in the inode allocation space reservation macro that
     caused regressions in generic/531.

   - Fix a potential livelock when dquot flush fails because the dquot
     buffer is locked.

   - Fix a miscalculation when reserving inode quota that could cause
     users to exceed a hardlimit.

   - Refactor struct xfs_dquot to use native types for incore fields
     instead of abusing the ondisk struct for this purpose. This will
     eventually enable proper y2038+ support, but for now it merely
     cleans up the quota function declarations.

   - Actually increment the quota softlimit warning counter so that soft
     failures turn into hard(er) failures when they exceed the softlimit
     warning counter limits set by the administrator.

   - Split incore dquot state flags into their own field and namespace,
     to avoid mixing them with quota type flags.

   - Create a new quota type flags namespace so that we can make it
     obvious when a quota function takes a quota type (user, group,
     project) as an argument.

   - Rename the ondisk dquot flags field to type, as that more
     accurately represents what we store in it.

   - Drop our bespoke memory allocation flags in favor of GFP_*.

   - Rearrange the xattr functions so that we no longer mix metadata
     updates and transaction management (e.g. rolling complex
     transactions) in the same functions. This work will prepare us for
     atomic xattr operations (itself a prerequisite for directory
     backrefs) in future release cycles.

   - Support FS_DAX_FL (aka FS_XFLAG_DAX) via GETFLAGS/SETFLAGS"

* tag 'xfs-5.9-merge-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (117 commits)
  fs/xfs: Support that ioctl(SETXFLAGS/GETXFLAGS) can set/get inode DAX on XFS.
  xfs: Lift -ENOSPC handler from xfs_attr_leaf_addname
  xfs: Simplify xfs_attr_node_addname
  xfs: Simplify xfs_attr_leaf_addname
  xfs: Add helper function xfs_attr_node_removename_rmt
  xfs: Add helper function xfs_attr_node_removename_setup
  xfs: Add remote block helper functions
  xfs: Add helper function xfs_attr_leaf_mark_incomplete
  xfs: Add helpers xfs_attr_is_shortform and xfs_attr_set_shortform
  xfs: Remove xfs_trans_roll in xfs_attr_node_removename
  xfs: Remove unneeded xfs_trans_roll_inode calls
  xfs: Add helper function xfs_attr_node_shrink
  xfs: Pull up xfs_attr_rmtval_invalidate
  xfs: Refactor xfs_attr_rmtval_remove
  xfs: Pull up trans roll in xfs_attr3_leaf_clearflag
  xfs: Factor out xfs_attr_rmtval_invalidate
  xfs: Pull up trans roll from xfs_attr3_leaf_setflag
  xfs: Refactor xfs_attr_try_sf_addname
  xfs: Split apart xfs_attr_leaf_addname
  xfs: Pull up trans handling in xfs_attr3_leaf_flipflags
  ...

1 files changed, 231 insertions, 184 deletions

diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index d5b7f03e93c8..04dc2be19c3a 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -23,6 +23,7 @@
 #include "xfs_trace.h"
 #include "xfs_log.h"
 #include "xfs_bmap_btree.h"
+#include "xfs_error.h"
 
 /*
  * Lock order:
@@ -66,39 +67,61 @@ xfs_qm_dqdestroy(
  */
 void
 xfs_qm_adjust_dqlimits(
-	struct xfs_mount	*mp,
 	struct xfs_dquot	*dq)
 {
+	struct xfs_mount	*mp = dq->q_mount;
 	struct xfs_quotainfo	*q = mp->m_quotainfo;
-	struct xfs_disk_dquot	*d = &dq->q_core;
 	struct xfs_def_quota	*defq;
 	int			prealloc = 0;
 
-	ASSERT(d->d_id);
+	ASSERT(dq->q_id);
 	defq = xfs_get_defquota(q, xfs_dquot_type(dq));
 
-	if (defq->bsoftlimit && !d->d_blk_softlimit) {
-		d->d_blk_softlimit = cpu_to_be64(defq->bsoftlimit);
+	if (!dq->q_blk.softlimit) {
+		dq->q_blk.softlimit = defq->blk.soft;
 		prealloc = 1;
 	}
-	if (defq->bhardlimit && !d->d_blk_hardlimit) {
-		d->d_blk_hardlimit = cpu_to_be64(defq->bhardlimit);
+	if (!dq->q_blk.hardlimit) {
+		dq->q_blk.hardlimit = defq->blk.hard;
 		prealloc = 1;
 	}
-	if (defq->isoftlimit && !d->d_ino_softlimit)
-		d->d_ino_softlimit = cpu_to_be64(defq->isoftlimit);
-	if (defq->ihardlimit && !d->d_ino_hardlimit)
-		d->d_ino_hardlimit = cpu_to_be64(defq->ihardlimit);
-	if (defq->rtbsoftlimit && !d->d_rtb_softlimit)
-		d->d_rtb_softlimit = cpu_to_be64(defq->rtbsoftlimit);
-	if (defq->rtbhardlimit && !d->d_rtb_hardlimit)
-		d->d_rtb_hardlimit = cpu_to_be64(defq->rtbhardlimit);
+	if (!dq->q_ino.softlimit)
+		dq->q_ino.softlimit = defq->ino.soft;
+	if (!dq->q_ino.hardlimit)
+		dq->q_ino.hardlimit = defq->ino.hard;
+	if (!dq->q_rtb.softlimit)
+		dq->q_rtb.softlimit = defq->rtb.soft;
+	if (!dq->q_rtb.hardlimit)
+		dq->q_rtb.hardlimit = defq->rtb.hard;
 
 	if (prealloc)
 		xfs_dquot_set_prealloc_limits(dq);
 }
 
 /*
+ * Determine if this quota counter is over either limit and set the quota
+ * timers as appropriate.
+ */
+static inline void
+xfs_qm_adjust_res_timer(
+	struct xfs_dquot_res	*res,
+	struct xfs_quota_limits	*qlim)
+{
+	ASSERT(res->hardlimit == 0 || res->softlimit <= res->hardlimit);
+
+	if ((res->softlimit && res->count > res->softlimit) ||
+	    (res->hardlimit && res->count > res->hardlimit)) {
+		if (res->timer == 0)
+			res->timer = ktime_get_real_seconds() + qlim->time;
+	} else {
+		if (res->timer == 0)
+			res->warnings = 0;
+		else
+			res->timer = 0;
+	}
+}
+
+/*
  * Check the limits and timers of a dquot and start or reset timers
  * if necessary.
  * This gets called even when quota enforcement is OFF, which makes our
@@ -113,96 +136,18 @@ xfs_qm_adjust_dqlimits(
  */
 void
 xfs_qm_adjust_dqtimers(
-	struct xfs_mount	*mp,
 	struct xfs_dquot	*dq)
 {
+	struct xfs_mount	*mp = dq->q_mount;
 	struct xfs_quotainfo	*qi = mp->m_quotainfo;
-	struct xfs_disk_dquot	*d = &dq->q_core;
 	struct xfs_def_quota	*defq;
 
-	ASSERT(d->d_id);
+	ASSERT(dq->q_id);
 	defq = xfs_get_defquota(qi, xfs_dquot_type(dq));
 
-#ifdef DEBUG
-	if (d->d_blk_hardlimit)
-		ASSERT(be64_to_cpu(d->d_blk_softlimit) <=
-		       be64_to_cpu(d->d_blk_hardlimit));
-	if (d->d_ino_hardlimit)
-		ASSERT(be64_to_cpu(d->d_ino_softlimit) <=
-		       be64_to_cpu(d->d_ino_hardlimit));
-	if (d->d_rtb_hardlimit)
-		ASSERT(be64_to_cpu(d->d_rtb_softlimit) <=
-		       be64_to_cpu(d->d_rtb_hardlimit));
-#endif
-
-	if (!d->d_btimer) {
-		if ((d->d_blk_softlimit &&
-		     (be64_to_cpu(d->d_bcount) >
-		      be64_to_cpu(d->d_blk_softlimit))) ||
-		    (d->d_blk_hardlimit &&
-		     (be64_to_cpu(d->d_bcount) >
-		      be64_to_cpu(d->d_blk_hardlimit)))) {
-			d->d_btimer = cpu_to_be32(ktime_get_real_seconds() +
-					defq->btimelimit);
-		} else {
-			d->d_bwarns = 0;
-		}
-	} else {
-		if ((!d->d_blk_softlimit ||
-		     (be64_to_cpu(d->d_bcount) <=
-		      be64_to_cpu(d->d_blk_softlimit))) &&
-		    (!d->d_blk_hardlimit ||
-		    (be64_to_cpu(d->d_bcount) <=
-		     be64_to_cpu(d->d_blk_hardlimit)))) {
-			d->d_btimer = 0;
-		}
-	}
-
-	if (!d->d_itimer) {
-		if ((d->d_ino_softlimit &&
-		     (be64_to_cpu(d->d_icount) >
-		      be64_to_cpu(d->d_ino_softlimit))) ||
-		    (d->d_ino_hardlimit &&
-		     (be64_to_cpu(d->d_icount) >
-		      be64_to_cpu(d->d_ino_hardlimit)))) {
-			d->d_itimer = cpu_to_be32(ktime_get_real_seconds() +
-					defq->itimelimit);
-		} else {
-			d->d_iwarns = 0;
-		}
-	} else {
-		if ((!d->d_ino_softlimit ||
-		     (be64_to_cpu(d->d_icount) <=
-		      be64_to_cpu(d->d_ino_softlimit)))  &&
-		    (!d->d_ino_hardlimit ||
-		     (be64_to_cpu(d->d_icount) <=
-		      be64_to_cpu(d->d_ino_hardlimit)))) {
-			d->d_itimer = 0;
-		}
-	}
-
-	if (!d->d_rtbtimer) {
-		if ((d->d_rtb_softlimit &&
-		     (be64_to_cpu(d->d_rtbcount) >
-		      be64_to_cpu(d->d_rtb_softlimit))) ||
-		    (d->d_rtb_hardlimit &&
-		     (be64_to_cpu(d->d_rtbcount) >
-		      be64_to_cpu(d->d_rtb_hardlimit)))) {
-			d->d_rtbtimer = cpu_to_be32(ktime_get_real_seconds() +
-					defq->rtbtimelimit);
-		} else {
-			d->d_rtbwarns = 0;
-		}
-	} else {
-		if ((!d->d_rtb_softlimit ||
-		     (be64_to_cpu(d->d_rtbcount) <=
-		      be64_to_cpu(d->d_rtb_softlimit))) &&
-		    (!d->d_rtb_hardlimit ||
-		     (be64_to_cpu(d->d_rtbcount) <=
-		      be64_to_cpu(d->d_rtb_hardlimit)))) {
-			d->d_rtbtimer = 0;
-		}
-	}
+	xfs_qm_adjust_res_timer(&dq->q_blk, &defq->blk);
+	xfs_qm_adjust_res_timer(&dq->q_ino, &defq->ino);
+	xfs_qm_adjust_res_timer(&dq->q_rtb, &defq->rtb);
 }
 
 /*
@@ -213,7 +158,7 @@ xfs_qm_init_dquot_blk(
 	struct xfs_trans	*tp,
 	struct xfs_mount	*mp,
 	xfs_dqid_t		id,
-	uint			type,
+	xfs_dqtype_t		type,
 	struct xfs_buf		*bp)
 {
 	struct xfs_quotainfo	*q = mp->m_quotainfo;
@@ -226,6 +171,24 @@ xfs_qm_init_dquot_blk(
 	ASSERT(tp);
 	ASSERT(xfs_buf_islocked(bp));
 
+	switch (type) {
+	case XFS_DQTYPE_USER:
+		qflag = XFS_UQUOTA_CHKD;
+		blftype = XFS_BLF_UDQUOT_BUF;
+		break;
+	case XFS_DQTYPE_PROJ:
+		qflag = XFS_PQUOTA_CHKD;
+		blftype = XFS_BLF_PDQUOT_BUF;
+		break;
+	case XFS_DQTYPE_GROUP:
+		qflag = XFS_GQUOTA_CHKD;
+		blftype = XFS_BLF_GDQUOT_BUF;
+		break;
+	default:
+		ASSERT(0);
+		return;
+	}
+
 	d = bp->b_addr;
 
 	/*
@@ -237,7 +200,7 @@ xfs_qm_init_dquot_blk(
 		d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
 		d->dd_diskdq.d_version = XFS_DQUOT_VERSION;
 		d->dd_diskdq.d_id = cpu_to_be32(curid);
-		d->dd_diskdq.d_flags = type;
+		d->dd_diskdq.d_type = type;
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {
 			uuid_copy(&d->dd_uuid, &mp->m_sb.sb_meta_uuid);
 			xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk),
@@ -245,17 +208,6 @@ xfs_qm_init_dquot_blk(
 		}
 	}
 
-	if (type & XFS_DQ_USER) {
-		qflag = XFS_UQUOTA_CHKD;
-		blftype = XFS_BLF_UDQUOT_BUF;
-	} else if (type & XFS_DQ_PROJ) {
-		qflag = XFS_PQUOTA_CHKD;
-		blftype = XFS_BLF_PDQUOT_BUF;
-	} else {
-		qflag = XFS_GQUOTA_CHKD;
-		blftype = XFS_BLF_GDQUOT_BUF;
-	}
-
 	xfs_trans_dquot_buf(tp, bp, blftype);
 
 	/*
@@ -290,8 +242,8 @@ xfs_dquot_set_prealloc_limits(struct xfs_dquot *dqp)
 {
 	uint64_t space;
 
-	dqp->q_prealloc_hi_wmark = be64_to_cpu(dqp->q_core.d_blk_hardlimit);
-	dqp->q_prealloc_lo_wmark = be64_to_cpu(dqp->q_core.d_blk_softlimit);
+	dqp->q_prealloc_hi_wmark = dqp->q_blk.hardlimit;
+	dqp->q_prealloc_lo_wmark = dqp->q_blk.softlimit;
 	if (!dqp->q_prealloc_lo_wmark) {
 		dqp->q_prealloc_lo_wmark = dqp->q_prealloc_hi_wmark;
 		do_div(dqp->q_prealloc_lo_wmark, 100);
@@ -321,14 +273,15 @@ xfs_dquot_disk_alloc(
 	struct xfs_trans	*tp = *tpp;
 	struct xfs_mount	*mp = tp->t_mountp;
 	struct xfs_buf		*bp;
-	struct xfs_inode	*quotip = xfs_quota_inode(mp, dqp->dq_flags);
+	xfs_dqtype_t		qtype = xfs_dquot_type(dqp);
+	struct xfs_inode	*quotip = xfs_quota_inode(mp, qtype);
 	int			nmaps = 1;
 	int			error;
 
 	trace_xfs_dqalloc(dqp);
 
 	xfs_ilock(quotip, XFS_ILOCK_EXCL);
-	if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
+	if (!xfs_this_quota_on(dqp->q_mount, qtype)) {
 		/*
 		 * Return if this type of quotas is turned off while we didn't
 		 * have an inode lock
@@ -365,8 +318,7 @@ xfs_dquot_disk_alloc(
 	 * Make a chunk of dquots out of this buffer and log
 	 * the entire thing.
 	 */
-	xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id),
-			      dqp->dq_flags & XFS_DQ_ALLTYPES, bp);
+	xfs_qm_init_dquot_blk(tp, mp, dqp->q_id, qtype, bp);
 	xfs_buf_set_ref(bp, XFS_DQUOT_REF);
 
 	/*
@@ -413,13 +365,14 @@ xfs_dquot_disk_read(
 {
 	struct xfs_bmbt_irec	map;
 	struct xfs_buf		*bp;
-	struct xfs_inode	*quotip = xfs_quota_inode(mp, dqp->dq_flags);
+	xfs_dqtype_t		qtype = xfs_dquot_type(dqp);
+	struct xfs_inode	*quotip = xfs_quota_inode(mp, qtype);
 	uint			lock_mode;
 	int			nmaps = 1;
 	int			error;
 
 	lock_mode = xfs_ilock_data_map_shared(quotip);
-	if (!xfs_this_quota_on(mp, dqp->dq_flags)) {
+	if (!xfs_this_quota_on(mp, qtype)) {
 		/*
 		 * Return if this type of quotas is turned off while we
 		 * didn't have the quota inode lock.
@@ -471,14 +424,14 @@ STATIC struct xfs_dquot *
 xfs_dquot_alloc(
 	struct xfs_mount	*mp,
 	xfs_dqid_t		id,
-	uint			type)
+	xfs_dqtype_t		type)
 {
 	struct xfs_dquot	*dqp;
 
-	dqp = kmem_zone_zalloc(xfs_qm_dqzone, 0);
+	dqp = kmem_cache_zalloc(xfs_qm_dqzone, GFP_KERNEL | __GFP_NOFAIL);
 
-	dqp->dq_flags = type;
-	dqp->q_core.d_id = cpu_to_be32(id);
+	dqp->q_type = type;
+	dqp->q_id = id;
 	dqp->q_mount = mp;
 	INIT_LIST_HEAD(&dqp->q_lru);
 	mutex_init(&dqp->q_qlock);
@@ -503,13 +456,13 @@ xfs_dquot_alloc(
 	 * quotas.
 	 */
 	switch (type) {
-	case XFS_DQ_USER:
+	case XFS_DQTYPE_USER:
 		/* uses the default lock class */
 		break;
-	case XFS_DQ_GROUP:
+	case XFS_DQTYPE_GROUP:
 		lockdep_set_class(&dqp->q_qlock, &xfs_dquot_group_class);
 		break;
-	case XFS_DQ_PROJ:
+	case XFS_DQTYPE_PROJ:
 		lockdep_set_class(&dqp->q_qlock, &xfs_dquot_project_class);
 		break;
 	default:
@@ -524,26 +477,91 @@ xfs_dquot_alloc(
 }
 
 /* Copy the in-core quota fields in from the on-disk buffer. */
-STATIC void
+STATIC int
 xfs_dquot_from_disk(
 	struct xfs_dquot	*dqp,
 	struct xfs_buf		*bp)
 {
 	struct xfs_disk_dquot	*ddqp = bp->b_addr + dqp->q_bufoffset;
 
+	/*
+	 * Ensure that we got the type and ID we were looking for.
+	 * Everything else was checked by the dquot buffer verifier.
+	 */
+	if ((ddqp->d_type & XFS_DQTYPE_REC_MASK) != xfs_dquot_type(dqp) ||
+	    be32_to_cpu(ddqp->d_id) != dqp->q_id) {
+		xfs_alert_tag(bp->b_mount, XFS_PTAG_VERIFIER_ERROR,
+			  "Metadata corruption detected at %pS, quota %u",
+			  __this_address, dqp->q_id);
+		xfs_alert(bp->b_mount, "Unmount and run xfs_repair");
+		return -EFSCORRUPTED;
+	}
+
 	/* copy everything from disk dquot to the incore dquot */
-	memcpy(&dqp->q_core, ddqp, sizeof(struct xfs_disk_dquot));
+	dqp->q_type = ddqp->d_type;
+	dqp->q_blk.hardlimit = be64_to_cpu(ddqp->d_blk_hardlimit);
+	dqp->q_blk.softlimit = be64_to_cpu(ddqp->d_blk_softlimit);
+	dqp->q_ino.hardlimit = be64_to_cpu(ddqp->d_ino_hardlimit);
+	dqp->q_ino.softlimit = be64_to_cpu(ddqp->d_ino_softlimit);
+	dqp->q_rtb.hardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit);
+	dqp->q_rtb.softlimit = be64_to_cpu(ddqp->d_rtb_softlimit);
+
+	dqp->q_blk.count = be64_to_cpu(ddqp->d_bcount);
+	dqp->q_ino.count = be64_to_cpu(ddqp->d_icount);
+	dqp->q_rtb.count = be64_to_cpu(ddqp->d_rtbcount);
+
+	dqp->q_blk.warnings = be16_to_cpu(ddqp->d_bwarns);
+	dqp->q_ino.warnings = be16_to_cpu(ddqp->d_iwarns);
+	dqp->q_rtb.warnings = be16_to_cpu(ddqp->d_rtbwarns);
+
+	dqp->q_blk.timer = be32_to_cpu(ddqp->d_btimer);
+	dqp->q_ino.timer = be32_to_cpu(ddqp->d_itimer);
+	dqp->q_rtb.timer = be32_to_cpu(ddqp->d_rtbtimer);
 
 	/*
 	 * Reservation counters are defined as reservation plus current usage
 	 * to avoid having to add every time.
 	 */
-	dqp->q_res_bcount = be64_to_cpu(ddqp->d_bcount);
-	dqp->q_res_icount = be64_to_cpu(ddqp->d_icount);
-	dqp->q_res_rtbcount = be64_to_cpu(ddqp->d_rtbcount);
+	dqp->q_blk.reserved = dqp->q_blk.count;
+	dqp->q_ino.reserved = dqp->q_ino.count;
+	dqp->q_rtb.reserved = dqp->q_rtb.count;
 
 	/* initialize the dquot speculative prealloc thresholds */
 	xfs_dquot_set_prealloc_limits(dqp);
+	return 0;
+}
+
+/* Copy the in-core quota fields into the on-disk buffer. */
+void
+xfs_dquot_to_disk(
+	struct xfs_disk_dquot	*ddqp,
+	struct xfs_dquot	*dqp)
+{
+	ddqp->d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
+	ddqp->d_version = XFS_DQUOT_VERSION;
+	ddqp->d_type = dqp->q_type;
+	ddqp->d_id = cpu_to_be32(dqp->q_id);
+	ddqp->d_pad0 = 0;
+	ddqp->d_pad = 0;
+
+	ddqp->d_blk_hardlimit = cpu_to_be64(dqp->q_blk.hardlimit);
+	ddqp->d_blk_softlimit = cpu_to_be64(dqp->q_blk.softlimit);
+	ddqp->d_ino_hardlimit = cpu_to_be64(dqp->q_ino.hardlimit);
+	ddqp->d_ino_softlimit = cpu_to_be64(dqp->q_ino.softlimit);
+	ddqp->d_rtb_hardlimit = cpu_to_be64(dqp->q_rtb.hardlimit);
+	ddqp->d_rtb_softlimit = cpu_to_be64(dqp->q_rtb.softlimit);
+
+	ddqp->d_bcount = cpu_to_be64(dqp->q_blk.count);
+	ddqp->d_icount = cpu_to_be64(dqp->q_ino.count);
+	ddqp->d_rtbcount = cpu_to_be64(dqp->q_rtb.count);
+
+	ddqp->d_bwarns = cpu_to_be16(dqp->q_blk.warnings);
+	ddqp->d_iwarns = cpu_to_be16(dqp->q_ino.warnings);
+	ddqp->d_rtbwarns = cpu_to_be16(dqp->q_rtb.warnings);
+
+	ddqp->d_btimer = cpu_to_be32(dqp->q_blk.timer);
+	ddqp->d_itimer = cpu_to_be32(dqp->q_ino.timer);
+	ddqp->d_rtbtimer = cpu_to_be32(dqp->q_rtb.timer);
 }
 
 /* Allocate and initialize the dquot buffer for this in-core dquot. */
@@ -592,7 +610,7 @@ static int
 xfs_qm_dqread(
 	struct xfs_mount	*mp,
 	xfs_dqid_t		id,
-	uint			type,
+	xfs_dqtype_t		type,
 	bool			can_alloc,
 	struct xfs_dquot	**dqpp)
 {
@@ -617,9 +635,11 @@ xfs_qm_dqread(
 	 * further.
 	 */
 	ASSERT(xfs_buf_islocked(bp));
-	xfs_dquot_from_disk(dqp, bp);
-
+	error = xfs_dquot_from_disk(dqp, bp);
 	xfs_buf_relse(bp);
+	if (error)
+		goto err;
+
 	*dqpp = dqp;
 	return error;
 
@@ -638,7 +658,7 @@ err:
 static int
 xfs_dq_get_next_id(
 	struct xfs_mount	*mp,
-	uint			type,
+	xfs_dqtype_t		type,
 	xfs_dqid_t		*id)
 {
 	struct xfs_inode	*quotip = xfs_quota_inode(mp, type);
@@ -706,7 +726,7 @@ restart:
 	}
 
 	xfs_dqlock(dqp);
-	if (dqp->dq_flags & XFS_DQ_FREEING) {
+	if (dqp->q_flags & XFS_DQFLAG_FREEING) {
 		xfs_dqunlock(dqp);
 		mutex_unlock(&qi->qi_tree_lock);
 		trace_xfs_dqget_freeing(dqp);
@@ -762,21 +782,21 @@ xfs_qm_dqget_cache_insert(
 static int
 xfs_qm_dqget_checks(
 	struct xfs_mount	*mp,
-	uint			type)
+	xfs_dqtype_t		type)
 {
 	if (WARN_ON_ONCE(!XFS_IS_QUOTA_RUNNING(mp)))
 		return -ESRCH;
 
 	switch (type) {
-	case XFS_DQ_USER:
+	case XFS_DQTYPE_USER:
 		if (!XFS_IS_UQUOTA_ON(mp))
 			return -ESRCH;
 		return 0;
-	case XFS_DQ_GROUP:
+	case XFS_DQTYPE_GROUP:
 		if (!XFS_IS_GQUOTA_ON(mp))
 			return -ESRCH;
 		return 0;
-	case XFS_DQ_PROJ:
+	case XFS_DQTYPE_PROJ:
 		if (!XFS_IS_PQUOTA_ON(mp))
 			return -ESRCH;
 		return 0;
@@ -794,7 +814,7 @@ int
 xfs_qm_dqget(
 	struct xfs_mount	*mp,
 	xfs_dqid_t		id,
-	uint			type,
+	xfs_dqtype_t		type,
 	bool			can_alloc,
 	struct xfs_dquot	**O_dqpp)
 {
@@ -844,7 +864,7 @@ int
 xfs_qm_dqget_uncached(
 	struct xfs_mount	*mp,
 	xfs_dqid_t		id,
-	uint			type,
+	xfs_dqtype_t		type,
 	struct xfs_dquot	**dqpp)
 {
 	int			error;
@@ -860,14 +880,14 @@ xfs_qm_dqget_uncached(
 xfs_dqid_t
 xfs_qm_id_for_quotatype(
 	struct xfs_inode	*ip,
-	uint			type)
+	xfs_dqtype_t		type)
 {
 	switch (type) {
-	case XFS_DQ_USER:
+	case XFS_DQTYPE_USER:
 		return i_uid_read(VFS_I(ip));
-	case XFS_DQ_GROUP:
+	case XFS_DQTYPE_GROUP:
 		return i_gid_read(VFS_I(ip));
-	case XFS_DQ_PROJ:
+	case XFS_DQTYPE_PROJ:
 		return ip->i_d.di_projid;
 	}
 	ASSERT(0);
@@ -882,7 +902,7 @@ xfs_qm_id_for_quotatype(
 int
 xfs_qm_dqget_inode(
 	struct xfs_inode	*ip,
-	uint			type,
+	xfs_dqtype_t		type,
 	bool			can_alloc,
 	struct xfs_dquot	**O_dqpp)
 {
@@ -968,7 +988,7 @@ int
 xfs_qm_dqget_next(
 	struct xfs_mount	*mp,
 	xfs_dqid_t		id,
-	uint			type,
+	xfs_dqtype_t		type,
 	struct xfs_dquot	**dqpp)
 {
 	struct xfs_dquot	*dqp;
@@ -1048,9 +1068,8 @@ xfs_qm_dqrele(
  * from the AIL if it has not been re-logged, and unlocking the dquot's
  * flush lock. This behavior is very similar to that of inodes..
  */
-STATIC void
+static void
 xfs_qm_dqflush_done(
-	struct xfs_buf		*bp,
 	struct xfs_log_item	*lip)
 {
 	struct xfs_dq_logitem	*qip = (struct xfs_dq_logitem *)lip;
@@ -1071,16 +1090,12 @@ xfs_qm_dqflush_done(
 	     test_bit(XFS_LI_FAILED, &lip->li_flags))) {
 
 		spin_lock(&ailp->ail_lock);
+		xfs_clear_li_failed(lip);
 		if (lip->li_lsn == qip->qli_flush_lsn) {
 			/* xfs_ail_update_finish() drops the AIL lock */
 			tail_lsn = xfs_ail_delete_one(ailp, lip);
 			xfs_ail_update_finish(ailp, tail_lsn);
 		} else {
-			/*
-			 * Clear the failed state since we are about to drop the
-			 * flush lock
-			 */
-			xfs_clear_li_failed(lip);
 			spin_unlock(&ailp->ail_lock);
 		}
 	}
@@ -1091,6 +1106,48 @@ xfs_qm_dqflush_done(
 	xfs_dqfunlock(dqp);
 }
 
+void
+xfs_dquot_done(
+	struct xfs_buf		*bp)
+{
+	struct xfs_log_item	*lip, *n;
+
+	list_for_each_entry_safe(lip, n, &bp->b_li_list, li_bio_list) {
+		list_del_init(&lip->li_bio_list);
+		xfs_qm_dqflush_done(lip);
+	}
+}
+
+/* Check incore dquot for errors before we flush. */
+static xfs_failaddr_t
+xfs_qm_dqflush_check(
+	struct xfs_dquot	*dqp)
+{
+	xfs_dqtype_t		type = xfs_dquot_type(dqp);
+
+	if (type != XFS_DQTYPE_USER &&
+	    type != XFS_DQTYPE_GROUP &&
+	    type != XFS_DQTYPE_PROJ)
+		return __this_address;
+
+	if (dqp->q_id == 0)
+		return NULL;
+
+	if (dqp->q_blk.softlimit && dqp->q_blk.count > dqp->q_blk.softlimit &&
+	    !dqp->q_blk.timer)
+		return __this_address;
+
+	if (dqp->q_ino.softlimit && dqp->q_ino.count > dqp->q_ino.softlimit &&
+	    !dqp->q_ino.timer)
+		return __this_address;
+
+	if (dqp->q_rtb.softlimit && dqp->q_rtb.count > dqp->q_rtb.softlimit &&
+	    !dqp->q_rtb.timer)
+		return __this_address;
+
+	return NULL;
+}
+
 /*
  * Write a modified dquot to disk.
  * The dquot must be locked and the flush lock too taken by caller.
@@ -1107,8 +1164,7 @@ xfs_qm_dqflush(
 	struct xfs_mount	*mp = dqp->q_mount;
 	struct xfs_log_item	*lip = &dqp->q_logitem.qli_item;
 	struct xfs_buf		*bp;
-	struct xfs_dqblk	*dqb;
-	struct xfs_disk_dquot	*ddqp;
+	struct xfs_dqblk	*dqblk;
 	xfs_failaddr_t		fa;
 	int			error;
 
@@ -1132,30 +1188,23 @@ xfs_qm_dqflush(
 	if (error)
 		goto out_abort;
 
-	/*
-	 * Calculate the location of the dquot inside the buffer.
-	 */
-	dqb = bp->b_addr + dqp->q_bufoffset;
-	ddqp = &dqb->dd_diskdq;
-
-	/* sanity check the in-core structure before we flush */
-	fa = xfs_dquot_verify(mp, &dqp->q_core, be32_to_cpu(dqp->q_core.d_id),
-			      0);
+	fa = xfs_qm_dqflush_check(dqp);
 	if (fa) {
 		xfs_alert(mp, "corrupt dquot ID 0x%x in memory at %pS",
-				be32_to_cpu(dqp->q_core.d_id), fa);
+				dqp->q_id, fa);
 		xfs_buf_relse(bp);
 		error = -EFSCORRUPTED;
 		goto out_abort;
 	}
 
-	/* This is the only portion of data that needs to persist */
-	memcpy(ddqp, &dqp->q_core, sizeof(struct xfs_disk_dquot));
+	/* Flush the incore dquot to the ondisk buffer. */
+	dqblk = bp->b_addr + dqp->q_bufoffset;
+	xfs_dquot_to_disk(&dqblk->dd_diskdq, dqp);
 
 	/*
 	 * Clear the dirty field and remember the flush lsn for later use.
 	 */
-	dqp->dq_flags &= ~XFS_DQ_DIRTY;
+	dqp->q_flags &= ~XFS_DQFLAG_DIRTY;
 
 	xfs_trans_ail_copy_lsn(mp->m_ail, &dqp->q_logitem.qli_flush_lsn,
 					&dqp->q_logitem.qli_item.li_lsn);
@@ -1170,17 +1219,17 @@ xfs_qm_dqflush(
 	 * of a dquot without an up-to-date CRC getting to disk.
 	 */
 	if (xfs_sb_version_hascrc(&mp->m_sb)) {
-		dqb->dd_lsn = cpu_to_be64(dqp->q_logitem.qli_item.li_lsn);
-		xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk),
+		dqblk->dd_lsn = cpu_to_be64(dqp->q_logitem.qli_item.li_lsn);
+		xfs_update_cksum((char *)dqblk, sizeof(struct xfs_dqblk),
 				 XFS_DQUOT_CRC_OFF);
 	}
 
 	/*
-	 * Attach an iodone routine so that we can remove this dquot from the
-	 * AIL and release the flush lock once the dquot is synced to disk.
+	 * Attach the dquot to the buffer so that we can remove this dquot from
+	 * the AIL and release the flush lock once the dquot is synced to disk.
 	 */
-	xfs_buf_attach_iodone(bp, xfs_qm_dqflush_done,
-				  &dqp->q_logitem.qli_item);
+	bp->b_flags |= _XBF_DQUOTS;
+	list_add_tail(&dqp->q_logitem.qli_item.li_bio_list, &bp->b_li_list);
 
 	/*
 	 * If the buffer is pinned then push on the log so we won't
@@ -1196,7 +1245,7 @@ xfs_qm_dqflush(
 	return 0;
 
 out_abort:
-	dqp->dq_flags &= ~XFS_DQ_DIRTY;
+	dqp->q_flags &= ~XFS_DQFLAG_DIRTY;
 	xfs_trans_ail_delete(lip, 0);
 	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
 out_unlock:
@@ -1217,8 +1266,7 @@ xfs_dqlock2(
 {
 	if (d1 && d2) {
 		ASSERT(d1 != d2);
-		if (be32_to_cpu(d1->q_core.d_id) >
-		    be32_to_cpu(d2->q_core.d_id)) {
+		if (d1->q_id > d2->q_id) {
 			mutex_lock(&d2->q_qlock);
 			mutex_lock_nested(&d1->q_qlock, XFS_QLOCK_NESTED);
 		} else {
@@ -1270,7 +1318,7 @@ xfs_qm_exit(void)
 int
 xfs_qm_dqiterate(
 	struct xfs_mount	*mp,
-	uint			dqtype,
+	xfs_dqtype_t		type,
 	xfs_qm_dqiterate_fn	iter_fn,
 	void			*priv)
 {
@@ -1279,16 +1327,15 @@ xfs_qm_dqiterate(
 	int			error;
 
 	do {
-		error = xfs_qm_dqget_next(mp, id, dqtype, &dq);
+		error = xfs_qm_dqget_next(mp, id, type, &dq);
 		if (error == -ENOENT)
 			return 0;
 		if (error)
 			return error;
 
-		error = iter_fn(dq, dqtype, priv);
-		id = be32_to_cpu(dq->q_core.d_id);
+		error = iter_fn(dq, type, priv);
+		id = dq->q_id;
 		xfs_qm_dqput(dq);
-		id++;
 	} while (error == 0 && id != 0);
 
 	return error;