1 files changed, 90 insertions, 97 deletions

diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 61ee110b47d7..23ba84ec919a 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -134,6 +134,7 @@ xfs_qm_dqpurge(
 
 	dqp->q_flags |= XFS_DQFLAG_FREEING;
 
+	xfs_qm_dqunpin_wait(dqp);
 	xfs_dqflock(dqp);
 
 	/*
@@ -148,17 +149,29 @@ xfs_qm_dqpurge(
 		 * We don't care about getting disk errors here. We need
 		 * to purge this dquot anyway, so we go ahead regardless.
 		 */
-		error = xfs_qm_dqflush(dqp, &bp);
+		error = xfs_dquot_use_attached_buf(dqp, &bp);
+		if (error == -EAGAIN) {
+			xfs_dqfunlock(dqp);
+			dqp->q_flags &= ~XFS_DQFLAG_FREEING;
+			goto out_unlock;
+		}
+		if (!bp)
+			goto out_funlock;
+
+		/*
+		 * dqflush completes dqflock on error, and the bwrite ioend
+		 * does it on success.
+		 */
+		error = xfs_qm_dqflush(dqp, bp);
 		if (!error) {
 			error = xfs_bwrite(bp);
 			xfs_buf_relse(bp);
-		} else if (error == -EAGAIN) {
-			dqp->q_flags &= ~XFS_DQFLAG_FREEING;
-			goto out_unlock;
 		}
 		xfs_dqflock(dqp);
 	}
+	xfs_dquot_detach_buf(dqp);
 
+out_funlock:
 	ASSERT(atomic_read(&dqp->q_pincount) == 0);
 	ASSERT(xlog_is_shutdown(dqp->q_logitem.qli_item.li_log) ||
 		!test_bit(XFS_LI_IN_AIL, &dqp->q_logitem.qli_item.li_flags));
@@ -218,10 +231,10 @@ xfs_qm_unmount_rt(
 
 	if (!rtg)
 		return;
-	if (rtg->rtg_inodes[XFS_RTGI_BITMAP])
-		xfs_qm_dqdetach(rtg->rtg_inodes[XFS_RTGI_BITMAP]);
-	if (rtg->rtg_inodes[XFS_RTGI_SUMMARY])
-		xfs_qm_dqdetach(rtg->rtg_inodes[XFS_RTGI_SUMMARY]);
+	if (rtg_bitmap(rtg))
+		xfs_qm_dqdetach(rtg_bitmap(rtg));
+	if (rtg_summary(rtg))
+		xfs_qm_dqdetach(rtg_summary(rtg));
 	xfs_rtgroup_rele(rtg);
 }
 
@@ -241,6 +254,10 @@ xfs_qm_destroy_quotainos(
 		xfs_irele(qi->qi_pquotaip);
 		qi->qi_pquotaip = NULL;
 	}
+	if (qi->qi_dirip) {
+		xfs_irele(qi->qi_dirip);
+		qi->qi_dirip = NULL;
+	}
 }
 
 /*
@@ -412,6 +429,8 @@ void
 xfs_qm_dqdetach(
 	xfs_inode_t	*ip)
 {
+	if (xfs_is_metadir_inode(ip))
+		return;
 	if (!(ip->i_udquot || ip->i_gdquot || ip->i_pdquot))
 		return;
 
@@ -447,6 +466,7 @@ xfs_qm_dquot_isolate(
 	struct xfs_dquot	*dqp = container_of(item,
 						struct xfs_dquot, q_lru);
 	struct xfs_qm_isolate	*isol = arg;
+	enum lru_status		ret = LRU_SKIP;
 
 	if (!xfs_dqlock_nowait(dqp))
 		goto out_miss_busy;
@@ -460,6 +480,16 @@ xfs_qm_dquot_isolate(
 		goto out_miss_unlock;
 
 	/*
+	 * If the dquot is pinned or dirty, rotate it to the end of the LRU to
+	 * give some time for it to be cleaned before we try to isolate it
+	 * again.
+	 */
+	ret = LRU_ROTATE;
+	if (XFS_DQ_IS_DIRTY(dqp) || atomic_read(&dqp->q_pincount) > 0) {
+		goto out_miss_unlock;
+	}
+
+	/*
 	 * This dquot has acquired a reference in the meantime remove it from
 	 * the freelist and try again.
 	 */
@@ -474,30 +504,15 @@ xfs_qm_dquot_isolate(
 	}
 
 	/*
-	 * If the dquot is dirty, flush it. If it's already being flushed, just
-	 * skip it so there is time for the IO to complete before we try to
-	 * reclaim it again on the next LRU pass.
+	 * The dquot may still be under IO, in which case the flush lock will be
+	 * held. If we can't get the flush lock now, just skip over the dquot as
+	 * if it was dirty.
 	 */
 	if (!xfs_dqflock_nowait(dqp))
 		goto out_miss_unlock;
 
-	if (XFS_DQ_IS_DIRTY(dqp)) {
-		struct xfs_buf	*bp = NULL;
-		int		error;
-
-		trace_xfs_dqreclaim_dirty(dqp);
-
-		/* we have to drop the LRU lock to flush the dquot */
-		spin_unlock(&lru->lock);
-
-		error = xfs_qm_dqflush(dqp, &bp);
-		if (error)
-			goto out_unlock_dirty;
-
-		xfs_buf_delwri_queue(bp, &isol->buffers);
-		xfs_buf_relse(bp);
-		goto out_unlock_dirty;
-	}
+	ASSERT(!XFS_DQ_IS_DIRTY(dqp));
+	xfs_dquot_detach_buf(dqp);
 	xfs_dqfunlock(dqp);
 
 	/*
@@ -518,13 +533,7 @@ out_miss_unlock:
 out_miss_busy:
 	trace_xfs_dqreclaim_busy(dqp);
 	XFS_STATS_INC(dqp->q_mount, xs_qm_dqreclaim_misses);
-	return LRU_SKIP;
-
-out_unlock_dirty:
-	trace_xfs_dqreclaim_busy(dqp);
-	XFS_STATS_INC(dqp->q_mount, xs_qm_dqreclaim_misses);
-	xfs_dqunlock(dqp);
-	return LRU_RETRY;
+	return ret;
 }
 
 static unsigned long
@@ -646,17 +655,13 @@ xfs_qm_init_timelimits(
 static int
 xfs_qm_load_metadir_qinos(
 	struct xfs_mount	*mp,
-	struct xfs_quotainfo	*qi,
-	struct xfs_inode	**dpp)
+	struct xfs_quotainfo	*qi)
 {
 	struct xfs_trans	*tp;
 	int			error;
 
-	error = xfs_trans_alloc_empty(mp, &tp);
-	if (error)
-		return error;
-
-	error = xfs_dqinode_load_parent(tp, dpp);
+	tp = xfs_trans_alloc_empty(mp);
+	error = xfs_dqinode_load_parent(tp, &qi->qi_dirip);
 	if (error == -ENOENT) {
 		/* no quota dir directory, but we'll create one later */
 		error = 0;
@@ -666,21 +671,21 @@ xfs_qm_load_metadir_qinos(
 		goto out_trans;
 
 	if (XFS_IS_UQUOTA_ON(mp)) {
-		error = xfs_dqinode_load(tp, *dpp, XFS_DQTYPE_USER,
+		error = xfs_dqinode_load(tp, qi->qi_dirip, XFS_DQTYPE_USER,
 				&qi->qi_uquotaip);
 		if (error && error != -ENOENT)
 			goto out_trans;
 	}
 
 	if (XFS_IS_GQUOTA_ON(mp)) {
-		error = xfs_dqinode_load(tp, *dpp, XFS_DQTYPE_GROUP,
+		error = xfs_dqinode_load(tp, qi->qi_dirip, XFS_DQTYPE_GROUP,
 				&qi->qi_gquotaip);
 		if (error && error != -ENOENT)
 			goto out_trans;
 	}
 
 	if (XFS_IS_PQUOTA_ON(mp)) {
-		error = xfs_dqinode_load(tp, *dpp, XFS_DQTYPE_PROJ,
+		error = xfs_dqinode_load(tp, qi->qi_dirip, XFS_DQTYPE_PROJ,
 				&qi->qi_pquotaip);
 		if (error && error != -ENOENT)
 			goto out_trans;
@@ -696,34 +701,40 @@ out_trans:
 STATIC int
 xfs_qm_create_metadir_qinos(
 	struct xfs_mount	*mp,
-	struct xfs_quotainfo	*qi,
-	struct xfs_inode	**dpp)
+	struct xfs_quotainfo	*qi)
 {
 	int			error;
 
-	if (!*dpp) {
-		error = xfs_dqinode_mkdir_parent(mp, dpp);
+	if (!qi->qi_dirip) {
+		error = xfs_dqinode_mkdir_parent(mp, &qi->qi_dirip);
 		if (error && error != -EEXIST)
 			return error;
+		/*
+		 * If the /quotas dirent points to an inode that isn't
+		 * loadable, qi_dirip will be NULL but mkdir_parent will return
+		 * -EEXIST.  In this case the metadir is corrupt, so bail out.
+		 */
+		if (XFS_IS_CORRUPT(mp, qi->qi_dirip == NULL))
+			return -EFSCORRUPTED;
 	}
 
 	if (XFS_IS_UQUOTA_ON(mp) && !qi->qi_uquotaip) {
-		error = xfs_dqinode_metadir_create(*dpp, XFS_DQTYPE_USER,
-				&qi->qi_uquotaip);
+		error = xfs_dqinode_metadir_create(qi->qi_dirip,
+				XFS_DQTYPE_USER, &qi->qi_uquotaip);
 		if (error)
 			return error;
 	}
 
 	if (XFS_IS_GQUOTA_ON(mp) && !qi->qi_gquotaip) {
-		error = xfs_dqinode_metadir_create(*dpp, XFS_DQTYPE_GROUP,
-				&qi->qi_gquotaip);
+		error = xfs_dqinode_metadir_create(qi->qi_dirip,
+				XFS_DQTYPE_GROUP, &qi->qi_gquotaip);
 		if (error)
 			return error;
 	}
 
 	if (XFS_IS_PQUOTA_ON(mp) && !qi->qi_pquotaip) {
-		error = xfs_dqinode_metadir_create(*dpp, XFS_DQTYPE_PROJ,
-				&qi->qi_pquotaip);
+		error = xfs_dqinode_metadir_create(qi->qi_dirip,
+				XFS_DQTYPE_PROJ, &qi->qi_pquotaip);
 		if (error)
 			return error;
 	}
@@ -768,7 +779,6 @@ xfs_qm_init_metadir_qinos(
 	struct xfs_mount	*mp)
 {
 	struct xfs_quotainfo	*qi = mp->m_quotainfo;
-	struct xfs_inode	*dp = NULL;
 	int			error;
 
 	if (!xfs_has_quota(mp)) {
@@ -777,20 +787,22 @@ xfs_qm_init_metadir_qinos(
 			return error;
 	}
 
-	error = xfs_qm_load_metadir_qinos(mp, qi, &dp);
+	error = xfs_qm_load_metadir_qinos(mp, qi);
 	if (error)
 		goto out_err;
 
-	error = xfs_qm_create_metadir_qinos(mp, qi, &dp);
+	error = xfs_qm_create_metadir_qinos(mp, qi);
 	if (error)
 		goto out_err;
 
-	xfs_irele(dp);
+	/* The only user of the quota dir inode is online fsck */
+#if !IS_ENABLED(CONFIG_XFS_ONLINE_SCRUB)
+	xfs_irele(qi->qi_dirip);
+	qi->qi_dirip = NULL;
+#endif
 	return 0;
 out_err:
 	xfs_qm_destroy_quotainos(mp->m_quotainfo);
-	if (dp)
-		xfs_irele(dp);
 	return error;
 }
 
@@ -1304,6 +1316,10 @@ xfs_qm_quotacheck_dqadjust(
 		return error;
 	}
 
+	error = xfs_dquot_attach_buf(NULL, dqp);
+	if (error)
+		return error;
+
 	trace_xfs_dqadjust(dqp);
 
 	/*
@@ -1446,7 +1462,6 @@ xfs_qm_flush_one(
 	struct xfs_dquot	*dqp,
 	void			*data)
 {
-	struct xfs_mount	*mp = dqp->q_mount;
 	struct list_head	*buffer_list = data;
 	struct xfs_buf		*bp = NULL;
 	int			error = 0;
@@ -1457,40 +1472,20 @@ xfs_qm_flush_one(
 	if (!XFS_DQ_IS_DIRTY(dqp))
 		goto out_unlock;
 
-	/*
-	 * The only way the dquot is already flush locked by the time quotacheck
-	 * gets here is if reclaim flushed it before the dqadjust walk dirtied
-	 * it for the final time. Quotacheck collects all dquot bufs in the
-	 * local delwri queue before dquots are dirtied, so reclaim can't have
-	 * possibly queued it for I/O. The only way out is to push the buffer to
-	 * cycle the flush lock.
-	 */
-	if (!xfs_dqflock_nowait(dqp)) {
-		/* buf is pinned in-core by delwri list */
-		error = xfs_buf_incore(mp->m_ddev_targp, dqp->q_blkno,
-				mp->m_quotainfo->qi_dqchunklen, 0, &bp);
-		if (error)
-			goto out_unlock;
-
-		if (!(bp->b_flags & _XBF_DELWRI_Q)) {
-			error = -EAGAIN;
-			xfs_buf_relse(bp);
-			goto out_unlock;
-		}
-		xfs_buf_unlock(bp);
-
-		xfs_buf_delwri_pushbuf(bp, buffer_list);
-		xfs_buf_rele(bp);
-
-		error = -EAGAIN;
-		goto out_unlock;
-	}
+	xfs_qm_dqunpin_wait(dqp);
+	xfs_dqflock(dqp);
 
-	error = xfs_qm_dqflush(dqp, &bp);
+	error = xfs_dquot_use_attached_buf(dqp, &bp);
 	if (error)
 		goto out_unlock;
+	if (!bp) {
+		error = -EFSCORRUPTED;
+		goto out_unlock;
+	}
 
-	xfs_buf_delwri_queue(bp, buffer_list);
+	error = xfs_qm_dqflush(dqp, bp);
+	if (!error)
+		xfs_buf_delwri_queue(bp, buffer_list);
 	xfs_buf_relse(bp);
 out_unlock:
 	xfs_dqunlock(dqp);
@@ -1665,7 +1660,8 @@ xfs_qm_mount_quotas(
 	 * immediately.  We only support rtquota if rtgroups are enabled to
 	 * avoid problems with older kernels.
 	 */
-	if (mp->m_sb.sb_rextents && !xfs_has_rtgroups(mp)) {
+	if (mp->m_sb.sb_rextents &&
+	    (!xfs_has_rtgroups(mp) || xfs_has_zoned(mp))) {
 		xfs_notice(mp, "Cannot turn on quotas for realtime filesystem");
 		mp->m_qflags = 0;
 		goto write_changes;
@@ -1756,10 +1752,7 @@ xfs_qm_qino_load(
 	struct xfs_inode	*dp = NULL;
 	int			error;
 
-	error = xfs_trans_alloc_empty(mp, &tp);
-	if (error)
-		return error;
-
+	tp = xfs_trans_alloc_empty(mp);
 	if (xfs_has_metadir(mp)) {
 		error = xfs_dqinode_load_parent(tp, &dp);
 		if (error)