1 files changed, 169 insertions, 112 deletions

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 516162be1398..f9e9ffe6fb46 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -149,13 +149,27 @@ xfs_alloc_compute_aligned(
 {
 	xfs_agblock_t	bno;
 	xfs_extlen_t	len;
+	xfs_extlen_t	diff;
 
 	/* Trim busy sections out of found extent */
 	xfs_extent_busy_trim(args, foundbno, foundlen, &bno, &len);
 
+	/*
+	 * If we have a largish extent that happens to start before min_agbno,
+	 * see if we can shift it into range...
+	 */
+	if (bno < args->min_agbno && bno + len > args->min_agbno) {
+		diff = args->min_agbno - bno;
+		if (len > diff) {
+			bno += diff;
+			len -= diff;
+		}
+	}
+
 	if (args->alignment > 1 && len >= args->minlen) {
 		xfs_agblock_t	aligned_bno = roundup(bno, args->alignment);
-		xfs_extlen_t	diff = aligned_bno - bno;
+
+		diff = aligned_bno - bno;
 
 		*resbno = aligned_bno;
 		*reslen = diff >= len ? 0 : len - diff;
@@ -795,9 +809,13 @@ xfs_alloc_find_best_extent(
 		 * The good extent is closer than this one.
 		 */
 		if (!dir) {
+			if (*sbnoa > args->max_agbno)
+				goto out_use_good;
 			if (*sbnoa >= args->agbno + gdiff)
 				goto out_use_good;
 		} else {
+			if (*sbnoa < args->min_agbno)
+				goto out_use_good;
 			if (*sbnoa <= args->agbno - gdiff)
 				goto out_use_good;
 		}
@@ -884,6 +902,17 @@ xfs_alloc_ag_vextent_near(
 	dofirst = prandom_u32() & 1;
 #endif
 
+	/* handle unitialized agbno range so caller doesn't have to */
+	if (!args->min_agbno && !args->max_agbno)
+		args->max_agbno = args->mp->m_sb.sb_agblocks - 1;
+	ASSERT(args->min_agbno <= args->max_agbno);
+
+	/* clamp agbno to the range if it's outside */
+	if (args->agbno < args->min_agbno)
+		args->agbno = args->min_agbno;
+	if (args->agbno > args->max_agbno)
+		args->agbno = args->max_agbno;
+
 restart:
 	bno_cur_lt = NULL;
 	bno_cur_gt = NULL;
@@ -976,6 +1005,8 @@ restart:
 						  &ltbnoa, &ltlena);
 			if (ltlena < args->minlen)
 				continue;
+			if (ltbnoa < args->min_agbno || ltbnoa > args->max_agbno)
+				continue;
 			args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
 			xfs_alloc_fix_len(args);
 			ASSERT(args->len >= args->minlen);
@@ -1096,11 +1127,11 @@ restart:
 			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
 			xfs_alloc_compute_aligned(args, ltbno, ltlen,
 						  &ltbnoa, &ltlena);
-			if (ltlena >= args->minlen)
+			if (ltlena >= args->minlen && ltbnoa >= args->min_agbno)
 				break;
 			if ((error = xfs_btree_decrement(bno_cur_lt, 0, &i)))
 				goto error0;
-			if (!i) {
+			if (!i || ltbnoa < args->min_agbno) {
 				xfs_btree_del_cursor(bno_cur_lt,
 						     XFS_BTREE_NOERROR);
 				bno_cur_lt = NULL;
@@ -1112,11 +1143,11 @@ restart:
 			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
 			xfs_alloc_compute_aligned(args, gtbno, gtlen,
 						  &gtbnoa, &gtlena);
-			if (gtlena >= args->minlen)
+			if (gtlena >= args->minlen && gtbnoa <= args->max_agbno)
 				break;
 			if ((error = xfs_btree_increment(bno_cur_gt, 0, &i)))
 				goto error0;
-			if (!i) {
+			if (!i || gtbnoa > args->max_agbno) {
 				xfs_btree_del_cursor(bno_cur_gt,
 						     XFS_BTREE_NOERROR);
 				bno_cur_gt = NULL;
@@ -1216,6 +1247,7 @@ restart:
 	ASSERT(ltnew >= ltbno);
 	ASSERT(ltnew + rlen <= ltbnoa + ltlena);
 	ASSERT(ltnew + rlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
+	ASSERT(ltnew >= args->min_agbno && ltnew <= args->max_agbno);
 	args->agbno = ltnew;
 
 	if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno, ltlen,
@@ -1825,11 +1857,11 @@ xfs_alloc_compute_maxlevels(
 xfs_extlen_t
 xfs_alloc_longest_free_extent(
 	struct xfs_mount	*mp,
-	struct xfs_perag	*pag)
+	struct xfs_perag	*pag,
+	xfs_extlen_t		need)
 {
-	xfs_extlen_t		need, delta = 0;
+	xfs_extlen_t		delta = 0;
 
-	need = XFS_MIN_FREELIST_PAG(pag, mp);
 	if (need > pag->pagf_flcount)
 		delta = need - pag->pagf_flcount;
 
@@ -1838,131 +1870,150 @@ xfs_alloc_longest_free_extent(
 	return pag->pagf_flcount > 0 || pag->pagf_longest > 0;
 }
 
+unsigned int
+xfs_alloc_min_freelist(
+	struct xfs_mount	*mp,
+	struct xfs_perag	*pag)
+{
+	unsigned int		min_free;
+
+	/* space needed by-bno freespace btree */
+	min_free = min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_BNOi] + 1,
+				       mp->m_ag_maxlevels);
+	/* space needed by-size freespace btree */
+	min_free += min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_CNTi] + 1,
+				       mp->m_ag_maxlevels);
+
+	return min_free;
+}
+
+/*
+ * Check if the operation we are fixing up the freelist for should go ahead or
+ * not. If we are freeing blocks, we always allow it, otherwise the allocation
+ * is dependent on whether the size and shape of free space available will
+ * permit the requested allocation to take place.
+ */
+static bool
+xfs_alloc_space_available(
+	struct xfs_alloc_arg	*args,
+	xfs_extlen_t		min_free,
+	int			flags)
+{
+	struct xfs_perag	*pag = args->pag;
+	xfs_extlen_t		longest;
+	int			available;
+
+	if (flags & XFS_ALLOC_FLAG_FREEING)
+		return true;
+
+	/* do we have enough contiguous free space for the allocation? */
+	longest = xfs_alloc_longest_free_extent(args->mp, pag, min_free);
+	if ((args->minlen + args->alignment + args->minalignslop - 1) > longest)
+		return false;
+
+	/* do have enough free space remaining for the allocation? */
+	available = (int)(pag->pagf_freeblks + pag->pagf_flcount -
+			  min_free - args->total);
+	if (available < (int)args->minleft)
+		return false;
+
+	return true;
+}
+
 /*
  * Decide whether to use this allocation group for this allocation.
  * If so, fix up the btree freelist's size.
  */
 STATIC int			/* error */
 xfs_alloc_fix_freelist(
-	xfs_alloc_arg_t	*args,	/* allocation argument structure */
-	int		flags)	/* XFS_ALLOC_FLAG_... */
+	struct xfs_alloc_arg	*args,	/* allocation argument structure */
+	int			flags)	/* XFS_ALLOC_FLAG_... */
 {
-	xfs_buf_t	*agbp;	/* agf buffer pointer */
-	xfs_agf_t	*agf;	/* a.g. freespace structure pointer */
-	xfs_buf_t	*agflbp;/* agfl buffer pointer */
-	xfs_agblock_t	bno;	/* freelist block */
-	xfs_extlen_t	delta;	/* new blocks needed in freelist */
-	int		error;	/* error result code */
-	xfs_extlen_t	longest;/* longest extent in allocation group */
-	xfs_mount_t	*mp;	/* file system mount point structure */
-	xfs_extlen_t	need;	/* total blocks needed in freelist */
-	xfs_perag_t	*pag;	/* per-ag information structure */
-	xfs_alloc_arg_t	targs;	/* local allocation arguments */
-	xfs_trans_t	*tp;	/* transaction pointer */
-
-	mp = args->mp;
+	struct xfs_mount	*mp = args->mp;
+	struct xfs_perag	*pag = args->pag;
+	struct xfs_trans	*tp = args->tp;
+	struct xfs_buf		*agbp = NULL;
+	struct xfs_buf		*agflbp = NULL;
+	struct xfs_alloc_arg	targs;	/* local allocation arguments */
+	xfs_agblock_t		bno;	/* freelist block */
+	xfs_extlen_t		need;	/* total blocks needed in freelist */
+	int			error;
 
-	pag = args->pag;
-	tp = args->tp;
 	if (!pag->pagf_init) {
-		if ((error = xfs_alloc_read_agf(mp, tp, args->agno, flags,
-				&agbp)))
-			return error;
+		error = xfs_alloc_read_agf(mp, tp, args->agno, flags, &agbp);
+		if (error)
+			goto out_no_agbp;
 		if (!pag->pagf_init) {
 			ASSERT(flags & XFS_ALLOC_FLAG_TRYLOCK);
 			ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING));
-			args->agbp = NULL;
-			return 0;
+			goto out_agbp_relse;
 		}
-	} else
-		agbp = NULL;
+	}
 
 	/*
-	 * If this is a metadata preferred pag and we are user data
-	 * then try somewhere else if we are not being asked to
-	 * try harder at this point
+	 * If this is a metadata preferred pag and we are user data then try
+	 * somewhere else if we are not being asked to try harder at this
+	 * point
 	 */
 	if (pag->pagf_metadata && args->userdata &&
 	    (flags & XFS_ALLOC_FLAG_TRYLOCK)) {
 		ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING));
-		args->agbp = NULL;
-		return 0;
+		goto out_agbp_relse;
 	}
 
-	if (!(flags & XFS_ALLOC_FLAG_FREEING)) {
-		/*
-		 * If it looks like there isn't a long enough extent, or enough
-		 * total blocks, reject it.
-		 */
-		need = XFS_MIN_FREELIST_PAG(pag, mp);
-		longest = xfs_alloc_longest_free_extent(mp, pag);
-		if ((args->minlen + args->alignment + args->minalignslop - 1) >
-				longest ||
-		    ((int)(pag->pagf_freeblks + pag->pagf_flcount -
-			   need - args->total) < (int)args->minleft)) {
-			if (agbp)
-				xfs_trans_brelse(tp, agbp);
-			args->agbp = NULL;
-			return 0;
-		}
-	}
+	need = xfs_alloc_min_freelist(mp, pag);
+	if (!xfs_alloc_space_available(args, need, flags))
+		goto out_agbp_relse;
 
 	/*
 	 * Get the a.g. freespace buffer.
 	 * Can fail if we're not blocking on locks, and it's held.
 	 */
-	if (agbp == NULL) {
-		if ((error = xfs_alloc_read_agf(mp, tp, args->agno, flags,
-				&agbp)))
-			return error;
-		if (agbp == NULL) {
+	if (!agbp) {
+		error = xfs_alloc_read_agf(mp, tp, args->agno, flags, &agbp);
+		if (error)
+			goto out_no_agbp;
+		if (!agbp) {
 			ASSERT(flags & XFS_ALLOC_FLAG_TRYLOCK);
 			ASSERT(!(flags & XFS_ALLOC_FLAG_FREEING));
-			args->agbp = NULL;
-			return 0;
-		}
-	}
-	/*
-	 * Figure out how many blocks we should have in the freelist.
-	 */
-	agf = XFS_BUF_TO_AGF(agbp);
-	need = XFS_MIN_FREELIST(agf, mp);
-	/*
-	 * If there isn't enough total or single-extent, reject it.
-	 */
-	if (!(flags & XFS_ALLOC_FLAG_FREEING)) {
-		delta = need > be32_to_cpu(agf->agf_flcount) ?
-			(need - be32_to_cpu(agf->agf_flcount)) : 0;
-		longest = be32_to_cpu(agf->agf_longest);
-		longest = (longest > delta) ? (longest - delta) :
-			(be32_to_cpu(agf->agf_flcount) > 0 || longest > 0);
-		if ((args->minlen + args->alignment + args->minalignslop - 1) >
-				longest ||
-		    ((int)(be32_to_cpu(agf->agf_freeblks) +
-		     be32_to_cpu(agf->agf_flcount) - need - args->total) <
-				(int)args->minleft)) {
-			xfs_trans_brelse(tp, agbp);
-			args->agbp = NULL;
-			return 0;
+			goto out_no_agbp;
 		}
 	}
+
+	/* If there isn't enough total space or single-extent, reject it. */
+	need = xfs_alloc_min_freelist(mp, pag);
+	if (!xfs_alloc_space_available(args, need, flags))
+		goto out_agbp_relse;
+
 	/*
 	 * Make the freelist shorter if it's too long.
+	 *
+	 * Note that from this point onwards, we will always release the agf and
+	 * agfl buffers on error. This handles the case where we error out and
+	 * the buffers are clean or may not have been joined to the transaction
+	 * and hence need to be released manually. If they have been joined to
+	 * the transaction, then xfs_trans_brelse() will handle them
+	 * appropriately based on the recursion count and dirty state of the
+	 * buffer.
+	 *
+	 * XXX (dgc): When we have lots of free space, does this buy us
+	 * anything other than extra overhead when we need to put more blocks
+	 * back on the free list? Maybe we should only do this when space is
+	 * getting low or the AGFL is more than half full?
 	 */
-	while (be32_to_cpu(agf->agf_flcount) > need) {
-		xfs_buf_t	*bp;
+	while (pag->pagf_flcount > need) {
+		struct xfs_buf	*bp;
 
 		error = xfs_alloc_get_freelist(tp, agbp, &bno, 0);
 		if (error)
-			return error;
-		if ((error = xfs_free_ag_extent(tp, agbp, args->agno, bno, 1, 1)))
-			return error;
+			goto out_agbp_relse;
+		error = xfs_free_ag_extent(tp, agbp, args->agno, bno, 1, 1);
+		if (error)
+			goto out_agbp_relse;
 		bp = xfs_btree_get_bufs(mp, tp, args->agno, bno, 0);
 		xfs_trans_binval(tp, bp);
 	}
-	/*
-	 * Initialize the args structure.
-	 */
+
 	memset(&targs, 0, sizeof(targs));
 	targs.tp = tp;
 	targs.mp = mp;
@@ -1971,21 +2022,20 @@ xfs_alloc_fix_freelist(
 	targs.alignment = targs.minlen = targs.prod = targs.isfl = 1;
 	targs.type = XFS_ALLOCTYPE_THIS_AG;
 	targs.pag = pag;
-	if ((error = xfs_alloc_read_agfl(mp, tp, targs.agno, &agflbp)))
-		return error;
-	/*
-	 * Make the freelist longer if it's too short.
-	 */
-	while (be32_to_cpu(agf->agf_flcount) < need) {
+	error = xfs_alloc_read_agfl(mp, tp, targs.agno, &agflbp);
+	if (error)
+		goto out_agbp_relse;
+
+	/* Make the freelist longer if it's too short. */
+	while (pag->pagf_flcount < need) {
 		targs.agbno = 0;
-		targs.maxlen = need - be32_to_cpu(agf->agf_flcount);
-		/*
-		 * Allocate as many blocks as possible at once.
-		 */
-		if ((error = xfs_alloc_ag_vextent(&targs))) {
-			xfs_trans_brelse(tp, agflbp);
-			return error;
-		}
+		targs.maxlen = need - pag->pagf_flcount;
+
+		/* Allocate as many blocks as possible at once. */
+		error = xfs_alloc_ag_vextent(&targs);
+		if (error)
+			goto out_agflbp_relse;
+
 		/*
 		 * Stop if we run out.  Won't happen if callers are obeying
 		 * the restrictions correctly.  Can happen for free calls
@@ -1994,9 +2044,7 @@ xfs_alloc_fix_freelist(
 		if (targs.agbno == NULLAGBLOCK) {
 			if (flags & XFS_ALLOC_FLAG_FREEING)
 				break;
-			xfs_trans_brelse(tp, agflbp);
-			args->agbp = NULL;
-			return 0;
+			goto out_agflbp_relse;
 		}
 		/*
 		 * Put each allocated block on the list.
@@ -2005,12 +2053,21 @@ xfs_alloc_fix_freelist(
 			error = xfs_alloc_put_freelist(tp, agbp,
 							agflbp, bno, 0);
 			if (error)
-				return error;
+				goto out_agflbp_relse;
 		}
 	}
 	xfs_trans_brelse(tp, agflbp);
 	args->agbp = agbp;
 	return 0;
+
+out_agflbp_relse:
+	xfs_trans_brelse(tp, agflbp);
+out_agbp_relse:
+	if (agbp)
+		xfs_trans_brelse(tp, agbp);
+out_no_agbp:
+	args->agbp = NULL;
+	return error;
 }
 
 /*