From 3ab78df2a59a485f479d26852a060acfd8c4ecd7 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 11:15:38 +1000
Subject: xfs: rework xfs_bmap_free callers to use xfs_defer_ops

Restructure everything that used xfs_bmap_free to use xfs_defer_ops
instead.  For now we'll just remove the old symbols and play some
cpp magic to make it work; in the next patch we'll actually rename
everything.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/libxfs/xfs_alloc.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 88c26b827a2d..5993f87eee3e 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -24,6 +24,7 @@
 #include "xfs_bit.h"
 #include "xfs_sb.h"
 #include "xfs_mount.h"
+#include "xfs_defer.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
 #include "xfs_alloc_btree.h"
-- 
cgit v1.2.3


From ba9e780246a15a35f8ebe5b60f4a11bb58e85bda Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 11:26:33 +1000
Subject: xfs: add tracepoints and error injection for deferred extent freeing

Add a couple of tracepoints for the deferred extent free operation and
a site for injecting errors while finishing the operation.  This makes
it easier to debug deferred ops and test log redo.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/libxfs/xfs_alloc.c | 7 +++++++
 fs/xfs/libxfs/xfs_bmap.c  | 2 ++
 fs/xfs/xfs_error.h        | 4 +++-
 fs/xfs/xfs_trace.h        | 5 ++++-
 4 files changed, 16 insertions(+), 2 deletions(-)

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 5993f87eee3e..22ac3f136f21 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2702,6 +2702,13 @@ xfs_free_extent(
 
 	ASSERT(len != 0);
 
+	trace_xfs_bmap_free_deferred(mp, agno, 0, agbno, len);
+
+	if (XFS_TEST_ERROR(false, mp,
+			XFS_ERRTAG_FREE_EXTENT,
+			XFS_RANDOM_FREE_EXTENT))
+		return -EIO;
+
 	error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
 	if (error)
 		return error;
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 540a6b7928af..8e14ff45c73a 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -596,6 +596,8 @@ xfs_bmap_add_free(
 	new = kmem_zone_alloc(xfs_bmap_free_item_zone, KM_SLEEP);
 	new->xefi_startblock = bno;
 	new->xefi_blockcount = (xfs_extlen_t)len;
+	trace_xfs_bmap_free_defer(mp, XFS_FSB_TO_AGNO(mp, bno), 0,
+			XFS_FSB_TO_AGBNO(mp, bno), len);
 	xfs_defer_add(dfops, XFS_DEFER_OPS_TYPE_FREE, &new->xefi_list);
 }
 
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index 2e4f67f68856..da6f951435c9 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -90,7 +90,8 @@ extern void xfs_verifier_error(struct xfs_buf *bp);
 #define XFS_ERRTAG_STRATCMPL_IOERR			19
 #define XFS_ERRTAG_DIOWRITE_IOERR			20
 #define XFS_ERRTAG_BMAPIFORMAT				21
-#define XFS_ERRTAG_MAX					22
+#define XFS_ERRTAG_FREE_EXTENT				22
+#define XFS_ERRTAG_MAX					23
 
 /*
  * Random factors for above tags, 1 means always, 2 means 1/2 time, etc.
@@ -117,6 +118,7 @@ extern void xfs_verifier_error(struct xfs_buf *bp);
 #define XFS_RANDOM_STRATCMPL_IOERR			(XFS_RANDOM_DEFAULT/10)
 #define XFS_RANDOM_DIOWRITE_IOERR			(XFS_RANDOM_DEFAULT/10)
 #define	XFS_RANDOM_BMAPIFORMAT				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_FREE_EXTENT				1
 
 #ifdef DEBUG
 extern int xfs_error_test_active;
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index a45b030d4592..939caf5f5a45 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -2417,9 +2417,12 @@ DEFINE_DEFER_PENDING_EVENT(xfs_defer_pending_cancel);
 DEFINE_DEFER_PENDING_EVENT(xfs_defer_pending_finish);
 DEFINE_DEFER_PENDING_EVENT(xfs_defer_pending_abort);
 
-DEFINE_PHYS_EXTENT_DEFERRED_EVENT(xfs_defer_phys_extent);
 DEFINE_MAP_EXTENT_DEFERRED_EVENT(xfs_defer_map_extent);
 
+#define DEFINE_BMAP_FREE_DEFERRED_EVENT DEFINE_PHYS_EXTENT_DEFERRED_EVENT
+DEFINE_BMAP_FREE_DEFERRED_EVENT(xfs_bmap_free_defer);
+DEFINE_BMAP_FREE_DEFERRED_EVENT(xfs_bmap_free_deferred);
+
 #endif /* _TRACE_XFS_H */
 
 #undef TRACE_INCLUDE_PATH
-- 
cgit v1.2.3


From b87049444ac4a6515ba0427d16a73438b646435b Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 11:30:32 +1000
Subject: xfs: introduce rmap btree definitions

Originally-From: Dave Chinner <dchinner@redhat.com>

Add new per-ag rmap btree definitions to the per-ag structures. The
rmap btree will sit in the empty slots on disk after the free space
btrees, and hence form a part of the array of space management
btrees. This requires the definition of the btree to be contiguous
with the free space btrees.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/libxfs/xfs_alloc.c  |  6 ++++++
 fs/xfs/libxfs/xfs_btree.c  |  4 ++--
 fs/xfs/libxfs/xfs_btree.h  |  3 +++
 fs/xfs/libxfs/xfs_format.h | 22 +++++++++++++++++-----
 fs/xfs/libxfs/xfs_types.h  |  4 ++--
 5 files changed, 30 insertions(+), 9 deletions(-)

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 22ac3f136f21..c02d8e8419ce 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2272,6 +2272,10 @@ xfs_agf_verify(
 	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) > XFS_BTREE_MAXLEVELS)
 		return false;
 
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb) &&
+	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) > XFS_BTREE_MAXLEVELS)
+		return false;
+
 	/*
 	 * during growfs operations, the perag is not fully initialised,
 	 * so we can't use it for any useful checking. growfs ensures we can't
@@ -2403,6 +2407,8 @@ xfs_alloc_read_agf(
 			be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNOi]);
 		pag->pagf_levels[XFS_BTNUM_CNTi] =
 			be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]);
+		pag->pagf_levels[XFS_BTNUM_RMAPi] =
+			be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAPi]);
 		spin_lock_init(&pag->pagb_lock);
 		pag->pagb_count = 0;
 		pag->pagb_tree = RB_ROOT;
diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index 4fc46529753d..4066b01a1807 100644
--- a/fs/xfs/libxfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -44,9 +44,9 @@ kmem_zone_t	*xfs_btree_cur_zone;
  * Btree magic numbers.
  */
 static const __uint32_t xfs_magics[2][XFS_BTNUM_MAX] = {
-	{ XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC,
+	{ XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, 0, XFS_BMAP_MAGIC, XFS_IBT_MAGIC,
 	  XFS_FIBT_MAGIC },
-	{ XFS_ABTB_CRC_MAGIC, XFS_ABTC_CRC_MAGIC,
+	{ XFS_ABTB_CRC_MAGIC, XFS_ABTC_CRC_MAGIC, XFS_RMAP_CRC_MAGIC,
 	  XFS_BMAP_CRC_MAGIC, XFS_IBT_CRC_MAGIC, XFS_FIBT_CRC_MAGIC }
 };
 #define xfs_btree_magic(cur) \
diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
index ac5cd6ef6859..25f1b69e7515 100644
--- a/fs/xfs/libxfs/xfs_btree.h
+++ b/fs/xfs/libxfs/xfs_btree.h
@@ -77,6 +77,7 @@ union xfs_btree_rec {
 #define	XFS_BTNUM_BMAP	((xfs_btnum_t)XFS_BTNUM_BMAPi)
 #define	XFS_BTNUM_INO	((xfs_btnum_t)XFS_BTNUM_INOi)
 #define	XFS_BTNUM_FINO	((xfs_btnum_t)XFS_BTNUM_FINOi)
+#define	XFS_BTNUM_RMAP	((xfs_btnum_t)XFS_BTNUM_RMAPi)
 
 /*
  * For logging record fields.
@@ -109,6 +110,7 @@ do {    \
 	case XFS_BTNUM_BMAP: __XFS_BTREE_STATS_INC(__mp, bmbt, stat); break; \
 	case XFS_BTNUM_INO: __XFS_BTREE_STATS_INC(__mp, ibt, stat); break; \
 	case XFS_BTNUM_FINO: __XFS_BTREE_STATS_INC(__mp, fibt, stat); break; \
+	case XFS_BTNUM_RMAP: break;	\
 	case XFS_BTNUM_MAX: ASSERT(0); /* fucking gcc */ ; break;	\
 	}       \
 } while (0)
@@ -129,6 +131,7 @@ do {    \
 		__XFS_BTREE_STATS_ADD(__mp, ibt, stat, val); break; \
 	case XFS_BTNUM_FINO:	\
 		__XFS_BTREE_STATS_ADD(__mp, fibt, stat, val); break; \
+	case XFS_BTNUM_RMAP: break; \
 	case XFS_BTNUM_MAX: ASSERT(0); /* fucking gcc */ ; break; \
 	}       \
 } while (0)
diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index adb204d40f22..74f4d3da598c 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -455,6 +455,7 @@ xfs_sb_has_compat_feature(
 }
 
 #define XFS_SB_FEAT_RO_COMPAT_FINOBT   (1 << 0)		/* free inode btree */
+#define XFS_SB_FEAT_RO_COMPAT_RMAPBT   (1 << 1)		/* reverse map btree */
 #define XFS_SB_FEAT_RO_COMPAT_ALL \
 		(XFS_SB_FEAT_RO_COMPAT_FINOBT)
 #define XFS_SB_FEAT_RO_COMPAT_UNKNOWN	~XFS_SB_FEAT_RO_COMPAT_ALL
@@ -538,6 +539,12 @@ static inline bool xfs_sb_version_hasmetauuid(struct xfs_sb *sbp)
 		(sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_META_UUID);
 }
 
+static inline bool xfs_sb_version_hasrmapbt(struct xfs_sb *sbp)
+{
+	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) &&
+		(sbp->sb_features_ro_compat & XFS_SB_FEAT_RO_COMPAT_RMAPBT);
+}
+
 /*
  * end of superblock version macros
  */
@@ -598,10 +605,10 @@ xfs_is_quota_inode(struct xfs_sb *sbp, xfs_ino_t ino)
 #define	XFS_AGI_GOOD_VERSION(v)	((v) == XFS_AGI_VERSION)
 
 /*
- * Btree number 0 is bno, 1 is cnt.  This value gives the size of the
+ * Btree number 0 is bno, 1 is cnt, 2 is rmap. This value gives the size of the
  * arrays below.
  */
-#define	XFS_BTNUM_AGF	((int)XFS_BTNUM_CNTi + 1)
+#define	XFS_BTNUM_AGF	((int)XFS_BTNUM_RMAPi + 1)
 
 /*
  * The second word of agf_levels in the first a.g. overlaps the EFS
@@ -618,12 +625,10 @@ typedef struct xfs_agf {
 	__be32		agf_seqno;	/* sequence # starting from 0 */
 	__be32		agf_length;	/* size in blocks of a.g. */
 	/*
-	 * Freespace information
+	 * Freespace and rmap information
 	 */
 	__be32		agf_roots[XFS_BTNUM_AGF];	/* root blocks */
-	__be32		agf_spare0;	/* spare field */
 	__be32		agf_levels[XFS_BTNUM_AGF];	/* btree levels */
-	__be32		agf_spare1;	/* spare field */
 
 	__be32		agf_flfirst;	/* first freelist block's index */
 	__be32		agf_fllast;	/* last freelist block's index */
@@ -1307,6 +1312,13 @@ typedef __be32 xfs_inobt_ptr_t;
 #define	XFS_IBT_BLOCK(mp)		((xfs_agblock_t)(XFS_CNT_BLOCK(mp) + 1))
 #define	XFS_FIBT_BLOCK(mp)		((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1))
 
+/*
+ * Reverse mapping btree format definitions
+ *
+ * There is a btree for the reverse map per allocation group
+ */
+#define	XFS_RMAP_CRC_MAGIC	0x524d4233	/* 'RMB3' */
+
 /*
  * The first data block of an AG depends on whether the filesystem was formatted
  * with the finobt feature. If so, account for the finobt reserved root btree
diff --git a/fs/xfs/libxfs/xfs_types.h b/fs/xfs/libxfs/xfs_types.h
index b79dc66b2ecd..3d503647f26b 100644
--- a/fs/xfs/libxfs/xfs_types.h
+++ b/fs/xfs/libxfs/xfs_types.h
@@ -108,8 +108,8 @@ typedef enum {
 } xfs_lookup_t;
 
 typedef enum {
-	XFS_BTNUM_BNOi, XFS_BTNUM_CNTi, XFS_BTNUM_BMAPi, XFS_BTNUM_INOi,
-	XFS_BTNUM_FINOi, XFS_BTNUM_MAX
+	XFS_BTNUM_BNOi, XFS_BTNUM_CNTi, XFS_BTNUM_RMAPi, XFS_BTNUM_BMAPi,
+	XFS_BTNUM_INOi, XFS_BTNUM_FINOi, XFS_BTNUM_MAX
 } xfs_btnum_t;
 
 struct xfs_name {
-- 
cgit v1.2.3


From 8018026ef29756af6144e2e2e8dffc9c2ed0d6f7 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 11:31:47 +1000
Subject: xfs: rmap btree add more reserved blocks

Originally-From: Dave Chinner <dchinner@redhat.com>

XFS reserves a small amount of space in each AG for the minimum
number of free blocks needed for operation. Adding the rmap btree
increases the number of reserved blocks, but it also increases the
complexity of the calculation as the free inode btree is optional
(like the rmbt).

Rather than calculate the prealloc blocks every time we need to
check it, add a function to calculate it at mount time and store it
in the struct xfs_mount, and convert the XFS_PREALLOC_BLOCKS macro
just to use the xfs-mount variable directly.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/libxfs/xfs_alloc.c  | 11 +++++++++++
 fs/xfs/libxfs/xfs_alloc.h  |  2 ++
 fs/xfs/libxfs/xfs_format.h |  9 +--------
 fs/xfs/xfs_fsops.c         |  6 +++---
 fs/xfs/xfs_mount.c         |  2 ++
 fs/xfs/xfs_mount.h         |  1 +
 6 files changed, 20 insertions(+), 11 deletions(-)

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index c02d8e8419ce..43c7e53a6a8e 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -50,6 +50,17 @@ STATIC int xfs_alloc_ag_vextent_size(xfs_alloc_arg_t *);
 STATIC int xfs_alloc_ag_vextent_small(xfs_alloc_arg_t *,
 		xfs_btree_cur_t *, xfs_agblock_t *, xfs_extlen_t *, int *);
 
+xfs_extlen_t
+xfs_prealloc_blocks(
+	struct xfs_mount	*mp)
+{
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
+		return XFS_RMAP_BLOCK(mp) + 1;
+	if (xfs_sb_version_hasfinobt(&mp->m_sb))
+		return XFS_FIBT_BLOCK(mp) + 1;
+	return XFS_IBT_BLOCK(mp) + 1;
+}
+
 /*
  * Lookup the record equal to [bno, len] in the btree given by cur.
  */
diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index cf268b2d0b6c..20b54aa87026 100644
--- a/fs/xfs/libxfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
@@ -232,4 +232,6 @@ int xfs_alloc_fix_freelist(struct xfs_alloc_arg *args, int flags);
 int xfs_free_extent_fix_freelist(struct xfs_trans *tp, xfs_agnumber_t agno,
 		struct xfs_buf **agbp);
 
+xfs_extlen_t xfs_prealloc_blocks(struct xfs_mount *mp);
+
 #endif	/* __XFS_ALLOC_H__ */
diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index 74f4d3da598c..3dbed011c29a 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -1319,18 +1319,11 @@ typedef __be32 xfs_inobt_ptr_t;
  */
 #define	XFS_RMAP_CRC_MAGIC	0x524d4233	/* 'RMB3' */
 
-/*
- * The first data block of an AG depends on whether the filesystem was formatted
- * with the finobt feature. If so, account for the finobt reserved root btree
- * block.
- */
-#define XFS_PREALLOC_BLOCKS(mp) \
+#define	XFS_RMAP_BLOCK(mp) \
 	(xfs_sb_version_hasfinobt(&((mp)->m_sb)) ? \
 	 XFS_FIBT_BLOCK(mp) + 1 : \
 	 XFS_IBT_BLOCK(mp) + 1)
 
-
-
 /*
  * BMAP Btree format definitions
  *
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 4c191d2194e8..0a3b822d9f8c 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -243,7 +243,7 @@ xfs_growfs_data_private(
 		agf->agf_flfirst = cpu_to_be32(1);
 		agf->agf_fllast = 0;
 		agf->agf_flcount = 0;
-		tmpsize = agsize - XFS_PREALLOC_BLOCKS(mp);
+		tmpsize = agsize - mp->m_ag_prealloc_blocks;
 		agf->agf_freeblks = cpu_to_be32(tmpsize);
 		agf->agf_longest = cpu_to_be32(tmpsize);
 		if (xfs_sb_version_hascrc(&mp->m_sb))
@@ -340,7 +340,7 @@ xfs_growfs_data_private(
 						agno, 0);
 
 		arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1);
-		arec->ar_startblock = cpu_to_be32(XFS_PREALLOC_BLOCKS(mp));
+		arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks);
 		arec->ar_blockcount = cpu_to_be32(
 			agsize - be32_to_cpu(arec->ar_startblock));
 
@@ -369,7 +369,7 @@ xfs_growfs_data_private(
 						agno, 0);
 
 		arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1);
-		arec->ar_startblock = cpu_to_be32(XFS_PREALLOC_BLOCKS(mp));
+		arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks);
 		arec->ar_blockcount = cpu_to_be32(
 			agsize - be32_to_cpu(arec->ar_startblock));
 		nfree += be32_to_cpu(arec->ar_blockcount);
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 5f3ba88985d5..24ae96b4078f 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -231,6 +231,8 @@ xfs_initialize_perag(
 
 	if (maxagi)
 		*maxagi = index;
+
+	mp->m_ag_prealloc_blocks = xfs_prealloc_blocks(mp);
 	return 0;
 
 out_unwind:
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index c1b798c72126..0537b1f28f90 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -119,6 +119,7 @@ typedef struct xfs_mount {
 	uint			m_ag_maxlevels;	/* XFS_AG_MAXLEVELS */
 	uint			m_bm_maxlevels[2]; /* XFS_BM_MAXLEVELS */
 	uint			m_in_maxlevels;	/* max inobt btree levels. */
+	xfs_extlen_t		m_ag_prealloc_blocks; /* reserved ag blocks */
 	struct radix_tree_root	m_perag_tree;	/* per-ag accounting info */
 	spinlock_t		m_perag_lock;	/* lock for m_perag_tree */
 	struct mutex		m_growlock;	/* growfs mutex */
-- 
cgit v1.2.3


From 340785cca16246f82ccaf11740d885017a9e9341 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 11:33:42 +1000
Subject: xfs: add owner field to extent allocation and freeing

For the rmap btree to work, we have to feed the extent owner
information to the the allocation and freeing functions. This
information is what will end up in the rmap btree that tracks
allocated extents. While we technically don't need the owner
information when freeing extents, passing it allows us to validate
that the extent we are removing from the rmap btree actually
belonged to the owner we expected it to belong to.

We also define a special set of owner values for internal metadata
that would otherwise have no owner. This allows us to tell the
difference between metadata owned by different per-ag btrees, as
well as static fs metadata (e.g. AG headers) and internal journal
blocks.

There are also a couple of special cases we need to take care of -
during EFI recovery, we don't actually know who the original owner
was, so we need to pass a wildcard to indicate that we aren't
checking the owner for validity. We also need special handling in
growfs, as we "free" the space in the last AG when extending it, but
because it's new space it has no actual owner...

While touching the xfs_bmap_add_free() function, re-order the
parameters to put the struct xfs_mount first.

Extend the owner field to include both the owner type and some sort
of index within the owner.  The index field will be used to support
reverse mappings when reflink is enabled.

When we're freeing extents from an EFI, we don't have the owner
information available (rmap updates have their own redo items).
xfs_free_extent therefore doesn't need to do an rmap update. Make
sure that the log replay code signals this correctly.

This is based upon a patch originally from Dave Chinner. It has been
extended to add more owner information with the intent of helping
recovery operations when things go wrong (e.g. offset of user data
block in a file).

[dchinner: de-shout the xfs_rmap_*_owner helpers]
[darrick: minor style fixes suggested by Christoph Hellwig]

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/libxfs/xfs_alloc.c        | 26 ++++++++++------
 fs/xfs/libxfs/xfs_alloc.h        |  8 +++--
 fs/xfs/libxfs/xfs_bmap.c         | 25 +++++++++++----
 fs/xfs/libxfs/xfs_bmap.h         |  4 ++-
 fs/xfs/libxfs/xfs_bmap_btree.c   |  7 ++++-
 fs/xfs/libxfs/xfs_format.h       | 28 +++++++++++++++++
 fs/xfs/libxfs/xfs_ialloc.c       |  8 +++--
 fs/xfs/libxfs/xfs_ialloc_btree.c |  8 ++++-
 fs/xfs/libxfs/xfs_rmap.h         | 67 ++++++++++++++++++++++++++++++++++++++++
 fs/xfs/xfs_extfree_item.c        |  6 +++-
 fs/xfs/xfs_fsops.c               | 17 +++++++---
 fs/xfs/xfs_trans.h               |  2 +-
 fs/xfs/xfs_trans_extfree.c       |  8 +++--
 13 files changed, 181 insertions(+), 33 deletions(-)
 create mode 100644 fs/xfs/libxfs/xfs_rmap.h

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 43c7e53a6a8e..6335b8b85349 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -36,6 +36,7 @@
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
+#include "xfs_rmap.h"
 
 struct workqueue_struct *xfs_alloc_wq;
 
@@ -1589,14 +1590,15 @@ error0:
 /*
  * Free the extent starting at agno/bno for length.
  */
-STATIC int			/* error */
+STATIC int
 xfs_free_ag_extent(
-	xfs_trans_t	*tp,	/* transaction pointer */
-	xfs_buf_t	*agbp,	/* buffer for a.g. freelist header */
-	xfs_agnumber_t	agno,	/* allocation group number */
-	xfs_agblock_t	bno,	/* starting block number */
-	xfs_extlen_t	len,	/* length of extent */
-	int		isfl)	/* set if is freelist blocks - no sb acctg */
+	xfs_trans_t		*tp,
+	xfs_buf_t		*agbp,
+	xfs_agnumber_t		agno,
+	xfs_agblock_t		bno,
+	xfs_extlen_t		len,
+	struct xfs_owner_info	*oinfo,
+	int			isfl)
 {
 	xfs_btree_cur_t	*bno_cur;	/* cursor for by-block btree */
 	xfs_btree_cur_t	*cnt_cur;	/* cursor for by-size btree */
@@ -2005,13 +2007,15 @@ xfs_alloc_fix_freelist(
 	 * back on the free list? Maybe we should only do this when space is
 	 * getting low or the AGFL is more than half full?
 	 */
+	xfs_rmap_ag_owner(&targs.oinfo, XFS_RMAP_OWN_AG);
 	while (pag->pagf_flcount > need) {
 		struct xfs_buf	*bp;
 
 		error = xfs_alloc_get_freelist(tp, agbp, &bno, 0);
 		if (error)
 			goto out_agbp_relse;
-		error = xfs_free_ag_extent(tp, agbp, args->agno, bno, 1, 1);
+		error = xfs_free_ag_extent(tp, agbp, args->agno, bno, 1,
+					   &targs.oinfo, 1);
 		if (error)
 			goto out_agbp_relse;
 		bp = xfs_btree_get_bufs(mp, tp, args->agno, bno, 0);
@@ -2021,6 +2025,7 @@ xfs_alloc_fix_freelist(
 	memset(&targs, 0, sizeof(targs));
 	targs.tp = tp;
 	targs.mp = mp;
+	xfs_rmap_ag_owner(&targs.oinfo, XFS_RMAP_OWN_AG);
 	targs.agbp = agbp;
 	targs.agno = args->agno;
 	targs.alignment = targs.minlen = targs.prod = targs.isfl = 1;
@@ -2709,7 +2714,8 @@ int				/* error */
 xfs_free_extent(
 	struct xfs_trans	*tp,	/* transaction pointer */
 	xfs_fsblock_t		bno,	/* starting block number of extent */
-	xfs_extlen_t		len)	/* length of extent */
+	xfs_extlen_t		len,	/* length of extent */
+	struct xfs_owner_info	*oinfo)	/* extent owner */
 {
 	struct xfs_mount	*mp = tp->t_mountp;
 	struct xfs_buf		*agbp;
@@ -2737,7 +2743,7 @@ xfs_free_extent(
 		agbno + len <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_length),
 				err);
 
-	error = xfs_free_ag_extent(tp, agbp, agno, agbno, len, 0);
+	error = xfs_free_ag_extent(tp, agbp, agno, agbno, len, oinfo, 0);
 	if (error)
 		goto err;
 
diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index 20b54aa87026..88053422fda0 100644
--- a/fs/xfs/libxfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
@@ -123,6 +123,7 @@ typedef struct xfs_alloc_arg {
 	char		isfl;		/* set if is freelist blocks - !acctg */
 	char		userdata;	/* mask defining userdata treatment */
 	xfs_fsblock_t	firstblock;	/* io first block allocated */
+	struct xfs_owner_info	oinfo;	/* owner of blocks being allocated */
 } xfs_alloc_arg_t;
 
 /*
@@ -208,9 +209,10 @@ xfs_alloc_vextent(
  */
 int				/* error */
 xfs_free_extent(
-	struct xfs_trans *tp,	/* transaction pointer */
-	xfs_fsblock_t	bno,	/* starting block number of extent */
-	xfs_extlen_t	len);	/* length of extent */
+	struct xfs_trans	*tp,	/* transaction pointer */
+	xfs_fsblock_t		bno,	/* starting block number of extent */
+	xfs_extlen_t		len,	/* length of extent */
+	struct xfs_owner_info	*oinfo);/* extent owner */
 
 int				/* error */
 xfs_alloc_lookup_ge(
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 8e14ff45c73a..919069fc7b23 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -46,6 +46,7 @@
 #include "xfs_symlink.h"
 #include "xfs_attr_leaf.h"
 #include "xfs_filestream.h"
+#include "xfs_rmap.h"
 
 
 kmem_zone_t		*xfs_bmap_free_item_zone;
@@ -571,10 +572,11 @@ xfs_bmap_validate_ret(
  */
 void
 xfs_bmap_add_free(
-	struct xfs_mount	*mp,		/* mount point structure */
-	struct xfs_defer_ops	*dfops,		/* list of extents */
-	xfs_fsblock_t		bno,		/* fs block number of extent */
-	xfs_filblks_t		len)		/* length of extent */
+	struct xfs_mount		*mp,
+	struct xfs_defer_ops		*dfops,
+	xfs_fsblock_t			bno,
+	xfs_filblks_t			len,
+	struct xfs_owner_info		*oinfo)
 {
 	struct xfs_extent_free_item	*new;		/* new element */
 #ifdef DEBUG
@@ -593,9 +595,14 @@ xfs_bmap_add_free(
 	ASSERT(agbno + len <= mp->m_sb.sb_agblocks);
 #endif
 	ASSERT(xfs_bmap_free_item_zone != NULL);
+
 	new = kmem_zone_alloc(xfs_bmap_free_item_zone, KM_SLEEP);
 	new->xefi_startblock = bno;
 	new->xefi_blockcount = (xfs_extlen_t)len;
+	if (oinfo)
+		new->xefi_oinfo = *oinfo;
+	else
+		xfs_rmap_skip_owner_update(&new->xefi_oinfo);
 	trace_xfs_bmap_free_defer(mp, XFS_FSB_TO_AGNO(mp, bno), 0,
 			XFS_FSB_TO_AGBNO(mp, bno), len);
 	xfs_defer_add(dfops, XFS_DEFER_OPS_TYPE_FREE, &new->xefi_list);
@@ -628,6 +635,7 @@ xfs_bmap_btree_to_extents(
 	xfs_mount_t		*mp;	/* mount point structure */
 	__be64			*pp;	/* ptr to block address */
 	struct xfs_btree_block	*rblock;/* root btree block */
+	struct xfs_owner_info	oinfo;
 
 	mp = ip->i_mount;
 	ifp = XFS_IFORK_PTR(ip, whichfork);
@@ -651,7 +659,8 @@ xfs_bmap_btree_to_extents(
 	cblock = XFS_BUF_TO_BLOCK(cbp);
 	if ((error = xfs_btree_check_block(cur, cblock, 0, cbp)))
 		return error;
-	xfs_bmap_add_free(mp, cur->bc_private.b.dfops, cbno, 1);
+	xfs_rmap_ino_bmbt_owner(&oinfo, ip->i_ino, whichfork);
+	xfs_bmap_add_free(mp, cur->bc_private.b.dfops, cbno, 1, &oinfo);
 	ip->i_d.di_nblocks--;
 	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
 	xfs_trans_binval(tp, cbp);
@@ -732,6 +741,7 @@ xfs_bmap_extents_to_btree(
 	memset(&args, 0, sizeof(args));
 	args.tp = tp;
 	args.mp = mp;
+	xfs_rmap_ino_bmbt_owner(&args.oinfo, ip->i_ino, whichfork);
 	args.firstblock = *firstblock;
 	if (*firstblock == NULLFSBLOCK) {
 		args.type = XFS_ALLOCTYPE_START_BNO;
@@ -878,6 +888,7 @@ xfs_bmap_local_to_extents(
 	memset(&args, 0, sizeof(args));
 	args.tp = tp;
 	args.mp = ip->i_mount;
+	xfs_rmap_ino_owner(&args.oinfo, ip->i_ino, whichfork, 0);
 	args.firstblock = *firstblock;
 	/*
 	 * Allocate a block.  We know we need only one, since the
@@ -3660,6 +3671,7 @@ xfs_bmap_btalloc(
 	args.tp = ap->tp;
 	args.mp = mp;
 	args.fsbno = ap->blkno;
+	xfs_rmap_skip_owner_update(&args.oinfo);
 
 	/* Trim the allocation back to the maximum an AG can fit. */
 	args.maxlen = MIN(ap->length, XFS_ALLOC_AG_MAX_USABLE(mp));
@@ -4839,6 +4851,7 @@ xfs_bmap_del_extent(
 		nblks = 0;
 		do_fx = 0;
 	}
+
 	/*
 	 * Set flag value to use in switch statement.
 	 * Left-contig is 2, right-contig is 1.
@@ -5026,7 +5039,7 @@ xfs_bmap_del_extent(
 	 */
 	if (do_fx)
 		xfs_bmap_add_free(mp, dfops, del->br_startblock,
-			del->br_blockcount);
+				del->br_blockcount, NULL);
 	/*
 	 * Adjust inode # blocks in the file.
 	 */
diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index dc01bb85a6e8..254034f96941 100644
--- a/fs/xfs/libxfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -67,6 +67,7 @@ struct xfs_extent_free_item
 	xfs_fsblock_t		xefi_startblock;/* starting fs block number */
 	xfs_extlen_t		xefi_blockcount;/* number of blocks in extent */
 	struct list_head	xefi_list;
+	struct xfs_owner_info	xefi_oinfo;	/* extent owner */
 };
 
 #define	XFS_BMAP_MAX_NMAP	4
@@ -165,7 +166,8 @@ void	xfs_bmap_trace_exlist(struct xfs_inode *ip, xfs_extnum_t cnt,
 int	xfs_bmap_add_attrfork(struct xfs_inode *ip, int size, int rsvd);
 void	xfs_bmap_local_to_extents_empty(struct xfs_inode *ip, int whichfork);
 void	xfs_bmap_add_free(struct xfs_mount *mp, struct xfs_defer_ops *dfops,
-			  xfs_fsblock_t bno, xfs_filblks_t len);
+			  xfs_fsblock_t bno, xfs_filblks_t len,
+			  struct xfs_owner_info *oinfo);
 void	xfs_bmap_compute_maxlevels(struct xfs_mount *mp, int whichfork);
 int	xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork);
diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c
index 0df5318cefff..9e34ca413e53 100644
--- a/fs/xfs/libxfs/xfs_bmap_btree.c
+++ b/fs/xfs/libxfs/xfs_bmap_btree.c
@@ -35,6 +35,7 @@
 #include "xfs_quota.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
+#include "xfs_rmap.h"
 
 /*
  * Determine the extent state.
@@ -447,6 +448,8 @@ xfs_bmbt_alloc_block(
 	args.mp = cur->bc_mp;
 	args.fsbno = cur->bc_private.b.firstblock;
 	args.firstblock = args.fsbno;
+	xfs_rmap_ino_bmbt_owner(&args.oinfo, cur->bc_private.b.ip->i_ino,
+			cur->bc_private.b.whichfork);
 
 	if (args.fsbno == NULLFSBLOCK) {
 		args.fsbno = be64_to_cpu(start->l);
@@ -526,8 +529,10 @@ xfs_bmbt_free_block(
 	struct xfs_inode	*ip = cur->bc_private.b.ip;
 	struct xfs_trans	*tp = cur->bc_tp;
 	xfs_fsblock_t		fsbno = XFS_DADDR_TO_FSB(mp, XFS_BUF_ADDR(bp));
+	struct xfs_owner_info	oinfo;
 
-	xfs_bmap_add_free(mp, cur->bc_private.b.dfops, fsbno, 1);
+	xfs_rmap_ino_bmbt_owner(&oinfo, ip->i_ino, cur->bc_private.b.whichfork);
+	xfs_bmap_add_free(mp, cur->bc_private.b.dfops, fsbno, 1, &oinfo);
 	ip->i_d.di_nblocks--;
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index 3dbed011c29a..eecfedf19be5 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -1319,6 +1319,34 @@ typedef __be32 xfs_inobt_ptr_t;
  */
 #define	XFS_RMAP_CRC_MAGIC	0x524d4233	/* 'RMB3' */
 
+/*
+ * Ownership info for an extent.  This is used to create reverse-mapping
+ * entries.
+ */
+#define XFS_OWNER_INFO_ATTR_FORK	(1 << 0)
+#define XFS_OWNER_INFO_BMBT_BLOCK	(1 << 1)
+struct xfs_owner_info {
+	uint64_t		oi_owner;
+	xfs_fileoff_t		oi_offset;
+	unsigned int		oi_flags;
+};
+
+/*
+ * Special owner types.
+ *
+ * Seeing as we only support up to 8EB, we have the upper bit of the owner field
+ * to tell us we have a special owner value. We use these for static metadata
+ * allocated at mkfs/growfs time, as well as for freespace management metadata.
+ */
+#define XFS_RMAP_OWN_NULL	(-1ULL)	/* No owner, for growfs */
+#define XFS_RMAP_OWN_UNKNOWN	(-2ULL)	/* Unknown owner, for EFI recovery */
+#define XFS_RMAP_OWN_FS		(-3ULL)	/* static fs metadata */
+#define XFS_RMAP_OWN_LOG	(-4ULL)	/* static fs metadata */
+#define XFS_RMAP_OWN_AG		(-5ULL)	/* AG freespace btree blocks */
+#define XFS_RMAP_OWN_INOBT	(-6ULL)	/* Inode btree blocks */
+#define XFS_RMAP_OWN_INODES	(-7ULL)	/* Inode chunk */
+#define XFS_RMAP_OWN_MIN	(-8ULL) /* guard */
+
 #define	XFS_RMAP_BLOCK(mp) \
 	(xfs_sb_version_hasfinobt(&((mp)->m_sb)) ? \
 	 XFS_FIBT_BLOCK(mp) + 1 : \
diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index f0e1f713b0be..51b4e0de1fdc 100644
--- a/fs/xfs/libxfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -40,6 +40,7 @@
 #include "xfs_icache.h"
 #include "xfs_trace.h"
 #include "xfs_log.h"
+#include "xfs_rmap.h"
 
 
 /*
@@ -615,6 +616,7 @@ xfs_ialloc_ag_alloc(
 	args.tp = tp;
 	args.mp = tp->t_mountp;
 	args.fsbno = NULLFSBLOCK;
+	xfs_rmap_ag_owner(&args.oinfo, XFS_RMAP_OWN_INODES);
 
 #ifdef DEBUG
 	/* randomly do sparse inode allocations */
@@ -1825,12 +1827,14 @@ xfs_difree_inode_chunk(
 	int		nextbit;
 	xfs_agblock_t	agbno;
 	int		contigblk;
+	struct xfs_owner_info	oinfo;
 	DECLARE_BITMAP(holemask, XFS_INOBT_HOLEMASK_BITS);
+	xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_INODES);
 
 	if (!xfs_inobt_issparse(rec->ir_holemask)) {
 		/* not sparse, calculate extent info directly */
 		xfs_bmap_add_free(mp, dfops, XFS_AGB_TO_FSB(mp, agno, sagbno),
-				  mp->m_ialloc_blks);
+				  mp->m_ialloc_blks, &oinfo);
 		return;
 	}
 
@@ -1874,7 +1878,7 @@ xfs_difree_inode_chunk(
 		ASSERT(agbno % mp->m_sb.sb_spino_align == 0);
 		ASSERT(contigblk % mp->m_sb.sb_spino_align == 0);
 		xfs_bmap_add_free(mp, dfops, XFS_AGB_TO_FSB(mp, agno, agbno),
-				  contigblk);
+				  contigblk, &oinfo);
 
 		/* reset range to current bit and carry on... */
 		startidx = endidx = nextbit;
diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c
index a48f4482004c..c83691edf6b2 100644
--- a/fs/xfs/libxfs/xfs_ialloc_btree.c
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.c
@@ -32,6 +32,7 @@
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
 #include "xfs_trans.h"
+#include "xfs_rmap.h"
 
 
 STATIC int
@@ -96,6 +97,7 @@ xfs_inobt_alloc_block(
 	memset(&args, 0, sizeof(args));
 	args.tp = cur->bc_tp;
 	args.mp = cur->bc_mp;
+	xfs_rmap_ag_owner(&args.oinfo, XFS_RMAP_OWN_INOBT);
 	args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.a.agno, sbno);
 	args.minlen = 1;
 	args.maxlen = 1;
@@ -125,8 +127,12 @@ xfs_inobt_free_block(
 	struct xfs_btree_cur	*cur,
 	struct xfs_buf		*bp)
 {
+	struct xfs_owner_info	oinfo;
+
+	xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_INOBT);
 	return xfs_free_extent(cur->bc_tp,
-			XFS_DADDR_TO_FSB(cur->bc_mp, XFS_BUF_ADDR(bp)), 1);
+			XFS_DADDR_TO_FSB(cur->bc_mp, XFS_BUF_ADDR(bp)), 1,
+			&oinfo);
 }
 
 STATIC int
diff --git a/fs/xfs/libxfs/xfs_rmap.h b/fs/xfs/libxfs/xfs_rmap.h
new file mode 100644
index 000000000000..b30e7ad6c7cc
--- /dev/null
+++ b/fs/xfs/libxfs/xfs_rmap.h
@@ -0,0 +1,67 @@
+/*
+ * Copyright (C) 2016 Oracle.  All Rights Reserved.
+ *
+ * Author: Darrick J. Wong <darrick.wong@oracle.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+#ifndef __XFS_RMAP_H__
+#define __XFS_RMAP_H__
+
+static inline void
+xfs_rmap_ag_owner(
+	struct xfs_owner_info	*oi,
+	uint64_t		owner)
+{
+	oi->oi_owner = owner;
+	oi->oi_offset = 0;
+	oi->oi_flags = 0;
+}
+
+static inline void
+xfs_rmap_ino_bmbt_owner(
+	struct xfs_owner_info	*oi,
+	xfs_ino_t		ino,
+	int			whichfork)
+{
+	oi->oi_owner = ino;
+	oi->oi_offset = 0;
+	oi->oi_flags = XFS_OWNER_INFO_BMBT_BLOCK;
+	if (whichfork == XFS_ATTR_FORK)
+		oi->oi_flags |= XFS_OWNER_INFO_ATTR_FORK;
+}
+
+static inline void
+xfs_rmap_ino_owner(
+	struct xfs_owner_info	*oi,
+	xfs_ino_t		ino,
+	int			whichfork,
+	xfs_fileoff_t		offset)
+{
+	oi->oi_owner = ino;
+	oi->oi_offset = offset;
+	oi->oi_flags = 0;
+	if (whichfork == XFS_ATTR_FORK)
+		oi->oi_flags |= XFS_OWNER_INFO_ATTR_FORK;
+}
+
+static inline void
+xfs_rmap_skip_owner_update(
+	struct xfs_owner_info	*oi)
+{
+	oi->oi_owner = XFS_RMAP_OWN_UNKNOWN;
+}
+
+#endif	/* __XFS_RMAP_H__ */
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 8c49f883934a..a8c461735ecb 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -27,6 +27,8 @@
 #include "xfs_buf_item.h"
 #include "xfs_extfree_item.h"
 #include "xfs_log.h"
+#include "xfs_btree.h"
+#include "xfs_rmap.h"
 
 
 kmem_zone_t	*xfs_efi_zone;
@@ -503,6 +505,7 @@ xfs_efi_recover(
 	int			error = 0;
 	xfs_extent_t		*extp;
 	xfs_fsblock_t		startblock_fsb;
+	struct xfs_owner_info	oinfo;
 
 	ASSERT(!test_bit(XFS_EFI_RECOVERED, &efip->efi_flags));
 
@@ -534,10 +537,11 @@ xfs_efi_recover(
 		return error;
 	efdp = xfs_trans_get_efd(tp, efip, efip->efi_format.efi_nextents);
 
+	xfs_rmap_skip_owner_update(&oinfo);
 	for (i = 0; i < efip->efi_format.efi_nextents; i++) {
 		extp = &(efip->efi_format.efi_extents[i]);
 		error = xfs_trans_free_extent(tp, efdp, extp->ext_start,
-					      extp->ext_len);
+					      extp->ext_len, &oinfo);
 		if (error)
 			goto abort_error;
 
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 0a3b822d9f8c..b625b61d1244 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -41,6 +41,7 @@
 #include "xfs_trace.h"
 #include "xfs_log.h"
 #include "xfs_filestream.h"
+#include "xfs_rmap.h"
 
 /*
  * File system operations
@@ -436,6 +437,8 @@ xfs_growfs_data_private(
 	 * There are new blocks in the old last a.g.
 	 */
 	if (new) {
+		struct xfs_owner_info	oinfo;
+
 		/*
 		 * Change the agi length.
 		 */
@@ -463,14 +466,20 @@ xfs_growfs_data_private(
 		       be32_to_cpu(agi->agi_length));
 
 		xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH);
+
 		/*
 		 * Free the new space.
+		 *
+		 * XFS_RMAP_OWN_NULL is used here to tell the rmap btree that
+		 * this doesn't actually exist in the rmap btree.
 		 */
-		error = xfs_free_extent(tp, XFS_AGB_TO_FSB(mp, agno,
-			be32_to_cpu(agf->agf_length) - new), new);
-		if (error) {
+		xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_NULL);
+		error = xfs_free_extent(tp,
+				XFS_AGB_TO_FSB(mp, agno,
+					be32_to_cpu(agf->agf_length) - new),
+				new, &oinfo);
+		if (error)
 			goto error0;
-		}
 	}
 
 	/*
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index a5fe1874f02f..9bc368bb5eb4 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -217,7 +217,7 @@ struct xfs_efd_log_item	*xfs_trans_get_efd(struct xfs_trans *,
 				  uint);
 int		xfs_trans_free_extent(struct xfs_trans *,
 				      struct xfs_efd_log_item *, xfs_fsblock_t,
-				      xfs_extlen_t);
+				      xfs_extlen_t, struct xfs_owner_info *);
 int		xfs_trans_commit(struct xfs_trans *);
 int		__xfs_trans_roll(struct xfs_trans **, struct xfs_inode *, int *);
 int		xfs_trans_roll(struct xfs_trans **, struct xfs_inode *);
diff --git a/fs/xfs/xfs_trans_extfree.c b/fs/xfs/xfs_trans_extfree.c
index b48490290b27..3aca3a7578c2 100644
--- a/fs/xfs/xfs_trans_extfree.c
+++ b/fs/xfs/xfs_trans_extfree.c
@@ -120,13 +120,14 @@ xfs_trans_free_extent(
 	struct xfs_trans	*tp,
 	struct xfs_efd_log_item	*efdp,
 	xfs_fsblock_t		start_block,
-	xfs_extlen_t		ext_len)
+	xfs_extlen_t		ext_len,
+	struct xfs_owner_info	*oinfo)
 {
 	uint			next_extent;
 	struct xfs_extent	*extp;
 	int			error;
 
-	error = xfs_free_extent(tp, start_block, ext_len);
+	error = xfs_free_extent(tp, start_block, ext_len, oinfo);
 
 	/*
 	 * Mark the transaction dirty, even on error. This ensures the
@@ -213,7 +214,8 @@ xfs_extent_free_finish_item(
 	free = container_of(item, struct xfs_extent_free_item, xefi_list);
 	error = xfs_trans_free_extent(tp, done_item,
 			free->xefi_startblock,
-			free->xefi_blockcount);
+			free->xefi_blockcount,
+			&free->xefi_oinfo);
 	kmem_free(free);
 	return error;
 }
-- 
cgit v1.2.3


From 673930c34a4500c616cf9b2bbe1ae131ead2e155 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 11:33:43 +1000
Subject: xfs: introduce rmap extent operation stubs

Originally-From: Dave Chinner <dchinner@redhat.com>

Add the stubs into the extent allocation and freeing paths that the
rmap btree implementation will hook into. While doing this, add the
trace points that will be used to track rmap btree extent
manipulations.

[darrick.wong@oracle.com: Extend the stubs to take full owner info.]

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/Makefile           |  1 +
 fs/xfs/libxfs/xfs_alloc.c | 19 ++++++++--
 fs/xfs/libxfs/xfs_rmap.c  | 89 +++++++++++++++++++++++++++++++++++++++++++++++
 fs/xfs/libxfs/xfs_rmap.h  | 11 ++++++
 fs/xfs/xfs_trace.h        | 77 ++++++++++++++++++++++++++++++++++++++++
 5 files changed, 195 insertions(+), 2 deletions(-)
 create mode 100644 fs/xfs/libxfs/xfs_rmap.c

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 01857b096cae..0498291829c6 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -52,6 +52,7 @@ xfs-y				+= $(addprefix libxfs/, \
 				   xfs_inode_fork.o \
 				   xfs_inode_buf.o \
 				   xfs_log_rlimit.o \
+				   xfs_rmap.o \
 				   xfs_sb.o \
 				   xfs_symlink_remote.o \
 				   xfs_trans_resv.o \
diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 6335b8b85349..dba7ce4b4be8 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -27,6 +27,7 @@
 #include "xfs_defer.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
+#include "xfs_rmap.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
@@ -36,7 +37,6 @@
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_log.h"
-#include "xfs_rmap.h"
 
 struct workqueue_struct *xfs_alloc_wq;
 
@@ -649,6 +649,14 @@ xfs_alloc_ag_vextent(
 	ASSERT(!args->wasfromfl || !args->isfl);
 	ASSERT(args->agbno % args->alignment == 0);
 
+	/* if not file data, insert new block into the reverse map btree */
+	if (args->oinfo.oi_owner != XFS_RMAP_OWN_UNKNOWN) {
+		error = xfs_rmap_alloc(args->tp, args->agbp, args->agno,
+				       args->agbno, args->len, &args->oinfo);
+		if (error)
+			return error;
+	}
+
 	if (!args->wasfromfl) {
 		error = xfs_alloc_update_counters(args->tp, args->pag,
 						  args->agbp,
@@ -1615,12 +1623,19 @@ xfs_free_ag_extent(
 	xfs_extlen_t	nlen;		/* new length of freespace */
 	xfs_perag_t	*pag;		/* per allocation group data */
 
+	bno_cur = cnt_cur = NULL;
 	mp = tp->t_mountp;
+
+	if (oinfo->oi_owner != XFS_RMAP_OWN_UNKNOWN) {
+		error = xfs_rmap_free(tp, agbp, agno, bno, len, oinfo);
+		if (error)
+			goto error0;
+	}
+
 	/*
 	 * Allocate and initialize a cursor for the by-block btree.
 	 */
 	bno_cur = xfs_allocbt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_BNO);
-	cnt_cur = NULL;
 	/*
 	 * Look for a neighboring block on the left (lower block numbers)
 	 * that is contiguous with this space.
diff --git a/fs/xfs/libxfs/xfs_rmap.c b/fs/xfs/libxfs/xfs_rmap.c
new file mode 100644
index 000000000000..b522bfc436a2
--- /dev/null
+++ b/fs/xfs/libxfs/xfs_rmap.c
@@ -0,0 +1,89 @@
+/*
+ * Copyright (c) 2014 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include "xfs.h"
+#include "xfs_fs.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
+#include "xfs_bit.h"
+#include "xfs_sb.h"
+#include "xfs_mount.h"
+#include "xfs_defer.h"
+#include "xfs_da_format.h"
+#include "xfs_da_btree.h"
+#include "xfs_btree.h"
+#include "xfs_trans.h"
+#include "xfs_alloc.h"
+#include "xfs_rmap.h"
+#include "xfs_trans_space.h"
+#include "xfs_trace.h"
+#include "xfs_error.h"
+#include "xfs_extent_busy.h"
+
+int
+xfs_rmap_free(
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp,
+	xfs_agnumber_t		agno,
+	xfs_agblock_t		bno,
+	xfs_extlen_t		len,
+	struct xfs_owner_info	*oinfo)
+{
+	struct xfs_mount	*mp = tp->t_mountp;
+	int			error = 0;
+
+	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
+		return 0;
+
+	trace_xfs_rmap_unmap(mp, agno, bno, len, false, oinfo);
+	if (1)
+		goto out_error;
+	trace_xfs_rmap_unmap_done(mp, agno, bno, len, false, oinfo);
+	return 0;
+
+out_error:
+	trace_xfs_rmap_unmap_error(mp, agno, error, _RET_IP_);
+	return error;
+}
+
+int
+xfs_rmap_alloc(
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp,
+	xfs_agnumber_t		agno,
+	xfs_agblock_t		bno,
+	xfs_extlen_t		len,
+	struct xfs_owner_info	*oinfo)
+{
+	struct xfs_mount	*mp = tp->t_mountp;
+	int			error = 0;
+
+	if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
+		return 0;
+
+	trace_xfs_rmap_map(mp, agno, bno, len, false, oinfo);
+	if (1)
+		goto out_error;
+	trace_xfs_rmap_map_done(mp, agno, bno, len, false, oinfo);
+	return 0;
+
+out_error:
+	trace_xfs_rmap_map_error(mp, agno, error, _RET_IP_);
+	return error;
+}
diff --git a/fs/xfs/libxfs/xfs_rmap.h b/fs/xfs/libxfs/xfs_rmap.h
index b30e7ad6c7cc..dbf0301e9ae1 100644
--- a/fs/xfs/libxfs/xfs_rmap.h
+++ b/fs/xfs/libxfs/xfs_rmap.h
@@ -64,4 +64,15 @@ xfs_rmap_skip_owner_update(
 	oi->oi_owner = XFS_RMAP_OWN_UNKNOWN;
 }
 
+/* Reverse mapping functions. */
+
+struct xfs_buf;
+
+int xfs_rmap_alloc(struct xfs_trans *tp, struct xfs_buf *agbp,
+		   xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		   struct xfs_owner_info *oinfo);
+int xfs_rmap_free(struct xfs_trans *tp, struct xfs_buf *agbp,
+		  xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
+		  struct xfs_owner_info *oinfo);
+
 #endif	/* __XFS_RMAP_H__ */
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index 939caf5f5a45..79e3812499ba 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -2423,6 +2423,83 @@ DEFINE_MAP_EXTENT_DEFERRED_EVENT(xfs_defer_map_extent);
 DEFINE_BMAP_FREE_DEFERRED_EVENT(xfs_bmap_free_defer);
 DEFINE_BMAP_FREE_DEFERRED_EVENT(xfs_bmap_free_deferred);
 
+/* rmap tracepoints */
+DECLARE_EVENT_CLASS(xfs_rmap_class,
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
+		 xfs_agblock_t agbno, xfs_extlen_t len, bool unwritten,
+		 struct xfs_owner_info *oinfo),
+	TP_ARGS(mp, agno, agbno, len, unwritten, oinfo),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, agbno)
+		__field(xfs_extlen_t, len)
+		__field(uint64_t, owner)
+		__field(uint64_t, offset)
+		__field(unsigned long, flags)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->agno = agno;
+		__entry->agbno = agbno;
+		__entry->len = len;
+		__entry->owner = oinfo->oi_owner;
+		__entry->offset = oinfo->oi_offset;
+		__entry->flags = oinfo->oi_flags;
+	),
+	TP_printk("dev %d:%d agno %u agbno %u len %u owner %lld offset %llu flags 0x%lx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno,
+		  __entry->agbno,
+		  __entry->len,
+		  __entry->owner,
+		  __entry->offset,
+		  __entry->flags)
+);
+#define DEFINE_RMAP_EVENT(name) \
+DEFINE_EVENT(xfs_rmap_class, name, \
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \
+		 xfs_agblock_t agbno, xfs_extlen_t len, bool unwritten, \
+		 struct xfs_owner_info *oinfo), \
+	TP_ARGS(mp, agno, agbno, len, unwritten, oinfo))
+
+/* simple AG-based error/%ip tracepoint class */
+DECLARE_EVENT_CLASS(xfs_ag_error_class,
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int error,
+		 unsigned long caller_ip),
+	TP_ARGS(mp, agno, error, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+		__field(int, error)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->agno = agno;
+		__entry->error = error;
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d agno %u error %d caller %ps",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno,
+		  __entry->error,
+		  (char *)__entry->caller_ip)
+);
+
+#define DEFINE_AG_ERROR_EVENT(name) \
+DEFINE_EVENT(xfs_ag_error_class, name, \
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int error, \
+		 unsigned long caller_ip), \
+	TP_ARGS(mp, agno, error, caller_ip))
+
+DEFINE_RMAP_EVENT(xfs_rmap_unmap);
+DEFINE_RMAP_EVENT(xfs_rmap_unmap_done);
+DEFINE_AG_ERROR_EVENT(xfs_rmap_unmap_error);
+DEFINE_RMAP_EVENT(xfs_rmap_map);
+DEFINE_RMAP_EVENT(xfs_rmap_map_done);
+DEFINE_AG_ERROR_EVENT(xfs_rmap_map_error);
+
 #endif /* _TRACE_XFS_H */
 
 #undef TRACE_INCLUDE_PATH
-- 
cgit v1.2.3


From 525488520ac69a3612dbceefa573b255a83005e9 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 11:38:24 +1000
Subject: xfs: rmap btree requires more reserved free space

Originally-From: Dave Chinner <dchinner@redhat.com>

The rmap btree is allocated from the AGFL, which means we have to
ensure ENOSPC is reported to userspace before we run out of free
space in each AG. The last allocation in an AG can cause a full
height rmap btree split, and that means we have to reserve at least
this many blocks *in each AG* to be placed on the AGFL at ENOSPC.
Update the various space calculation functions to handle this.

Also, because the macros are now executing conditional code and are
called quite frequently, convert them to functions that initialise
variables in the struct xfs_mount, use the new variables everywhere
and document the calculations better.

[darrick.wong@oracle.com: don't reserve blocks if !rmap]
[dchinner@redhat.com: update m_ag_max_usable after growfs]

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/libxfs/xfs_alloc.c | 69 +++++++++++++++++++++++++++++++++++++++++++++++
 fs/xfs/libxfs/xfs_alloc.h | 41 ++++------------------------
 fs/xfs/libxfs/xfs_bmap.c  |  2 +-
 fs/xfs/libxfs/xfs_sb.c    |  2 ++
 fs/xfs/xfs_discard.c      |  2 +-
 fs/xfs/xfs_fsops.c        |  5 ++--
 fs/xfs/xfs_log_recover.c  |  1 +
 fs/xfs/xfs_mount.c        |  2 +-
 fs/xfs/xfs_mount.h        |  2 ++
 fs/xfs/xfs_super.c        |  2 +-
 10 files changed, 86 insertions(+), 42 deletions(-)

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index dba7ce4b4be8..73ab1ea1270d 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -62,6 +62,70 @@ xfs_prealloc_blocks(
 	return XFS_IBT_BLOCK(mp) + 1;
 }
 
+/*
+ * In order to avoid ENOSPC-related deadlock caused by out-of-order locking of
+ * AGF buffer (PV 947395), we place constraints on the relationship among
+ * actual allocations for data blocks, freelist blocks, and potential file data
+ * bmap btree blocks. However, these restrictions may result in no actual space
+ * allocated for a delayed extent, for example, a data block in a certain AG is
+ * allocated but there is no additional block for the additional bmap btree
+ * block due to a split of the bmap btree of the file. The result of this may
+ * lead to an infinite loop when the file gets flushed to disk and all delayed
+ * extents need to be actually allocated. To get around this, we explicitly set
+ * aside a few blocks which will not be reserved in delayed allocation.
+ *
+ * When rmap is disabled, we need to reserve 4 fsbs _per AG_ for the freelist
+ * and 4 more to handle a potential split of the file's bmap btree.
+ *
+ * When rmap is enabled, we must also be able to handle two rmap btree inserts
+ * to record both the file data extent and a new bmbt block.  The bmbt block
+ * might not be in the same AG as the file data extent.  In the worst case
+ * the bmap btree splits multiple levels and all the new blocks come from
+ * different AGs, so set aside enough to handle rmap btree splits in all AGs.
+ */
+unsigned int
+xfs_alloc_set_aside(
+	struct xfs_mount	*mp)
+{
+	unsigned int		blocks;
+
+	blocks = 4 + (mp->m_sb.sb_agcount * XFS_ALLOC_AGFL_RESERVE);
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
+		blocks += mp->m_sb.sb_agcount * mp->m_rmap_maxlevels;
+	return blocks;
+}
+
+/*
+ * When deciding how much space to allocate out of an AG, we limit the
+ * allocation maximum size to the size the AG. However, we cannot use all the
+ * blocks in the AG - some are permanently used by metadata. These
+ * blocks are generally:
+ *	- the AG superblock, AGF, AGI and AGFL
+ *	- the AGF (bno and cnt) and AGI btree root blocks, and optionally
+ *	  the AGI free inode and rmap btree root blocks.
+ *	- blocks on the AGFL according to xfs_alloc_set_aside() limits
+ *	- the rmapbt root block
+ *
+ * The AG headers are sector sized, so the amount of space they take up is
+ * dependent on filesystem geometry. The others are all single blocks.
+ */
+unsigned int
+xfs_alloc_ag_max_usable(
+	struct xfs_mount	*mp)
+{
+	unsigned int		blocks;
+
+	blocks = XFS_BB_TO_FSB(mp, XFS_FSS_TO_BB(mp, 4)); /* ag headers */
+	blocks += XFS_ALLOC_AGFL_RESERVE;
+	blocks += 3;			/* AGF, AGI btree root blocks */
+	if (xfs_sb_version_hasfinobt(&mp->m_sb))
+		blocks++;		/* finobt root block */
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
+		blocks++; 		/* rmap root block */
+
+	return mp->m_sb.sb_agblocks - blocks;
+}
+
 /*
  * Lookup the record equal to [bno, len] in the btree given by cur.
  */
@@ -1904,6 +1968,11 @@ xfs_alloc_min_freelist(
 	/* space needed by-size freespace btree */
 	min_free += min_t(unsigned int, pag->pagf_levels[XFS_BTNUM_CNTi] + 1,
 				       mp->m_ag_maxlevels);
+	/* space needed reverse mapping used space btree */
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
+		min_free += min_t(unsigned int,
+				  pag->pagf_levels[XFS_BTNUM_RMAPi] + 1,
+				  mp->m_rmap_maxlevels);
 
 	return min_free;
 }
diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index 88053422fda0..360f9e425536 100644
--- a/fs/xfs/libxfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
@@ -55,42 +55,6 @@ typedef unsigned int xfs_alloctype_t;
 #define	XFS_ALLOC_FLAG_TRYLOCK	0x00000001  /* use trylock for buffer locking */
 #define	XFS_ALLOC_FLAG_FREEING	0x00000002  /* indicate caller is freeing extents*/
 
-/*
- * In order to avoid ENOSPC-related deadlock caused by
- * out-of-order locking of AGF buffer (PV 947395), we place
- * constraints on the relationship among actual allocations for
- * data blocks, freelist blocks, and potential file data bmap
- * btree blocks. However, these restrictions may result in no
- * actual space allocated for a delayed extent, for example, a data
- * block in a certain AG is allocated but there is no additional
- * block for the additional bmap btree block due to a split of the
- * bmap btree of the file. The result of this may lead to an
- * infinite loop in xfssyncd when the file gets flushed to disk and
- * all delayed extents need to be actually allocated. To get around
- * this, we explicitly set aside a few blocks which will not be
- * reserved in delayed allocation. Considering the minimum number of
- * needed freelist blocks is 4 fsbs _per AG_, a potential split of file's bmap
- * btree requires 1 fsb, so we set the number of set-aside blocks
- * to 4 + 4*agcount.
- */
-#define XFS_ALLOC_SET_ASIDE(mp)  (4 + ((mp)->m_sb.sb_agcount * 4))
-
-/*
- * When deciding how much space to allocate out of an AG, we limit the
- * allocation maximum size to the size the AG. However, we cannot use all the
- * blocks in the AG - some are permanently used by metadata. These
- * blocks are generally:
- *	- the AG superblock, AGF, AGI and AGFL
- *	- the AGF (bno and cnt) and AGI btree root blocks
- *	- 4 blocks on the AGFL according to XFS_ALLOC_SET_ASIDE() limits
- *
- * The AG headers are sector sized, so the amount of space they take up is
- * dependent on filesystem geometry. The others are all single blocks.
- */
-#define XFS_ALLOC_AG_MAX_USABLE(mp)	\
-	((mp)->m_sb.sb_agblocks - XFS_BB_TO_FSB(mp, XFS_FSS_TO_BB(mp, 4)) - 7)
-
-
 /*
  * Argument structure for xfs_alloc routines.
  * This is turned into a structure to avoid having 20 arguments passed
@@ -133,6 +97,11 @@ typedef struct xfs_alloc_arg {
 #define XFS_ALLOC_INITIAL_USER_DATA	(1 << 1)/* special case start of file */
 #define XFS_ALLOC_USERDATA_ZERO		(1 << 2)/* zero extent on allocation */
 
+/* freespace limit calculations */
+#define XFS_ALLOC_AGFL_RESERVE	4
+unsigned int xfs_alloc_set_aside(struct xfs_mount *mp);
+unsigned int xfs_alloc_ag_max_usable(struct xfs_mount *mp);
+
 xfs_extlen_t xfs_alloc_longest_free_extent(struct xfs_mount *mp,
 		struct xfs_perag *pag, xfs_extlen_t need);
 unsigned int xfs_alloc_min_freelist(struct xfs_mount *mp,
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 919069fc7b23..c4b89218c7eb 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -3674,7 +3674,7 @@ xfs_bmap_btalloc(
 	xfs_rmap_skip_owner_update(&args.oinfo);
 
 	/* Trim the allocation back to the maximum an AG can fit. */
-	args.maxlen = MIN(ap->length, XFS_ALLOC_AG_MAX_USABLE(mp));
+	args.maxlen = MIN(ap->length, mp->m_ag_max_usable);
 	args.firstblock = *ap->firstblock;
 	blen = 0;
 	if (nullfb) {
diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c
index fb33e70e5c9a..0e3d4f5ec33c 100644
--- a/fs/xfs/libxfs/xfs_sb.c
+++ b/fs/xfs/libxfs/xfs_sb.c
@@ -745,6 +745,8 @@ xfs_sb_mount_common(
 		mp->m_ialloc_min_blks = sbp->sb_spino_align;
 	else
 		mp->m_ialloc_min_blks = mp->m_ialloc_blks;
+	mp->m_alloc_set_aside = xfs_alloc_set_aside(mp);
+	mp->m_ag_max_usable = xfs_alloc_ag_max_usable(mp);
 }
 
 /*
diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c
index 272c3f8b6f7d..4ff499aa7338 100644
--- a/fs/xfs/xfs_discard.c
+++ b/fs/xfs/xfs_discard.c
@@ -179,7 +179,7 @@ xfs_ioc_trim(
 	 * matter as trimming blocks is an advisory interface.
 	 */
 	if (range.start >= XFS_FSB_TO_B(mp, mp->m_sb.sb_dblocks) ||
-	    range.minlen > XFS_FSB_TO_B(mp, XFS_ALLOC_AG_MAX_USABLE(mp)) ||
+	    range.minlen > XFS_FSB_TO_B(mp, mp->m_ag_max_usable) ||
 	    range.len < mp->m_sb.sb_blocksize)
 		return -EINVAL;
 
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 27e6e8cdd431..83304d54e6d4 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -584,6 +584,7 @@ xfs_growfs_data_private(
 	} else
 		mp->m_maxicount = 0;
 	xfs_set_low_space_thresholds(mp);
+	mp->m_alloc_set_aside = xfs_alloc_set_aside(mp);
 
 	/* update secondary superblocks. */
 	for (agno = 1; agno < nagcount; agno++) {
@@ -721,7 +722,7 @@ xfs_fs_counts(
 	cnt->allocino = percpu_counter_read_positive(&mp->m_icount);
 	cnt->freeino = percpu_counter_read_positive(&mp->m_ifree);
 	cnt->freedata = percpu_counter_read_positive(&mp->m_fdblocks) -
-							XFS_ALLOC_SET_ASIDE(mp);
+						mp->m_alloc_set_aside;
 
 	spin_lock(&mp->m_sb_lock);
 	cnt->freertx = mp->m_sb.sb_frextents;
@@ -809,7 +810,7 @@ xfs_reserve_blocks(
 	error = -ENOSPC;
 	do {
 		free = percpu_counter_sum(&mp->m_fdblocks) -
-							XFS_ALLOC_SET_ASIDE(mp);
+						mp->m_alloc_set_aside;
 		if (!free)
 			break;
 
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index af608aade6df..f64d4d06cd95 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -5008,6 +5008,7 @@ xlog_do_recover(
 		xfs_warn(mp, "Failed post-recovery per-ag init: %d", error);
 		return error;
 	}
+	mp->m_alloc_set_aside = xfs_alloc_set_aside(mp);
 
 	xlog_recover_check_summary(log);
 
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 0041866d51ca..faeead671f9f 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -1221,7 +1221,7 @@ xfs_mod_fdblocks(
 		batch = XFS_FDBLOCKS_BATCH;
 
 	__percpu_counter_add(&mp->m_fdblocks, delta, batch);
-	if (__percpu_counter_compare(&mp->m_fdblocks, XFS_ALLOC_SET_ASIDE(mp),
+	if (__percpu_counter_compare(&mp->m_fdblocks, mp->m_alloc_set_aside,
 				     XFS_FDBLOCKS_BATCH) >= 0) {
 		/* we had space! */
 		return 0;
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 0ed0f29ddaf5..b36676cde103 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -123,6 +123,8 @@ typedef struct xfs_mount {
 	uint			m_in_maxlevels;	/* max inobt btree levels. */
 	uint			m_rmap_maxlevels; /* max rmap btree levels */
 	xfs_extlen_t		m_ag_prealloc_blocks; /* reserved ag blocks */
+	uint			m_alloc_set_aside; /* space we can't use */
+	uint			m_ag_max_usable; /* max space per AG */
 	struct radix_tree_root	m_perag_tree;	/* per-ag accounting info */
 	spinlock_t		m_perag_lock;	/* lock for m_perag_tree */
 	struct mutex		m_growlock;	/* growfs mutex */
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 95a11beb48c6..449cadf149f9 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -1075,7 +1075,7 @@ xfs_fs_statfs(
 	statp->f_blocks = sbp->sb_dblocks - lsize;
 	spin_unlock(&mp->m_sb_lock);
 
-	statp->f_bfree = fdblocks - XFS_ALLOC_SET_ASIDE(mp);
+	statp->f_bfree = fdblocks - mp->m_alloc_set_aside;
 	statp->f_bavail = statp->f_bfree;
 
 	fakeinos = statp->f_bfree << sbp->sb_inopblog;
-- 
cgit v1.2.3


From 04f130605ff6fb01a93a0885607921df9c463eed Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 12:19:53 +1000
Subject: xfs: don't update rmapbt when fixing agfl

Allow a caller of xfs_alloc_fix_freelist to disable rmapbt updates
when fixing the AG freelist.  xfs_repair needs this during phase 5
to be able to adjust the freelist while it's reconstructing the rmap
btree; the missing entries will be added back at the very end of
phase 5 once the AGFL contents settle down.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/libxfs/xfs_alloc.c | 18 ++++++++++++++----
 fs/xfs/libxfs/xfs_alloc.h |  3 +++
 2 files changed, 17 insertions(+), 4 deletions(-)

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 73ab1ea1270d..a0dc2999778d 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2090,9 +2090,21 @@ xfs_alloc_fix_freelist(
 	 * 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?
+	 *
+	 * The NOSHRINK flag prevents the AGFL from being shrunk if it's too
+	 * big; the NORMAP flag prevents AGFL expand/shrink operations from
+	 * updating the rmapbt.  Both flags are used in xfs_repair while we're
+	 * rebuilding the rmapbt, and neither are used by the kernel.  They're
+	 * both required to ensure that rmaps are correctly recorded for the
+	 * regenerated AGFL, bnobt, and cntbt.  See repair/phase5.c and
+	 * repair/rmap.c in xfsprogs for details.
 	 */
-	xfs_rmap_ag_owner(&targs.oinfo, XFS_RMAP_OWN_AG);
-	while (pag->pagf_flcount > need) {
+	memset(&targs, 0, sizeof(targs));
+	if (flags & XFS_ALLOC_FLAG_NORMAP)
+		xfs_rmap_skip_owner_update(&targs.oinfo);
+	else
+		xfs_rmap_ag_owner(&targs.oinfo, XFS_RMAP_OWN_AG);
+	while (!(flags & XFS_ALLOC_FLAG_NOSHRINK) && pag->pagf_flcount > need) {
 		struct xfs_buf	*bp;
 
 		error = xfs_alloc_get_freelist(tp, agbp, &bno, 0);
@@ -2106,10 +2118,8 @@ xfs_alloc_fix_freelist(
 		xfs_trans_binval(tp, bp);
 	}
 
-	memset(&targs, 0, sizeof(targs));
 	targs.tp = tp;
 	targs.mp = mp;
-	xfs_rmap_ag_owner(&targs.oinfo, XFS_RMAP_OWN_AG);
 	targs.agbp = agbp;
 	targs.agno = args->agno;
 	targs.alignment = targs.minlen = targs.prod = targs.isfl = 1;
diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index 360f9e425536..6fe2d6b7cfe9 100644
--- a/fs/xfs/libxfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
@@ -54,6 +54,9 @@ typedef unsigned int xfs_alloctype_t;
  */
 #define	XFS_ALLOC_FLAG_TRYLOCK	0x00000001  /* use trylock for buffer locking */
 #define	XFS_ALLOC_FLAG_FREEING	0x00000002  /* indicate caller is freeing extents*/
+#define	XFS_ALLOC_FLAG_NORMAP	0x00000004  /* don't modify the rmapbt */
+#define	XFS_ALLOC_FLAG_NOSHRINK	0x00000008  /* don't shrink the freelist */
+
 
 /*
  * Argument structure for xfs_alloc routines.
-- 
cgit v1.2.3


From 3481b68285238054be519ad0c8cad5cc2425e26c Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Date: Wed, 3 Aug 2016 12:31:07 +1000
Subject: xfs: move (and rename) the deferred bmap-free tracepoints

Rename the deferred bmap-free to extent_free and make them only
trigger when we're really running deferred ops.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
---
 fs/xfs/libxfs/xfs_alloc.c  | 2 --
 fs/xfs/xfs_trans_extfree.c | 7 +++++++
 2 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'fs/xfs/libxfs/xfs_alloc.c')

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index a0dc2999778d..776ae2f325d1 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2819,8 +2819,6 @@ xfs_free_extent(
 
 	ASSERT(len != 0);
 
-	trace_xfs_bmap_free_deferred(mp, agno, 0, agbno, len);
-
 	if (XFS_TEST_ERROR(false, mp,
 			XFS_ERRTAG_FREE_EXTENT,
 			XFS_RANDOM_FREE_EXTENT))
diff --git a/fs/xfs/xfs_trans_extfree.c b/fs/xfs/xfs_trans_extfree.c
index 7d1e84c1b848..459ddec137a4 100644
--- a/fs/xfs/xfs_trans_extfree.c
+++ b/fs/xfs/xfs_trans_extfree.c
@@ -21,6 +21,7 @@
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
+#include "xfs_bit.h"
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_trans.h"
@@ -28,6 +29,7 @@
 #include "xfs_extfree_item.h"
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
+#include "xfs_trace.h"
 
 /*
  * This routine is called to allocate an "extent free done"
@@ -68,10 +70,15 @@ xfs_trans_free_extent(
 	xfs_extlen_t		ext_len,
 	struct xfs_owner_info	*oinfo)
 {
+	struct xfs_mount	*mp = tp->t_mountp;
 	uint			next_extent;
+	xfs_agnumber_t		agno = XFS_FSB_TO_AGNO(mp, start_block);
+	xfs_agblock_t		agbno = XFS_FSB_TO_AGBNO(mp, start_block);
 	struct xfs_extent	*extp;
 	int			error;
 
+	trace_xfs_bmap_free_deferred(tp->t_mountp, agno, 0, agbno, ext_len);
+
 	error = xfs_free_extent(tp, start_block, ext_len, oinfo);
 
 	/*
-- 
cgit v1.2.3