4.13-stable updates for XFS

4.13-stable updates for XFS

[Christoph Hellwig]

Hi all,

below are backports up to 4.14-rc5 for XFS.

[PATCH 01/18] iomap_dio_rw: Allocate AIO completion queue before submitting dio

[Christoph Hellwig]

From: Chandan Rajendra <chandan@linux.vnet.ibm.com>

commit 546e7be8244dc050effef0555df5b8d94d10dafc upstream.

Executing xfs/104 test in a loop on Linux-v4.13 kernel on a ppc64
machine can cause the following NULL pointer dereference,

.queue_work_on+0x4c/0x80
.iomap_dio_bio_end_io+0xbc/0x1f0
.bio_endio+0x118/0x1f0
.blk_update_request+0xd0/0x470
.blk_mq_end_request+0x24/0xc0
.lo_complete_rq+0x40/0xe0
.__blk_mq_complete_request_remote+0x28/0x40
.flush_smp_call_function_queue+0xc4/0x1e0
.smp_ipi_demux_relaxed+0x8c/0x100
.icp_hv_ipi_action+0x54/0xa0
.__handle_irq_event_percpu+0x84/0x2c0
.handle_irq_event_percpu+0x28/0x80
.handle_percpu_irq+0x78/0xc0
.generic_handle_irq+0x40/0x70
.__do_irq+0x88/0x200
.call_do_irq+0x14/0x24
.do_IRQ+0x84/0x130

This occurs due to the following sequence of events,

1. Allocate dio for Direct I/O write.
2. Invoke iomap_apply() until iov_iter_count() bytes have been submitted.
   - Assume that we have submitted atleast one bio. Hence iomap_dio->ref value
     will be >= 2.
   - If during the second iteration, iomap_apply() ends up returning -ENOSPC, we would
     break out of the loop and since the 'ret' value is a negative number we
     end up not allocating memory for super_block->s_dio_done_wq.
3. Meanwhile, iomap_dio_bio_end_io() is invoked for bios that have been
   submitted and here the code ends up dereferencing the NULL pointer stored
   at super_block->s_dio_done_wq.

This commit fixes the bug by allocating memory for
super_block->s_dio_done_wq before iomap_apply() is invoked.

Reported-by: Eryu Guan <eguan@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Tested-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/iomap.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/fs/iomap.c b/fs/iomap.c
index 59cc98ad7577..b8a91064abc6 100644
--- a/fs/iomap.c
+++ b/fs/iomap.c
@@ -993,6 +993,13 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	WARN_ON_ONCE(ret);
 	ret = 0;
 
+	if (iov_iter_rw(iter) == WRITE && !is_sync_kiocb(iocb) &&
+	    !inode->i_sb->s_dio_done_wq) {
+		ret = sb_init_dio_done_wq(inode->i_sb);
+		if (ret < 0)
+			goto out_free_dio;
+	}
+
 	inode_dio_begin(inode);
 
 	blk_start_plug(&plug);
@@ -1015,13 +1022,6 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 	if (ret < 0)
 		iomap_dio_set_error(dio, ret);
 
-	if (ret >= 0 && iov_iter_rw(iter) == WRITE && !is_sync_kiocb(iocb) &&
-			!inode->i_sb->s_dio_done_wq) {
-		ret = sb_init_dio_done_wq(inode->i_sb);
-		if (ret < 0)
-			iomap_dio_set_error(dio, ret);
-	}
-
 	if (!atomic_dec_and_test(&dio->ref)) {
 		if (!is_sync_kiocb(iocb))
 			return -EIOCBQUEUED;
-- 
2.14.2

[PATCH 02/18] xfs: don't unconditionally clear the reflink flag on zero-block files

[Christoph Hellwig]

From: "Darrick J. Wong" <darrick.wong@oracle.com>

commit cc6f77710a6de6210f9feda7cd53e2f5ee7a7e69 upstream.

If we have speculative cow preallocations hanging around in the cow
fork, don't let a truncate operation clear the reflink flag because if
we do then there's a chance we'll forget to free those extents when we
destroy the incore inode.

Reported-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_inode.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 97045e8dfed5..cd2e5b122b01 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1623,10 +1623,12 @@ xfs_itruncate_extents(
 		goto out;
 
 	/*
-	 * Clear the reflink flag if we truncated everything.
+	 * Clear the reflink flag if there are no data fork blocks and
+	 * there are no extents staged in the cow fork.
 	 */
-	if (ip->i_d.di_nblocks == 0 && xfs_is_reflink_inode(ip)) {
-		ip->i_d.di_flags2 &= ~XFS_DIFLAG2_REFLINK;
+	if (xfs_is_reflink_inode(ip) && ip->i_cnextents == 0) {
+		if (ip->i_d.di_nblocks == 0)
+			ip->i_d.di_flags2 &= ~XFS_DIFLAG2_REFLINK;
 		xfs_inode_clear_cowblocks_tag(ip);
 	}
 
-- 
2.14.2

[PATCH 03/18] xfs: evict CoW fork extents when performing finsert/fcollapse

[Christoph Hellwig]

From: "Darrick J. Wong" <darrick.wong@oracle.com>

commit 3af423b03435c81036fa710623d3ae92fbe346a3 upstream.

When we perform an finsert/fcollapse operation, cancel all the CoW
extents for the affected file offset range so that they don't end up
pointing to the wrong blocks.

Reported-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_bmap_util.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 3e9b7a4fb8fd..1de4e96df15a 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -1459,7 +1459,19 @@ xfs_shift_file_space(
 		return error;
 
 	/*
-	 * The extent shiting code works on extent granularity. So, if
+	 * Clean out anything hanging around in the cow fork now that
+	 * we've flushed all the dirty data out to disk to avoid having
+	 * CoW extents at the wrong offsets.
+	 */
+	if (xfs_is_reflink_inode(ip)) {
+		error = xfs_reflink_cancel_cow_range(ip, offset, NULLFILEOFF,
+				true);
+		if (error)
+			return error;
+	}
+
+	/*
+	 * The extent shifting code works on extent granularity. So, if
 	 * stop_fsb is not the starting block of extent, we need to split
 	 * the extent at stop_fsb.
 	 */
-- 
2.14.2

[PATCH 04/18] fs/xfs: Use %pS printk format for direct addresses

[Christoph Hellwig]

From: Helge Deller <deller@gmx.de>

commit e150dcd459e1b441eaf08f341a986f04e61bf3b8 upstream.

Use the %pS instead of the %pF printk format specifier for printing symbols
from direct addresses. This is needed for the ia64, ppc64 and parisc64
architectures.

Signed-off-by: Helge Deller <deller@gmx.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_error.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 2f4feb959bfb..028e50a36f95 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -344,7 +344,7 @@ xfs_verifier_error(
 {
 	struct xfs_mount *mp = bp->b_target->bt_mount;
 
-	xfs_alert(mp, "Metadata %s detected at %pF, %s block 0x%llx",
+	xfs_alert(mp, "Metadata %s detected at %pS, %s block 0x%llx",
 		  bp->b_error == -EFSBADCRC ? "CRC error" : "corruption",
 		  __return_address, bp->b_ops->name, bp->b_bn);
 
-- 
2.14.2

[PATCH 05/18] xfs: report zeroed or not correctly in xfs_zero_range()

[Christoph Hellwig]

From: Eryu Guan <eguan@redhat.com>

commit d20a5e3851969fa685f118a80e4df670255a4e8d upstream.

The 'did_zero' param of xfs_zero_range() was not passed to
iomap_zero_range() correctly. This was introduced by commit
7bb41db3ea16 ("xfs: handle 64-bit length in xfs_iozero"), and found
by code inspection.

Signed-off-by: Eryu Guan <eguan@redhat.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index c4893e226fd8..812cd17b331f 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -58,7 +58,7 @@ xfs_zero_range(
 	xfs_off_t		count,
 	bool			*did_zero)
 {
-	return iomap_zero_range(VFS_I(ip), pos, count, NULL, &xfs_iomap_ops);
+	return iomap_zero_range(VFS_I(ip), pos, count, did_zero, &xfs_iomap_ops);
 }
 
 int
-- 
2.14.2

[PATCH 06/18] xfs: update i_size after unwritten conversion in dio completion

[Christoph Hellwig]

From: Eryu Guan <eguan@redhat.com>

commit ee70daaba82d70766d0723b743d9fdeb3b06102a upstream.

Since commit d531d91d6990 ("xfs: always use unwritten extents for
direct I/O writes"), we start allocating unwritten extents for all
direct writes to allow appending aio in XFS.

But for dio writes that could extend file size we update the in-core
inode size first, then convert the unwritten extents to real
allocations at dio completion time in xfs_dio_write_end_io(). Thus a
racing direct read could see the new i_size and find the unwritten
extents first and read zeros instead of actual data, if the direct
writer also takes a shared iolock.

Fix it by updating the in-core inode size after the unwritten extent
conversion. To do this, introduce a new boolean argument to
xfs_iomap_write_unwritten() to tell if we want to update in-core
i_size or not.

Suggested-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_aops.c  |  3 ++-
 fs/xfs/xfs_file.c  | 33 +++++++++++++++++++--------------
 fs/xfs/xfs_iomap.c |  7 +++++--
 fs/xfs/xfs_iomap.h |  2 +-
 fs/xfs/xfs_pnfs.c  |  2 +-
 5 files changed, 28 insertions(+), 19 deletions(-)

diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index f9efd67f6fa1..78aee670ad61 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -330,7 +330,8 @@ xfs_end_io(
 		error = xfs_reflink_end_cow(ip, offset, size);
 		break;
 	case XFS_IO_UNWRITTEN:
-		error = xfs_iomap_write_unwritten(ip, offset, size);
+		/* writeback should never update isize */
+		error = xfs_iomap_write_unwritten(ip, offset, size, false);
 		break;
 	default:
 		ASSERT(!xfs_ioend_is_append(ioend) || ioend->io_append_trans);
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 812cd17b331f..ad5100ce8c44 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -432,7 +432,6 @@ xfs_dio_write_end_io(
 	struct inode		*inode = file_inode(iocb->ki_filp);
 	struct xfs_inode	*ip = XFS_I(inode);
 	loff_t			offset = iocb->ki_pos;
-	bool			update_size = false;
 	int			error = 0;
 
 	trace_xfs_end_io_direct_write(ip, offset, size);
@@ -443,6 +442,21 @@ xfs_dio_write_end_io(
 	if (size <= 0)
 		return size;
 
+	if (flags & IOMAP_DIO_COW) {
+		error = xfs_reflink_end_cow(ip, offset, size);
+		if (error)
+			return error;
+	}
+
+	/*
+	 * Unwritten conversion updates the in-core isize after extent
+	 * conversion but before updating the on-disk size. Updating isize any
+	 * earlier allows a racing dio read to find unwritten extents before
+	 * they are converted.
+	 */
+	if (flags & IOMAP_DIO_UNWRITTEN)
+		return xfs_iomap_write_unwritten(ip, offset, size, true);
+
 	/*
 	 * We need to update the in-core inode size here so that we don't end up
 	 * with the on-disk inode size being outside the in-core inode size. We
@@ -457,20 +471,11 @@ xfs_dio_write_end_io(
 	spin_lock(&ip->i_flags_lock);
 	if (offset + size > i_size_read(inode)) {
 		i_size_write(inode, offset + size);
-		update_size = true;
-	}
-	spin_unlock(&ip->i_flags_lock);
-
-	if (flags & IOMAP_DIO_COW) {
-		error = xfs_reflink_end_cow(ip, offset, size);
-		if (error)
-			return error;
-	}
-
-	if (flags & IOMAP_DIO_UNWRITTEN)
-		error = xfs_iomap_write_unwritten(ip, offset, size);
-	else if (update_size)
+		spin_unlock(&ip->i_flags_lock);
 		error = xfs_setfilesize(ip, offset, size);
+	} else {
+		spin_unlock(&ip->i_flags_lock);
+	}
 
 	return error;
 }
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 813394c62849..2cef389d8955 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -829,7 +829,8 @@ int
 xfs_iomap_write_unwritten(
 	xfs_inode_t	*ip,
 	xfs_off_t	offset,
-	xfs_off_t	count)
+	xfs_off_t	count,
+	bool		update_isize)
 {
 	xfs_mount_t	*mp = ip->i_mount;
 	xfs_fileoff_t	offset_fsb;
@@ -840,6 +841,7 @@ xfs_iomap_write_unwritten(
 	xfs_trans_t	*tp;
 	xfs_bmbt_irec_t imap;
 	struct xfs_defer_ops dfops;
+	struct inode	*inode = VFS_I(ip);
 	xfs_fsize_t	i_size;
 	uint		resblks;
 	int		error;
@@ -899,7 +901,8 @@ xfs_iomap_write_unwritten(
 		i_size = XFS_FSB_TO_B(mp, offset_fsb + count_fsb);
 		if (i_size > offset + count)
 			i_size = offset + count;
-
+		if (update_isize && i_size > i_size_read(inode))
+			i_size_write(inode, i_size);
 		i_size = xfs_new_eof(ip, i_size);
 		if (i_size) {
 			ip->i_d.di_size = i_size;
diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h
index 00db3ecea084..ee535065c5d0 100644
--- a/fs/xfs/xfs_iomap.h
+++ b/fs/xfs/xfs_iomap.h
@@ -27,7 +27,7 @@ int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
 			struct xfs_bmbt_irec *, int);
 int xfs_iomap_write_allocate(struct xfs_inode *, int, xfs_off_t,
 			struct xfs_bmbt_irec *);
-int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, xfs_off_t);
+int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, xfs_off_t, bool);
 
 void xfs_bmbt_to_iomap(struct xfs_inode *, struct iomap *,
 		struct xfs_bmbt_irec *);
diff --git a/fs/xfs/xfs_pnfs.c b/fs/xfs/xfs_pnfs.c
index 2f2dc3c09ad0..4246876df7b7 100644
--- a/fs/xfs/xfs_pnfs.c
+++ b/fs/xfs/xfs_pnfs.c
@@ -274,7 +274,7 @@ xfs_fs_commit_blocks(
 					(end - 1) >> PAGE_SHIFT);
 		WARN_ON_ONCE(error);
 
-		error = xfs_iomap_write_unwritten(ip, start, length);
+		error = xfs_iomap_write_unwritten(ip, start, length, false);
 		if (error)
 			goto out_drop_iolock;
 	}
-- 
2.14.2

[PATCH 07/18] xfs: perag initialization should only touch m_ag_max_usable for AG 0

[Christoph Hellwig]

From: "Darrick J. Wong" <darrick.wong@oracle.com>

commit 9789dd9e1d939232e8ff4c50ef8e75aa6781b3fb upstream.

We call __xfs_ag_resv_init to make a per-AG reservation for each AG.
This makes the reservation per-AG, not per-filesystem.  Therefore, it
is incorrect to adjust m_ag_max_usable for each AG.  Adjust it only
when we're reserving AG 0's blocks so that we only do it once per fs.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
---
 fs/xfs/libxfs/xfs_ag_resv.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c
index b008ff3250eb..df3e600835e8 100644
--- a/fs/xfs/libxfs/xfs_ag_resv.c
+++ b/fs/xfs/libxfs/xfs_ag_resv.c
@@ -156,7 +156,8 @@ __xfs_ag_resv_free(
 	trace_xfs_ag_resv_free(pag, type, 0);
 
 	resv = xfs_perag_resv(pag, type);
-	pag->pag_mount->m_ag_max_usable += resv->ar_asked;
+	if (pag->pag_agno == 0)
+		pag->pag_mount->m_ag_max_usable += resv->ar_asked;
 	/*
 	 * AGFL blocks are always considered "free", so whatever
 	 * was reserved at mount time must be given back at umount.
@@ -216,7 +217,14 @@ __xfs_ag_resv_init(
 		return error;
 	}
 
-	mp->m_ag_max_usable -= ask;
+	/*
+	 * Reduce the maximum per-AG allocation length by however much we're
+	 * trying to reserve for an AG.  Since this is a filesystem-wide
+	 * counter, we only make the adjustment for AG 0.  This assumes that
+	 * there aren't any AGs hungrier for per-AG reservation than AG 0.
+	 */
+	if (pag->pag_agno == 0)
+		mp->m_ag_max_usable -= ask;
 
 	resv = xfs_perag_resv(pag, type);
 	resv->ar_asked = ask;
-- 
2.14.2

[PATCH 08/18] xfs: Capture state of the right inode in xfs_iflush_done

[Christoph Hellwig]

From: Carlos Maiolino <cmaiolino@redhat.com>

commit 842f6e9f786226c58fcbd5ef80eadca72fdfe652 upstream.

My previous patch: d3a304b6292168b83b45d624784f973fdc1ca674 check for
XFS_LI_FAILED flag xfs_iflush done, so the failed item can be properly
resubmitted.

In the loop scanning other inodes being completed, it should check the
current item for the XFS_LI_FAILED, and not the initial one.

The state of the initial inode is checked after the loop ends

Kudos to Eric for catching this.

Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_inode_item.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 6d0f74ec31e8..a705f34b58fa 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -745,7 +745,7 @@ xfs_iflush_done(
 		 */
 		iip = INODE_ITEM(blip);
 		if ((iip->ili_logged && blip->li_lsn == iip->ili_flush_lsn) ||
-		    lip->li_flags & XFS_LI_FAILED)
+		    (blip->li_flags & XFS_LI_FAILED))
 			need_ail++;
 
 		blip = next;
-- 
2.14.2

[PATCH 09/18] xfs: always swap the cow forks when swapping extents

[Christoph Hellwig]

From: "Darrick J. Wong" <darrick.wong@oracle.com>

commit 52bfcdd7adbc26639bc7b2356ab9a3f5dad68ad6 upstream.

Since the CoW fork exists as a secondary data structure to the data
fork, we must always swap cow forks during swapext.  We also need to
swap the extent counts and reset the cowblocks tags.

Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_bmap_util.c | 24 ++++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 1de4e96df15a..3f97a4122b17 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -2121,11 +2121,31 @@ xfs_swap_extents(
 		ip->i_d.di_flags2 |= tip->i_d.di_flags2 & XFS_DIFLAG2_REFLINK;
 		tip->i_d.di_flags2 &= ~XFS_DIFLAG2_REFLINK;
 		tip->i_d.di_flags2 |= f & XFS_DIFLAG2_REFLINK;
+	}
+
+	/* Swap the cow forks. */
+	if (xfs_sb_version_hasreflink(&mp->m_sb)) {
+		xfs_extnum_t	extnum;
+
+		ASSERT(ip->i_cformat == XFS_DINODE_FMT_EXTENTS);
+		ASSERT(tip->i_cformat == XFS_DINODE_FMT_EXTENTS);
+
+		extnum = ip->i_cnextents;
+		ip->i_cnextents = tip->i_cnextents;
+		tip->i_cnextents = extnum;
+
 		cowfp = ip->i_cowfp;
 		ip->i_cowfp = tip->i_cowfp;
 		tip->i_cowfp = cowfp;
-		xfs_inode_set_cowblocks_tag(ip);
-		xfs_inode_set_cowblocks_tag(tip);
+
+		if (ip->i_cowfp && ip->i_cnextents)
+			xfs_inode_set_cowblocks_tag(ip);
+		else
+			xfs_inode_clear_cowblocks_tag(ip);
+		if (tip->i_cowfp && tip->i_cnextents)
+			xfs_inode_set_cowblocks_tag(tip);
+		else
+			xfs_inode_clear_cowblocks_tag(tip);
 	}
 
 	xfs_trans_log_inode(tp, ip,  src_log_flags);
-- 
2.14.2

[PATCH 10/18] xfs: handle racy AIO in xfs_reflink_end_cow

[Christoph Hellwig]

commit e12199f85d0ad1b04ce6c425ad93cd847fe930bb upstream.

If we got two AIO writes into a COW area the second one might not have any
COW extents left to convert.  Handle that case gracefully instead of
triggering an assert or accessing beyond the bounds of the extent list.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_reflink.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c
index f45fbf0db9bb..8c8390a7f121 100644
--- a/fs/xfs/xfs_reflink.c
+++ b/fs/xfs/xfs_reflink.c
@@ -735,7 +735,13 @@ xfs_reflink_end_cow(
 	/* If there is a hole at end_fsb - 1 go to the previous extent */
 	if (!xfs_iext_lookup_extent(ip, ifp, end_fsb - 1, &idx, &got) ||
 	    got.br_startoff > end_fsb) {
-		ASSERT(idx > 0);
+		/*
+		 * In case of racing, overlapping AIO writes no COW extents
+		 * might be left by the time I/O completes for the loser of
+		 * the race.  In that case we are done.
+		 */
+		if (idx <= 0)
+			goto out_cancel;
 		xfs_iext_get_extent(ifp, --idx, &got);
 	}
 
@@ -807,6 +813,7 @@ xfs_reflink_end_cow(
 
 out_defer:
 	xfs_defer_cancel(&dfops);
+out_cancel:
 	xfs_trans_cancel(tp);
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 out:
-- 
2.14.2

[PATCH 11/18] xfs: Don't log uninitialised fields in inode structures

[Christoph Hellwig]

From: Dave Chinner <dchinner@redhat.com>

commit 20413e37d71befd02b5846acdaf5e2564dd1c38e upstream.

Prevent kmemcheck from throwing warnings about reading uninitialised
memory when formatting inodes into the incore log buffer. There are
several issues here - we don't always log all the fields in the
inode log format item, and we never log the inode the
di_next_unlinked field.

In the case of the inode log format item, this is exacerbated
by the old xfs_inode_log_format structure padding issue. Hence make
the padded, 64 bit aligned version of the structure the one we always
use for formatting the log and get rid of the 64 bit variant. This
means we'll always log the 64-bit version and so recovery only needs
to convert from the unpadded 32 bit version from older 32 bit
kernels.

Signed-Off-By: Dave Chinner <dchinner@redhat.com>
Tested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/libxfs/xfs_log_format.h | 27 +++++----------
 fs/xfs/xfs_inode_item.c        | 79 ++++++++++++++++++++++--------------------
 fs/xfs/xfs_ondisk.h            |  2 +-
 3 files changed, 50 insertions(+), 58 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h
index 8372e9bcd7b6..71de185735e0 100644
--- a/fs/xfs/libxfs/xfs_log_format.h
+++ b/fs/xfs/libxfs/xfs_log_format.h
@@ -270,6 +270,7 @@ typedef struct xfs_inode_log_format {
 	uint32_t		ilf_fields;	/* flags for fields logged */
 	uint16_t		ilf_asize;	/* size of attr d/ext/root */
 	uint16_t		ilf_dsize;	/* size of data/ext/root */
+	uint32_t		ilf_pad;	/* pad for 64 bit boundary */
 	uint64_t		ilf_ino;	/* inode number */
 	union {
 		uint32_t	ilfu_rdev;	/* rdev value for dev inode*/
@@ -280,29 +281,17 @@ typedef struct xfs_inode_log_format {
 	int32_t			ilf_boffset;	/* off of inode in buffer */
 } xfs_inode_log_format_t;
 
-typedef struct xfs_inode_log_format_32 {
-	uint16_t		ilf_type;	/* inode log item type */
-	uint16_t		ilf_size;	/* size of this item */
-	uint32_t		ilf_fields;	/* flags for fields logged */
-	uint16_t		ilf_asize;	/* size of attr d/ext/root */
-	uint16_t		ilf_dsize;	/* size of data/ext/root */
-	uint64_t		ilf_ino;	/* inode number */
-	union {
-		uint32_t	ilfu_rdev;	/* rdev value for dev inode*/
-		uuid_t		ilfu_uuid;	/* mount point value */
-	} ilf_u;
-	int64_t			ilf_blkno;	/* blkno of inode buffer */
-	int32_t			ilf_len;	/* len of inode buffer */
-	int32_t			ilf_boffset;	/* off of inode in buffer */
-} __attribute__((packed)) xfs_inode_log_format_32_t;
-
-typedef struct xfs_inode_log_format_64 {
+/*
+ * Old 32 bit systems will log in this format without the 64 bit
+ * alignment padding. Recovery will detect this and convert it to the
+ * correct format.
+ */
+struct xfs_inode_log_format_32 {
 	uint16_t		ilf_type;	/* inode log item type */
 	uint16_t		ilf_size;	/* size of this item */
 	uint32_t		ilf_fields;	/* flags for fields logged */
 	uint16_t		ilf_asize;	/* size of attr d/ext/root */
 	uint16_t		ilf_dsize;	/* size of data/ext/root */
-	uint32_t		ilf_pad;	/* pad for 64 bit boundary */
 	uint64_t		ilf_ino;	/* inode number */
 	union {
 		uint32_t	ilfu_rdev;	/* rdev value for dev inode*/
@@ -311,7 +300,7 @@ typedef struct xfs_inode_log_format_64 {
 	int64_t			ilf_blkno;	/* blkno of inode buffer */
 	int32_t			ilf_len;	/* len of inode buffer */
 	int32_t			ilf_boffset;	/* off of inode in buffer */
-} xfs_inode_log_format_64_t;
+} __attribute__((packed));
 
 
 /*
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index a705f34b58fa..9bbc2d7cc8cb 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -364,6 +364,9 @@ xfs_inode_to_log_dinode(
 	to->di_dmstate = from->di_dmstate;
 	to->di_flags = from->di_flags;
 
+	/* log a dummy value to ensure log structure is fully initialised */
+	to->di_next_unlinked = NULLAGINO;
+
 	if (from->di_version == 3) {
 		to->di_changecount = inode->i_version;
 		to->di_crtime.t_sec = from->di_crtime.t_sec;
@@ -404,6 +407,11 @@ xfs_inode_item_format_core(
  * the second with the on-disk inode structure, and a possible third and/or
  * fourth with the inode data/extents/b-tree root and inode attributes
  * data/extents/b-tree root.
+ *
+ * Note: Always use the 64 bit inode log format structure so we don't
+ * leave an uninitialised hole in the format item on 64 bit systems. Log
+ * recovery on 32 bit systems handles this just fine, so there's no reason
+ * for not using an initialising the properly padded structure all the time.
  */
 STATIC void
 xfs_inode_item_format(
@@ -412,8 +420,8 @@ xfs_inode_item_format(
 {
 	struct xfs_inode_log_item *iip = INODE_ITEM(lip);
 	struct xfs_inode	*ip = iip->ili_inode;
-	struct xfs_inode_log_format *ilf;
 	struct xfs_log_iovec	*vecp = NULL;
+	struct xfs_inode_log_format *ilf;
 
 	ASSERT(ip->i_d.di_version > 1);
 
@@ -425,7 +433,17 @@ xfs_inode_item_format(
 	ilf->ilf_boffset = ip->i_imap.im_boffset;
 	ilf->ilf_fields = XFS_ILOG_CORE;
 	ilf->ilf_size = 2; /* format + core */
-	xlog_finish_iovec(lv, vecp, sizeof(struct xfs_inode_log_format));
+
+	/*
+	 * make sure we don't leak uninitialised data into the log in the case
+	 * when we don't log every field in the inode.
+	 */
+	ilf->ilf_dsize = 0;
+	ilf->ilf_asize = 0;
+	ilf->ilf_pad = 0;
+	uuid_copy(&ilf->ilf_u.ilfu_uuid, &uuid_null);
+
+	xlog_finish_iovec(lv, vecp, sizeof(*ilf));
 
 	xfs_inode_item_format_core(ip, lv, &vecp);
 	xfs_inode_item_format_data_fork(iip, ilf, lv, &vecp);
@@ -855,44 +873,29 @@ xfs_istale_done(
 }
 
 /*
- * convert an xfs_inode_log_format struct from either 32 or 64 bit versions
- * (which can have different field alignments) to the native version
+ * convert an xfs_inode_log_format struct from the old 32 bit version
+ * (which can have different field alignments) to the native 64 bit version
  */
 int
 xfs_inode_item_format_convert(
-	xfs_log_iovec_t		*buf,
-	xfs_inode_log_format_t	*in_f)
+	struct xfs_log_iovec		*buf,
+	struct xfs_inode_log_format	*in_f)
 {
-	if (buf->i_len == sizeof(xfs_inode_log_format_32_t)) {
-		xfs_inode_log_format_32_t *in_f32 = buf->i_addr;
-
-		in_f->ilf_type = in_f32->ilf_type;
-		in_f->ilf_size = in_f32->ilf_size;
-		in_f->ilf_fields = in_f32->ilf_fields;
-		in_f->ilf_asize = in_f32->ilf_asize;
-		in_f->ilf_dsize = in_f32->ilf_dsize;
-		in_f->ilf_ino = in_f32->ilf_ino;
-		/* copy biggest field of ilf_u */
-		uuid_copy(&in_f->ilf_u.ilfu_uuid, &in_f32->ilf_u.ilfu_uuid);
-		in_f->ilf_blkno = in_f32->ilf_blkno;
-		in_f->ilf_len = in_f32->ilf_len;
-		in_f->ilf_boffset = in_f32->ilf_boffset;
-		return 0;
-	} else if (buf->i_len == sizeof(xfs_inode_log_format_64_t)){
-		xfs_inode_log_format_64_t *in_f64 = buf->i_addr;
-
-		in_f->ilf_type = in_f64->ilf_type;
-		in_f->ilf_size = in_f64->ilf_size;
-		in_f->ilf_fields = in_f64->ilf_fields;
-		in_f->ilf_asize = in_f64->ilf_asize;
-		in_f->ilf_dsize = in_f64->ilf_dsize;
-		in_f->ilf_ino = in_f64->ilf_ino;
-		/* copy biggest field of ilf_u */
-		uuid_copy(&in_f->ilf_u.ilfu_uuid, &in_f64->ilf_u.ilfu_uuid);
-		in_f->ilf_blkno = in_f64->ilf_blkno;
-		in_f->ilf_len = in_f64->ilf_len;
-		in_f->ilf_boffset = in_f64->ilf_boffset;
-		return 0;
-	}
-	return -EFSCORRUPTED;
+	struct xfs_inode_log_format_32	*in_f32 = buf->i_addr;
+
+	if (buf->i_len != sizeof(*in_f32))
+		return -EFSCORRUPTED;
+
+	in_f->ilf_type = in_f32->ilf_type;
+	in_f->ilf_size = in_f32->ilf_size;
+	in_f->ilf_fields = in_f32->ilf_fields;
+	in_f->ilf_asize = in_f32->ilf_asize;
+	in_f->ilf_dsize = in_f32->ilf_dsize;
+	in_f->ilf_ino = in_f32->ilf_ino;
+	/* copy biggest field of ilf_u */
+	uuid_copy(&in_f->ilf_u.ilfu_uuid, &in_f32->ilf_u.ilfu_uuid);
+	in_f->ilf_blkno = in_f32->ilf_blkno;
+	in_f->ilf_len = in_f32->ilf_len;
+	in_f->ilf_boffset = in_f32->ilf_boffset;
+	return 0;
 }
diff --git a/fs/xfs/xfs_ondisk.h b/fs/xfs/xfs_ondisk.h
index 0c381d71b242..0492436a053f 100644
--- a/fs/xfs/xfs_ondisk.h
+++ b/fs/xfs/xfs_ondisk.h
@@ -134,7 +134,7 @@ xfs_check_ondisk_structs(void)
 	XFS_CHECK_STRUCT_SIZE(struct xfs_icreate_log,		28);
 	XFS_CHECK_STRUCT_SIZE(struct xfs_ictimestamp,		8);
 	XFS_CHECK_STRUCT_SIZE(struct xfs_inode_log_format_32,	52);
-	XFS_CHECK_STRUCT_SIZE(struct xfs_inode_log_format_64,	56);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_inode_log_format,	56);
 	XFS_CHECK_STRUCT_SIZE(struct xfs_qoff_logformat,	20);
 	XFS_CHECK_STRUCT_SIZE(struct xfs_trans_header,		16);
 }
-- 
2.14.2

[PATCH 12/18] xfs: move more RT specific code under CONFIG_XFS_RT

[Christoph Hellwig]

From: Dave Chinner <dchinner@redhat.com>

commit bb9c2e5433250f5b477035dc478314f8e6dd5e36 upstream.

Various utility functions and interfaces that iterate internal
devices try to reference the realtime device even when RT support is
not compiled into the kernel.

Make sure this code is excluded from the CONFIG_XFS_RT=n build,
and where appropriate stub functions to return fatal errors if
they ever get called when RT support is not present.

Signed-Off-By: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_bmap_util.c |  2 ++
 fs/xfs/xfs_bmap_util.h | 13 +++++++++++++
 fs/xfs/xfs_fsmap.c     | 12 ++++++++++++
 3 files changed, 27 insertions(+)

diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 3f97a4122b17..799c69a72f2c 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -84,6 +84,7 @@ xfs_zero_extent(
 		GFP_NOFS, 0);
 }
 
+#ifdef CONFIG_XFS_RT
 int
 xfs_bmap_rtalloc(
 	struct xfs_bmalloca	*ap)	/* bmap alloc argument struct */
@@ -190,6 +191,7 @@ xfs_bmap_rtalloc(
 	}
 	return 0;
 }
+#endif /* CONFIG_XFS_RT */
 
 /*
  * Check if the endoff is outside the last extent. If so the caller will grow
diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h
index 0cede1043571..fb8d0c7d1db8 100644
--- a/fs/xfs/xfs_bmap_util.h
+++ b/fs/xfs/xfs_bmap_util.h
@@ -28,7 +28,20 @@ struct xfs_mount;
 struct xfs_trans;
 struct xfs_bmalloca;
 
+#ifdef CONFIG_XFS_RT
 int	xfs_bmap_rtalloc(struct xfs_bmalloca *ap);
+#else /* !CONFIG_XFS_RT */
+/*
+ * Attempts to allocate RT extents when RT is disable indicates corruption and
+ * should trigger a shutdown.
+ */
+static inline int
+xfs_bmap_rtalloc(struct xfs_bmalloca *ap)
+{
+	return -EFSCORRUPTED;
+}
+#endif /* CONFIG_XFS_RT */
+
 int	xfs_bmap_eof(struct xfs_inode *ip, xfs_fileoff_t endoff,
 		     int whichfork, int *eof);
 int	xfs_bmap_punch_delalloc_range(struct xfs_inode *ip,
diff --git a/fs/xfs/xfs_fsmap.c b/fs/xfs/xfs_fsmap.c
index 814ed729881d..560e0b40ac1b 100644
--- a/fs/xfs/xfs_fsmap.c
+++ b/fs/xfs/xfs_fsmap.c
@@ -521,6 +521,7 @@ __xfs_getfsmap_rtdev(
 	return query_fn(tp, info);
 }
 
+#ifdef CONFIG_XFS_RT
 /* Actually query the realtime bitmap. */
 STATIC int
 xfs_getfsmap_rtdev_rtbitmap_query(
@@ -561,6 +562,7 @@ xfs_getfsmap_rtdev_rtbitmap(
 	return __xfs_getfsmap_rtdev(tp, keys, xfs_getfsmap_rtdev_rtbitmap_query,
 			info);
 }
+#endif /* CONFIG_XFS_RT */
 
 /* Execute a getfsmap query against the regular data device. */
 STATIC int
@@ -795,7 +797,15 @@ xfs_getfsmap_check_keys(
 	return false;
 }
 
+/*
+ * There are only two devices if we didn't configure RT devices at build time.
+ */
+#ifdef CONFIG_XFS_RT
 #define XFS_GETFSMAP_DEVS	3
+#else
+#define XFS_GETFSMAP_DEVS	2
+#endif /* CONFIG_XFS_RT */
+
 /*
  * Get filesystem's extents as described in head, and format for
  * output.  Calls formatter to fill the user's buffer until all
@@ -853,10 +863,12 @@ xfs_getfsmap(
 		handlers[1].dev = new_encode_dev(mp->m_logdev_targp->bt_dev);
 		handlers[1].fn = xfs_getfsmap_logdev;
 	}
+#ifdef CONFIG_XFS_RT
 	if (mp->m_rtdev_targp) {
 		handlers[2].dev = new_encode_dev(mp->m_rtdev_targp->bt_dev);
 		handlers[2].fn = xfs_getfsmap_rtdev_rtbitmap;
 	}
+#endif /* CONFIG_XFS_RT */
 
 	xfs_sort(handlers, XFS_GETFSMAP_DEVS, sizeof(struct xfs_getfsmap_dev),
 			xfs_getfsmap_dev_compare);
-- 
2.14.2

[PATCH 13/18] xfs: don't change inode mode if ACL update fails

[Christoph Hellwig]

From: Dave Chinner <dchinner@redhat.com>

commit 67f2ffe31d1a683170c2ba0ecc643e42a5fdd397 upstream.

If we get ENOSPC half way through setting the ACL, the inode mode
can still be changed even though the ACL does not exist. Reorder the
operation to only change the mode of the inode if the ACL is set
correctly.

Whilst this does not fix the problem with crash consistency (that requires
attribute addition to be a deferred op) it does prevent ENOSPC and other
non-fatal errors setting an xattr to be handled sanely.

This fixes xfstests generic/449.

Signed-Off-By: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_acl.c | 22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 7034e17535de..3354140de07e 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -247,6 +247,8 @@ xfs_set_mode(struct inode *inode, umode_t mode)
 int
 xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 {
+	umode_t mode;
+	bool set_mode = false;
 	int error = 0;
 
 	if (!acl)
@@ -257,16 +259,24 @@ xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 		return error;
 
 	if (type == ACL_TYPE_ACCESS) {
-		umode_t mode;
-
 		error = posix_acl_update_mode(inode, &mode, &acl);
 		if (error)
 			return error;
-		error = xfs_set_mode(inode, mode);
-		if (error)
-			return error;
+		set_mode = true;
 	}
 
  set_acl:
-	return __xfs_set_acl(inode, acl, type);
+	error =  __xfs_set_acl(inode, acl, type);
+	if (error)
+		return error;
+
+	/*
+	 * We set the mode after successfully updating the ACL xattr because the
+	 * xattr update can fail at ENOSPC and we don't want to change the mode
+	 * if the ACL update hasn't been applied.
+	 */
+	if (set_mode)
+		error = xfs_set_mode(inode, mode);
+
+	return error;
 }
-- 
2.14.2

[PATCH 14/18] xfs: reinit btree pointer on attr tree inactivation walk

[Christoph Hellwig]

From: Brian Foster <bfoster@redhat.com>

commit f35c5e10c6ed6ba52a8dd8573924a80b6a02f03f upstream.

xfs_attr3_root_inactive() walks the attr fork tree to invalidate the
associated blocks. xfs_attr3_node_inactive() recursively descends
from internal blocks to leaf blocks, caching block address values
along the way to revisit parent blocks, locate the next entry and
descend down that branch of the tree.

The code that attempts to reread the parent block is unsafe because
it assumes that the local xfs_da_node_entry pointer remains valid
after an xfs_trans_brelse() and re-read of the parent buffer. Under
heavy memory pressure, it is possible that the buffer has been
reclaimed and reallocated by the time the parent block is reread.
This means that 'btree' can point to an invalid memory address, lead
to a random/garbage value for child_fsb and cause the subsequent
read of the attr fork to go off the rails and return a NULL buffer
for an attr fork offset that is most likely not allocated.

Note that this problem can be manufactured by setting
XFS_ATTR_BTREE_REF to 0 to prevent LRU caching of attr buffers,
creating a file with a multi-level attr fork and removing it to
trigger inactivation.

To address this problem, reinit the node/btree pointers to the
parent buffer after it has been re-read. This ensures btree points
to a valid record and allows the walk to proceed.

Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_attr_inactive.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/xfs/xfs_attr_inactive.c b/fs/xfs/xfs_attr_inactive.c
index be0b79d8900f..c6643004e583 100644
--- a/fs/xfs/xfs_attr_inactive.c
+++ b/fs/xfs/xfs_attr_inactive.c
@@ -302,6 +302,8 @@ xfs_attr3_node_inactive(
 						 &bp, XFS_ATTR_FORK);
 			if (error)
 				return error;
+			node = bp->b_addr;
+			btree = dp->d_ops->node_tree_p(node);
 			child_fsb = be32_to_cpu(btree[i + 1].before);
 			xfs_trans_brelse(*trans, bp);
 		}
-- 
2.14.2

[PATCH 15/18] xfs: handle error if xfs_btree_get_bufs fails

[Christoph Hellwig]

From: Eric Sandeen <sandeen@sandeen.net>

commit 93e8befc17f6d6ea92b0aee3741ceac8bca4590f upstream.

Jason reported that a corrupted filesystem failed to replay
the log with a metadata block out of bounds warning:

XFS (dm-2): _xfs_buf_find: Block out of range: block 0x80270fff8, EOFS 0x9c40000

_xfs_buf_find() and xfs_btree_get_bufs() return NULL if
that happens, and then when xfs_alloc_fix_freelist() calls
xfs_trans_binval() on that NULL bp, we oops with:

BUG: unable to handle kernel NULL pointer dereference at 00000000000000f8

We don't handle _xfs_buf_find errors very well, every
caller higher up the stack gets to guess at why it failed.
But we should at least handle it somehow, so return
EFSCORRUPTED here.

Reported-by: Jason L Tibbitts III <tibbs@math.uh.edu>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/libxfs/xfs_alloc.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 744dcaec34cc..f965ce832bc0 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -1584,6 +1584,10 @@ xfs_alloc_ag_vextent_small(
 
 				bp = xfs_btree_get_bufs(args->mp, args->tp,
 					args->agno, fbno, 0);
+				if (!bp) {
+					error = -EFSCORRUPTED;
+					goto error0;
+				}
 				xfs_trans_binval(args->tp, bp);
 			}
 			args->len = 1;
@@ -2141,6 +2145,10 @@ xfs_alloc_fix_freelist(
 		if (error)
 			goto out_agbp_relse;
 		bp = xfs_btree_get_bufs(mp, tp, args->agno, bno, 0);
+		if (!bp) {
+			error = -EFSCORRUPTED;
+			goto out_agbp_relse;
+		}
 		xfs_trans_binval(tp, bp);
 	}
 
-- 
2.14.2

[PATCH 16/18] xfs: cancel dirty pages on invalidation

[Christoph Hellwig]

From: Dave Chinner <dchinner@redhat.com>

commit 793d7dbe6d82a50b9d14bf992b9eaacb70a11ce6 upstream.

Recently we've had warnings arise from the vm handing us pages
without bufferheads attached to them. This should not ever occur
in XFS, but we don't defend against it properly if it does. The only
place where we remove bufferheads from a page is in
xfs_vm_releasepage(), but we can't tell the difference here between
"page is dirty so don't release" and "page is dirty but is being
invalidated so release it".

In some places that are invalidating pages ask for pages to be
released and follow up afterward calling ->releasepage by checking
whether the page was dirty and then aborting the invalidation. This
is a possible vector for releasing buffers from a page but then
leaving it in the mapping, so we really do need to avoid dirty pages
in xfs_vm_releasepage().

To differentiate between invalidated pages and normal pages, we need
to clear the page dirty flag when invalidating the pages. This can
be done through xfs_vm_invalidatepage(), and will result
xfs_vm_releasepage() seeing the page as clean which matches the
bufferhead state on the page after calling block_invalidatepage().

Hence we can re-add the page dirty check in xfs_vm_releasepage to
catch the case where we might be releasing a page that is actually
dirty and so should not have the bufferheads on it removed. This
will remove one possible vector of "dirty page with no bufferheads"
and so help narrow down the search for the root cause of that
problem.

Signed-Off-By: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_aops.c | 34 ++++++++++++++++++++++------------
 1 file changed, 22 insertions(+), 12 deletions(-)

diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 78aee670ad61..f425bbcd49b0 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -722,6 +722,14 @@ xfs_vm_invalidatepage(
 {
 	trace_xfs_invalidatepage(page->mapping->host, page, offset,
 				 length);
+
+	/*
+	 * If we are invalidating the entire page, clear the dirty state from it
+	 * so that we can check for attempts to release dirty cached pages in
+	 * xfs_vm_releasepage().
+	 */
+	if (offset == 0 && length >= PAGE_SIZE)
+		cancel_dirty_page(page);
 	block_invalidatepage(page, offset, length);
 }
 
@@ -1177,25 +1185,27 @@ xfs_vm_releasepage(
 	 * mm accommodates an old ext3 case where clean pages might not have had
 	 * the dirty bit cleared. Thus, it can send actual dirty pages to
 	 * ->releasepage() via shrink_active_list(). Conversely,
-	 * block_invalidatepage() can send pages that are still marked dirty
-	 * but otherwise have invalidated buffers.
+	 * block_invalidatepage() can send pages that are still marked dirty but
+	 * otherwise have invalidated buffers.
 	 *
 	 * We want to release the latter to avoid unnecessary buildup of the
-	 * LRU, skip the former and warn if we've left any lingering
-	 * delalloc/unwritten buffers on clean pages. Skip pages with delalloc
-	 * or unwritten buffers and warn if the page is not dirty. Otherwise
-	 * try to release the buffers.
+	 * LRU, so xfs_vm_invalidatepage() clears the page dirty flag on pages
+	 * that are entirely invalidated and need to be released.  Hence the
+	 * only time we should get dirty pages here is through
+	 * shrink_active_list() and so we can simply skip those now.
+	 *
+	 * warn if we've left any lingering delalloc/unwritten buffers on clean
+	 * or invalidated pages we are about to release.
 	 */
+	if (PageDirty(page))
+		return 0;
+
 	xfs_count_page_state(page, &delalloc, &unwritten);
 
-	if (delalloc) {
-		WARN_ON_ONCE(!PageDirty(page));
+	if (WARN_ON_ONCE(delalloc))
 		return 0;
-	}
-	if (unwritten) {
-		WARN_ON_ONCE(!PageDirty(page));
+	if (WARN_ON_ONCE(unwritten))
 		return 0;
-	}
 
 	return try_to_free_buffers(page);
 }
-- 
2.14.2

[PATCH 17/18] xfs: trim writepage mapping to within eof

[Christoph Hellwig]

From: Brian Foster <bfoster@redhat.com>

commit 40214d128e07dd21bb07a8ed6a7fe2f911281ab2 upstream.

The writeback rework in commit fbcc02561359 ("xfs: Introduce
writeback context for writepages") introduced a subtle change in
behavior with regard to the block mapping used across the
->writepages() sequence. The previous xfs_cluster_write() code would
only flush pages up to EOF at the time of the writepage, thus
ensuring that any pages due to file-extending writes would be
handled on a separate cycle and with a new, updated block mapping.

The updated code establishes a block mapping in xfs_writepage_map()
that could extend beyond EOF if the file has post-eof preallocation.
Because we now use the generic writeback infrastructure and pass the
cached mapping to each writepage call, there is no implicit EOF
limit in place. If eofblocks trimming occurs during ->writepages(),
any post-eof portion of the cached mapping becomes invalid. The
eofblocks code has no means to serialize against writeback because
there are no pages associated with post-eof blocks. Therefore if an
eofblocks trim occurs and is followed by a file-extending buffered
write, not only has the mapping become invalid, but we could end up
writing a page to disk based on the invalid mapping.

Consider the following sequence of events:

- A buffered write creates a delalloc extent and post-eof
  speculative preallocation.
- Writeback starts and on the first writepage cycle, the delalloc
  extent is converted to real blocks (including the post-eof blocks)
  and the mapping is cached.
- The file is closed and xfs_release() trims post-eof blocks. The
  cached writeback mapping is now invalid.
- Another buffered write appends the file with a delalloc extent.
- The concurrent writeback cycle picks up the just written page
  because the writeback range end is LLONG_MAX. xfs_writepage_map()
  attributes it to the (now invalid) cached mapping and writes the
  data to an incorrect location on disk (and where the file offset is
  still backed by a delalloc extent).

This problem is reproduced by xfstests test generic/464, which
triggers racing writes, appends, open/closes and writeback requests.

To address this problem, trim the mapping used during writeback to
within EOF when the mapping is validated. This ensures the mapping
is revalidated for any pages encountered beyond EOF as of the time
the current mapping was cached or last validated.

Reported-by: Eryu Guan <eguan@redhat.com>
Diagnosed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/libxfs/xfs_bmap.c | 11 +++++++++++
 fs/xfs/libxfs/xfs_bmap.h |  1 +
 fs/xfs/xfs_aops.c        | 13 +++++++++++++
 3 files changed, 25 insertions(+)

diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 6f2a5baded76..5c6eb19664f2 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -3860,6 +3860,17 @@ xfs_trim_extent(
 	}
 }
 
+/* trim extent to within eof */
+void
+xfs_trim_extent_eof(
+	struct xfs_bmbt_irec	*irec,
+	struct xfs_inode	*ip)
+
+{
+	xfs_trim_extent(irec, 0, XFS_B_TO_FSB(ip->i_mount,
+					      i_size_read(VFS_I(ip))));
+}
+
 /*
  * Trim the returned map to the required bounds
  */
diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index 851982a5dfbc..502e0d8fb4ff 100644
--- a/fs/xfs/libxfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -208,6 +208,7 @@ void	xfs_bmap_trace_exlist(struct xfs_inode *ip, xfs_extnum_t cnt,
 
 void	xfs_trim_extent(struct xfs_bmbt_irec *irec, xfs_fileoff_t bno,
 		xfs_filblks_t len);
+void	xfs_trim_extent_eof(struct xfs_bmbt_irec *, struct xfs_inode *);
 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,
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index f425bbcd49b0..41b767ecfe50 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -433,6 +433,19 @@ xfs_imap_valid(
 {
 	offset >>= inode->i_blkbits;
 
+	/*
+	 * We have to make sure the cached mapping is within EOF to protect
+	 * against eofblocks trimming on file release leaving us with a stale
+	 * mapping. Otherwise, a page for a subsequent file extending buffered
+	 * write could get picked up by this writeback cycle and written to the
+	 * wrong blocks.
+	 *
+	 * Note that what we really want here is a generic mapping invalidation
+	 * mechanism to protect us from arbitrary extent modifying contexts, not
+	 * just eofblocks.
+	 */
+	xfs_trim_extent_eof(imap, XFS_I(inode));
+
 	return offset >= imap->br_startoff &&
 		offset < imap->br_startoff + imap->br_blockcount;
 }
-- 
2.14.2

[PATCH 18/18] xfs: move two more RT specific functions into CONFIG_XFS_RT

[Christoph Hellwig]

From: Arnd Bergmann <arnd@arndb.de>

commit 785545c8982604fe3ba79d16409e83993be77d5e upstream.

The last cleanup introduced two harmless warnings:

fs/xfs/xfs_fsmap.c:480:1: warning: '__xfs_getfsmap_rtdev' defined but not used
fs/xfs/xfs_fsmap.c:372:1: warning: 'xfs_getfsmap_rtdev_rtbitmap_helper' defined but not used

This moves those two functions as well.

Fixes: bb9c2e543325 ("xfs: move more RT specific code under CONFIG_XFS_RT")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_fsmap.c | 48 ++++++++++++++++++++++++------------------------
 1 file changed, 24 insertions(+), 24 deletions(-)

diff --git a/fs/xfs/xfs_fsmap.c b/fs/xfs/xfs_fsmap.c
index 560e0b40ac1b..43cfc07996a4 100644
--- a/fs/xfs/xfs_fsmap.c
+++ b/fs/xfs/xfs_fsmap.c
@@ -367,29 +367,6 @@ xfs_getfsmap_datadev_helper(
 	return xfs_getfsmap_helper(cur->bc_tp, info, rec, rec_daddr);
 }
 
-/* Transform a rtbitmap "record" into a fsmap */
-STATIC int
-xfs_getfsmap_rtdev_rtbitmap_helper(
-	struct xfs_trans		*tp,
-	struct xfs_rtalloc_rec		*rec,
-	void				*priv)
-{
-	struct xfs_mount		*mp = tp->t_mountp;
-	struct xfs_getfsmap_info	*info = priv;
-	struct xfs_rmap_irec		irec;
-	xfs_daddr_t			rec_daddr;
-
-	rec_daddr = XFS_FSB_TO_BB(mp, rec->ar_startblock);
-
-	irec.rm_startblock = rec->ar_startblock;
-	irec.rm_blockcount = rec->ar_blockcount;
-	irec.rm_owner = XFS_RMAP_OWN_NULL;	/* "free" */
-	irec.rm_offset = 0;
-	irec.rm_flags = 0;
-
-	return xfs_getfsmap_helper(tp, info, &irec, rec_daddr);
-}
-
 /* Transform a bnobt irec into a fsmap */
 STATIC int
 xfs_getfsmap_datadev_bnobt_helper(
@@ -475,6 +452,30 @@ xfs_getfsmap_logdev(
 	return xfs_getfsmap_helper(tp, info, &rmap, 0);
 }
 
+#ifdef CONFIG_XFS_RT
+/* Transform a rtbitmap "record" into a fsmap */
+STATIC int
+xfs_getfsmap_rtdev_rtbitmap_helper(
+	struct xfs_trans		*tp,
+	struct xfs_rtalloc_rec		*rec,
+	void				*priv)
+{
+	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_getfsmap_info	*info = priv;
+	struct xfs_rmap_irec		irec;
+	xfs_daddr_t			rec_daddr;
+
+	rec_daddr = XFS_FSB_TO_BB(mp, rec->ar_startblock);
+
+	irec.rm_startblock = rec->ar_startblock;
+	irec.rm_blockcount = rec->ar_blockcount;
+	irec.rm_owner = XFS_RMAP_OWN_NULL;	/* "free" */
+	irec.rm_offset = 0;
+	irec.rm_flags = 0;
+
+	return xfs_getfsmap_helper(tp, info, &irec, rec_daddr);
+}
+
 /* Execute a getfsmap query against the realtime device. */
 STATIC int
 __xfs_getfsmap_rtdev(
@@ -521,7 +522,6 @@ __xfs_getfsmap_rtdev(
 	return query_fn(tp, info);
 }
 
-#ifdef CONFIG_XFS_RT
 /* Actually query the realtime bitmap. */
 STATIC int
 xfs_getfsmap_rtdev_rtbitmap_query(
-- 
2.14.2

Re: 4.13-stable updates for XFS

[Greg KH]

On Thu, Oct 19, 2017 at 04:21:05PM +0200, Christoph Hellwig wrote:
> Hi all,
> 
> below are backports up to 4.14-rc5 for XFS.

All now queued up, thanks.

greg k-h