Skip to contentSkip to conversation

FAILED: patch "[PATCH] xfs: bounds-check buffer log item's dirty bitmap" failed to apply to 6.12-stable tree

PATCH New [email protected] 2 participants 4 messages Aug 20 – Aug 23

Subsystems (2)
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The patch below does not apply to the 6.12-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y
git checkout FETCH_HEAD
git cherry-pick -x 813f8136a2ce1fee266d02a7df73db6e8a541604
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026082044-shawl-cushy-402b@gregkh' --subject-prefix 'PATCH 6.12.y' 'HEAD^..'

Possible dependencies:

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 813f8136a2ce1fee266d02a7df73db6e8a541604 Mon Sep 17 00:00:00 2001
From: Ibrahim Hashimov <[email protected]>
Date: Wed, 15 Jul 2026 09:17:23 +0200
Subject: [PATCH] xfs: bounds-check buffer log item's dirty bitmap

xlog_recover_do_reg_buffer() replays each dirty region described by a buffer log item's bitmap into the buffer read for that item:

	memcpy(xfs_buf_offset(bp, (uint)bit << XFS_BLF_SHIFT),
		item->ri_buf[i].iov_base,
		nbits << XFS_BLF_SHIFT);

The destination offset (bit/nbits, from the logged dirty bitmap) and the buffer size (from the logged blf_len) are both attacker-controlled and otherwise unrelated, yet the only thing bounding the copy is an ASSERT(), which compiles away on production kernels. A crafted image logging a small blf_len together with a bitmap bit past the end of that buffer drives the memcpy() past the buffer's allocation, corrupting adjacent kernel heap during mount-time log recovery. This is reachable by anyone who can get a crafted image mounted -- the malicious-filesystem threat model XFS already guards against elsewhere.

Turn the ASSERT() into a real XFS_IS_CORRUPT() check that aborts recovery of the buffer with -EFSCORRUPTED, consistent with the validate-and-fail idiom already used in xlog_recover_do_inode_buffer() and xfs_dquot_item_recover.c. xlog_recover_do_reg_buffer() therefore becomes STATIC int and its three callers propagate the error.

Found and confirmed with KASAN on a CONFIG_XFS_DEBUG=n build: the crafted image trips a slab-out-of-bounds write before this change and fails recovery cleanly with -EFSCORRUPTED after it.

fs/xfs/xfs_buf_item_recover.c +41 -16
index 02b95b89d1b5..240deb3f7827 100644
@@ -461,7 +461,7 @@ xlog_recover_validate_buf_type(
  * given buffer.  The bitmap in the buf log format structure indicates
  * where to place the logged data.
  */
-STATIC void
+STATIC int
 xlog_recover_do_reg_buffer(
 	struct xfs_mount		*mp,
 	struct xlog_recover_item	*item,
@@ -489,8 +489,24 @@ xlog_recover_do_reg_buffer(
 		ASSERT(nbits > 0);
 		ASSERT(item->ri_buf[i].iov_base != NULL);
 		ASSERT(item->ri_buf[i].iov_len % XFS_BLF_CHUNK == 0);
-		ASSERT(BBTOB(bp->b_length) >=
-		       ((uint)bit << XFS_BLF_SHIFT) + (nbits << XFS_BLF_SHIFT));
+		/*
+		 * The bitmap is only trustworthy to the extent that it
+		 * describes a region that actually fits inside the buffer we
+		 * read in based on the (attacker-controlled) blf_len.  Do not
+		 * rely on an ASSERT() for this -- it compiles away entirely on
+		 * non-DEBUG kernels, which is exactly where this matters, so
+		 * validate it for real and abort recovery of this buffer rather
+		 * than copying past the end of it.
+		 */
+		if (XFS_IS_CORRUPT(mp, BBTOB(bp->b_length) <
+				((uint)bit << XFS_BLF_SHIFT) +
+				(nbits << XFS_BLF_SHIFT))) {
+			xfs_alert(mp,
+	"Bad buffer log item dirty bitmap (bit %d, nbits %d) for %d-byte buffer at daddr 0x%llx.",
+				bit, nbits, BBTOB(bp->b_length),
+				xfs_buf_daddr(bp));
+			return -EFSCORRUPTED;
+		}
 
 		/*
 		 * The dirty regions logged in the buffer, even though
@@ -544,6 +560,7 @@ xlog_recover_do_reg_buffer(
 	ASSERT(i == item->ri_total);
 
 	xlog_recover_validate_buf_type(mp, bp, buf_f, current_lsn);
+	return 0;
 }
 
 /*
@@ -552,10 +569,10 @@ xlog_recover_do_reg_buffer(
  * (ie. USR or GRP), then just toss this buffer away; don't recover it.
  * Else, treat it as a regular buffer and do recovery.
  *
- * Return false if the buffer was tossed and true if we recovered the buffer to
- * indicate to the caller if the buffer needs writing.
+ * Return 0 if the buffer was not recovered (tossed), 1 if it was recovered and
+ * needs writing, or a negative errno if recovery of the buffer failed.
  */
-STATIC bool
+STATIC int
 xlog_recover_do_dquot_buffer(
 	struct xfs_mount		*mp,
 	struct xlog			*log,
@@ -564,6 +581,7 @@ xlog_recover_do_dquot_buffer(
 	struct xfs_buf_log_format	*buf_f)
 {
 	uint			type;
+	int			error;
 
 	trace_xfs_log_recover_buf_dquot_buf(log, buf_f);
 
@@ -571,7 +589,7 @@ xlog_recover_do_dquot_buffer(
 	 * Filesystems are required to send in quota flags at mount time.
 	 */
 	if (!mp->m_qflags)
-		return false;
+		return 0;
 
 	type = 0;
 	if (buf_f->blf_flags & XFS_BLF_UDQUOT_BUF)
@@ -584,10 +602,12 @@ xlog_recover_do_dquot_buffer(
 	 * This type of quotas was turned off, so ignore this buffer
 	 */
 	if (log->l_quotaoffs_flag & type)
-		return false;
+		return 0;
 
-	xlog_recover_do_reg_buffer(mp, item, bp, buf_f, NULLCOMMITLSN);
-	return true;
+	error = xlog_recover_do_reg_buffer(mp, item, bp, buf_f, NULLCOMMITLSN);
+	if (error)
+		return error;
+	return 1;
 }
 
 /*
@@ -724,7 +744,9 @@ xlog_recover_do_primary_sb_buffer(
 	xfs_rgnumber_t			orig_rgcount = mp->m_sb.sb_rgcount;
 	int				error;
 
-	xlog_recover_do_reg_buffer(mp, item, bp, buf_f, current_lsn);
+	error = xlog_recover_do_reg_buffer(mp, item, bp, buf_f, current_lsn);
+	if (error)
+		return error;
 
 	if (orig_agcount == 0) {
 		xfs_alert(mp, "Trying to grow file system without AGs");
@@ -1081,11 +1103,11 @@ xlog_recover_buf_commit_pass2(
 			goto out_release;
 	} else if (buf_f->blf_flags &
 		  (XFS_BLF_UDQUOT_BUF|XFS_BLF_PDQUOT_BUF|XFS_BLF_GDQUOT_BUF)) {
-		bool	dirty;
-
-		dirty = xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
-		if (!dirty)
+		error = xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
+		if (error <= 0)
 			goto out_release;
+		/* write dirty buffer */
+		error = 0;
 	} else if ((xfs_blft_from_flags(buf_f) & XFS_BLFT_SB_BUF) &&
 			xfs_buf_daddr(bp) == 0) {
 		error = xlog_recover_do_primary_sb_buffer(mp, item, bp, buf_f,
@@ -1105,7 +1127,10 @@ xlog_recover_buf_commit_pass2(
 			xfs_buf_relse(rtsb_bp);
 		}
 	} else {
-		xlog_recover_do_reg_buffer(mp, item, bp, buf_f, current_lsn);
+		error = xlog_recover_do_reg_buffer(mp, item, bp, buf_f,
+						   current_lsn);
+		if (error)
+			goto out_release;
 	}
 
 	/*
maintainer
#1 lore raw

PATCH 1/3 xfs: namespace the maximum length/refcount symbols

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit 70fcf6866578e69635399e806273376f5e0b8e2b ]

Actually namespace these variables properly, so that readers can tell that this is an XFS symbol, and that it's for the refcount functionality.

Stable-dep-of: 813f8136a2ce ("xfs: bounds-check buffer log item's dirty bitmap")

---
 fs/xfs/libxfs/xfs_format.h     |  4 ++--
 fs/xfs/libxfs/xfs_refcount.c   | 18 +++++++++---------
 fs/xfs/scrub/refcount.c        |  2 +-
 fs/xfs/scrub/refcount_repair.c |  4 ++--
 4 files changed, 14 insertions(+), 14 deletions(-)
fs/xfs/libxfs/xfs_format.h +2 -2
index e1bfee0c3b1a8..0922008296193 100644
@@ -1625,8 +1625,8 @@ struct xfs_refcount_key {
 	__be32		rc_startblock;	/* starting block number */
 };
 
-#define MAXREFCOUNT	((xfs_nlink_t)~0U)
-#define MAXREFCEXTLEN	((xfs_extlen_t)~0U)
+#define XFS_REFC_REFCOUNT_MAX	((xfs_nlink_t)~0U)
+#define XFS_REFC_LEN_MAX	((xfs_extlen_t)~0U)
 
 /* btree pointer type */
 typedef __be32 xfs_refcount_ptr_t;
fs/xfs/libxfs/xfs_refcount.c +9 -9
index e26a2472c25d1..b30850049fd18 100644
@@ -128,7 +128,7 @@ xfs_refcount_check_irec(
 	struct xfs_perag		*pag,
 	const struct xfs_refcount_irec	*irec)
 {
-	if (irec->rc_blockcount == 0 || irec->rc_blockcount > MAXREFCEXTLEN)
+	if (irec->rc_blockcount == 0 || irec->rc_blockcount > XFS_REFC_LEN_MAX)
 		return __this_address;
 
 	if (!xfs_refcount_check_domain(irec))
@@ -138,7 +138,7 @@ xfs_refcount_check_irec(
 	if (!xfs_verify_agbext(pag, irec->rc_startblock, irec->rc_blockcount))
 		return __this_address;
 
-	if (irec->rc_refcount == 0 || irec->rc_refcount > MAXREFCOUNT)
+	if (irec->rc_refcount == 0 || irec->rc_refcount > XFS_REFC_REFCOUNT_MAX)
 		return __this_address;
 
 	return NULL;
@@ -853,9 +853,9 @@ xfs_refc_merge_refcount(
 	const struct xfs_refcount_irec	*irec,
 	enum xfs_refc_adjust_op		adjust)
 {
-	/* Once a record hits MAXREFCOUNT, it is pinned there forever */
-	if (irec->rc_refcount == MAXREFCOUNT)
-		return MAXREFCOUNT;
+	/* Once a record hits XFS_REFC_REFCOUNT_MAX, it is pinned forever */
+	if (irec->rc_refcount == XFS_REFC_REFCOUNT_MAX)
+		return XFS_REFC_REFCOUNT_MAX;
 	return irec->rc_refcount + adjust;
 }
 
@@ -898,7 +898,7 @@ xfs_refc_want_merge_center(
 	 * hence we need to catch u32 addition overflows here.
 	 */
 	ulen += cleft->rc_blockcount + right->rc_blockcount;
-	if (ulen >= MAXREFCEXTLEN)
+	if (ulen >= XFS_REFC_LEN_MAX)
 		return false;
 
 	*ulenp = ulen;
@@ -933,7 +933,7 @@ xfs_refc_want_merge_left(
 	 * hence we need to catch u32 addition overflows here.
 	 */
 	ulen += cleft->rc_blockcount;
-	if (ulen >= MAXREFCEXTLEN)
+	if (ulen >= XFS_REFC_LEN_MAX)
 		return false;
 
 	return true;
@@ -967,7 +967,7 @@ xfs_refc_want_merge_right(
 	 * hence we need to catch u32 addition overflows here.
 	 */
 	ulen += cright->rc_blockcount;
-	if (ulen >= MAXREFCEXTLEN)
+	if (ulen >= XFS_REFC_LEN_MAX)
 		return false;
 
 	return true;
@@ -1196,7 +1196,7 @@ xfs_refcount_adjust_extents(
 		 * Adjust the reference count and either update the tree
 		 * (incr) or free the blocks (decr).
 		 */
-		if (ext.rc_refcount == MAXREFCOUNT)
+		if (ext.rc_refcount == XFS_REFC_REFCOUNT_MAX)
 			goto skip;
 		ext.rc_refcount += adj;
 		trace_xfs_refcount_modify_extent(cur, &ext);
fs/xfs/scrub/refcount.c +1 -1
index cccf39d917a09..dc3a75ac26cec 100644
@@ -421,7 +421,7 @@ xchk_refcount_mergeable(
 	if (r1->rc_refcount != r2->rc_refcount)
 		return false;
 	if ((unsigned long long)r1->rc_blockcount + r2->rc_blockcount >
-			MAXREFCEXTLEN)
+			XFS_REFC_LEN_MAX)
 		return false;
 
 	return true;
fs/xfs/scrub/refcount_repair.c +2 -2
index 1512e7363de12..8cf3e84e83a39 100644
@@ -177,7 +177,7 @@ xrep_refc_stash(
 	if (xchk_should_terminate(sc, &error))
 		return error;
 
-	irec.rc_refcount = min_t(uint64_t, MAXREFCOUNT, refcount);
+	irec.rc_refcount = min_t(uint64_t, XFS_REFC_REFCOUNT_MAX, refcount);
 
 	error = xrep_refc_check_ext(rr->sc, &irec);
 	if (error)
@@ -416,7 +416,7 @@ xrep_refc_find_refcounts(
 	/*
 	 * Set up a bag to store all the rmap records that we're tracking to
 	 * generate a reference count record.  If the size of the bag exceeds
-	 * MAXREFCOUNT, we clamp rc_refcount.
+	 * XFS_REFC_REFCOUNT_MAX, we clamp rc_refcount.
 	 */
 	error = rcbag_init(sc->mp, sc->xmbtp, &rcstack);
 	if (error)
Signature · 1 line
2.53.0
maintainer
#2 lore raw

PATCH 2/3 xfs: don't use a xfs_log_iovec for ri_buf in log recovery

From: Christoph Hellwig <[email protected]>

[ Upstream commit ded74fddcaf685a9440c5612f7831d0c4c1473ca ]

ri_buf just holds a pointer/len pair and is not a log iovec used for writing to the log. Switch to use a kvec instead.

Stable-dep-of: 813f8136a2ce ("xfs: bounds-check buffer log item's dirty bitmap")

---
 fs/xfs/libxfs/xfs_log_recover.h |  4 +--
 fs/xfs/xfs_attr_item.c          | 32 ++++++++++++------------
 fs/xfs/xfs_bmap_item.c          | 18 +++++++-------
 fs/xfs/xfs_buf_item.c           |  8 +++---
 fs/xfs/xfs_buf_item.h           |  2 +-
 fs/xfs/xfs_buf_item_recover.c   | 38 ++++++++++++++---------------
 fs/xfs/xfs_dquot_item_recover.c | 20 +++++++--------
 fs/xfs/xfs_exchmaps_item.c      |  8 +++---
 fs/xfs/xfs_extfree_item.c       | 43 ++++++++++++++++++---------------
 fs/xfs/xfs_icreate_item.c       |  2 +-
 fs/xfs/xfs_inode_item.c         |  6 ++---
 fs/xfs/xfs_inode_item.h         |  4 +--
 fs/xfs/xfs_inode_item_recover.c | 26 ++++++++++----------
 fs/xfs/xfs_log_recover.c        | 16 ++++++------
 fs/xfs/xfs_refcount_item.c      | 16 ++++++------
 fs/xfs/xfs_rmap_item.c          | 16 ++++++------
 fs/xfs/xfs_trans.h              |  1 -
 17 files changed, 131 insertions(+), 129 deletions(-)
fs/xfs/libxfs/xfs_log_recover.h +2 -2
index 521d327e4c89e..97892ea598282 100644
@@ -98,7 +98,7 @@ struct xlog_recover_item {
 	struct list_head	ri_list;
 	int			ri_cnt;	/* count of regions found */
 	int			ri_total;	/* total regions */
-	struct xfs_log_iovec	*ri_buf;	/* ptr to regions buffer */
+	struct kvec		*ri_buf;	/* ptr to regions buffer */
 	const struct xlog_recover_item_ops *ri_ops;
 };
 
@@ -111,7 +111,7 @@ struct xlog_recover {
 	struct list_head	r_itemq;	/* q for items */
 };
 
-#define ITEM_TYPE(i)	(*(unsigned short *)(i)->ri_buf[0].i_addr)
+#define ITEM_TYPE(i)	(*(unsigned short *)(i)->ri_buf[0].iov_base)
 
 #define	XLOG_RECOVER_CRCPASS	0
 #define	XLOG_RECOVER_PASS1	1
fs/xfs/xfs_attr_item.c +16 -16
index b81eafc9543b1..1c886bd50df18 100644
@@ -952,50 +952,50 @@ static inline void *
 xfs_attri_validate_name_iovec(
 	struct xfs_mount		*mp,
 	struct xfs_attri_log_format     *attri_formatp,
-	const struct xfs_log_iovec	*iovec,
+	const struct kvec		*iovec,
 	unsigned int			name_len)
 {
-	if (iovec->i_len != xlog_calc_iovec_len(name_len)) {
+	if (iovec->iov_len != xlog_calc_iovec_len(name_len)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
 				attri_formatp, sizeof(*attri_formatp));
 		return NULL;
 	}
 
-	if (!xfs_attr_namecheck(attri_formatp->alfi_attr_filter, iovec->i_addr,
+	if (!xfs_attr_namecheck(attri_formatp->alfi_attr_filter, iovec->iov_base,
 				name_len)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
 				attri_formatp, sizeof(*attri_formatp));
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				iovec->i_addr, iovec->i_len);
+				iovec->iov_base, iovec->iov_len);
 		return NULL;
 	}
 
-	return iovec->i_addr;
+	return iovec->iov_base;
 }
 
 static inline void *
 xfs_attri_validate_value_iovec(
 	struct xfs_mount		*mp,
 	struct xfs_attri_log_format     *attri_formatp,
-	const struct xfs_log_iovec	*iovec,
+	const struct kvec		*iovec,
 	unsigned int			value_len)
 {
-	if (iovec->i_len != xlog_calc_iovec_len(value_len)) {
+	if (iovec->iov_len != xlog_calc_iovec_len(value_len)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
 				attri_formatp, sizeof(*attri_formatp));
 		return NULL;
 	}
 
 	if ((attri_formatp->alfi_attr_filter & XFS_ATTR_PARENT) &&
-	    !xfs_parent_valuecheck(mp, iovec->i_addr, value_len)) {
+	    !xfs_parent_valuecheck(mp, iovec->iov_base, value_len)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
 				attri_formatp, sizeof(*attri_formatp));
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				iovec->i_addr, iovec->i_len);
+				iovec->iov_base, iovec->iov_len);
 		return NULL;
 	}
 
-	return iovec->i_addr;
+	return iovec->iov_base;
 }
 
 STATIC int
@@ -1022,13 +1022,13 @@ xlog_recover_attri_commit_pass2(
 
 	/* Validate xfs_attri_log_format before the large memory allocation */
 	len = sizeof(struct xfs_attri_log_format);
-	if (item->ri_buf[i].i_len != len) {
+	if (item->ri_buf[i].iov_len != len) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
-	attri_formatp = item->ri_buf[i].i_addr;
+	attri_formatp = item->ri_buf[i].iov_base;
 	if (!xfs_attri_validate(mp, attri_formatp)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
 				attri_formatp, len);
@@ -1217,10 +1217,10 @@ xlog_recover_attrd_commit_pass2(
 {
 	struct xfs_attrd_log_format	*attrd_formatp;
 
-	attrd_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_attrd_log_format)) {
+	attrd_formatp = item->ri_buf[0].iov_base;
+	if (item->ri_buf[0].iov_len != sizeof(struct xfs_attrd_log_format)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, log->l_mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
fs/xfs/xfs_bmap_item.c +9 -9
index 5f10a61c9b25a..4fb5eb431c1a9 100644
@@ -653,24 +653,24 @@ xlog_recover_bui_commit_pass2(
 	struct xfs_bui_log_format	*bui_formatp;
 	size_t				len;
 
-	bui_formatp = item->ri_buf[0].i_addr;
+	bui_formatp = item->ri_buf[0].iov_base;
 
-	if (item->ri_buf[0].i_len < xfs_bui_log_format_sizeof(0)) {
+	if (item->ri_buf[0].iov_len < xfs_bui_log_format_sizeof(0)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
 	if (bui_formatp->bui_nextents != XFS_BUI_MAX_FAST_EXTENTS) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
 	len = xfs_bui_log_format_sizeof(bui_formatp->bui_nextents);
-	if (item->ri_buf[0].i_len != len) {
+	if (item->ri_buf[0].iov_len != len) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
@@ -704,10 +704,10 @@ xlog_recover_bud_commit_pass2(
 {
 	struct xfs_bud_log_format	*bud_formatp;
 
-	bud_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_bud_log_format)) {
+	bud_formatp = item->ri_buf[0].iov_base;
+	if (item->ri_buf[0].iov_len != sizeof(struct xfs_bud_log_format)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, log->l_mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
fs/xfs/xfs_buf_item.c +4 -4
index f4b9cd959b687..b746a1a0f8dd9 100644
@@ -35,16 +35,16 @@ static inline struct xfs_buf_log_item *BUF_ITEM(struct xfs_log_item *lip)
 /* Is this log iovec plausibly large enough to contain the buffer log format? */
 bool
 xfs_buf_log_check_iovec(
-	struct xfs_log_iovec		*iovec)
+	struct kvec			*iovec)
 {
-	struct xfs_buf_log_format	*blfp = iovec->i_addr;
+	struct xfs_buf_log_format	*blfp = iovec->iov_base;
 	char				*bmp_end;
 	char				*item_end;
 
-	if (offsetof(struct xfs_buf_log_format, blf_data_map) > iovec->i_len)
+	if (offsetof(struct xfs_buf_log_format, blf_data_map) > iovec->iov_len)
 		return false;
 
-	item_end = (char *)iovec->i_addr + iovec->i_len;
+	item_end = (char *)iovec->iov_base + iovec->iov_len;
 	bmp_end = (char *)&blfp->blf_data_map[blfp->blf_map_size];
 	return bmp_end <= item_end;
 }
fs/xfs/xfs_buf_item.h +1 -1
index 4d8a6aece995d..7ddda0aa9dac2 100644
@@ -67,7 +67,7 @@ static inline void xfs_buf_dquot_io_fail(struct xfs_buf *bp)
 }
 #endif /* CONFIG_XFS_QUOTA */
 void	xfs_buf_iodone(struct xfs_buf *);
-bool	xfs_buf_log_check_iovec(struct xfs_log_iovec *iovec);
+bool	xfs_buf_log_check_iovec(struct kvec *iovec);
 
 extern struct kmem_cache	*xfs_buf_item_cache;
 
fs/xfs/xfs_buf_item_recover.c +19 -19
index 0185c92df8c2e..5dca2dbf866b2 100644
@@ -157,7 +157,7 @@ STATIC enum xlog_recover_reorder
 xlog_recover_buf_reorder(
 	struct xlog_recover_item	*item)
 {
-	struct xfs_buf_log_format	*buf_f = item->ri_buf[0].i_addr;
+	struct xfs_buf_log_format	*buf_f = item->ri_buf[0].iov_base;
 
 	if (buf_f->blf_flags & XFS_BLF_CANCEL)
 		return XLOG_REORDER_CANCEL_LIST;
@@ -171,7 +171,7 @@ xlog_recover_buf_ra_pass2(
 	struct xlog                     *log,
 	struct xlog_recover_item        *item)
 {
-	struct xfs_buf_log_format	*buf_f = item->ri_buf[0].i_addr;
+	struct xfs_buf_log_format	*buf_f = item->ri_buf[0].iov_base;
 
 	xlog_buf_readahead(log, buf_f->blf_blkno, buf_f->blf_len, NULL);
 }
@@ -185,11 +185,11 @@ xlog_recover_buf_commit_pass1(
 	struct xlog			*log,
 	struct xlog_recover_item	*item)
 {
-	struct xfs_buf_log_format	*bf = item->ri_buf[0].i_addr;
+	struct xfs_buf_log_format	*bf = item->ri_buf[0].iov_base;
 
 	if (!xfs_buf_log_check_iovec(&item->ri_buf[0])) {
-		xfs_err(log->l_mp, "bad buffer log item size (%d)",
-				item->ri_buf[0].i_len);
+		xfs_err(log->l_mp, "bad buffer log item size (%zd)",
+				item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
@@ -470,8 +470,8 @@ xlog_recover_do_reg_buffer(
 		nbits = xfs_contig_bits(buf_f->blf_data_map,
 					buf_f->blf_map_size, bit);
 		ASSERT(nbits > 0);
-		ASSERT(item->ri_buf[i].i_addr != NULL);
-		ASSERT(item->ri_buf[i].i_len % XFS_BLF_CHUNK == 0);
+		ASSERT(item->ri_buf[i].iov_base != NULL);
+		ASSERT(item->ri_buf[i].iov_len % XFS_BLF_CHUNK == 0);
 		ASSERT(BBTOB(bp->b_length) >=
 		       ((uint)bit << XFS_BLF_SHIFT) + (nbits << XFS_BLF_SHIFT));
 
@@ -483,8 +483,8 @@ xlog_recover_do_reg_buffer(
 		 * the log. Hence we need to trim nbits back to the length of
 		 * the current region being copied out of the log.
 		 */
-		if (item->ri_buf[i].i_len < (nbits << XFS_BLF_SHIFT))
-			nbits = item->ri_buf[i].i_len >> XFS_BLF_SHIFT;
+		if (item->ri_buf[i].iov_len < (nbits << XFS_BLF_SHIFT))
+			nbits = item->ri_buf[i].iov_len >> XFS_BLF_SHIFT;
 
 		/*
 		 * Do a sanity check if this is a dquot buffer. Just checking
@@ -494,18 +494,18 @@ xlog_recover_do_reg_buffer(
 		fa = NULL;
 		if (buf_f->blf_flags &
 		   (XFS_BLF_UDQUOT_BUF|XFS_BLF_PDQUOT_BUF|XFS_BLF_GDQUOT_BUF)) {
-			if (item->ri_buf[i].i_addr == NULL) {
+			if (item->ri_buf[i].iov_base == NULL) {
 				xfs_alert(mp,
 					"XFS: NULL dquot in %s.", __func__);
 				goto next;
 			}
-			if (item->ri_buf[i].i_len < size_disk_dquot) {
+			if (item->ri_buf[i].iov_len < size_disk_dquot) {
 				xfs_alert(mp,
-					"XFS: dquot too small (%d) in %s.",
-					item->ri_buf[i].i_len, __func__);
+					"XFS: dquot too small (%zd) in %s.",
+					item->ri_buf[i].iov_len, __func__);
 				goto next;
 			}
-			fa = xfs_dquot_verify(mp, item->ri_buf[i].i_addr, -1);
+			fa = xfs_dquot_verify(mp, item->ri_buf[i].iov_base, -1);
 			if (fa) {
 				xfs_alert(mp,
 	"dquot corrupt at %pS trying to replay into block 0x%llx",
@@ -516,7 +516,7 @@ xlog_recover_do_reg_buffer(
 
 		memcpy(xfs_buf_offset(bp,
 			(uint)bit << XFS_BLF_SHIFT),	/* dest */
-			item->ri_buf[i].i_addr,		/* source */
+			item->ri_buf[i].iov_base,		/* source */
 			nbits<<XFS_BLF_SHIFT);		/* length */
  next:
 		i++;
@@ -652,8 +652,8 @@ xlog_recover_do_inode_buffer(
 		if (next_unlinked_offset < reg_buf_offset)
 			continue;
 
-		ASSERT(item->ri_buf[item_index].i_addr != NULL);
-		ASSERT((item->ri_buf[item_index].i_len % XFS_BLF_CHUNK) == 0);
+		ASSERT(item->ri_buf[item_index].iov_base != NULL);
+		ASSERT((item->ri_buf[item_index].iov_len % XFS_BLF_CHUNK) == 0);
 		ASSERT((reg_buf_offset + reg_buf_bytes) <= BBTOB(bp->b_length));
 
 		/*
@@ -661,7 +661,7 @@ xlog_recover_do_inode_buffer(
 		 * current di_next_unlinked field.  Extract its value
 		 * and copy it to the buffer copy.
 		 */
-		logged_nextp = item->ri_buf[item_index].i_addr +
+		logged_nextp = item->ri_buf[item_index].iov_base +
 				next_unlinked_offset - reg_buf_offset;
 		if (XFS_IS_CORRUPT(mp, *logged_nextp == 0)) {
 			xfs_alert(mp,
@@ -951,7 +951,7 @@ xlog_recover_buf_commit_pass2(
 	struct xlog_recover_item	*item,
 	xfs_lsn_t			current_lsn)
 {
-	struct xfs_buf_log_format	*buf_f = item->ri_buf[0].i_addr;
+	struct xfs_buf_log_format	*buf_f = item->ri_buf[0].iov_base;
 	struct xfs_mount		*mp = log->l_mp;
 	struct xfs_buf			*bp;
 	int				error;
fs/xfs/xfs_dquot_item_recover.c +10 -10
index adda4e0dc41c4..590412c8e6150 100644
@@ -34,10 +34,10 @@ xlog_recover_dquot_ra_pass2(
 	if (mp->m_qflags == 0)
 		return;
 
-	recddq = item->ri_buf[1].i_addr;
+	recddq = item->ri_buf[1].iov_base;
 	if (recddq == NULL)
 		return;
-	if (item->ri_buf[1].i_len < sizeof(struct xfs_disk_dquot))
+	if (item->ri_buf[1].iov_len < sizeof(struct xfs_disk_dquot))
 		return;
 
 	type = recddq->d_type & XFS_DQTYPE_REC_MASK;
@@ -45,7 +45,7 @@ xlog_recover_dquot_ra_pass2(
 	if (log->l_quotaoffs_flag & type)
 		return;
 
-	dq_f = item->ri_buf[0].i_addr;
+	dq_f = item->ri_buf[0].iov_base;
 	ASSERT(dq_f);
 	ASSERT(dq_f->qlf_len == 1);
 
@@ -79,14 +79,14 @@ xlog_recover_dquot_commit_pass2(
 	if (mp->m_qflags == 0)
 		return 0;
 
-	recddq = item->ri_buf[1].i_addr;
+	recddq = item->ri_buf[1].iov_base;
 	if (recddq == NULL) {
 		xfs_alert(log->l_mp, "NULL dquot in %s.", __func__);
 		return -EFSCORRUPTED;
 	}
-	if (item->ri_buf[1].i_len < sizeof(struct xfs_disk_dquot)) {
-		xfs_alert(log->l_mp, "dquot too small (%d) in %s.",
-			item->ri_buf[1].i_len, __func__);
+	if (item->ri_buf[1].iov_len < sizeof(struct xfs_disk_dquot)) {
+		xfs_alert(log->l_mp, "dquot too small (%zd) in %s.",
+			item->ri_buf[1].iov_len, __func__);
 		return -EFSCORRUPTED;
 	}
 
@@ -108,7 +108,7 @@ xlog_recover_dquot_commit_pass2(
 	 * The other possibility, of course, is that the quota subsystem was
 	 * removed since the last mount - ENOSYS.
 	 */
-	dq_f = item->ri_buf[0].i_addr;
+	dq_f = item->ri_buf[0].iov_base;
 	ASSERT(dq_f);
 	fa = xfs_dquot_verify(mp, recddq, dq_f->qlf_id);
 	if (fa) {
@@ -147,7 +147,7 @@ xlog_recover_dquot_commit_pass2(
 		}
 	}
 
-	memcpy(ddq, recddq, item->ri_buf[1].i_len);
+	memcpy(ddq, recddq, item->ri_buf[1].iov_len);
 	if (xfs_has_crc(mp)) {
 		xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk),
 				 XFS_DQUOT_CRC_OFF);
@@ -192,7 +192,7 @@ xlog_recover_quotaoff_commit_pass1(
 	struct xlog			*log,
 	struct xlog_recover_item	*item)
 {
-	struct xfs_qoff_logformat	*qoff_f = item->ri_buf[0].i_addr;
+	struct xfs_qoff_logformat	*qoff_f = item->ri_buf[0].iov_base;
 	ASSERT(qoff_f);
 
 	/*
fs/xfs/xfs_exchmaps_item.c +4 -4
index 264a121c5e16d..229cbe0adf17f 100644
@@ -558,12 +558,12 @@ xlog_recover_xmi_commit_pass2(
 	size_t				len;
 
 	len = sizeof(struct xfs_xmi_log_format);
-	if (item->ri_buf[0].i_len != len) {
+	if (item->ri_buf[0].iov_len != len) {
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
 	}
 
-	xmi_formatp = item->ri_buf[0].i_addr;
+	xmi_formatp = item->ri_buf[0].iov_base;
 	if (xmi_formatp->__pad != 0) {
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
@@ -598,8 +598,8 @@ xlog_recover_xmd_commit_pass2(
 {
 	struct xfs_xmd_log_format	*xmd_formatp;
 
-	xmd_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_xmd_log_format)) {
+	xmd_formatp = item->ri_buf[0].iov_base;
+	if (item->ri_buf[0].iov_len != sizeof(struct xfs_xmd_log_format)) {
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp);
 		return -EFSCORRUPTED;
 	}
fs/xfs/xfs_extfree_item.c +23 -20

91 lines not on this page · raw message

fs/xfs/xfs_icreate_item.c +1 -1
index 4345db501714e..f83ec2bd0583c 100644
@@ -158,7 +158,7 @@ xlog_recover_icreate_commit_pass2(
 	int				nbufs;
 	int				i;
 
-	icl = (struct xfs_icreate_log *)item->ri_buf[0].i_addr;
+	icl = (struct xfs_icreate_log *)item->ri_buf[0].iov_base;
 	if (icl->icl_type != XFS_LI_ICREATE) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad type");
 		return -EINVAL;
fs/xfs/xfs_inode_item.c +3 -3
index 5eccc54fbde2e..e273a2d527c43 100644
@@ -1180,12 +1180,12 @@ xfs_iflush_shutdown_abort(
  */
 int
 xfs_inode_item_format_convert(
-	struct xfs_log_iovec		*buf,
+	struct kvec			*buf,
 	struct xfs_inode_log_format	*in_f)
 {
-	struct xfs_inode_log_format_32	*in_f32 = buf->i_addr;
+	struct xfs_inode_log_format_32	*in_f32 = buf->iov_base;
 
-	if (buf->i_len != sizeof(*in_f32)) {
+	if (buf->iov_len != sizeof(*in_f32)) {
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
 		return -EFSCORRUPTED;
 	}
fs/xfs/xfs_inode_item.h +2 -2
index 377e060078044..ba92ce11a0111 100644
@@ -46,8 +46,8 @@ extern void xfs_inode_item_init(struct xfs_inode *, struct xfs_mount *);
 extern void xfs_inode_item_destroy(struct xfs_inode *);
 extern void xfs_iflush_abort(struct xfs_inode *);
 extern void xfs_iflush_shutdown_abort(struct xfs_inode *);
-extern int xfs_inode_item_format_convert(xfs_log_iovec_t *,
-					 struct xfs_inode_log_format *);
+int xfs_inode_item_format_convert(struct kvec *buf,
+		struct xfs_inode_log_format *in_f);
 
 extern struct kmem_cache	*xfs_ili_cache;
 
fs/xfs/xfs_inode_item_recover.c +13 -13
index dbdab4ce7c44c..23b145f3ac319 100644
@@ -28,13 +28,13 @@ xlog_recover_inode_ra_pass2(
 	struct xlog                     *log,
 	struct xlog_recover_item        *item)
 {
-	if (item->ri_buf[0].i_len == sizeof(struct xfs_inode_log_format)) {
-		struct xfs_inode_log_format	*ilfp = item->ri_buf[0].i_addr;
+	if (item->ri_buf[0].iov_len == sizeof(struct xfs_inode_log_format)) {
+		struct xfs_inode_log_format	*ilfp = item->ri_buf[0].iov_base;
 
 		xlog_buf_readahead(log, ilfp->ilf_blkno, ilfp->ilf_len,
 				   &xfs_inode_buf_ra_ops);
 	} else {
-		struct xfs_inode_log_format_32	*ilfp = item->ri_buf[0].i_addr;
+		struct xfs_inode_log_format_32	*ilfp = item->ri_buf[0].iov_base;
 
 		xlog_buf_readahead(log, ilfp->ilf_blkno, ilfp->ilf_len,
 				   &xfs_inode_buf_ra_ops);
@@ -288,8 +288,8 @@ xlog_recover_inode_commit_pass2(
 	int				need_free = 0;
 	xfs_failaddr_t			fa;
 
-	if (item->ri_buf[0].i_len == sizeof(struct xfs_inode_log_format)) {
-		in_f = item->ri_buf[0].i_addr;
+	if (item->ri_buf[0].iov_len == sizeof(struct xfs_inode_log_format)) {
+		in_f = item->ri_buf[0].iov_base;
 	} else {
 		in_f = kmalloc(sizeof(struct xfs_inode_log_format),
 				GFP_KERNEL | __GFP_NOFAIL);
@@ -328,7 +328,7 @@ xlog_recover_inode_commit_pass2(
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
-	ldip = item->ri_buf[1].i_addr;
+	ldip = item->ri_buf[1].iov_base;
 	if (XFS_IS_CORRUPT(mp, ldip->di_magic != XFS_DINODE_MAGIC)) {
 		xfs_alert(mp,
 			"%s: Bad inode log record, rec ptr "PTR_FMT", ino %lld",
@@ -433,12 +433,12 @@ xlog_recover_inode_commit_pass2(
 		goto out_release;
 	}
 	isize = xfs_log_dinode_size(mp);
-	if (unlikely(item->ri_buf[1].i_len > isize)) {
+	if (unlikely(item->ri_buf[1].iov_len > isize)) {
 		XFS_CORRUPTION_ERROR("Bad log dinode size", XFS_ERRLEVEL_LOW,
 				     mp, ldip, sizeof(*ldip));
 		xfs_alert(mp,
-			"Bad inode 0x%llx log dinode size 0x%x",
-			in_f->ilf_ino, item->ri_buf[1].i_len);
+			"Bad inode 0x%llx log dinode size 0x%zx",
+			in_f->ilf_ino, item->ri_buf[1].iov_len);
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
@@ -461,8 +461,8 @@ xlog_recover_inode_commit_pass2(
 
 	if (in_f->ilf_size == 2)
 		goto out_owner_change;
-	len = item->ri_buf[2].i_len;
-	src = item->ri_buf[2].i_addr;
+	len = item->ri_buf[2].iov_len;
+	src = item->ri_buf[2].iov_base;
 	ASSERT(in_f->ilf_size <= 4);
 	ASSERT((in_f->ilf_size == 3) || (fields & XFS_ILOG_AFORK));
 	ASSERT(!(fields & XFS_ILOG_DFORK) ||
@@ -499,8 +499,8 @@ xlog_recover_inode_commit_pass2(
 		} else {
 			attr_index = 2;
 		}
-		len = item->ri_buf[attr_index].i_len;
-		src = item->ri_buf[attr_index].i_addr;
+		len = item->ri_buf[attr_index].iov_len;
+		src = item->ri_buf[attr_index].iov_base;
 		ASSERT(len == xlog_calc_iovec_len(in_f->ilf_asize));
 
 		switch (in_f->ilf_fields & XFS_ILOG_AFORK) {
fs/xfs/xfs_log_recover.c +8 -8
index 1f9e53e480c58..4884bbf5df80e 100644
@@ -2135,15 +2135,15 @@ xlog_recover_add_to_cont_trans(
 	item = list_entry(trans->r_itemq.prev, struct xlog_recover_item,
 			  ri_list);
 
-	old_ptr = item->ri_buf[item->ri_cnt-1].i_addr;
-	old_len = item->ri_buf[item->ri_cnt-1].i_len;
+	old_ptr = item->ri_buf[item->ri_cnt-1].iov_base;
+	old_len = item->ri_buf[item->ri_cnt-1].iov_len;
 
 	ptr = kvrealloc(old_ptr, len + old_len, GFP_KERNEL);
 	if (!ptr)
 		return -ENOMEM;
 	memcpy(&ptr[old_len], dp, len);
-	item->ri_buf[item->ri_cnt-1].i_len += len;
-	item->ri_buf[item->ri_cnt-1].i_addr = ptr;
+	item->ri_buf[item->ri_cnt-1].iov_len += len;
+	item->ri_buf[item->ri_cnt-1].iov_base = ptr;
 	trace_xfs_log_recover_item_add_cont(log, trans, item, 0);
 	return 0;
 }
@@ -2227,7 +2227,7 @@ xlog_recover_add_to_trans(
 		}
 
 		item->ri_total = in_f->ilf_size;
-		item->ri_buf = kzalloc(item->ri_total * sizeof(xfs_log_iovec_t),
+		item->ri_buf = kcalloc(item->ri_total, sizeof(*item->ri_buf),
 				GFP_KERNEL | __GFP_NOFAIL);
 	}
 
@@ -2241,8 +2241,8 @@ xlog_recover_add_to_trans(
 	}
 
 	/* Description region is ri_buf[0] */
-	item->ri_buf[item->ri_cnt].i_addr = ptr;
-	item->ri_buf[item->ri_cnt].i_len  = len;
+	item->ri_buf[item->ri_cnt].iov_base = ptr;
+	item->ri_buf[item->ri_cnt].iov_len  = len;
 	item->ri_cnt++;
 	trace_xfs_log_recover_item_add(log, trans, item, 0);
 	return 0;
@@ -2266,7 +2266,7 @@ xlog_recover_free_trans(
 		/* Free the regions in the item. */
 		list_del(&item->ri_list);
 		for (i = 0; i < item->ri_cnt; i++)
-			kvfree(item->ri_buf[i].i_addr);
+			kvfree(item->ri_buf[i].iov_base);
 		/* Free the item itself */
 		kfree(item->ri_buf);
 		kfree(item);
fs/xfs/xfs_refcount_item.c +8 -8
index 27398512b179b..181499ffdacd3 100644
@@ -587,18 +587,18 @@ xlog_recover_cui_commit_pass2(
 	struct xfs_cui_log_format	*cui_formatp;
 	size_t				len;
 
-	cui_formatp = item->ri_buf[0].i_addr;
+	cui_formatp = item->ri_buf[0].iov_base;
 
-	if (item->ri_buf[0].i_len < xfs_cui_log_format_sizeof(0)) {
+	if (item->ri_buf[0].iov_len < xfs_cui_log_format_sizeof(0)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
 	len = xfs_cui_log_format_sizeof(cui_formatp->cui_nextents);
-	if (item->ri_buf[0].i_len != len) {
+	if (item->ri_buf[0].iov_len != len) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
@@ -632,10 +632,10 @@ xlog_recover_cud_commit_pass2(
 {
 	struct xfs_cud_log_format	*cud_formatp;
 
-	cud_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_cud_log_format)) {
+	cud_formatp = item->ri_buf[0].iov_base;
+	if (item->ri_buf[0].iov_len != sizeof(struct xfs_cud_log_format)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, log->l_mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
fs/xfs/xfs_rmap_item.c +8 -8
index 88b5580e1e19f..2134340e23e69 100644
@@ -638,18 +638,18 @@ xlog_recover_rui_commit_pass2(
 	struct xfs_rui_log_format	*rui_formatp;
 	size_t				len;
 
-	rui_formatp = item->ri_buf[0].i_addr;
+	rui_formatp = item->ri_buf[0].iov_base;
 
-	if (item->ri_buf[0].i_len < xfs_rui_log_format_sizeof(0)) {
+	if (item->ri_buf[0].iov_len < xfs_rui_log_format_sizeof(0)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
 	len = xfs_rui_log_format_sizeof(rui_formatp->rui_nextents);
-	if (item->ri_buf[0].i_len != len) {
+	if (item->ri_buf[0].iov_len != len) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
-				item->ri_buf[0].i_addr, item->ri_buf[0].i_len);
+				item->ri_buf[0].iov_base, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
@@ -683,10 +683,10 @@ xlog_recover_rud_commit_pass2(
 {
 	struct xfs_rud_log_format	*rud_formatp;
 
-	rud_formatp = item->ri_buf[0].i_addr;
-	if (item->ri_buf[0].i_len != sizeof(struct xfs_rud_log_format)) {
+	rud_formatp = item->ri_buf[0].iov_base;
+	if (item->ri_buf[0].iov_len != sizeof(struct xfs_rud_log_format)) {
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, log->l_mp,
-				rud_formatp, item->ri_buf[0].i_len);
+				rud_formatp, item->ri_buf[0].iov_len);
 		return -EFSCORRUPTED;
 	}
 
fs/xfs/xfs_trans.h +0 -1
index f06cc0f41665a..02b1c030cf178 100644
@@ -15,7 +15,6 @@ struct xfs_efd_log_item;
 struct xfs_efi_log_item;
 struct xfs_inode;
 struct xfs_item_ops;
-struct xfs_log_iovec;
 struct xfs_mount;
 struct xfs_trans;
 struct xfs_trans_res;
Signature · 1 line
2.53.0
maintainer
#3 lore raw

PATCH 3/3 xfs: bounds-check buffer log item's dirty bitmap

In reply to Sasha Levin

From: Ibrahim Hashimov <[email protected]>

[ Upstream commit 813f8136a2ce1fee266d02a7df73db6e8a541604 ]

xlog_recover_do_reg_buffer() replays each dirty region described by a buffer log item's bitmap into the buffer read for that item:

	memcpy(xfs_buf_offset(bp, (uint)bit << XFS_BLF_SHIFT),
		item->ri_buf[i].iov_base,
		nbits << XFS_BLF_SHIFT);

The destination offset (bit/nbits, from the logged dirty bitmap) and the buffer size (from the logged blf_len) are both attacker-controlled and otherwise unrelated, yet the only thing bounding the copy is an ASSERT(), which compiles away on production kernels. A crafted image logging a small blf_len together with a bitmap bit past the end of that buffer drives the memcpy() past the buffer's allocation, corrupting adjacent kernel heap during mount-time log recovery. This is reachable by anyone who can get a crafted image mounted -- the malicious-filesystem threat model XFS already guards against elsewhere.

Turn the ASSERT() into a real XFS_IS_CORRUPT() check that aborts recovery of the buffer with -EFSCORRUPTED, consistent with the validate-and-fail idiom already used in xlog_recover_do_inode_buffer() and xfs_dquot_item_recover.c. xlog_recover_do_reg_buffer() therefore becomes STATIC int and its three callers propagate the error.

Found and confirmed with KASAN on a CONFIG_XFS_DEBUG=n build: the crafted image trips a slab-out-of-bounds write before this change and fails recovery cleanly with -EFSCORRUPTED after it.

---
 fs/xfs/xfs_buf_item_recover.c | 57 +++++++++++++++++++++++++----------
 1 file changed, 41 insertions(+), 16 deletions(-)
fs/xfs/xfs_buf_item_recover.c +41 -16
index 5dca2dbf866b2..5917a5d0ef83a 100644
@@ -444,7 +444,7 @@ xlog_recover_validate_buf_type(
  * given buffer.  The bitmap in the buf log format structure indicates
  * where to place the logged data.
  */
-STATIC void
+STATIC int
 xlog_recover_do_reg_buffer(
 	struct xfs_mount		*mp,
 	struct xlog_recover_item	*item,
@@ -472,8 +472,24 @@ xlog_recover_do_reg_buffer(
 		ASSERT(nbits > 0);
 		ASSERT(item->ri_buf[i].iov_base != NULL);
 		ASSERT(item->ri_buf[i].iov_len % XFS_BLF_CHUNK == 0);
-		ASSERT(BBTOB(bp->b_length) >=
-		       ((uint)bit << XFS_BLF_SHIFT) + (nbits << XFS_BLF_SHIFT));
+		/*
+		 * The bitmap is only trustworthy to the extent that it
+		 * describes a region that actually fits inside the buffer we
+		 * read in based on the (attacker-controlled) blf_len.  Do not
+		 * rely on an ASSERT() for this -- it compiles away entirely on
+		 * non-DEBUG kernels, which is exactly where this matters, so
+		 * validate it for real and abort recovery of this buffer rather
+		 * than copying past the end of it.
+		 */
+		if (XFS_IS_CORRUPT(mp, BBTOB(bp->b_length) <
+				((uint)bit << XFS_BLF_SHIFT) +
+				(nbits << XFS_BLF_SHIFT))) {
+			xfs_alert(mp,
+	"Bad buffer log item dirty bitmap (bit %d, nbits %d) for %d-byte buffer at daddr 0x%llx.",
+				bit, nbits, BBTOB(bp->b_length),
+				xfs_buf_daddr(bp));
+			return -EFSCORRUPTED;
+		}
 
 		/*
 		 * The dirty regions logged in the buffer, even though
@@ -527,6 +543,7 @@ xlog_recover_do_reg_buffer(
 	ASSERT(i == item->ri_total);
 
 	xlog_recover_validate_buf_type(mp, bp, buf_f, current_lsn);
+	return 0;
 }
 
 /*
@@ -535,10 +552,10 @@ xlog_recover_do_reg_buffer(
  * (ie. USR or GRP), then just toss this buffer away; don't recover it.
  * Else, treat it as a regular buffer and do recovery.
  *
- * Return false if the buffer was tossed and true if we recovered the buffer to
- * indicate to the caller if the buffer needs writing.
+ * Return 0 if the buffer was not recovered (tossed), 1 if it was recovered and
+ * needs writing, or a negative errno if recovery of the buffer failed.
  */
-STATIC bool
+STATIC int
 xlog_recover_do_dquot_buffer(
 	struct xfs_mount		*mp,
 	struct xlog			*log,
@@ -547,6 +564,7 @@ xlog_recover_do_dquot_buffer(
 	struct xfs_buf_log_format	*buf_f)
 {
 	uint			type;
+	int			error;
 
 	trace_xfs_log_recover_buf_dquot_buf(log, buf_f);
 
@@ -554,7 +572,7 @@ xlog_recover_do_dquot_buffer(
 	 * Filesystems are required to send in quota flags at mount time.
 	 */
 	if (!mp->m_qflags)
-		return false;
+		return 0;
 
 	type = 0;
 	if (buf_f->blf_flags & XFS_BLF_UDQUOT_BUF)
@@ -567,10 +585,12 @@ xlog_recover_do_dquot_buffer(
 	 * This type of quotas was turned off, so ignore this buffer
 	 */
 	if (log->l_quotaoffs_flag & type)
-		return false;
+		return 0;
 
-	xlog_recover_do_reg_buffer(mp, item, bp, buf_f, NULLCOMMITLSN);
-	return true;
+	error = xlog_recover_do_reg_buffer(mp, item, bp, buf_f, NULLCOMMITLSN);
+	if (error)
+		return error;
+	return 1;
 }
 
 /*
@@ -706,7 +726,9 @@ xlog_recover_do_primary_sb_buffer(
 	xfs_agnumber_t			orig_agcount = mp->m_sb.sb_agcount;
 	int				error;
 
-	xlog_recover_do_reg_buffer(mp, item, bp, buf_f, current_lsn);
+	error = xlog_recover_do_reg_buffer(mp, item, bp, buf_f, current_lsn);
+	if (error)
+		return error;
 
 	if (orig_agcount == 0) {
 		xfs_alert(mp, "Trying to grow file system without AGs");
@@ -1026,11 +1048,11 @@ xlog_recover_buf_commit_pass2(
 			goto out_release;
 	} else if (buf_f->blf_flags &
 		  (XFS_BLF_UDQUOT_BUF|XFS_BLF_PDQUOT_BUF|XFS_BLF_GDQUOT_BUF)) {
-		bool	dirty;
-
-		dirty = xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
-		if (!dirty)
+		error = xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
+		if (error <= 0)
 			goto out_release;
+		/* write dirty buffer */
+		error = 0;
 	} else if ((xfs_blft_from_flags(buf_f) & XFS_BLFT_SB_BUF) &&
 			xfs_buf_daddr(bp) == 0) {
 		error = xlog_recover_do_primary_sb_buffer(mp, item, bp, buf_f,
@@ -1038,7 +1060,10 @@ xlog_recover_buf_commit_pass2(
 		if (error)
 			goto out_writebuf;
 	} else {
-		xlog_recover_do_reg_buffer(mp, item, bp, buf_f, current_lsn);
+		error = xlog_recover_do_reg_buffer(mp, item, bp, buf_f,
+						   current_lsn);
+		if (error)
+			goto out_release;
 	}
 
 	/*
Signature · 1 line
2.53.0