| Message ID | 20200814080833.84760-2-chandanrlinux@gmail.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | Bail out if transaction can cause extent count to overflow | expand |
> +int > +xfs_iext_count_may_overflow( > + struct xfs_inode *ip, > + int whichfork, > + int nr_to_add) > +{ > + struct xfs_ifork *ifp; > + uint64_t max_exts = 0; > + uint64_t nr_exts; > + > + switch (whichfork) { > + case XFS_DATA_FORK: > + max_exts = MAXEXTNUM; > + break; > + > + case XFS_ATTR_FORK: > + max_exts = MAXAEXTNUM; > + break; > + > + default: > + ASSERT(0); > + break; > + } > + > + ifp = XFS_IFORK_PTR(ip, whichfork); > + nr_exts = ifp->if_nextents + nr_to_add; > + > + if (nr_exts > max_exts) > + return -EFBIG; > + > + return 0; > +} Maybe it's just me, but I would structure this very different (just cosmetic differences, though). First add a: static inline uint32_t xfs_max_extents(int whichfork) { return XFS_ATTR_FORK ? MAXAEXTNUM : MAXEXTNUM; } to have a single place that determines the max number of extents. And the simplify the helper down to: int xfs_iext_count_may_overflow( struct xfs_inode *ip, int whichfork, int nr_to_add) { struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); uint64_t max_exts = xfs_max_extents(whichfork); uint64_t nr_exts; if (check_add_overflow(ifp->if_nextents, nr_to_add, &nr_exts) || nr_exts > max_exts)) return -EFBIG; return 0; } which actually might be small enough for an inline function now.
On Monday 17 August 2020 12:21:23 PM IST Christoph Hellwig wrote: > > +int > > +xfs_iext_count_may_overflow( > > + struct xfs_inode *ip, > > + int whichfork, > > + int nr_to_add) > > +{ > > + struct xfs_ifork *ifp; > > + uint64_t max_exts = 0; > > + uint64_t nr_exts; > > + > > + switch (whichfork) { > > + case XFS_DATA_FORK: > > + max_exts = MAXEXTNUM; > > + break; > > + > > + case XFS_ATTR_FORK: > > + max_exts = MAXAEXTNUM; > > + break; > > + > > + default: > > + ASSERT(0); > > + break; > > + } > > + > > + ifp = XFS_IFORK_PTR(ip, whichfork); > > + nr_exts = ifp->if_nextents + nr_to_add; > > + > > + if (nr_exts > max_exts) > > + return -EFBIG; > > + > > + return 0; > > +} > > Maybe it's just me, but I would structure this very different (just > cosmetic differences, though). First add a: > > static inline uint32_t xfs_max_extents(int whichfork) > { > return XFS_ATTR_FORK ? MAXAEXTNUM : MAXEXTNUM; > } > > to have a single place that determines the max number of extents. > > And the simplify the helper down to: > > int > xfs_iext_count_may_overflow( > struct xfs_inode *ip, > int whichfork, > int nr_to_add) > { > struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); > uint64_t max_exts = xfs_max_extents(whichfork); > uint64_t nr_exts; > > if (check_add_overflow(ifp->if_nextents, nr_to_add, &nr_exts) || > nr_exts > max_exts)) > return -EFBIG; > return 0; > } > > which actually might be small enough for an inline function now. > I agree. I will make the suggested changes in the next version.
diff --git a/fs/xfs/libxfs/xfs_trans_resv.c b/fs/xfs/libxfs/xfs_trans_resv.c index d1a0848cb52e..d21990d9df7a 100644 --- a/fs/xfs/libxfs/xfs_trans_resv.c +++ b/fs/xfs/libxfs/xfs_trans_resv.c @@ -832,6 +832,39 @@ xfs_calc_sb_reservation( return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize); } +int +xfs_iext_count_may_overflow( + struct xfs_inode *ip, + int whichfork, + int nr_to_add) +{ + struct xfs_ifork *ifp; + uint64_t max_exts = 0; + uint64_t nr_exts; + + switch (whichfork) { + case XFS_DATA_FORK: + max_exts = MAXEXTNUM; + break; + + case XFS_ATTR_FORK: + max_exts = MAXAEXTNUM; + break; + + default: + ASSERT(0); + break; + } + + ifp = XFS_IFORK_PTR(ip, whichfork); + nr_exts = ifp->if_nextents + nr_to_add; + + if (nr_exts > max_exts) + return -EFBIG; + + return 0; +} + void xfs_trans_resv_calc( struct xfs_mount *mp, diff --git a/fs/xfs/libxfs/xfs_trans_resv.h b/fs/xfs/libxfs/xfs_trans_resv.h index 7241ab28cf84..9d71b51990ac 100644 --- a/fs/xfs/libxfs/xfs_trans_resv.h +++ b/fs/xfs/libxfs/xfs_trans_resv.h @@ -93,5 +93,7 @@ struct xfs_trans_resv { void xfs_trans_resv_calc(struct xfs_mount *mp, struct xfs_trans_resv *resp); uint xfs_allocfree_log_count(struct xfs_mount *mp, uint num_ops); +int xfs_iext_count_may_overflow(struct xfs_inode *ip, int whichfork, + int nr_exts); #endif /* __XFS_TRANS_RESV_H__ */
XFS does not check for possible overflow of per-inode extent counter fields when adding extents to either data or attr fork. For e.g. 1. Insert 5 million xattrs (each having a value size of 255 bytes) and then delete 50% of them in an alternating manner. 2. On a 4k block sized XFS filesystem instance, the above causes 98511 extents to be created in the attr fork of the inode. xfsaild/loop0 2035 [003] 9643.390490: probe:xfs_iflush_int: (ffffffffac6225c0) if_nextents=98511 inode=131 3. The incore inode fork extent counter is a signed 32-bit quantity. However the on-disk extent counter is an unsigned 16-bit quantity and hence cannot hold 98511 extents. 4. The following incorrect value is stored in the attr extent counter, # xfs_db -f -c 'inode 131' -c 'print core.naextents' /dev/loop0 core.naextents = -32561 This commit adds a new helper function (i.e. xfs_iext_count_may_overflow()) to check for overflow of the per-inode data and xattr extent counters. Future patches will use this function to make sure that an FS operation will not cause the extent counter to overflow. Suggested-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Chandan Babu R <chandanrlinux@gmail.com> --- fs/xfs/libxfs/xfs_trans_resv.c | 33 +++++++++++++++++++++++++++++++++ fs/xfs/libxfs/xfs_trans_resv.h | 2 ++ 2 files changed, 35 insertions(+)