kernel/git/stable/linux-stable.git/fs/ceph, branch linux-4.19.y

ceph: remove the incorrect Fw reference check when dirtying pages

2024-11-08T15:19:10+00:00

[ Upstream commit c08dfb1b49492c09cf13838c71897493ea3b424e ] When doing the direct-io reads it will also try to mark pages dirty, but for the read path it won't hold the Fw caps and there is case will it get the Fw reference. Fixes: 5dda377cf0a6 ("ceph: set i_head_snapc when getting CEPH_CAP_FILE_WR reference") Signed-off-by: Xiubo Li Reviewed-by: Patrick Donnelly Signed-off-by: Ilya Dryomov Signed-off-by: Sasha Levin

ceph: fix deadlock or deadcode of misusing dget()

2024-02-23T07:12:51+00:00

[ Upstream commit b493ad718b1f0357394d2cdecbf00a44a36fa085 ] The lock order is incorrect between denty and its parent, we should always make sure that the parent get the lock first. But since this deadcode is never used and the parent dir will always be set from the callers, let's just remove it. Link: https://lore.kernel.org/r/20231116081919.GZ1957730@ZenIV Reported-by: Al Viro Signed-off-by: Xiubo Li Reviewed-by: Jeff Layton Signed-off-by: Ilya Dryomov Signed-off-by: Sasha Levin

ceph: fix incorrect revoked caps assert in ceph_fill_file_size()

2023-10-25T09:16:31+00:00

commit 15c0a870dc44ed14e01efbdd319d232234ee639f upstream. When truncating the inode the MDS will acquire the xlock for the ifile Locker, which will revoke the 'Frwsxl' caps from the clients. But when the client just releases and flushes the 'Fw' caps to MDS, for exmaple, and once the MDS receives the caps flushing msg it just thought the revocation has finished. Then the MDS will continue truncating the inode and then issued the truncate notification to all the clients. While just before the clients receives the cap flushing ack they receive the truncation notification, the clients will detecte that the 'issued | dirty' is still holding the 'Fw' caps. Cc: stable@vger.kernel.org Link: https://tracker.ceph.com/issues/56693 Fixes: b0d7c2231015 ("ceph: introduce i_truncate_mutex") Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Signed-off-by: Ilya Dryomov Signed-off-by: Greg Kroah-Hartman

ceph: don't let check_caps skip sending responses for revoke msgs

2023-08-11T09:45:22+00:00

commit 257e6172ab36ebbe295a6c9ee9a9dd0fe54c1dc2 upstream. If a client sends out a cap update dropping caps with the prior 'seq' just before an incoming cap revoke request, then the client may drop the revoke because it believes it's already released the requested capabilities. This causes the MDS to wait indefinitely for the client to respond to the revoke. It's therefore always a good idea to ack the cap revoke request with the bumped up 'seq'. Cc: stable@vger.kernel.org Link: https://tracker.ceph.com/issues/61782 Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Reviewed-by: Patrick Donnelly Signed-off-by: Ilya Dryomov Signed-off-by: Greg Kroah-Hartman

ceph: fix use-after-free bug for inodes when flushing capsnaps

2023-06-14T08:57:14+00:00

commit 409e873ea3c1fd3079909718bbeb06ac1ec7f38b upstream. There is a race between capsnaps flush and removing the inode from 'mdsc->snap_flush_list' list: == Thread A == == Thread B == ceph_queue_cap_snap() -> allocate 'capsnapA' ->ihold('&ci->vfs_inode') ->add 'capsnapA' to 'ci->i_cap_snaps' ->add 'ci' to 'mdsc->snap_flush_list' ... == Thread C == ceph_flush_snaps() ->__ceph_flush_snaps() ->__send_flush_snap() handle_cap_flushsnap_ack() ->iput('&ci->vfs_inode') this also will release 'ci' ... == Thread D == ceph_handle_snap() ->flush_snaps() ->iterate 'mdsc->snap_flush_list' ->get the stale 'ci' ->remove 'ci' from ->ihold(&ci->vfs_inode) this 'mdsc->snap_flush_list' will WARNING To fix this we will increase the inode's i_count ref when adding 'ci' to the 'mdsc->snap_flush_list' list. [ idryomov: need_put int -> bool ] Cc: stable@vger.kernel.org Link: https://bugzilla.redhat.com/show_bug.cgi?id=2209299 Signed-off-by: Xiubo Li Reviewed-by: Milind Changire Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov Signed-off-by: Greg Kroah-Hartman

ceph: force updating the msg pointer in non-split case

2023-05-30T11:42:11+00:00

commit 4cafd0400bcb6187c0d4ab4d4b0229a89ac4f8c2 upstream. When the MClientSnap reqeust's op is not CEPH_SNAP_OP_SPLIT the request may still contain a list of 'split_realms', and we need to skip it anyway. Or it will be parsed as a corrupt snaptrace. Cc: stable@vger.kernel.org Link: https://tracker.ceph.com/issues/61200 Reported-by: Frank Schilder Signed-off-by: Xiubo Li Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov Signed-off-by: Greg Kroah-Hartman

ceph: avoid putting the realm twice when decoding snaps fails

2022-12-08T10:18:31+00:00

[ Upstream commit 51884d153f7ec85e18d607b2467820a90e0f4359 ] When decoding the snaps fails it maybe leaving the 'first_realm' and 'realm' pointing to the same snaprealm memory. And then it'll put it twice and could cause random use-after-free, BUG_ON, etc issues. Cc: stable@vger.kernel.org Link: https://tracker.ceph.com/issues/57686 Signed-off-by: Xiubo Li Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov Signed-off-by: Sasha Levin

ceph: do not update snapshot context when there is no new snapshot

2022-12-08T10:18:31+00:00

[ Upstream commit 2e586641c950e7f3e7e008404bd783a466b9b590 ] We will only track the uppest parent snapshot realm from which we need to rebuild the snapshot contexts _downward_ in hierarchy. For all the others having no new snapshot we will do nothing. This fix will avoid calling ceph_queue_cap_snap() on some inodes inappropriately. For example, with the code in mainline, suppose there are 2 directory hierarchies (with 6 directories total), like this: /dir_X1/dir_X2/dir_X3/ /dir_Y1/dir_Y2/dir_Y3/ Firstly, make a snapshot under /dir_X1/dir_X2/.snap/snap_X2, then make a root snapshot under /.snap/root_snap. Every time we make snapshots under /dir_Y1/..., the kclient will always try to rebuild the snap context for snap_X2 realm and finally will always try to queue cap snaps for dir_Y2 and dir_Y3, which makes no sense. That's because the snap_X2's seq is 2 and root_snap's seq is 3. So when creating a new snapshot under /dir_Y1/... the new seq will be 4, and the mds will send the kclient a snapshot backtrace in _downward_ order: seqs 4, 3. When ceph_update_snap_trace() is called, it will always rebuild the from the last realm, that's the root_snap. So later when rebuilding the snap context, the current logic will always cause it to rebuild the snap_X2 realm and then try to queue cap snaps for all the inodes related in that realm, even though it's not necessary. This is accompanied by a lot of these sorts of dout messages: "ceph: queue_cap_snap 00000000a42b796b nothing dirty|writing" Fix the logic to avoid this situation. Also, the 'invalidate' word is not precise here. In actuality, it will cause a rebuild of the existing snapshot contexts or just build non-existent ones. Rename it to 'rebuild_snapcs'. URL: https://tracker.ceph.com/issues/44100 Signed-off-by: Xiubo Li Reviewed-by: Jeff Layton Signed-off-by: Ilya Dryomov Stable-dep-of: 51884d153f7e ("ceph: avoid putting the realm twice when decoding snaps fails") Signed-off-by: Sasha Levin

ceph: don't truncate file in atomic_open

2022-10-26T11:19:18+00:00

commit 7cb9994754f8a36ae9e5ec4597c5c4c2d6c03832 upstream. Clear O_TRUNC from the flags sent in the MDS create request. `atomic_open' is called before permission check. We should not do any modification to the file here. The caller will do the truncation afterward. Fixes: 124e68e74099 ("ceph: file operations") Signed-off-by: Hu Weiwen Reviewed-by: Xiubo Li Signed-off-by: Ilya Dryomov [Xiubo: fixed a trivial conflict for 5.10 backport] Signed-off-by: Xiubo Li Signed-off-by: Greg Kroah-Hartman

ceph: allow ceph.dir.rctime xattr to be updatable

2022-06-14T14:59:38+00:00

[ Upstream commit d7a2dc523085f8b8c60548ceedc696934aefeb0e ] `rctime' has been a pain point in cephfs due to its buggy nature - inconsistent values reported and those sorts. Fixing rctime is non-trivial needing an overall redesign of the entire nested statistics infrastructure. As a workaround, PR http://github.com/ceph/ceph/pull/37938 allows this extended attribute to be manually set. This allows users to "fixup" inconsistent rctime values. While this sounds messy, its probably the wisest approach allowing users/scripts to workaround buggy rctime values. The above PR enables Ceph MDS to allow manually setting rctime extended attribute with the corresponding user-land changes. We may as well allow the same to be done via kclient for parity. Signed-off-by: Venky Shankar Reviewed-by: Xiubo Li Signed-off-by: Ilya Dryomov Signed-off-by: Sasha Levin