linux-stable.git/fs/ceph, branch v4.9.166

ceph: avoid repeatedly adding inode to mdsc->snap_flush_list

2019-02-27T09:06:58+00:00

commit 04242ff3ac0abbaa4362f97781dac268e6c3541a upstream.

Otherwise, mdsc->snap_flush_list may get corrupted.

Cc: stable@vger.kernel.org
Signed-off-by: "Yan, Zheng" 
Reviewed-by: Ilya Dryomov 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Greg Kroah-Hartman

ceph: don't update importing cap's mseq when handing cap export

2019-01-13T09:03:54+00:00

commit 3c1392d4c49962a31874af14ae9ff289cb2b3851 upstream.

Updating mseq makes client think importer mds has accepted all prior
cap messages and importer mds knows what caps client wants. Actually
some cap messages may have been dropped because of mseq mismatch.

If mseq is left untouched, importing cap's mds_wanted later will get
reset by cap import message.

Cc: stable@vger.kernel.org
Signed-off-by: "Yan, Zheng" 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Greg Kroah-Hartman

libceph: add authorizer challenge

2018-12-08T12:05:10+00:00

commit 6daca13d2e72bedaaacfc08f873114c9307d5aea upstream.

When a client authenticates with a service, an authorizer is sent with
a nonce to the service (ceph_x_authorize_[ab]) and the service responds
with a mutation of that nonce (ceph_x_authorize_reply).  This lets the
client verify the service is who it says it is but it doesn't protect
against a replay: someone can trivially capture the exchange and reuse
the same authorizer to authenticate themselves.

Allow the service to reject an initial authorizer with a random
challenge (ceph_x_authorize_challenge).  The client then has to respond
with an updated authorizer proving they are able to decrypt the
service's challenge and that the new authorizer was produced for this
specific connection instance.

The accepting side requires this challenge and response unconditionally
if the client side advertises they have CEPHX_V2 feature bit.

This addresses CVE-2018-1128.

Link: http://tracker.ceph.com/issues/24836
Signed-off-by: Ilya Dryomov 
Reviewed-by: Sage Weil 
Signed-off-by: Ben Hutchings 
Signed-off-by: Greg Kroah-Hartman

libceph: drop len argument of *verify_authorizer_reply()

2018-12-08T12:05:09+00:00

commit 0dde584882ade13dc9708d611fbf69b0ae8a9e48 upstream.

The length of the reply is protocol-dependent - for cephx it's
ceph_x_authorize_reply.  Nothing sensible can be passed from the
messenger layer anyway.

Signed-off-by: Ilya Dryomov 
Reviewed-by: Sage Weil 
Signed-off-by: Ben Hutchings 
Signed-off-by: Greg Kroah-Hartman

Revert "ceph: fix dentry leak in splice_dentry()"

2018-11-21T08:25:59+00:00

commit efe328230dc01aa0b1269aad0b5fae73eea4677a upstream.

This reverts commit 8b8f53af1ed9df88a4c0fbfdf3db58f62060edf3.

splice_dentry() is used by three places. For two places, req->r_dentry
is passed to splice_dentry(). In the case of error, req->r_dentry does
not get updated. So splice_dentry() should not drop reference.

Cc: stable@vger.kernel.org # 4.18+
Signed-off-by: "Yan, Zheng" 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Greg Kroah-Hartman

ceph: fix dentry leak in splice_dentry()

2018-08-24T11:12:33+00:00

[ Upstream commit 8b8f53af1ed9df88a4c0fbfdf3db58f62060edf3 ]

In any case, d_splice_alias() does not drop reference of original
dentry.

Signed-off-by: "Yan, Zheng" 
Reviewed-by: Jeff Layton 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

ceph: fix dentry leak when failing to init debugfs

2018-05-30T05:50:32+00:00

[ Upstream commit 18106734b512664a8541026519ce4b862498b6c3 ]

When failing from ceph_fs_debugfs_init() in ceph_real_mount(),
there is lack of dput of root_dentry and it causes slab errors,
so change the calling order of ceph_fs_debugfs_init() and
open_root_dentry() and do some cleanups to avoid this issue.

Signed-off-by: Chengguang Xu 
Reviewed-by: "Yan, Zheng" 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

ceph: only dirty ITER_IOVEC pages for direct read

2018-04-08T10:12:42+00:00

commit 85784f9395987a422fa04263e7c0fb13da11eb5c upstream.

If a page is already locked, attempting to dirty it leads to a deadlock
in lock_page().  This is what currently happens to ITER_BVEC pages when
a dio-enabled loop device is backed by ceph:

  $ losetup --direct-io /dev/loop0 /mnt/cephfs/img
  $ xfs_io -c 'pread 0 4k' /dev/loop0

Follow other file systems and only dirty ITER_IOVEC pages.

Cc: stable@kernel.org
Signed-off-by: "Yan, Zheng" 
Reviewed-by: Ilya Dryomov 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Greg Kroah-Hartman

ceph: drop negative child dentries before try pruning inode's alias

2017-12-20T09:07:16+00:00

commit 040d786032bf59002d374b86d75b04d97624005c upstream.

Negative child dentry holds reference on inode's alias, it makes
d_prune_aliases() do nothing.

Signed-off-by: "Yan, Zheng" 
Reviewed-by: Jeff Layton 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Greg Kroah-Hartman

ceph: unlock dangling spinlock in try_flush_caps()

2017-11-02T08:49:13+00:00

commit 6c2838fbdedb9b72a81c931d49e56b229b6cdbca upstream.

sparse warns:

  fs/ceph/caps.c:2042:9: warning: context imbalance in 'try_flush_caps' - wrong count at exit

We need to exit this function with the lock unlocked, but a couple of
cases leave it locked.

Signed-off-by: Jeff Layton 
Reviewed-by: "Yan, Zheng" 
Reviewed-by: Ilya Dryomov 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Greg Kroah-Hartman