kernel.git/net/ceph/messenger.c, branch linux-5.2.y

Merge tag 'afs-fixes-20190516' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs

2019-05-17T00:00:13Z

Pull misc AFS fixes from David Howells: "This fixes a set of miscellaneous issues in the afs filesystem, including: - leak of keys on file close. - broken error handling in xattr functions. - missing locking when updating VL server list. - volume location server DNS lookup whereby preloaded cells may not ever get a lookup and regular DNS lookups to maintain server lists consume power unnecessarily. - incorrect error propagation and handling in the fileserver iteration code causes operations to sometimes apparently succeed. - interruption of server record check/update side op during fileserver iteration causes uninterruptible main operations to fail unexpectedly. - callback promise expiry time miscalculation. - over invalidation of the callback promise on directories. - double locking on callback break waking up file locking waiters. - double increment of the vnode callback break counter. Note that it makes some changes outside of the afs code, including: - an extra parameter to dns_query() to allow the dns_resolver key just accessed to be immediately invalidated. AFS is caching the results itself, so the key can be discarded. - an interruptible version of wait_var_event(). - an rxrpc function to allow the maximum lifespan to be set on a call. - a way for an rxrpc call to be marked as non-interruptible" * tag 'afs-fixes-20190516' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs: afs: Fix double inc of vnode->cb_break afs: Fix lock-wait/callback-break double locking afs: Don't invalidate callback if AFS_VNODE_DIR_VALID not set afs: Fix calculation of callback expiry time afs: Make dynamic root population wait uninterruptibly for proc_cells_lock afs: Make some RPC operations non-interruptible rxrpc: Allow the kernel to mark a call as being non-interruptible afs: Fix error propagation from server record check/update afs: Fix the maximum lifespan of VL and probe calls rxrpc: Provide kernel interface to set max lifespan on a call afs: Fix "kAFS: AFS vnode with undefined type 0" afs: Fix cell DNS lookup Add wait_var_event_interruptible() dns_resolver: Allow used keys to be invalidated afs: Fix afs_cell records to always have a VL server list record afs: Fix missing lock when replacing VL server list afs: Fix afs_xattr_get_yfs() to not try freeing an error value afs: Fix incorrect error handling in afs_xattr_get_acl() afs: Fix key leak in afs_release() and afs_evict_inode()

dns_resolver: Allow used keys to be invalidated

2019-05-15T16:35:54Z

Allow used DNS resolver keys to be invalidated after use if the caller is doing its own caching of the results. This reduces the amount of resources required. Fix AFS to invalidate DNS results to kill off permanent failure records that get lodged in the resolver keyring and prevent future lookups from happening. Fixes: 0a5143f2f89c ("afs: Implement VL server rotation") Signed-off-by: David Howells

libceph: make ceph_pr_addr take an struct ceph_entity_addr pointer

2019-05-07T17:43:05Z

GCC9 is throwing a lot of warnings about unaligned accesses by callers of ceph_pr_addr. All of the current callers are passing a pointer to the sockaddr inside struct ceph_entity_addr. Fix it to take a pointer to a struct ceph_entity_addr instead, and then have the function make a copy of the sockaddr before printing it. Signed-off-by: Jeff Layton Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov

libceph: fix unaligned accesses in ceph_entity_addr handling

2019-05-07T17:43:05Z

GCC9 is throwing a lot of warnings about unaligned access. This patch fixes some of them by changing most of the sockaddr handling functions to take a pointer to struct ceph_entity_addr instead of struct sockaddr_storage. The lower functions can then make copies or do unaligned accesses as needed. Signed-off-by: Jeff Layton Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov

libceph: fix breakage caused by multipage bvecs

2019-03-25T21:28:07Z

A bvec can now consist of multiple physically contiguous pages. This means that bvec_iter_advance() can move to a different page while staying in the same bvec (i.e. ->bi_bvec_done != 0). The messenger works in terms of segments which can now be defined as the smaller of a bvec and a page. The "more bytes to process in this segment" condition holds only if bvec_iter_advance() leaves us in the same bvec _and_ in the same page. On next bvec (possibly in the same page) and on next page (possibly in the same bvec) we may need to set ->last_piece. Signed-off-by: Ilya Dryomov

libceph: handle an empty authorize reply

2019-02-18T17:05:33Z

The authorize reply can be empty, for example when the ticket used to build the authorizer is too old and TAG_BADAUTHORIZER is returned from the service. Calling ->verify_authorizer_reply() results in an attempt to decrypt and validate (somewhat) random data in au->buf (most likely the signature block from calc_signature()), which fails and ends up in con_fault_finish() with !con->auth_retry. The ticket isn't invalidated and the connection is retried again and again until a new ticket is obtained from the monitor: libceph: osd2 192.168.122.1:6809 bad authorize reply libceph: osd2 192.168.122.1:6809 bad authorize reply libceph: osd2 192.168.122.1:6809 bad authorize reply libceph: osd2 192.168.122.1:6809 bad authorize reply Let TAG_BADAUTHORIZER handler kick in and increment con->auth_retry. Cc: stable@vger.kernel.org Fixes: 5c056fdc5b47 ("libceph: verify authorize reply on connect") Link: https://tracker.ceph.com/issues/20164 Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil

libceph: avoid KEEPALIVE_PENDING races in ceph_con_keepalive()

2019-01-21T13:53:12Z

con_fault() can transition the connection into STANDBY right after ceph_con_keepalive() clears STANDBY in clear_standby(): libceph user thread ceph-msgr worker ceph_con_keepalive() mutex_lock(&con->mutex) clear_standby(con) mutex_unlock(&con->mutex) mutex_lock(&con->mutex) con_fault() ... if KEEPALIVE_PENDING isn't set set state to STANDBY ... mutex_unlock(&con->mutex) set KEEPALIVE_PENDING set WRITE_PENDING This triggers warnings in clear_standby() when either ceph_con_send() or ceph_con_keepalive() get to clearing STANDBY next time. I don't see a reason to condition queue_con() call on the previous value of KEEPALIVE_PENDING, so move the setting of KEEPALIVE_PENDING into the critical section -- unlike WRITE_PENDING, KEEPALIVE_PENDING could have been a non-atomic flag. Reported-by: syzbot+acdeb633f6211ccdf886@syzkaller.appspotmail.com Signed-off-by: Ilya Dryomov Tested-by: Myungho Jung

libceph: switch more to bool in ceph_tcp_sendmsg()

2018-12-26T14:56:04Z

Unlike in ceph_tcp_sendpage(), it's a bool. Signed-off-by: Ilya Dryomov

libceph: use MSG_SENDPAGE_NOTLAST with ceph_tcp_sendpage()

2018-12-26T14:56:04Z

Prevent do_tcp_sendpages() from calling tcp_push() (at least) once per page. Instead, arrange for tcp_push() to be called (at least) once per data payload. This results in more MSS-sized packets and fewer packets overall (5-10% reduction in my tests with typical OSD request sizes). See commits 2f5338442425 ("tcp: allow splice() to build full TSO packets"), 35f9c09fe9c7 ("tcp: tcp_sendpages() should call tcp_push() once") and ae62ca7b0321 ("tcp: fix MSG_SENDPAGE_NOTLAST logic") for details. Here is an example of a packet size histogram for 128K OSD requests (MSS = 1448, top 5): Before: SIZE COUNT 1448 777700 952 127915 1200 39238 1219 9806 21 5675 After: SIZE COUNT 1448 897280 21 6201 1019 2797 643 2739 376 2479 We could do slightly better by explicitly corking the socket but it's not clear it's worth it. Signed-off-by: Ilya Dryomov

libceph: use sock_no_sendpage() as a fallback in ceph_tcp_sendpage()

2018-12-26T14:56:04Z

sock_no_sendpage() makes the code cleaner. Also, don't set MSG_EOR. sendpage doesn't act on MSG_EOR on its own, it just honors the setting from the preceding sendmsg call by looking at ->eor in tcp_skb_can_collapse_to(). Signed-off-by: Ilya Dryomov