linux/fs, branch v3.4.26

libceph: Unlock unprocessed pages in start_read() error path

2013-01-17T16:51:21Z

Function start_read() can get an error before processing all pages. It must not only release the remaining pages, but unlock them too. This fixes http://tracker.newdream.net/issues/3370 Signed-off-by: David Zafman Reviewed-by: Alex Elder (cherry picked from commit 8884d53dd63b1d9315b343564fcbe1ede004a99e) Signed-off-by: Greg Kroah-Hartman

ceph: call handle_cap_grant() for cap import message

2013-01-17T16:51:21Z

If client sends cap message that requests new max size during exporting caps, the exporting MDS will drop the message quietly. So the client may wait for the reply that updates the max size forever. call handle_cap_grant() for cap import message can avoid this issue. Signed-off-by: Yan, Zheng Signed-off-by: Sage Weil (cherry picked from commit 0e5e1774a92e6fe9c511585de8f078b4c4c68dbb) Signed-off-by: Greg Kroah-Hartman

ceph: Fix __ceph_do_pending_vmtruncate

2013-01-17T16:51:21Z

we should set i_truncate_pending to 0 after page cache is truncated to i_truncate_size Signed-off-by: Yan, Zheng Signed-off-by: Sage Weil (cherry picked from commit a85f50b6ef93fbbb2ae932ce9b2376509d172796) Signed-off-by: Greg Kroah-Hartman

ceph: Don't add dirty inode to dirty list if caps is in migration

2013-01-17T16:51:21Z

Add dirty inode to cap_dirty_migrating list instead, this can avoid ceph_flush_dirty_caps() entering infinite loop. Signed-off-by: Yan, Zheng Signed-off-by: Sage Weil (cherry picked from commit 0685235ffd9dbdb9ccbda587f8a3c83ad1d5a921) Signed-off-by: Greg Kroah-Hartman

ceph: Fix infinite loop in __wake_requests

2013-01-17T16:51:21Z

__wake_requests() will enter infinite loop if we use it to wake requests in the session->s_waiting list. __wake_requests() deletes requests from the list and __do_request() adds requests back to the list. Signed-off-by: Yan, Zheng Signed-off-by: Sage Weil (cherry picked from commit ed75ec2cd19b47efcd292b6e23f58e56f4c5bc34) Signed-off-by: Greg Kroah-Hartman

ceph: Don't update i_max_size when handling non-auth cap

2013-01-17T16:51:21Z

The cap from non-auth mds doesn't have a meaningful max_size value. Signed-off-by: Yan, Zheng Signed-off-by: Sage Weil (cherry picked from commit 5e62ad30157d0da04cf40c6d1a2f4bc840948b9c) Signed-off-by: Greg Kroah-Hartman

libceph: remove 'osdtimeout' option

2013-01-17T16:51:20Z

This would reset a connection with any OSD that had an outstanding request that was taking more than N seconds. The idea was that if the OSD was buggy, the client could compensate by resending the request. In reality, this only served to hide server bugs, and we haven't actually seen such a bug in quite a while. Moreover, the userspace client code never did this. More importantly, often the request is taking a long time because the OSD is trying to recover, or overloaded, and killing the connection and retrying would only make the situation worse by giving the OSD more work to do. Signed-off-by: Sage Weil Reviewed-by: Alex Elder (cherry picked from commit 83aff95eb9d60aff5497e9f44a2ae906b86d8e88) Signed-off-by: Greg Kroah-Hartman

ceph: propagate layout error on osd request creation

2013-01-17T16:51:18Z

If we are creating an osd request and get an invalid layout, return an EINVAL to the caller. We switch up the return to have an error code instead of NULL implying -ENOMEM. Signed-off-by: Sage Weil Reviewed-by: Alex Elder (cherry picked from commit 6816282dab3a72efe8c0d182c1bc2960d87f4322) Signed-off-by: Greg Kroah-Hartman

ceph: tolerate (and warn on) extraneous dentry from mds

2013-01-17T16:51:18Z

If the MDS gives us a dentry and we weren't prepared to handle it, WARN_ON_ONCE instead of crashing. Reported-by: Yan, Zheng Signed-off-by: Sage Weil Reviewed-by: Alex Elder (cherry picked from commit 6c5e50fa614fea5325a2973be06f7ec6f1055316) Signed-off-by: Greg Kroah-Hartman

ceph: close old con before reopening on mds reconnect

2013-01-17T16:51:17Z

When we detect a mds session reset, close the old ceph_connection before reopening it. This ensures we clean up the old socket properly and keep the ceph_connection state correct. Signed-off-by: Sage Weil Reviewed-by: Alex Elder Reviewed-by: Yehuda Sadeh (cherry picked from commit a53aab645c82f0146e35684b34692c69b5118121) Signed-off-by: Greg Kroah-Hartman