linux/fs/nfs/nfs4proc.c, branch v3.7-rc5

NFS4: nfs4_opendata_access should return errno

2012-11-02T22:51:54Z

Return errno - not an NFS4ERR_. This worked because NFS4ERR_ACCESS == EACCES. Signed-off-by: Weston Andros Adamson Signed-off-by: Trond Myklebust

NFSv4: Initialise the NFSv4.1 slot table highest_used_slotid correctly

2012-11-01T16:02:03Z

Signed-off-by: Trond Myklebust

NFS: add nfs_sb_deactive_async to avoid deadlock

2012-10-31T20:26:26Z

Use nfs_sb_deactive_async instead of nfs_sb_deactive when in a workqueue context. This avoids a deadlock where rpc_shutdown_client loops forever in a workqueue kworker context, trying to kill all RPC tasks associated with the client, while one or more of these tasks have already been assigned to the same kworker (and will never run rpc_exit_task). This approach is needed because RPC tasks that have already been assigned to a kworker by queue_work cannot be canceled, as explained in the comment for workqueue.c:insert_wq_barrier. Signed-off-by: Weston Andros Adamson [Trond: add module_get/put.] Signed-off-by: Trond Myklebust

NFSv4: nfs4_locku_done must release the sequence id

2012-10-31T19:10:04Z

If the state recovery machinery is triggered by the call to nfs4_async_handle_error() then we can deadlock. Signed-off-by: Trond Myklebust Cc: stable@vger.kernel.org

NFSv4.1: We must release the sequence id when we fail to get a session slot

2012-10-31T19:08:18Z

If we do not release the sequence id in cases where we fail to get a session slot, then we can deadlock if we hit a recovery scenario. Signed-off-by: Trond Myklebust Cc: stable@vger.kernel.org

NFS: Wait for session recovery to finish before returning

2012-10-31T17:13:28Z

Currently, we will schedule session recovery and then return to the caller of nfs4_handle_exception. This works for most cases, but causes a hang on the following test case: Client Server ------ ------ Open file over NFS v4.1 Write to file Expire client Try to lock file The server will return NFS4ERR_BADSESSION, prompting the client to schedule recovery. However, the client will continue placing lock attempts and the open recovery never seems to be scheduled. The simplest solution is to wait for session recovery to run before retrying the lock. Signed-off-by: Bryan Schumaker Signed-off-by: Trond Myklebust Cc: stable@vger.kernel.org

NFSv4 set open access operation call flag in nfs4_init_opendata_res

2012-10-04T00:10:28Z

nfs4_open_recover_helper zeros the nfs4_opendata result structures, removing the result access_request information which leads to an XDR decode error. Move the setting of the result access_request field to nfs4_init_opendata_res which sets all the other required nfs4_opendata result fields and is shared between the open and recover open paths. Signed-off-by: Andy Adamson Signed-off-by: Trond Myklebust

NFSv4 reduce attribute requests for open reclaim

2012-10-03T01:12:25Z

We currently make no distinction in attribute requests between normal OPENs and OPEN with CLAIM_PREVIOUS. This offers more possibility of failures in the GETATTR response which foils OPEN reclaim attempts. Reduce the requested attributes to the bare minimum needed to update the reclaim open stateid and split nfs4_opendata_to_nfs4_state processing accordingly. Signed-off-by: Andy Adamson Signed-off-by: Trond Myklebust

NFSv4: nfs4_open_done first must check that GETATTR decoded a file type

2012-10-03T00:09:00Z

...before it can check the validity of that file type. Signed-off-by: Trond Myklebust

NFSv4: don't put ACCESS in OPEN compound if O_EXCL

2012-10-02T21:56:19Z

Don't put an ACCESS op in OPEN compound if O_EXCL, because ACCESS will return permission denied for all bits until close. Fixes a regression due to commit 6168f62c (NFSv4: Add ACCESS operation to OPEN compound) Signed-off-by: Weston Andros Adamson Signed-off-by: Trond Myklebust