- 31 Jul, 2014 20 commits
-
-
Jeff Layton authored
We don't want to rely on the client_mutex for protection in the case of NFSv4 open owners. Instead, we add a mutex that will only be taken for NFSv4.0 state mutating operations, and that will be released once the entire compound is done. Also, ensure that nfsd4_cstate_assign_replay/nfsd4_cstate_clear_replay take a reference to the stateowner when they are using it for NFSv4.0 open and lock replay caching. Signed-off-by:
Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by:
Jeff Layton <jlayton@primarydata.com> Signed-off-by:
J. Bruce Fields <bfields@redhat.com>
-
Jeff Layton authored
The way stateowners are managed today is somewhat awkward. They need to be explicitly destroyed, even though the stateids reference them. This will be particularly problematic when we remove the client_mutex. We may create a new stateowner and attempt to open a file or set a lock, and have that fail. In the meantime, another RPC may come in that uses that same stateowner and succeed. We can't have the first task tearing down the stateowner in that situation. To fix this, we need to change how stateowners are tracked altogether. Refcount them and only destroy them once all stateids that reference them have been destroyed. This patch starts by adding the refcounting necessary to do that. Signed-off-by:
-
Trond Myklebust authored
Allow nfs4_find_stateid_by_type to take the stateid reference, while still holding the &cl->cl_lock. Necessary step toward client_mutex removal. Signed-off-by:
-
Trond Myklebust authored
Allow nfs4_lookup_stateid to take the stateid reference, instead of having all the callers do so. Signed-off-by:
-
Trond Myklebust authored
Allow nfs4_preprocess_seqid_op to take the stateid reference, instead of having all the callers do so. Signed-off-by:
-
Trond Myklebust authored
Ensure that all the callers put the open stateid after use. Necessary step toward client_mutex removal. Signed-off-by:
-
Trond Myklebust authored
Ensure that nfsd4_open_confirm() keeps a reference to the open stateid until it is done working with it. Necessary step toward client_mutex removal. Signed-off-by:
-
Trond Myklebust authored
Prepare nfsd4_close for a future where nfs4_preprocess_seqid_op() hands it a fully referenced open stateid. Necessary step toward client_mutex removal. Signed-off-by:
-
Trond Myklebust authored
Ensure that nfsd4_process_open2() keeps a reference to the open stateid until it is done working with it. Necessary step toward client_mutex removal. Signed-off-by:
-
Trond Myklebust authored
Ensure that nfsd4_process_open2() keeps a reference to the delegation stateid until it is done working with it. Necessary step toward client_mutex removal. Signed-off-by:
-
Trond Myklebust authored
Ensure that nfs4_open_delegation() keeps a reference to the delegation stateid until it is done working with it. Necessary step toward client_mutex removal. Signed-off-by:
-
Trond Myklebust authored
Ensure that nfsd4_locku() keeps a reference to the lock stateid until it is done working with it. Necessary step toward client_mutex removal. Signed-off-by:
-
Trond Myklebust authored
Ensure that nfsd4_lock() references the lock stateid while it is manipulating it. Not currently necessary, but will be once the client_mutex is removed. Signed-off-by:
-
Jeff Layton authored
Hold the cl_lock over the bulk of these functions. In addition to ensuring that they aren't freed prematurely, this will also help prevent a potential race that could be introduced later. Once we remove the client_mutex, it'll be possible for FREE_STATEID and CLOSE to race and for both to try to put the "persistent" reference to the stateid. Signed-off-by:
-
Jeff Layton authored
Preparation for removal of the client_mutex. Currently, no lock aside from the client_mutex is held when calling find_lock_state. Ensure that the cl_lock is held by adding a lockdep assertion. Once we remove the client_mutex, it'll be possible for another thread to race in and insert a lock state for the same file after we search but before we insert a new one. Ensure that doesn't happen by redoing the search after allocating a new stid that we plan to insert. If one is found just put the one that was allocated. Signed-off-by:
-
Jeff Layton authored
Change to using the clp->cl_lock for this. For now, there's a lot of cl_lock thrashing, but in later patches we'll eliminate that and close the potential races that can occur when releasing the cl_lock while walking the lists. For now, the client_mutex prevents those races. Signed-off-by:
-
Jeff Layton authored
Releasing locks when we unhash the stateid instead of doing so only when the stateid is actually released will be problematic in later patches when we need to protect the unhashing with spinlocks. Move it into the sc_free operation instead. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Jeff Layton authored
Currently, this is serialized by the client_mutex, which is slated for removal. Add finer-grained locking here. Also, do some cleanup around find_stateid to prepare for taking references. Signed-off-by:
Benny Halevy <bhalevy@primarydata.com> Signed-off-by:
-
Trond Myklebust authored
All stateids are associated with a nfs4_file. Let's consolidate. Replace delegation->dl_file with the dl_stid.sc_file, and nfs4_ol_stateid->st_file with st_stid.sc_file. Signed-off-by:
-
Trond Myklebust authored
When we remove the client_mutex, we'll need to be able to ensure that these objects aren't destroyed while we're not holding locks. Add a ->free() callback to the struct nfs4_stid, so that we can release a reference to the stid without caring about the contents. Signed-off-by:
-
- 30 Jul, 2014 1 commit
-
-
Jeff Layton authored
It's possible for nfsd to fail opening a file that it has just created. When that happens, we throw a WARN but it doesn't include any info about the error code. Print the status code to give us a bit more info. Our QA group hit some of these warnings under some very heavy stress testing. My suspicion is that they hit the file-max limit, but it's hard to know for sure. Go ahead and add a -ENFILE mapping to nfserr_serverfault to make the error more distinct (and correct). Signed-off-by:
-
- 29 Jul, 2014 12 commits
-
-
Trond Myklebust authored
Signed-off-by:
-
Trond Myklebust authored
If requests are queued in the socket inbuffer waiting for an svc_tcp_has_wspace() requirement to be satisfied, then we do not want to clear the SOCK_NOSPACE flag until we've satisfied that requirement. Signed-off-by:
-
Trond Myklebust authored
Ensure that all calls to svc_xprt_enqueue() except svc_xprt_received() check the value of XPT_BUSY, before attempting to grab spinlocks etc. This is to avoid situations such as the following "perf" trace, which shows heavy contention on the pool spinlock: 54.15% nfsd [kernel.kallsyms] [k] _raw_spin_lock_bh | --- _raw_spin_lock_bh | |--71.43%-- svc_xprt_enqueue | | | |--50.31%-- svc_reserve | | | |--31.35%-- svc_xprt_received | | | |--18.34%-- svc_tcp_data_ready ... Signed-off-by: -
Jeff Layton authored
Now that the nfs4_file has a filehandle in it, we no longer need to keep a per-delegation copy of it. Switch to using the one in the nfs4_file instead. Signed-off-by:
-
Jeff Layton authored
The state lock can be fairly heavily contended, and there's no reason that nfs4_file lookups and delegation_blocked should be mutually exclusive. Let's give the new block_delegation code its own spinlock. It does mean that we'll need to take a different lock in the delegation break code, but that's not generally as critical to performance. Cc: Neil Brown <neilb@suse.de> Signed-off-by:
-
Jeff Layton authored
Move the alloc_init_deleg call into nfs4_set_delegation and change the function to return a pointer to the delegation or an IS_ERR return. This allows us to skip allocating a delegation if the file has already experienced a lease conflict. Signed-off-by:
-
Jeff Layton authored
No need to pass in a net pointer since we can derive that. Signed-off-by:
-
Jeff Layton authored
Signed-off-by:
-
Trond Myklebust authored
We want to convert to an atomic type so that we don't need to lock across the call to alloc_init_deleg(). Then convert to a long type so that we match the size of 'max_delegations'. None of this is a problem today, but it will be once we remove client_mutex protection. Signed-off-by:
-
Jeff Layton authored
Currently, both destroy_revoked_delegation and revoke_delegation manipulate the cl_revoked list without any locking aside from the client_mutex. Ensure that the clp->cl_lock is held when manipulating it, except for the list walking in destroy_client. At that point, the client should no longer be in use, and so it should be safe to walk the list without any locking. That also means that we don't need to do the list_splice_init there either. Also, the fact that revoke_delegation deletes dl_recall_lru list_head without any locking makes it difficult to know whether it's doing so safely in all cases. Move the list_del_init calls into the callers, and add a WARN_ON in the event that t's passed a delegation that has a non-empty list_head. Signed-off-by:
-
Jeff Layton authored
Ensure that the delegations cannot be found by the laundromat etc once we add them to the various 'revoke' lists. Signed-off-by:
-
Trond Myklebust authored
Don't allow stateids to clear the open file pointer until they are being destroyed. In a later patches we'll want to rely on the fact that we have a valid file pointer when dealing with the stateid and this will save us from having to do a lot of NULL pointer checks before doing so. Also, move to allocating stateids with kzalloc and get rid of the explicit zeroing of fields. Signed-off-by:
-
- 23 Jul, 2014 7 commits
-
-
Jeff Layton authored
Remove the fi_inode field in struct nfs4_file in order to remove the possibility of struct nfs4_file pinning the inode when it does not have any open state. The only place we still need to get to an inode is in check_for_locks, so change it to use find_any_file and use the inode from any that it finds. If it doesn't find one, then just assume there aren't any. Signed-off-by:
-
Trond Myklebust authored
...instead of just checking the inode that corresponds to it. Signed-off-by:
-
Trond Myklebust authored
This makes more sense anyway since an inode pointer value can change even when the filehandle doesn't. Signed-off-by:
-
Trond Myklebust authored
For use when we may not have a struct inode. Signed-off-by:
-
Himangi Saraogi authored
Replace a comma between expression statements by a semicolon. This changes the semantics of the code, but given the current indentation appears to be what is intended. A simplified version of the Coccinelle semantic patch that performs this transformation is as follows: // <smpl> @r@ expression e1,e2; @@ e1 -, +; e2; // </smpl> Signed-off-by:
Himangi Saraogi <himangi774@gmail.com> Acked-by:
Julia Lawall <julia.lawall@lip6.fr> Signed-off-by:
-
Chuck Lever authored
The RDMA credit limit controls how many concurrent RPCs are allowed per connection. An NFS/RDMA client and server exchange their credit limits in the RPC/RDMA headers. The Linux client and the Solaris client and server allow 32 credits. The Linux server allows only 16, which limits its performance. Set the server's default credit limit to 32, like the other well- known implementations, so the out-of-the-shrinkwrap performance of the Linux server is better. Signed-off-by:
Chuck Lever <chuck.lever@oracle.com> Signed-off-by:
-
Jeff Layton authored
Open stateids must be initialized with the st_access_bmap and st_deny_bmap set to 0, so that nfs4_get_vfs_file can properly record their state in old_access_bmap and old_deny_bmap. This bug was introduced in commit baeb4ff0 (nfsd: make deny mode enforcement more efficient and close races in it) and was causing the refcounts to end up incorrect when nfs4_get_vfs_file returned an error after bumping the refcounts. This made it impossible to unmount the underlying filesystem after running pynfs tests that involve deny modes. Signed-off-by:
-
