summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* LOCKD: Make nlmsvc_traverse_shares return voidJ. Bruce Fields2006-03-203-7/+3
| | | | | | | The nlmsvc_traverse_shares return value is always zero, hence useless. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* LOCKD: nlmsvc_traverse_blocks return is unusedJ. Bruce Fields2006-03-203-5/+5
| | | | | | | Note that we never return non-zero. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* SUNRPC,RPCSEC_GSS: fix krb5 sequence numbers.J. Bruce Fields2006-03-203-6/+16
| | | | | | | Use a spinlock to ensure unique sequence numbers when creating krb5 gss tokens. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Dont list system.nfs4_acl for filesystems that don't support it.J. Bruce Fields2006-03-201-0/+2
| | | | | | | | | | Thanks to Frank Filz for pointing out that we list system.nfs4_acl extended attribute even on filesystems where we don't actually support nfs4_acl. This is inconsistent with the e.g. ext3 POSIX ACL behaviour, and seems to annoy cp. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* SUNRPC,RPCSEC_GSS: remove unnecessary kmalloc of a checksumJ. Bruce Fields2006-03-205-17/+12
| | | | | | | | | | | | | | Remove unnecessary kmalloc of temporary space to hold the md5 result; it's small enough to just put on the stack. This code may be called to process rpc's necessary to perform writes, so there's a potential deadlock whenever we kmalloc() here. After this a couple kmalloc()'s still remain, to be removed soon. This also fixes a rare double-free on error noticed by coverity. Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* SUNRPC: Ensure rpc_call_async() always calls tk_ops->rpc_release()Trond Myklebust2006-03-203-21/+26
| | | | | | | Currently this will not happen if we exit before rpc_new_task() was called. Also fix up rpc_run_task() to do the same (for consistency). Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* SUNRPC: Fix memory barriers for req->rq_receivedTrond Myklebust2006-03-202-1/+7
| | | | | | | We need to ensure that all writes to the XDR buffers are done before req->rq_received is visible to other processors. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Fix a race in nfs_sync_inode()Trond Myklebust2006-03-203-30/+56
| | | | | | | | | | | | | | | | | | | | | | | | | Kudos to Neil Brown for spotting the problem: "in nfs_sync_inode, there is effectively the sequence: nfs_wait_on_requests nfs_flush_inode nfs_commit_inode This seems a bit racy to me as if the only requests are on the ->commit list, and nfs_commit_inode is called separately after nfs_wait_on_requests completes, and before nfs_commit_inode start (say: by nfs_write_inode) then none of these function will return >0, yet there will be some pending request that aren't waited for." The solution is to search for requests to wait upon, search for dirty requests, and search for uncommitted requests while holding the nfsi->req_lock The patch also cleans up nfs_sync_inode(), getting rid of the redundant FLUSH_WAIT flag. It turns out that we were always setting it. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Clean up nfs_flush_list()Trond Myklebust2006-03-201-33/+29
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Fix a race with PG_private and nfs_release_page()Trond Myklebust2006-03-202-5/+7
| | | | | | | | | | | | | | We don't need to set PG_private for readahead pages, since they never get unlocked while I/O is in progress. However there is a small race in nfs_readpage_release() whereby the page may be unlocked, and have PG_private set. Fix is to have PG_private set only for the case of writes... Also fix a bug in nfs_clear_page_writeback(): Don't attempt to clear the radix_tree tag if we've already deleted the radix tree entry. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Ensure the callback daemon flushes signalsTrond Myklebust2006-03-201-6/+13
| | | | | | | | | | If the callback daemon is signalled, but is unable to exit because it still has users, then we need to flush signals. If not, then svc_recv() can never sleep, and so we hang. If we flush signals, then we also have to be prepared to resend them when we want the thread to exit. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* SUNRPC: Fix a 'Busy inodes' error in rpc_pipefsTrond Myklebust2006-03-204-14/+35
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS, NLM: Allow blocking locks to respect signalsTrond Myklebust2006-03-201-6/+0
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Make nfs_fhget() return appropriate error valuesTrond Myklebust2006-03-203-14/+13
| | | | | | | Currently it returns NULL, which usually gets interpreted as ENOMEM. In fact it can mean a host of issues. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Fix an oops in nfs4_fill_superTrond Myklebust2006-03-202-8/+8
| | | | | | | | | | The mount statistics patches introduced a call to nfs_free_iostats that is not only redundant, but actually causes an oops. Also fix a memory leak due to the lack of a call to nfs_free_iostats on unmount. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: blocks should hold a reference to the nlm_fileTrond Myklebust2006-03-201-35/+46
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: SETCLIENTID_CONFIRM should handle NFS4ERR_DELAY/NFS4ERR_RESOURCETrond Myklebust2006-03-202-2/+20
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Send the delegation stateid for SETATTR callsTrond Myklebust2006-03-203-14/+33
| | | | | | | In the case where we hold a delegation stateid, use that in for inside SETATTR calls. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Ensure nfs_callback_down() calls svc_destroy()Trond Myklebust2006-03-201-0/+1
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: Fix a typo in nlmsvc_grant_release()Trond Myklebust2006-03-201-1/+3
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: Add helper for *_RES callbacksTrond Myklebust2006-03-204-192/+129
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NLM: Add nlmclnt_release_callTrond Myklebust2006-03-205-203/+135
| | | | | | Add a helper function to simplify the freeing of NLM client requests. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NLM: Fix nlmclnt_test to not copy private part of locksTrond Myklebust2006-03-202-2/+7
| | | | | | | The struct file_lock does not carry a properly initialised lock, so don't copy it as if it were. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NLM: Simplify client locksTrond Myklebust2006-03-203-54/+35
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: O_DIRECT needs to use a completionTrond Myklebust2006-03-201-5/+5
| | | | | | | | Now that we have aio writes, it is possible for dreq->outstanding to be zero, but for the I/O not to have completed. Convert struct nfs_direct_req to use a completion to signal when the I/O is done. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Clean up nfs_get_user_pagesTrond Myklebust2006-03-201-30/+29
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: fix compiler warnings on 64-bit platformsChuck Lever2006-03-201-2/+2
| | | | | | | | | | Introduced by NFS aio+dio patches. Test plan: Compile kernel with CONFIG_NFS enabled on 64-bit hardware. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* SUNRPC: fix compile warnings on 64-bit platformsChuck Lever2006-03-201-1/+1
| | | | | | | | | | Introduced by NFS metrics patch. Test plan: Compile kernel with CONFIG_NFS enabled on a 64-bit platform. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NLM: nlmclnt_cancel_callback should accept NLM_LCK_DENIED errorsTrond Myklebust2006-03-201-0/+1
| | | | | | | NLM_LCK_DENIED is a valid error return for an NLM_CANCEL call by the client. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: Fix Oopses due to list manipulation errors.Trond Myklebust2006-03-203-14/+15
| | | | | | | | The patch "stop abusing file_lock_list introduces a couple of bugs since the locks may be copied and need to be removed from the lists when they are destroyed. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: stop abusing file_lock_listChristoph Hellwig2006-03-207-63/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | Currently lockd directly access the file_lock_list from fs/locks.c. It does so to mark locks granted or reclaimable. This is very suboptimal, because a) lockd needs to poke into locks.c internals, and b) it needs to iterate over all locks in the system for marking locks granted or reclaimable. This patch adds lists for granted and reclaimable locks to the nlm_host structure instead, and adds locks to those. nlmclnt_lock: now adds the lock to h_granted instead of setting the NFS_LCK_GRANTED, still O(1) nlmclnt_mark_reclaim: goes away completely, replaced by a list_splice_init. Complexity reduced from O(locks in the system) to O(1) reclaimer: iterates over h_reclaim now, complexity reduced from O(locks in the system) to O(locks per nlm_host) Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: Make lockd use rpc_new_client() instead of rpc_create_clientTrond Myklebust2006-03-201-1/+2
| | | | | | | | When doing NLM_GRANTED requests, lockd may end up blocking if we use rpc_create_client() due to the synchronous call to rpc_ping(). Instead, use rpc_new_client(). Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: Make nlmsvc_create_block() use nlmsvc_lookup_host()Trond Myklebust2006-03-203-6/+3
| | | | | | | Currently it uses nlmclnt_lookup_host(), which puts the resulting host structure on a different list. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: Clean up of the server-side GRANTED codeTrond Myklebust2006-03-203-48/+53
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: Add refcounting to struct nlm_blockTrond Myklebust2006-03-202-50/+47
| | | | | | | Otherwise, the block may disappear from underneath us when in nlmsvc_retry_blocked. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: Fix server-side lock blocking codeTrond Myklebust2006-03-201-33/+40
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* lockd: posix_test_lock() should not call locks_copy_lock()Trond Myklebust2006-03-201-5/+13
| | | | | | | | The caller of posix_test_lock() should never need to look at the lock private data, so do not copy that information. This also means that there is no need to call the fl_release_private methods. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Uninline nfs_writedata_(alloc|free) and nfs_readdata_(alloc|free)Trond Myklebust2006-03-203-65/+66
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Debugging code for nfs_direct_(read|write)_schedule()Trond Myklebust2006-03-201-0/+4
| | | | | | Make sure that we're doing our list accounting correctly. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: O_DIRECT async IO may lose contextTrond Myklebust2006-03-201-18/+12
| | | | | | | | | | | | The struct nfs_direct_req currently keeps a pointer to the file descriptor without referencing it. This may cause problems if the parent process is killed. The nfs_open_context should normally have all the information that we're currently using the filp for, and unlike fput(), is safe to release from an rpciod process context. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* nfs: Use UNSTABLE + COMMIT for NFS O_DIRECT writesTrond Myklebust2006-03-202-25/+200
| | | | | | | | | | | | | | | | | | Currently NFS O_DIRECT writes use FILE_SYNC so that a COMMIT is not necessary. This simplifies the internal logic, but this could be a difficult workload for some servers. Instead, let's send UNSTABLE writes, and after they all complete, send a COMMIT for the dirty range. After the COMMIT returns successfully, then do the wake_up or fire off aio_complete(). Test plan: Async direct I/O tests against Solaris (or any server that requires committed unstable writes). Reboot server during test. Based on an earlier patch by Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Make nfs_commit_alloc() externTrond Myklebust2006-03-202-2/+7
| | | | | | We need to use nfs_commit_alloc() in fs/nfs/direct.c. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: fix data_update accounting in NFS direct I/O pathChuck Lever2006-03-201-0/+3
| | | | | | | | | | | ^C against "iozone -I" is hitting the assertion in nfs_clear_inode(). Test plan: "iozone -i0 -I -a -c" against a slow server, then control C. This should not cause an oops. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Replace atomic_t variables in nfs_direct_req with a single spin lockChuck Lever2006-03-201-34/+47
| | | | | | | | | | | | | | | | Three atomic_t variables cause a lot of bus locking. Because they are all used in the same places in the code, just use a single spin lock. Now that the atomic_t variables are gone, we can remove the request size limitation since the code no longer depends on the limited width of atomic_t on some platforms. Test plan: Compile with CONFIG_NFS and CONFIG_NFS_DIRECTIO enabled. Millions of fsx operations, iozone, OraSim. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: clean up comments and tab damage in direct.cChuck Lever2006-03-201-23/+31
| | | | | | | | | | | Clean up tab damage and comments. Replace "file_offset" with more commonly used "pos". Test plan: Compile with CONFIG_NFS and CONFIG_NFS_DIRECTIO enabled. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: support EIOCBQUEUED return in direct write pathChuck Lever2006-03-201-7/+13
| | | | | | | | | | | | | | | | | For async iocb's, the NFS direct write path now returns EIOCBQUEUED, and calls aio_complete when all the requested writes are finished. The synchronous part of the NFS direct write path behaves exactly as it was before. Shared mapped NFS files will have some coherency difficulties when accessed concurrently with aio+dio. Will need to explore how this is handled in the local file system case. Test plan: aio-stress with "-O". OraSim. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: make iocb available everywhere in direct write pathChuck Lever2006-03-201-32/+14
| | | | | | | | | | | | Pass the iocb argument all the way down to the direct write request scheduler, and make it available in nfs_direct_write_result. Test plan: Compile the kernel with CONFIG_NFS and CONFIG_NFS_DIRECTIO enabled. Millions of fsx-odirect ops. OraSim. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: remove support for multi-segment iovs in the direct write pathChuck Lever2006-03-201-51/+17
| | | | | | | | | | | | | Eliminate the persistent use of automatic storage in all parts of the NFS client's direct write path to pave the way for introducing support for aio against files opened with the O_DIRECT flag. Test plan: Compile the kernel with CONFIG_NFS and CONFIG_NFS_DIRECTIO enabled. Millions of fsx-odirect ops. OraSim. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: make direct write path generate write requests concurrentlyChuck Lever2006-03-203-83/+162
| | | | | | | | | | | | | | | | | Duplicate infrastructure from direct read path that will allow write path to generate multiple write requests concurrently. This will enable us to add support for aio in this path. Temporarily we will lose the ability to do UNSTABLE writes followed by a COMMIT in the direct write path. However, all applications I am aware of that use NFS O_DIRECT currently write in relatively small chunks, so this should not be inconvenient in any way. Test plan: Millions of fsx-odirect ops. OraSim. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: create common routine for handling direct I/O completionChuck Lever2006-03-201-20/+26
| | | | | | | | | | Factor out the common piece of completing an NFS direct I/O request. Test plan: Compile kernel with CONFIG_NFS and CONFIG_NFS_DIRECTIO enabled. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
OpenPOWER on IntegriCloud