summaryrefslogtreecommitdiffstats
path: root/net/sunrpc
Commit message (Collapse)AuthorAgeFilesLines
...
| | * | xprtrdma: Fix MR list handlingChuck Lever2019-10-241-20/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Close some holes introduced by commit 6dc6ec9e04c4 ("xprtrdma: Cache free MRs in each rpcrdma_req") that could result in list corruption. In addition, the result that is tabulated in @count is no longer used, so @count is removed. Fixes: 6dc6ec9e04c4 ("xprtrdma: Cache free MRs in each rpcrdma_req") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| | * | xprtrdma: Close window between waking RPC senders and posting ReceivesChuck Lever2019-10-243-4/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A recent clean up attempted to separate Receive handling and RPC Reply processing, in the name of clean layering. Unfortunately, we can't do this because the Receive Queue has to be refilled _after_ the most recent credit update from the responder is parsed from the transport header, but _before_ we wake up the next RPC sender. That is right in the middle of rpcrdma_reply_handler(). Usually this isn't a problem because current responder implementations don't vary their credit grant. The one exception is when a connection is established: the grant goes from one to a much larger number on the first Receive. The requester MUST post enough Receives right then so that any outstanding requests can be sent without risking RNR and connection loss. Fixes: 6ceea36890a0 ("xprtrdma: Refactor Receive accounting") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| | * | xprtrdma: Initialize rb_credits in one placeChuck Lever2019-10-244-16/+38
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Clean up/code de-duplication. Nit: RPC_CWNDSHIFT is incorrect as the initial value for xprt->cwnd. This mistake does not appear to have operational consequences, since the cwnd value is replaced with a valid value upon the first Receive completion. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| | * | xprtrdma: Connection becomes unstable after a reconnectChuck Lever2019-10-242-0/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is because xprt_request_get_cong() is allowing more than one RPC Call to be transmitted before the first Receive on the new connection. The first Receive fills the Receive Queue based on the server's credit grant. Before that Receive, there is only a single Receive WR posted because the client doesn't know the server's credit grant. Solution is to clear rq_cong on all outstanding rpc_rqsts when the the cwnd is reset. This is because an RPC/RDMA credit is good for one connection instance only. Fixes: 75891f502f5f ("SUNRPC: Support for congestion control ... ") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| | * | xprtrdma: Add unique trace points for posting Local Invalidate WRsChuck Lever2019-10-241-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When adding frwr_unmap_async way back when, I re-used the existing trace_xprtrdma_post_send() trace point to record the return code of ib_post_send. Unfortunately there are some cases where re-using that trace point causes a crash. Instead, construct a trace point specific to posting Local Invalidate WRs that will always be safe to use in that context, and will act as a trace log eye-catcher for Local Invalidation. Fixes: 847568942f93 ("xprtrdma: Remove fr_state") Fixes: d8099feda483 ("xprtrdma: Reduce context switching due ... ") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Tested-by: Bill Baker <bill.baker@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| | * | SUNRPC: Add trace points to observe transport congestion controlChuck Lever2019-10-241-9/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To help debug problems with RPC/RDMA credit management, replace dprintk() call sites in the transport send lock paths with trace events. Similar trace points are defined for the non-congestion paths. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| | * | SUNRPC: Eliminate log noise in call_reserveresultChuck Lever2019-10-241-12/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Sep 11 16:35:20 manet kernel: call_reserveresult: unrecognized error -512, exiting Diagnostic error messages such as this likely have no value for NFS client administrators. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | | SUNRPC: Avoid RPC delays when exiting suspendTrond Myklebust2019-11-061-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Jon Hunter: "I have been tracking down another suspend/NFS related issue where again I am seeing random delays exiting suspend. The delays can be up to a couple minutes in the worst case and this is causing a suspend test we have to fail." Change the use of a deferrable work to a standard delayed one. Reported-by: Jon Hunter <jonathanh@nvidia.com> Tested-by: Jon Hunter <jonathanh@nvidia.com> Fixes: 7e0a0e38fcfea ("SUNRPC: Replace the queue timer with a delayed work function") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
| * | | NFSv4.1: Don't rebind to the same source port when reconnecting to the serverTrond Myklebust2019-11-032-2/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | NFSv2, v3 and NFSv4 servers often have duplicate replay caches that look at the source port when deciding whether or not an RPC call is a replay of a previous call. This requires clients to perform strange TCP gymnastics in order to ensure that when they reconnect to the server, they bind to the same source port. NFSv4.1 and NFSv4.2 have sessions that provide proper replay semantics, that do not look at the source port of the connection. This patch therefore ensures they can ignore the rebind requirement. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
* | | | kernel/notifier.c: remove blocking_notifier_chain_cond_register()Xiaoming Ni2019-12-041-1/+1
|/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | blocking_notifier_chain_cond_register() does not consider system_booting state, which is the only difference between this function and blocking_notifier_cain_register(). This can be a bug and is a piece of duplicate code. Delete blocking_notifier_chain_cond_register() Link: http://lkml.kernel.org/r/1568861888-34045-4-git-send-email-nixiaoming@huawei.com Signed-off-by: Xiaoming Ni <nixiaoming@huawei.com> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Anna Schumaker <anna.schumaker@netapp.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: Chuck Lever <chuck.lever@oracle.com> Cc: David S. Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@kernel.org> Cc: J. Bruce Fields <bfields@fieldses.org> Cc: Jeff Layton <jlayton@kernel.org> Cc: Nadia Derbey <Nadia.Derbey@bull.net> Cc: "Paul E. McKenney" <paulmck@kernel.org> Cc: Sam Protsenko <semen.protsenko@linaro.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Trond Myklebust <trond.myklebust@hammerspace.com> Cc: Vasily Averin <vvs@virtuozzo.com> Cc: Viresh Kumar <viresh.kumar@linaro.org> Cc: YueHaibing <yuehaibing@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | | SUNRPC: Destroy the back channel when we destroy the host transportTrond Myklebust2019-10-302-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we're destroying the host transport mechanism, we should ensure that we do not leak memory by failing to release any back channel slots that might still exist. Reported-by: Neil Brown <neilb@suse.de> Reported-by: kbuild test robot <lkp@intel.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
* | | SUNRPC: The RDMA back channel mustn't disappear while requests are outstandingTrond Myklebust2019-10-301-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If there are RDMA back channel requests being processed by the server threads, then we should hold a reference to the transport to ensure it doesn't get freed from underneath us. Reported-by: Neil Brown <neilb@suse.de> Fixes: 63cae47005af ("xprtrdma: Handle incoming backward direction RPC calls") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
* | | SUNRPC: The TCP back channel mustn't disappear while requests are outstandingTrond Myklebust2019-10-301-2/+3
|/ / | | | | | | | | | | | | | | | | | | | | If there are TCP back channel requests being processed by the server threads, then we should hold a reference to the transport to ensure it doesn't get freed from underneath us. Reported-by: Neil Brown <neilb@suse.de> Fixes: 2ea24497a1b3 ("SUNRPC: RPC callbacks may be split across several..") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
* / SUNRPC: fix race to sk_err after xs_error_reportBenjamin Coddington2019-10-101-9/+8
|/ | | | | | | | | | | | | | | | | Since commit 4f8943f80883 ("SUNRPC: Replace direct task wakeups from softirq context") there has been a race to the value of the sk_err if both XPRT_SOCK_WAKE_ERROR and XPRT_SOCK_WAKE_DISCONNECT are set. In that case, we may end up losing the sk_err value that existed when xs_error_report was called. Fix this by reverting to the previous behavior: instead of using SO_ERROR to retrieve the value at a later time (which might also return sk_err_soft), copy the sk_err value onto struct sock_xprt, and use that value to wake pending tasks. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Fixes: 4f8943f80883 ("SUNRPC: Replace direct task wakeups from softirq context") Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
* Merge tag 'nfsd-5.4' of git://linux-nfs.org/~bfields/linuxLinus Torvalds2019-09-275-30/+26
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull nfsd updates from Bruce Fields: "Highlights: - Add a new knfsd file cache, so that we don't have to open and close on each (NFSv2/v3) READ or WRITE. This can speed up read and write in some cases. It also replaces our readahead cache. - Prevent silent data loss on write errors, by treating write errors like server reboots for the purposes of write caching, thus forcing clients to resend their writes. - Tweak the code that allocates sessions to be more forgiving, so that NFSv4.1 mounts are less likely to hang when a server already has a lot of clients. - Eliminate an arbitrary limit on NFSv4 ACL sizes; they should now be limited only by the backend filesystem and the maximum RPC size. - Allow the server to enforce use of the correct kerberos credentials when a client reclaims state after a reboot. And some miscellaneous smaller bugfixes and cleanup" * tag 'nfsd-5.4' of git://linux-nfs.org/~bfields/linux: (34 commits) sunrpc: clean up indentation issue nfsd: fix nfs read eof detection nfsd: Make nfsd_reset_boot_verifier_locked static nfsd: degraded slot-count more gracefully as allocation nears exhaustion. nfsd: handle drc over-allocation gracefully. nfsd: add support for upcall version 2 nfsd: add a "GetVersion" upcall for nfsdcld nfsd: Reset the boot verifier on all write I/O errors nfsd: Don't garbage collect files that might contain write errors nfsd: Support the server resetting the boot verifier nfsd: nfsd_file cache entries should be per net namespace nfsd: eliminate an unnecessary acl size limit Deprecate nfsd fault injection nfsd: remove duplicated include from filecache.c nfsd: Fix the documentation for svcxdr_tmpalloc() nfsd: Fix up some unused variable warnings nfsd: close cached files prior to a REMOVE or RENAME that would replace target nfsd: rip out the raparms cache nfsd: have nfsd_test_lock use the nfsd_file cache nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache ...
| * sunrpc: clean up indentation issueColin Ian King2019-09-251-2/+2
| | | | | | | | | | | | | | | | There are statements that are indented incorrectly, remove the extraneous spacing. Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
| * sunrpc: add a new cache_detail operation for when a cache is flushedJeff Layton2019-08-191-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the exports table is changed, exportfs will usually write a new time to the "flush" file in the nfsd.export cache procfile. This tells the kernel to flush any entries that are older than that value. This gives us a mechanism to tell whether an unexport might have occurred. Add a new ->flush cache_detail operation that is called after flushing the cache whenever someone writes to a "flush" file. Signed-off-by: Jeff Layton <jeff.layton@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
| * svcrdma: Use llist for managing cache of recv_ctxtsChuck Lever2019-08-192-16/+11
| | | | | | | | | | | | | | | | Use a wait-free mechanism for managing the svc_rdma_recv_ctxts free list. Subsequently, sc_recv_lock can be eliminated. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
| * svcrdma: Remove svc_rdma_wqChuck Lever2019-08-192-8/+2
| | | | | | | | | | | | | | Clean up: the system workqueue will work just as well. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
| * SUNRPC: Track writers of the 'channel' file to improve cache_listeners_existDave Wysochanski2019-07-301-4/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The sunrpc cache interface is susceptible to being fooled by a rogue process just reading a 'channel' file. If this happens the kernel may think a valid daemon exists to service the cache when it does not. For example, the following may fool the kernel: cat /proc/net/rpc/auth.unix.gid/channel Change the tracking of readers to writers when considering whether a listener exists as all valid daemon processes either open a channel file O_RDWR or O_WRONLY. While this does not prevent a rogue process from "stealing" a message from the kernel, it does at least improve the kernels perception of whether a valid process servicing the cache exists. Signed-off-by: Dave Wysochanski <dwysocha@redhat.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
* | Merge tag 'nfs-for-5.4-1' of git://git.linux-nfs.org/projects/anna/linux-nfsLinus Torvalds2019-09-2612-384/+388
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull NFS client updates from Anna Schumaker: "Stable bugfixes: - Dequeue the request from the receive queue while we're re-encoding # v4.20+ - Fix buffer handling of GSS MIC without slack # 5.1 Features: - Increase xprtrdma maximum transport header and slot table sizes - Add support for nfs4_call_sync() calls using a custom rpc_task_struct - Optimize the default readahead size - Enable pNFS filelayout LAYOUTGET on OPEN Other bugfixes and cleanups: - Fix possible null-pointer dereferences and memory leaks - Various NFS over RDMA cleanups - Various NFS over RDMA comment updates - Don't receive TCP data into a reset request buffer - Don't try to parse incomplete RPC messages - Fix congestion window race with disconnect - Clean up pNFS return-on-close error handling - Fixes for NFS4ERR_OLD_STATEID handling" * tag 'nfs-for-5.4-1' of git://git.linux-nfs.org/projects/anna/linux-nfs: (53 commits) pNFS/filelayout: enable LAYOUTGET on OPEN NFS: Optimise the default readahead size NFSv4: Handle NFS4ERR_OLD_STATEID in LOCKU NFSv4: Handle NFS4ERR_OLD_STATEID in CLOSE/OPEN_DOWNGRADE NFSv4: Fix OPEN_DOWNGRADE error handling pNFS: Handle NFS4ERR_OLD_STATEID on layoutreturn by bumping the state seqid NFSv4: Add a helper to increment stateid seqids NFSv4: Handle RPC level errors in LAYOUTRETURN NFSv4: Handle NFS4ERR_DELAY correctly in return-on-close NFSv4: Clean up pNFS return-on-close error handling pNFS: Ensure we do clear the return-on-close layout stateid on fatal errors NFS: remove unused check for negative dentry NFSv3: use nfs_add_or_obtain() to create and reference inodes NFS: Refactor nfs_instantiate() for dentry referencing callers SUNRPC: Fix congestion window race with disconnect SUNRPC: Don't try to parse incomplete RPC messages SUNRPC: Rename xdr_buf_read_netobj to xdr_buf_read_mic SUNRPC: Fix buffer handling of GSS MIC without slack SUNRPC: RPC level errors should always set task->tk_rpc_status SUNRPC: Don't receive TCP data into a request buffer that has been reset ...
| * | SUNRPC: Fix congestion window race with disconnectChuck Lever2019-09-201-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the congestion window closes just as the transport disconnects, a reconnect is never driven because: 1. The XPRT_CONG_WAIT flag prevents tasks from taking the write lock 2. There's no wake-up of the first task on the xprt->sending queue To address this, clear the congestion wait flag as part of completing a disconnect. Fixes: 75891f502f5f ("SUNRPC: Support for congestion control ... ") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | SUNRPC: Don't try to parse incomplete RPC messagesTrond Myklebust2019-09-201-1/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the copy of the RPC reply into our buffers did not complete, and we could end up with a truncated message. In that case, just resend the call. Fixes: a0584ee9aed80 ("SUNRPC: Use struct xdr_stream when decoding...") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | SUNRPC: Rename xdr_buf_read_netobj to xdr_buf_read_micBenjamin Coddington2019-09-202-23/+31
| | | | | | | | | | | | | | | | | | | | | | | | Let the name reflect the single use. The function now assumes the GSS MIC is the last object in the buffer. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | SUNRPC: Fix buffer handling of GSS MIC without slackBenjamin Coddington2019-09-201-9/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The GSS Message Integrity Check data for krb5i may lie partially in the XDR reply buffer's pages and tail. If so, we try to copy the entire MIC into free space in the tail. But as the estimations of the slack space required for authentication and verification have improved there may be less free space in the tail to complete this copy -- see commit 2c94b8eca1a2 ("SUNRPC: Use au_rslack when computing reply buffer size"). In fact, there may only be room in the tail for a single copy of the MIC, and not part of the MIC and then another complete copy. The real world failure reported is that `ls` of a directory on NFS may sometimes return -EIO, which can be traced back to xdr_buf_read_netobj() failing to find available free space in the tail to copy the MIC. Fix this by checking for the case of the MIC crossing the boundaries of head, pages, and tail. If so, shift the buffer until the MIC is contained completely within the pages or tail. This allows the remainder of the function to create a sub buffer that directly address the complete MIC. Signed-off-by: Benjamin Coddington <bcodding@redhat.com> Cc: stable@vger.kernel.org # v5.1 Reviewed-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | SUNRPC: RPC level errors should always set task->tk_rpc_statusTrond Myklebust2019-09-172-4/+7
| | | | | | | | | | | | | | | | | | | | | | | | Ensure that we set task->tk_rpc_status for all RPC level errors so that the caller can distinguish between those and server reply status errors. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | SUNRPC: Don't receive TCP data into a request buffer that has been resetTrond Myklebust2019-09-171-1/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If we've removed the request from the receive list, and have added it back after resetting the request receive buffer, then we should only receive message data if it is a new reply (i.e. if transport->recv.copied is zero). Fixes: 277e4ab7d530b ("SUNRPC: Simplify TCP receive code by switching...") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | SUNRPC: Dequeue the request from the receive queue while we're re-encodingTrond Myklebust2019-09-172-26/+34
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Ensure that we dequeue the request from the transport receive queue while we're re-encoding to prevent issues like use-after-free when we release the bvec. Fixes: 7536908982047 ("SUNRPC: Ensure the bvecs are reset when we re-encode...") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Cc: stable@vger.kernel.org # v4.20+ Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Send Queue size grows after a reconnectChuck Lever2019-08-261-12/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Eli Dorfman reports that after a series of idle disconnects, an RPC/RDMA transport becomes unusable (rdma_create_qp returns -ENOMEM). Problem was tracked down to increasing Send Queue size after each reconnect. The rdma_create_qp() API does not promise to leave its @qp_init_attr parameter unaltered. In fact, some drivers do modify one or more of its fields. Thus our calls to rdma_create_qp must use a fresh copy of ib_qp_init_attr each time. This fix is appropriate for kernels dating back to late 2007, though it will have to be adapted, as the connect code has changed over the years. Reported-by: Eli Dorfman <eli@vastdata.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Clear xprt->reestablish_timeout on closeChuck Lever2019-08-263-4/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | Ensure that the re-establishment delay does not grow exponentially on each good reconnect. This probably should have been part of commit 675dd90ad093 ("xprtrdma: Modernize ops->connect"). Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Recycle MRs after disconnectChuck Lever2019-08-263-9/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The optimization done in "xprtrdma: Simplify rpcrdma_mr_pop" was a bit too optimistic. MRs left over after a reconnect still need to be recycled, not added back to the free list, since they could be in flight or actually fully registered. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Optimize rpcrdma_post_recvs()Chuck Lever2019-08-211-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | Micro-optimization: In rpcrdma_post_recvs, since commit e340c2d6ef2a ("xprtrdma: Reduce the doorbell rate (Receive)"), the common case is to return without doing anything. Found with perf. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Inline XDR chunk encoder functionsChuck Lever2019-08-211-9/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Micro-optimization: Save the cost of three function calls during transport header encoding. These were "noinline" before to generate more meaningful call stacks during debugging, but this code is now pretty stable. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Fix bc_max_slots return valueChuck Lever2019-08-211-3/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For the moment the returned value just happens to be correct because the current backchannel server implementation does not vary the number of credits it offers. The spec does permit this value to change during the lifetime of a connection, however. The actual maximum is fixed for all RPC/RDMA transports, because each transport instance has to pre-allocate the resources for processing BC requests. That's the value that should be returned. Fixes: 7402a4fedc2b ("SUNRPC: Fix up backchannel slot table ... ") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Clean up xprt_rdma_set_connect_timeout()Chuck Lever2019-08-211-4/+4
| | | | | | | | | | | | | | | | | | | | | Clean up: The function name should match the documenting comment. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Use an llist to manage free rpcrdma_repsChuck Lever2019-08-212-59/+53
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | rpcrdma_rep objects are removed from their free list by only a single thread: the Receive completion handler. Thus that free list can be converted to an llist, where a single-threaded consumer and a multi-threaded producer (rpcrdma_buffer_put) can both access the llist without the need for any serialization. This eliminates spin lock contention between the Receive completion handler and rpcrdma_buffer_get, and makes the rep consumer wait- free. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Remove rpcrdma_buffer::rb_mrlockChuck Lever2019-08-213-18/+16
| | | | | | | | | | | | | | | | | | | | | | | | Clean up: Now that the free list is used sparingly, get rid of the separate spin lock protecting it. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Cache free MRs in each rpcrdma_reqChuck Lever2019-08-214-12/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of a globally-contended MR free list, cache MRs in each rpcrdma_req as they are released. This means acquiring and releasing an MR will be lock-free in the common case, even outside the transport send lock. The original idea of per-rpcrdma_req MR free lists was suggested by Shirley Ma <shirley.ma@oracle.com> several years ago. I just now figured out how to make that idea work with on-demand MR allocation. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Ensure creating an MR does not trigger FS writebackChuck Lever2019-08-202-4/+5
| | | | | | | | | | | | | | | | | | | | | Probably would be good to also pass GFP flags to ib_alloc_mr. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Move rpcrdma_mr_get out of frwr_mapChuck Lever2019-08-204-43/+35
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Refactor: Retrieve an MR and handle error recovery entirely in rpc_rdma.c, as this is not a device-specific function. Note that since commit 89f90fe1ad8b ("SUNRPC: Allow calls to xprt_transmit() to drain the entire transmit queue"), the xprt_transmit function handles the cond_resched. The transport no longer has to do this itself. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Combine rpcrdma_mr_put and rpcrdma_mr_unmap_and_putChuck Lever2019-08-203-28/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Clean up. There is only one remaining rpcrdma_mr_put call site, and it can be directly replaced with unmap_and_put because mr->mr_dir is set to DMA_NONE just before the call. Now all the call sites do a DMA unmap, and we can just rename mr_unmap_and_put to mr_put, which nicely matches mr_get. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Simplify rpcrdma_mr_popChuck Lever2019-08-204-21/+11
| | | | | | | | | | | | | | | | | | | | | | | | Clean up: rpcrdma_mr_pop call sites check if the list is empty first. Let's replace the list_empty with less costly logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Toggle XPRT_CONGESTED in xprtrdma's slot methodsChuck Lever2019-08-201-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 48be539dd44a ("xprtrdma: Introduce ->alloc_slot call-out for xprtrdma") added a separate alloc_slot and free_slot to the RPC/RDMA transport. Later, commit 75891f502f5f ("SUNRPC: Support for congestion control when queuing is enabled") modified the generic alloc/free_slot methods, but neglected the methods in xprtrdma. Found via code review. Fixes: 75891f502f5f ("SUNRPC: Support for congestion control ... ") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Rename rpcrdma_buffer::rb_allChuck Lever2019-08-202-19/+9
| | | | | | | | | | | | | | | | | | | | | | | | Clean up: There are other "all" list heads. For code clarity distinguish this one as for use only for MRs by renaming it. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Rename CQE field in Receive trace pointsChuck Lever2019-08-201-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | Make the field name the same for all trace points that handle pointers to struct rpcrdma_rep. That makes it easy to grep for matching rep points in trace output. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Boost maximum transport header sizeChuck Lever2019-08-202-14/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Although I haven't seen any performance results that justify it, I've received several complaints that NFS/RDMA no longer supports a maximum rsize and wsize of 1MB. These days it is somewhat smaller. To simplify the logic that determines whether a chunk list is necessary, the implementation uses a fixed maximum size of the transport header. Currently that maximum size is 256 bytes, one quarter of the default inline threshold size for RPC/RDMA v1. Since commit a78868497c2e ("xprtrdma: Reduce max_frwr_depth"), the size of chunks is also smaller to take advantage of inline page lists in device internal MR data structures. The combination of these two design choices has reduced the maximum NFS rsize and wsize that can be used for most RNIC/HCAs. Increasing the maximum transport header size and the maximum number of RDMA segments it can contain increases the negotiated maximum rsize/wsize on common RNIC/HCAs. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Fix calculation of ri_max_segs againChuck Lever2019-08-201-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 302d3deb206 ("xprtrdma: Prevent inline overflow") added this calculation back in 2016, but got it wrong. I tested only the lower bound, which is why there is a max_t there. The upper bound should be rounded up too. Now, when using DIV_ROUND_UP, that takes care of the lower bound as well. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Update obsolete commentChuck Lever2019-08-201-3/+0
| | | | | | | | | | | | | | | | | | | | | | | | Comment was made obsolete by commit 8cec3dba76a4 ("xprtrdma: rpcrdma_regbuf alignment"). Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | xprtrdma: Refresh the documenting comment in frwr_ops.cChuck Lever2019-08-201-48/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Things have changed since this comment was written. In particular, the reworking of connection closing, on-demand creation of MRs, and the removal of fr_state all mean that deferring MR recovery to frwr_map is no longer needed. The description is obsolete. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
| * | SUNRPC: Inline xdr_commit_encodeChuck Lever2019-08-201-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Micro-optimization: For xdr_commit_encode call sites in net/sunrpc/xdr.c, eliminate the extra calling sequence. On my client, this change saves about a microsecond for every 30 calls to xdr_reserve_space(). Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-16 10:36:05 +0000