- 17 Nov, 2017 35 commits
-
-
Trond Myklebust authored
Signed-off-by:
Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by:
Anna Schumaker <Anna.Schumaker@Netapp.com>
-
Trond Myklebust authored
Ben Coddington has noted the following race between OPEN and CLOSE on a single client. Process 1 Process 2 Server ========= ========= ====== 1) OPEN file 2) OPEN file 3) Process OPEN (1) seqid=1 4) Process OPEN (2) seqid=2 5) Reply OPEN (2) 6) Receive reply (2) 7) new stateid, seqid=2 8) CLOSE file, using stateid w/ seqid=2 9) Reply OPEN (1) 10( Process CLOSE (8) 11) Reply CLOSE (8) 12) Forget stateid file closed 13) Receive reply (7) 14) Forget stateid file closed. 15) Receive reply (1). 16) New stateid seqid=1 is really the same stateid that was closed. IOW: the reply to the first OPEN is delayed. Since "Process 2" does not wait before closing the file, and it does not cache the closed stateid, then when the delayed reply is finally received, it is treated as setting up a new stateid by the client. The fix is to ensure that the client processes the OPEN and CLOSE calls in the same order in which the server processed them. This commit ensures that we examine the seqid of the stateid returned by OPEN. If it is a new stateid, we assume the seqid must be equal to the value 1, and that each state transition increments the seqid value by 1 (See RFC7530, Section 9.1.4.2, and RFC5661, Section 8.2.2). If the tracker sees that an OPEN returns with a seqid that is greater than the cached seqid + 1, then it bumps a flag to ensure that the caller waits for the RPCs carrying the missing seqids to complete. Note that there can still be pathologies where the server crashes before it can even send us the missing seqids. Since the OPEN call is still holding a slot when it waits here, that could cause the recovery to stall forever. To avoid that, we time out after a 5 second wait. Reported-by:Benjamin Coddington <bcodding@redhat.com> Signed-off-by:
-
Chuck Lever authored
Display information about the RPC procedure being requested in the trace log. This sometimes critical information cannot always be derived from other RPC trace entries. Signed-off-by:
Chuck Lever <chuck.lever@oracle.com> Signed-off-by:
-
Chuck Lever authored
The rpc_task_begin trace point always display a task ID of zero. Move the trace point call site so that it picks up the new task ID. Signed-off-by:
-
Chuck Lever authored
mount.nf-11159 8.... 905.248380: xprt_transmit: [FAILED TO PARSE] xid=351291440 status=0 addr=192.168.2.5 port=20049 mount.nf-11159 8.... 905.248381: rpc_task_sleep: task:6210@1 flags=0e80 state=0005 status=0 timeout=60000 queue=xprt_pending kworker/-1591 1.... 905.248419: xprt_lookup_rqst: [FAILED TO PARSE] xid=351291440 status=0 addr=192.168.2.5 port=20049 kworker/-1591 1.... 905.248423: xprt_complete_rqst: [FAILED TO PARSE] xid=351291440 status=24 addr=192.168.2.5 port=20049 Byte swapping is not available during trace-cmd report. Signed-off-by:
-
Gustavo A. R. Silva authored
In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Signed-off-by:
Gustavo A. R. Silva <garsilva@embeddedor.com> Signed-off-by:
-
Thomas Meyer authored
Bool initializations should use true and false. Bool tests don't need comparisons. Signed-off-by:
Thomas Meyer <thomas@m3y3r.de> Signed-off-by:
-
Anna Schumaker authored
There isn't an obvious way to acquire and release the RCU lock during a tracepoint, so we can't use the rpc_peeraddr2str() function here. Instead, rely on the client's cl_hostname, which should have similar enough information without needing an rcu_dereference(). Reported-by:
Dave Jones <davej@codemonkey.org.uk> Cc: stable@vger.kernel.org # v3.12 Signed-off-by:
-
Chuck Lever authored
Credit work contributed by Oracle engineers since 2014. Signed-off-by:
-
Chuck Lever authored
Clean up. This include should have been removed by commit 23826c7a ("xprtrdma: Serialize credit accounting again"). Signed-off-by:
Devesh Sharma <devesh.sharma@broadcom.com> Signed-off-by:
-
Chuck Lever authored
Clean up: C-structure style XDR encoding and decoding logic has been replaced over the past several merge windows on both the client and server. These data structures are no longer used. Signed-off-by:
-
Chuck Lever authored
Lift the Send and LocalInv completion handlers out of soft IRQ mode to make room for other work. Also, move the Send CQ to a different CPU than the CPU where the Receive CQ is running, for improved scalability. Signed-off-by:
-
Elena Reshetova authored
atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable nfs_client.cl_count is used as pure reference counter. Convert it to refcount_t and fix up the operations. Suggested-by:
Kees Cook <keescook@chromium.org> Reviewed-by:
David Windsor <dwindsor@gmail.com> Reviewed-by:
Hans Liljestrand <ishkamiel@gmail.com> Signed-off-by:
Elena Reshetova <elena.reshetova@intel.com> Signed-off-by:
-
Elena Reshetova authored
atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable nfs_lock_context.count is used as pure reference counter. Convert it to refcount_t and fix up the operations. Suggested-by:
-
Elena Reshetova authored
atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable nfs4_lock_state.ls_count is used as pure reference counter. Convert it to refcount_t and fix up the operations. Suggested-by:
-
Elena Reshetova authored
atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable nfs_cache_defer_req.count is used as pure reference counter. Convert it to refcount_t and fix up the operations. Suggested-by:
-
Elena Reshetova authored
atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable nfs4_ff_layout_mirror.ref is used as pure reference counter. Convert it to refcount_t and fix up the operations. Suggested-by:
-
Elena Reshetova authored
atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable pnfs_layout_hdr.plh_refcount is used as pure reference counter. Convert it to refcount_t and fix up the operations. Suggested-by:
-
Elena Reshetova authored
refcount_t type and corresponding API should be used instead of atomic_t when the variable is used as a reference counter. This allows to avoid accidental refcounter overflows that might lead to use-after-free situations. Signed-off-by:
-
Elena Reshetova authored
atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable nfs4_pnfs_ds.ds_count is used as pure reference counter. Convert it to refcount_t and fix up the operations. Suggested-by:
-
Trond Myklebust authored
If the previous request on a slot was interrupted before it was processed by the server, then our slot sequence number may be out of whack, and so we try the next operation using the old sequence number. The problem with this, is that not all servers check to see that the client is replaying the same operations as previously when they decide to go to the replay cache, and so instead of the expected error of NFS4ERR_SEQ_FALSE_RETRY, we get a replay of the old reply, which could (if the operations match up) be mistaken by the client for a new reply. To fix this, we attempt to send a COMPOUND containing only the SEQUENCE op in order to resync our slot sequence number. Cc: Olga Kornievskaia <olga.kornievskaia@gmail.com> [olga.kornievskaia@gmail.com: fix an Oops] Signed-off-by:
-
Chuck Lever authored
The sendctx circular queue now guarantees that xprtrdma cannot overflow the Send Queue, so remove the remaining bits of the original Send WQE counting mechanism. Signed-off-by:
-
Chuck Lever authored
When an RPC Call includes a file data payload, that payload can come from pages in the page cache, or a user buffer (for direct I/O). If the payload can fit inline, xprtrdma includes it in the Send using a scatter-gather technique. xprtrdma mustn't allow the RPC consumer to re-use the memory where that payload resides before the Send completes. Otherwise, the new contents of that memory would be exposed by an HCA retransmit of the Send operation. So, block RPC completion on Send completion, but only in the case where a separate file data payload is part of the Send. This prevents the reuse of that memory while it is still part of a Send operation without an undue cost to other cases. Waiting is avoided in the common case because typically the Send will have completed long before the RPC Reply arrives. These days, an RPC timeout will trigger a disconnect, which tears down the QP. The disconnect flushes all waiting Sends. This bounds the amount of time the reply handler has to wait for a Send completion. Signed-off-by:
-
Chuck Lever authored
Invoke a common routine for releasing hardware resources (for example, invalidating MRs). This needs to be done whether an RPC Reply has arrived or the RPC was terminated early. Signed-off-by:
-
Chuck Lever authored
We have one boolean flag in rpcrdma_req today. I'd like to add more flags, so convert that boolean to a bit flag. Signed-off-by:
-
Chuck Lever authored
Problem statement: Recently Sagi Grimberg <sagi@grimberg.me> observed that kernel RDMA- enabled storage initiators don't handle delayed Send completion correctly. If Send completion is delayed beyond the end of a ULP transaction, the ULP may release resources that are still being used by the HCA to complete a long-running Send operation. This is a common design trait amongst our initiators. Most Send operations are faster than the ULP transaction they are part of. Waiting for a completion for these is typically unnecessary. Infrequently, a network partition or some other problem crops up where an ordering problem can occur. In NFS parlance, the RPC Reply arrives and completes the RPC, but the HCA is still retrying the Send WR that conveyed the RPC Call. In this case, the HCA can try to use memory that has been invalidated or DMA unmapped, and the connection is lost. If that memory has been re-used for something else (possibly not related to NFS), and the Send retransmission exposes that data on the wire. Thus we cannot assume that it is safe to release Send-related resources just because a ULP reply has arrived. After some analysis, we have determined that the completion housekeeping will not be difficult for xprtrdma: - Inline Send buffers are registered via the local DMA key, and are already left DMA mapped for the lifetime of a transport connection, thus no additional handling is necessary for those - Gathered Sends involving page cache pages _will_ need to DMA unmap those pages after the Send completes. But like inline send buffers, they are registered via the local DMA key, and thus will not need to be invalidated In addition, RPC completion will need to wait for Send completion in the latter case. However, nearly always, the Send that conveys the RPC Call will have completed long before the RPC Reply arrives, and thus no additional latency will be accrued. Design notes: In this patch, the rpcrdma_sendctx object is introduced, and a lock-free circular queue is added to manage a set of them per transport. The RPC client's send path already prevents sending more than one RPC Call at the same time. This allows us to treat the consumer side of the queue (rpcrdma_sendctx_get_locked) as if there is a single consumer thread. The producer side of the queue (rpcrdma_sendctx_put_locked) is invoked only from the Send completion handler, which is a single thread of execution (soft IRQ). The only care that needs to be taken is with the tail index, which is shared between the producer and consumer. Only the producer updates the tail index. The consumer compares the head with the tail to ensure that the a sendctx that is in use is never handed out again (or, expressed more conventionally, the queue is empty). When the sendctx queue empties completely, there are enough Sends outstanding that posting more Send operations can result in a Send Queue overflow. In this case, the ULP is told to wait and try again. This introduces strong Send Queue accounting to xprtrdma. As a final touch, Jason Gunthorpe <jgunthorpe@obsidianresearch.com> suggested a mechanism that does not require signaling every Send. We signal once every N Sends, and perform SGE unmapping of N Send operations during that one completion. Reported-by:
Sagi Grimberg <sagi@grimberg.me> Suggested-by:
Jason Gunthorpe <jgunthorpe@obsidianresearch.com> Signed-off-by:
-
Chuck Lever authored
Commit 655fec69 ("xprtrdma: Use gathered Send for large inline messages") assumed that, since the zeroeth element of the Send SGE array always pointed to req->rl_rdmabuf, it needed to be initialized just once. This was a valid assumption because the Send SGE array and rl_rdmabuf both live in the same rpcrdma_req. In a subsequent patch, the Send SGE array will be separated from the rpcrdma_req, so the zeroeth element of the SGE array needs to be initialized every time. Signed-off-by:
-
Chuck Lever authored
Clean up: Make rpcrdma_prepare_send_sges() return a negative errno instead of a bool. Soon callers will want distinct treatments of different types of failures. Signed-off-by:
-
Chuck Lever authored
When this function fails, it needs to undo the DMA mappings it's done so far. Otherwise these are leaked. Signed-off-by:
-
Chuck Lever authored
Clean up. rpcrdma_prepare_hdr_sge() sets num_sge to one, then rpcrdma_prepare_msg_sges() sets num_sge again to the count of SGEs it added, plus one for the header SGE just mapped in rpcrdma_prepare_hdr_sge(). This is confusing, and nails in an assumption about when these functions are called. Instead, maintain a running count that both functions can update with just the number of SGEs they have added to the SGE array. Signed-off-by:
-
Chuck Lever authored
We need to decode and save the incoming rdma_credits field _after_ we know that the direction of the message is "forward direction Reply". Otherwise, the credits value in reverse direction Calls is also used to update the forward direction credits. It is safe to decode the rdma_credits field in rpcrdma_reply_handler now that rpcrdma_reply_handler is single-threaded. Receives complete in the same order as they were sent on the NFS server. Signed-off-by:
-
Chuck Lever authored
I noticed that the soft IRQ thread looked pretty busy under heavy I/O workloads. perf suggested one area that was expensive was the queue_work() call in rpcrdma_wc_receive. That gave me some ideas. Instead of scheduling a separate worker to process RPC Replies, promote the Receive completion handler to IB_POLL_WORKQUEUE, and invoke rpcrdma_reply_handler directly. Note that the poll workqueue is single-threaded. In order to keep memory invalidation from serializing all RPC Replies, handle any necessary invalidation tasks in a separate multi-threaded workqueue. This provides a two-tier scheme, similar to OS I/O interrupt handlers: A fast interrupt handler that schedules the slow handler and re-enables the interrupt, and a slower handler that is invoked for any needed heavy lifting. Benefits include: - One less context switch for RPCs that don't register memory - Receive completion handling is moved out of soft IRQ context to make room for other users of soft IRQ - The same CPU core now DMA syncs and XDR decodes the Receive buffer Signed-off-by:
-
Chuck Lever authored
Clean up: I'd like to be able to invoke the tail of rpcrdma_reply_handler in two different places. Split the tail out into its own helper function. Signed-off-by:
-
Chuck Lever authored
Clean up: Make it easier to pass the decoded XID, vers, credits, and proc fields around by moving these variables into struct rpcrdma_rep. Note: the credits field will be handled in a subsequent patch. Signed-off-by:
-
Chuck Lever authored
A reply with an unrecognized value in the version field means the transport header is potentially garbled and therefore all the fields are untrustworthy. Fixes: 59aa1f9a ("xprtrdma: Properly handle RDMA_ERROR ... ") Signed-off-by:
-
- 16 Oct, 2017 5 commits
-
-
Chuck Lever authored
Clean up: There are no remaining callers of this method. Signed-off-by:
-
Chuck Lever authored
The "safe" version of ro_unmap is used here to avoid waiting unnecessarily. However: - It is safe to wait. After all, we have to wait anyway when using FMR to register memory. - This case is rare: it occurs only after a reconnect. By switching this call site to ro_unmap_sync, the final use of ro_unmap_safe is removed. Signed-off-by:
-
Chuck Lever authored
In current kernels, waiting in xprt_release appears to be safe to do. I had erroneously believed that for ASYNC RPCs, waiting of any kind in xprt_release->xprt_rdma_free would result in deadlock. I've done injection testing and consulted with Trond to confirm that waiting in the RPC release path is safe. For the very few times where RPC resources haven't yet been released earlier by the reply handler, it is safe to wait synchronously in xprt_rdma_free for invalidation rather than defering it to MR recovery. Note: When the QP is error state, posting a LocalInvalidate should flush and mark the MR as bad. There is no way the remote HCA can access that MR via a QP in error state, so it is effectively already inaccessible and thus safe for the Upper Layer to access. The next time the MR is used it should be recognized and cleaned up properly by frwr_op_map. Signed-off-by:
-
NeilBrown authored
Commit f5a73672 ("NFS: allow close-to-open cache semantics to apply to root of NFS filesystem") added a call to __nfs_revalidate_inode() to nfs_opendir to as the lookup process wouldn't reliable do this. Subsequent commit a3fbbde7 ("VFS: we need to set LOOKUP_JUMPED on mountpoint crossing") make this unnecessary. So remove the unnecessary code. Signed-off-by:
NeilBrown <neilb@suse.com> Signed-off-by:
-
NeilBrown authored
For correct close-to-open semantics, NFS must validate the change attribute of a directory (or file) on open. Since commit ecf3d1f1 ("vfs: kill FS_REVAL_DOT by adding a d_weak_revalidate dentry op"), open() of "." or a path ending ".." is not revalidated reliably (except when that direct is a mount point). Prior to that commit, "." was revalidated using nfs_lookup_revalidate() which checks the LOOKUP_OPEN flag and forces revalidation if the flag is set. Since that commit, nfs_weak_revalidate() is used for NFSv3 (which ignores the flags) and nothing is used for NFSv4. This is fixed by using nfs_lookup_verify_inode() in nfs_weak_revalidate(). This does the revalidation exactly when needed. Also, add a definition of .d_weak_revalidate for NFSv4. The incorrect behavior is easily demonstrated by running "echo *" in some non-mountpoint NFS directory while watching network traffic. Without this patch, "echo *" sometimes doesn't produce any traffic. With the patch it always does. Fixes: ecf3d1f1 ("vfs: kill FS_REVAL_DOT by adding a d_weak_revalidate dentry op") cc: stable@vger.kernel.org (3.9+) Signed-off-by:
-
