summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2020-04-03NFS/pnfs: Reference the layout cred in pnfs_prepare_layoutreturn()Trond Myklebust
When we're sending a layoutreturn, ensure that we reference the layout cred atomically with the copy of the stateid. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-03NFS/pnfs: Fix dereference of layout cred in pnfs_layoutcommit_inode()Trond Myklebust
Ensure that the dereference of the layout cred is atomic with the stateid. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-03NFS: Beware when dereferencing the delegation credTrond Myklebust
When we look up the delegation cred, we are usually doing so in conjunction with a read of the stateid, and we want to ensure that the look up is atomic with that read. Fixes: 57f188e04773 ("NFSv4: nfs_update_inplace_delegation() should update delegation cred") [sfr@canb.auug.org.au: Fixed up borken Fixes: line from Trond :-)] Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-02NFS: Add a module parameter to set nfs_mountpoint_expiry_timeoutTrond Myklebust
Setting nfs_mountpoint_expiry_timeout() to a negative value stops mountpoint expiration, while setting it to a positive value restarts the scheduler. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-02NFS: finish_automount() requires us to hold 2 refs to the mount recordTrond Myklebust
We must not return from nfs_d_automount() without holding 2 references to the mount record. Doing so, will trigger the BUG() in finish_automount(). Also ensure that we don't try to reschedule the automount timer with a negative or zero timeout value. Fixes: 22a1ae9a93fb ("NFS: If nfs_mountpoint_expiry_timeout < 0, do not expire submounts") Cc: stable@vger.kernel.org # v5.5+ Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-02NFS: Fix a few constant_table array definitionsScott Mayhew
nfs_vers_tokens, nfs_xprt_protocol_tokens, and nfs_secflavor_tokens were all missing an empty item at the end of the array, allowing lookup_constant() to potentially walk off the end and trigger and oops. Reported-by: Olga Kornievskaia <aglo@umich.edu> Signed-off-by: Scott Mayhew <smayhew@redhat.com> Fixes: e38bb238ed8c ("NFS: Convert mount option parsing to use functionality from fs_parser.h") Cc: stable@vger.kernel.org # v5.6 Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Try to join page groups before an O_DIRECT retransmissionTrond Myklebust
If we have to retransmit requests, try to join their page groups first. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Refactor nfs_lock_and_join_requests()Trond Myklebust
Refactor nfs_lock_and_join_requests() in order to separate out the subrequest merging into its own function nfs_lock_and_join_group() that can be used by O_DIRECT. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Reverse the submission order of requests in __nfs_pageio_add_request()Trond Myklebust
If we have to split the request up into subrequests, we have to submit the request pointed to by the function call parameter last, in case there is an error or other issue that causes us to exit before the last request is submitted. The reason is that the caller is expected to perform cleanup in those cases. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Clean up nfs_lock_and_join_requests()Trond Myklebust
Clean up nfs_lock_and_join_requests() to simplify the calculation of the range covered by the page group, taking into account the presence of mirrors. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Remove the redundant function nfs_pgio_has_mirroring()Trond Myklebust
We need to trust that desc->pg_mirror_idx is set correctly, whether or not mirroring is enabled. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Fix memory leaks in nfs_pageio_stop_mirroring()Trond Myklebust
If we just set the mirror count to 1 without first clearing out the mirrors, we can leak queued up requests. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Fix a request reference leak in nfs_direct_write_clear_reqs()Trond Myklebust
nfs_direct_write_scan_commit_list() will lock the request and bump the reference count, but we also need to account for the reference that was taken when we initially added the request to the commit list. Fixes: fb5f7f20cdb9 ("NFS: commit errors should be fatal") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Fix use-after-free issues in nfs_pageio_add_request()Trond Myklebust
We need to ensure that we create the mirror requests before calling nfs_pageio_add_request_mirror() on the request we are adding. Otherwise, we can end up with a use-after-free if the call to nfs_pageio_add_request_mirror() triggers I/O. Fixes: c917cfaf9bbe ("NFS: Fix up NFS I/O subrequest creation") Cc: stable@vger.kernel.org Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Fix races nfs_page_group_destroy() vs nfs_destroy_unlinked_subrequests()Trond Myklebust
When a subrequest is being detached from the subgroup, we want to ensure that it is not holding the group lock, or in the process of waiting for the group lock. Fixes: 5b2b5187fa85 ("NFS: Fix nfs_page_group_destroy() and nfs_lock_and_join_requests() race cases") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-04-01NFS: Fix a page leak in nfs_destroy_unlinked_subrequests()Trond Myklebust
When we detach a subrequest from the list, we must also release the reference it holds to the parent. Fixes: 5b2b5187fa85 ("NFS: Fix nfs_page_group_destroy() and nfs_lock_and_join_requests() race cases") Cc: stable@vger.kernel.org # v4.14+ Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-28Merge tag 'nfs-rdma-for-5.7-1' of ↵Trond Myklebust
git://git.linux-nfs.org/projects/anna/linux-nfs NFSoRDMA Client Updates for Linux 5.7 New Features: - Allow one active connection and several zombie connections to prevent blocking if the remote server is unresponsive. Bugfixes and Cleanups: - Enhance MR-related trace points - Refactor connection set-up and disconnect functions - Make Protection Domains per-connection instead of per-transport - Merge struct rpcrdma_ia into rpcrdma_ep
2020-03-28NFS: Remove unused FLUSH_SYNC support in nfs_initiate_pgio()Trond Myklebust
If the FLUSH_SYNC flag is set, nfs_initiate_pgio() will currently wait for completion, and then return the status of the I/O operation. What we actually want to report in nfs_pageio_doio() is whether or not the RPC call was launched successfully, whereas actual I/O status is intended handled in the reply callbacks. Since FLUSH_SYNC is never set by any of the callers anyway, let's just remove that code altogether. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27pNFS/flexfiles: Specify the layout segment range in LAYOUTGETTrond Myklebust
Move from requesting only full file layout segments, to requesting layout segments that match our I/O size. This means the server is still free to return a full file layout, but we will no longer error out if it does not. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27pNFS/flexfiles: remove requirement for whole file layoutsTrond Myklebust
Remove the requirement that the server always sends whole file layouts. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27pNFS/flexfiles: Check the layout segment range before doing I/OTrond Myklebust
When starting to read or write with a layout segment, check that the range matches our request. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27pNFS/flexfile: Don't merge layout segments if the mirrors don't matchTrond Myklebust
Check that the number of mirrors, and the mirror information matches before deciding to merge layout segments in pNFS/flexfiles. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS/pNFS: Fix pnfs_layout_mark_request_commit() invalid layout segment handlingTrond Myklebust
Fix up pnfs_layout_mark_request_commit() to alway reschedule the write if the layout segment is invalid. Also minor cleanup. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS/pNFS: Simplify bucket layout segment reference countingTrond Myklebust
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS/pNFS: Clean up pNFS commit operationsTrond Myklebust
Move the pNFS commit related operations into a separate structure that can be carried by the pnfs_ds_commit_info. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS: Remove bucket array from struct pnfs_ds_commit_infoTrond Myklebust
Remove the unused bucket array in struct pnfs_ds_commit_info. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS/pNFS: Add a helper pnfs_generic_search_commit_reqs()Trond Myklebust
Lift filelayout_search_commit_reqs() into the generic pnfs/nfs code, and add support for commit arrays. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27pNFS: Enable per-layout segment commit structuresTrond Myklebust
Enable adding and lookup of per-layout segment commits in filelayout and flexfilelayout. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27pNFS: Add infrastructure for cleaning up per-layout commit structuresTrond Myklebust
Ensure that both the file and flexfiles layout types clean up when freeing the layout segments. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS/pNFS: Support commit arrays in nfs_clear_pnfs_ds_commit_verifiers()Trond Myklebust
Add support for scanning the full list of per-layout segment commit arrays to nfs_clear_pnfs_ds_commit_verifiers(). Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS: Fix O_DIRECT commit verifier handlingTrond Myklebust
Instead of trying to save the commit verifiers and checking them against previous writes, adopt the same strategy as for buffered writes, of just checking the verifiers at commit time. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS: commit errors should be fatalTrond Myklebust
Fix the O_DIRECT code to avoid retries if the COMMIT fails with a fatal error. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFS/pNFS: Allow O_DIRECT to release the DS commitinfoTrond Myklebust
Add a pNFS callback to allow the O_DIRECT code to release the DS commitinfo when freeing the dreq. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27pNFS: Support per-layout segment commits in pnfs_generic_commit_pagelist()Trond Myklebust
Add support for scanning the full list of per-layout segment commit arrays to pnfs_generic_commit_pagelist(). Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27pNFS: Support per-layout segment commits in pnfs_generic_recover_commit_reqs()Trond Myklebust
Add support for scanning the full list of per-layout segment commit arrays to pnfs_generic_recover_commit_reqs(). Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFSv4/pNFS: Scan the full list of commit arrays when committingTrond Myklebust
Add support for scanning the full list of per-layout segment commit arrays to pnfs_generic_scan_commit_lists() Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27NFSv4/pnfs: Support a list of commit arrays in struct pnfs_ds_commit_infoTrond Myklebust
When we have multiple layout segments with different lists of mirrored data, we need to track the commits on a per layout segment basis. This patch adds a list to support this tracking in struct pnfs_ds_commit_info. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
2020-03-27xprtrdma: kmalloc rpcrdma_ep separate from rpcrdma_xprtChuck Lever
Change the rpcrdma_xprt_disconnect() function so that it no longer waits for the DISCONNECTED event. This prevents blocking if the remote is unresponsive. In rpcrdma_xprt_disconnect(), the transport's rpcrdma_ep is detached. Upon return from rpcrdma_xprt_disconnect(), the transport (r_xprt) is ready immediately for a new connection. The RDMA_CM_DEVICE_REMOVAL and RDMA_CM_DISCONNECTED events are now handled almost identically. However, because the lifetimes of rpcrdma_xprt structures and rpcrdma_ep structures are now independent, creating an rpcrdma_ep needs to take a module ref count. The ep now owns most of the hardware resources for a transport. Also, a kref is needed to ensure that rpcrdma_ep sticks around long enough for the cm_event_handler to finish. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Extract sockaddr from struct rdma_cm_idChuck Lever
rpcrdma_cm_event_handler() is always passed an @id pointer that is valid. However, in a subsequent patch, we won't be able to extract an r_xprt in every case. So instead of using the r_xprt's presentation address strings, extract them from struct rdma_cm_id. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Merge struct rpcrdma_ia into struct rpcrdma_epChuck Lever
I eventually want to allocate rpcrdma_ep separately from struct rpcrdma_xprt so that on occasion there can be more than one ep per xprt. The new struct rpcrdma_ep will contain all the fields currently in rpcrdma_ia and in rpcrdma_ep. This is all the device and CM settings for the connection, in addition to per-connection settings negotiated with the remote. Take this opportunity to rename the existing ep fields from rep_* to re_* to disambiguate these from struct rpcrdma_rep. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Disconnect on flushed completionChuck Lever
Completion errors after a disconnect often occur much sooner than a CM_DISCONNECT event. Use this to try to detect connection loss more quickly. Note that other kernel ULPs do take care to disconnect explicitly when a WR is flushed. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Remove rpcrdma_ia::ri_flagsChuck Lever
Clean up: The upper layer serializes calls to xprt_rdma_close, so there is no need for an atomic bit operation, saving 8 bytes in rpcrdma_ia. This enables merging rpcrdma_ia_remove directly into the disconnect logic. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Invoke rpcrdma_ia_open in the connect workerChuck Lever
Move rdma_cm_id creation into rpcrdma_ep_create() so that it is now responsible for allocating all per-connection hardware resources. With this clean-up, all three arms of the switch statement in rpcrdma_ep_connect are exactly the same now, thus the switch can be removed. Because device removal behaves a little differently than disconnection, there is a little more work to be done before rpcrdma_ep_destroy() can release the connection's rdma_cm_id. So it is not quite symmetrical with rpcrdma_ep_create() yet. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Allocate Protection Domain in rpcrdma_ep_create()Chuck Lever
Make a Protection Domain (PD) a per-connection resource rather than a per-transport resource. In other words, when the connection terminates, the PD is destroyed. Thus there is one less HW resource that remains allocated to a transport after a connection is closed. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Refactor rpcrdma_ep_connect() and rpcrdma_ep_disconnect()Chuck Lever
Clean up: Simplify the synopses of functions in the connect and disconnect paths in preparation for combining the rpcrdma_ia and struct rpcrdma_ep structures. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Clean up the post_send pathChuck Lever
Clean up: Simplify the synopses of functions in the post_send path by combining the struct rpcrdma_ia and struct rpcrdma_ep arguments. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Refactor frwr_init_mr()Chuck Lever
Clean up: prepare for combining the rpcrdma_ia and rpcrdma_ep structures. Take the opportunity to rename the function to be consistent with the "subsystem _ object _ verb" naming scheme. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Invoke rpcrdma_ep_create() in the connect workerChuck Lever
Refactor rpcrdma_ep_create(), rpcrdma_ep_disconnect(), and rpcrdma_ep_destroy(). rpcrdma_ep_create will be invoked at connect time instead of at transport set-up time. It will be responsible for allocating per- connection resources. In this patch it allocates the CQs and creates a QP. More to come. rpcrdma_ep_destroy() is the inverse functionality that is invoked at disconnect time. It will be responsible for releasing the CQs and QP. These changes should be safe to do because both connect and disconnect is guaranteed to be serialized by the transport send lock. This takes us another step closer to resolving the address and route only at connect time so that connection failover to another device will work correctly. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-27xprtrdma: Enhance MR-related trace pointsChuck Lever
Two changes: - Show the number of SG entries that were mapped. This helps debug DMA-related problems. - Record the MR's resource ID instead of its memory address. This groups each MR with its associated rdma-tool output, and reduces needless exposure of memory addresses. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2020-03-26pNFS: Add a helper to allocate the array of bucketsTrond Myklebust
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-03-15 13:20:23 +0000