summaryrefslogtreecommitdiffstats
path: root/fs/cifs/smbdirect.c (follow)
Commit message (Collapse)AuthorAgeFilesLines
* cifs: fix potential deadlock in direct reclaimVincent Whitchurch2022-06-011-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The srv_mutex is used during writeback so cifs should ensure that allocations done when that mutex is held are done with GFP_NOFS, to avoid having direct reclaim ending up waiting for the same mutex and causing a deadlock. This is detected by lockdep with the splat below: ====================================================== WARNING: possible circular locking dependency detected 5.18.0 #70 Not tainted ------------------------------------------------------ kswapd0/49 is trying to acquire lock: ffff8880195782e0 (&tcp_ses->srv_mutex){+.+.}-{3:3}, at: compound_send_recv but task is already holding lock: ffffffffa98e66c0 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (fs_reclaim){+.+.}-{0:0}: fs_reclaim_acquire kmem_cache_alloc_trace __request_module crypto_alg_mod_lookup crypto_alloc_tfm_node crypto_alloc_shash cifs_alloc_hash smb311_crypto_shash_allocate smb311_update_preauth_hash compound_send_recv cifs_send_recv SMB2_negotiate smb2_negotiate cifs_negotiate_protocol cifs_get_smb_ses cifs_mount cifs_smb3_do_mount smb3_get_tree vfs_get_tree path_mount __x64_sys_mount do_syscall_64 entry_SYSCALL_64_after_hwframe -> #0 (&tcp_ses->srv_mutex){+.+.}-{3:3}: __lock_acquire lock_acquire __mutex_lock mutex_lock_nested compound_send_recv cifs_send_recv SMB2_write smb2_sync_write cifs_write cifs_writepage_locked cifs_writepage shrink_page_list shrink_lruvec shrink_node balance_pgdat kswapd kthread ret_from_fork other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(fs_reclaim); lock(&tcp_ses->srv_mutex); lock(fs_reclaim); lock(&tcp_ses->srv_mutex); *** DEADLOCK *** 1 lock held by kswapd0/49: #0: ffffffffa98e66c0 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat stack backtrace: CPU: 2 PID: 49 Comm: kswapd0 Not tainted 5.18.0 #70 Call Trace: <TASK> dump_stack_lvl dump_stack print_circular_bug.cold check_noncircular __lock_acquire lock_acquire __mutex_lock mutex_lock_nested compound_send_recv cifs_send_recv SMB2_write smb2_sync_write cifs_write cifs_writepage_locked cifs_writepage shrink_page_list shrink_lruvec shrink_node balance_pgdat kswapd kthread ret_from_fork </TASK> Fix this by using the memalloc_nofs_save/restore APIs around the places where the srv_mutex is held. Do this in a wrapper function for the lock/unlock of the srv_mutex, and rename the srv_mutex to avoid missing call sites in the conversion. Note that there is another lockdep warning involving internal crypto locks, which was masked by this problem and is visible after this fix, see the discussion in this thread: https://lore.kernel.org/all/20220523123755.GA13668@axis.com/ Link: https://lore.kernel.org/r/CANT5p=rqcYfYMVHirqvdnnca4Mo+JQSw5Qu12v=kPfpk5yhhmg@mail.gmail.com/ Reported-by: Shyam Prasad N <nspmangalore@gmail.com> Suggested-by: Lars Persson <larper@axis.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Enzo Matsumiya <ematsumiya@suse.de> Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* Merge tag '5.19-rc-smb3-client-fixes-updated' of ↵Linus Torvalds2022-05-281-1/+1
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.samba.org/sfrench/cifs-2.6 Pull cifs client updates from Steve French: - multichannel fixes to improve reconnect after network failure - improved caching of root directory contents (extending benefit of directory leases) - two DFS fixes - three fixes for improved debugging - an NTLMSSP fix for mounts t0 older servers - new mount parm to allow disabling creating sparse files - various cleanup fixes and minor fixes pointed out by coverity * tag '5.19-rc-smb3-client-fixes-updated' of git://git.samba.org/sfrench/cifs-2.6: (24 commits) smb3: remove unneeded null check in cifs_readdir cifs: fix ntlmssp on old servers cifs: cache the dirents for entries in a cached directory cifs: avoid parallel session setups on same channel cifs: use new enum for ses_status cifs: do not use tcpStatus after negotiate completes smb3: add mount parm nosparse smb3: don't set rc when used and unneeded in query_info_compound smb3: check for null tcon cifs: fix minor compile warning Add various fsctl structs Add defines for various newer FSCTLs smb3: add trace point for oplock not found cifs: return the more nuanced writeback error on close() smb3: add trace point for lease not found issue cifs: smbd: fix typo in comment cifs: set the CREATE_NOT_FILE when opening the directory in use_cached_dir() cifs: check for smb1 in open_cached_dir() cifs: move definition of cifs_fattr earlier in cifsglob.h cifs: print TIDs as hex ...
| * cifs: smbd: fix typo in commentJulia Lawall2022-05-221-1/+1
| | | | | | | | | | | | | | | | Spelling mistake (triple letters) in comment. Detected with the help of Coccinelle. Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: Steve French <stfrench@microsoft.com>
* | RDMA: Split kernel-only global device caps from uverbs device capsJason Gunthorpe2022-04-061-1/+1
|/ | | | | | | | | | | | | | | | | | Split out flags from ib_device::device_cap_flags that are only used internally to the kernel into kernel_cap_flags that is not part of the uapi. This limits the device_cap_flags to being the same bitmap that will be copied to userspace. This cleanly splits out the uverbs flags from the kernel flags to avoid confusion in the flags bitmap. Add some short comments describing which each of the kernel flags is connected to. Remove unused kernel flags. Link: https://lore.kernel.org/r/0-v2-22c19e565eef+139a-kern_caps_jgg@nvidia.com Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
* smbdirect: missing rc checks while waiting for rdma eventsSteve French2021-06-221-2/+12
| | | | | | | | | | There were two places where we weren't checking for error (e.g. ERESTARTSYS) while waiting for rdma resolution. Addresses-Coverity: 1462165 ("Unchecked return value") Reviewed-by: Tom Talpey <tom@talpey.com> Reviewed-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: Fix fall-through warnings for ClangGustavo A. R. Silva2020-12-141-0/+1
| | | | | | | | | | In preparation to enable -Wimplicit-fallthrough for Clang, fix multiple warnings by explicitly adding multiple break/goto statements instead of just letting the code fall through to the next case. Link: https://github.com/KSPP/linux/issues/115 Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: Standardize logging outputJoe Perches2020-06-011-99/+66
| | | | | | | | | | | | | | | | | | | | | | | | | | | Use pr_fmt to standardize all logging for fs/cifs. Some logging output had no CIFS: specific prefix. Now all output has one of three prefixes: o CIFS: o CIFS: VFS: o Root-CIFS: Miscellanea: o Convert printks to pr_<level> o Neaten macro definitions o Remove embedded CIFS: prefixes from formats o Convert "illegal" to "invalid" o Coalesce formats o Add missing '\n' format terminations o Consolidate multiple cifs_dbg continuations into single calls o More consistent use of upper case first word output logging o Multiline statement argument alignment and wrapping Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Do not schedule work to send immediate packet on every receiveLong Li2020-04-071-51/+10
| | | | | | | | | | | Immediate packets should only be sent to peer when there are new receive credits made available. New credits show up on freeing receive buffer, not on receiving data. Fix this by avoid unnenecessary work schedules. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Properly process errors on ib_post_sendLong Li2020-04-071-123/+97
| | | | | | | | | | When processing errors from ib_post_send(), the transport state needs to be rolled back to the condition before the error. Refactor the old code to make it easy to roll back on IB errors, and fix this. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Update receive credits before sending and deal with credits roll ↵Long Li2020-04-071-7/+18
| | | | | | | | | | | | back on failure before sending Recevie credits should be updated before sending the packet, not before a work is scheduled. Also, the value needs roll back if something fails and cannot send. Signed-off-by: Long Li <longli@microsoft.com> Reported-by: kbuild test robot <lkp@intel.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Check send queue size before posting a sendLong Li2020-04-071-1/+10
| | | | | | | | | | | Sometimes the remote peer may return more send credits than the send queue depth. If all the send credits are used to post senasd, we may overflow the send queue. Fix this by checking the send queue size before posting a send. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Merge code to track pending packetsLong Li2020-04-071-30/+10
| | | | | | | | | | | As an optimization, SMBD tries to track two types of packets: packets with payload and without payload. There is no obvious benefit or performance gain to separately track two types of packets. Just treat them as pending packets and merge the tracking code. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Check and extend sender credits in interrupt contextLong Li2020-03-291-23/+15
| | | | | | | | | When a RDMA packet is received and server is extending send credits, we should check and unblock senders immediately in IRQ context. Doing it in a worker queue causes unnecessary delay and doesn't save much CPU on the receive path. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Calculate the correct maximum packet size for segmented ↵Long Li2020-03-291-2/+1
| | | | | | | | | | | | | SMBDirect send/receive The packet size needs to take account of SMB2 header size and possible encryption header size. This is only done when signing is used and it is for RDMA send/receive, not read/write. Also remove the dead SMBD code in smb2_negotiate_r(w)size. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Only queue work for error recovery on memory registrationLong Li2019-11-251-11/+15
| | | | | | | | | | | It's not necessary to queue invalidated memory registration to work queue, as all we need to do is to unmap the SG and make it usable again. This can save CPU cycles in normal data paths as memory registration errors are rare and normally only happens during reconnection. Signed-off-by: Long Li <longli@microsoft.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Return -ECONNABORTED when trasnport is not in connected stateLong Li2019-11-251-1/+1
| | | | | | | | The transport should return this error so the upper layer will reconnect. Signed-off-by: Long Li <longli@microsoft.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Add messages on RDMA session destroy and reconnectionLong Li2019-11-251-2/+4
| | | | | | | | Log these activities to help production support. Signed-off-by: Long Li <longli@microsoft.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Return -EINVAL when the number of iovs exceeds SMBDIRECT_MAX_SGELong Li2019-11-251-1/+1
| | | | | | | | | | While it's not friendly to fail user processes that issue more iovs than we support, at least we should return the correct error code so the user process gets a chance to retry with smaller number of iovs. Signed-off-by: Long Li <longli@microsoft.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
* rdma: Enable ib_alloc_cq to spread work over a device's comp_vectorsChuck Lever2019-08-051-4/+6
| | | | | | | | | | | | | | | | | | | | | | | | Send and Receive completion is handled on a single CPU selected at the time each Completion Queue is allocated. Typically this is when an initiator instantiates an RDMA transport, or when a target accepts an RDMA connection. Some ULPs cannot open a connection per CPU to spread completion workload across available CPUs and MSI vectors. For such ULPs, provide an API that allows the RDMA core to select a completion vector based on the device's complement of available comp_vecs. ULPs that invoke ib_alloc_cq() with only comp_vector 0 are converted to use the new API so that their completion workloads interfere less with each other. Suggested-by: Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Cc: <linux-cifs@vger.kernel.org> Cc: <v9fs-developer@lists.sourceforge.net> Link: https://lore.kernel.org/r/20190729171923.13428.52555.stgit@manet.1015granger.net Signed-off-by: Doug Ledford <dledford@redhat.com>
* treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 157Thomas Gleixner2019-05-301-10/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Based on 3 normalized pattern(s): this program is free software you can redistribute it and or modify it under the terms of the gnu general public license as published by the free software foundation either version 2 of the license or at your option any later version this program is distributed in the hope that it will be useful but without any warranty without even the implied warranty of merchantability or fitness for a particular purpose see the gnu general public license for more details this program is free software you can redistribute it and or modify it under the terms of the gnu general public license as published by the free software foundation either version 2 of the license or at your option any later version [author] [kishon] [vijay] [abraham] [i] [kishon]@[ti] [com] this program is distributed in the hope that it will be useful but without any warranty without even the implied warranty of merchantability or fitness for a particular purpose see the gnu general public license for more details this program is free software you can redistribute it and or modify it under the terms of the gnu general public license as published by the free software foundation either version 2 of the license or at your option any later version [author] [graeme] [gregory] [gg]@[slimlogic] [co] [uk] [author] [kishon] [vijay] [abraham] [i] [kishon]@[ti] [com] [based] [on] [twl6030]_[usb] [c] [author] [hema] [hk] [hemahk]@[ti] [com] this program is distributed in the hope that it will be useful but without any warranty without even the implied warranty of merchantability or fitness for a particular purpose see the gnu general public license for more details extracted by the scancode license scanner the SPDX license identifier GPL-2.0-or-later has been chosen to replace the boilerplate/reference in 1105 file(s). Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Allison Randal <allison@lohutok.net> Reviewed-by: Richard Fontana <rfontana@redhat.com> Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org> Cc: linux-spdx@vger.kernel.org Link: https://lkml.kernel.org/r/20190527070033.202006027@linutronix.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* cifs:smbd Use the correct DMA direction when sending dataLong Li2019-05-141-3/+5
| | | | | | | | | | When sending data, use the DMA_TO_DEVICE to map buffers. Also log the number of requests in a compounding request from upper layer. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com> Acked-by: Pavel Shilovsky <pshilov@microsoft.com> Acked-by: Ronnie Sahlberg <lsahlber@redhat.com>
* cifs: smbd: take an array of reqeusts when sending upper layer dataLong Li2019-05-081-27/+28
| | | | | | | | | | | To support compounding, __smb_send_rqst() now sends an array of requests to the transport layer. Change smbd_send() to take an array of requests, and send them in as few packets as possible. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com> CC: Stable <stable@vger.kernel.org>
* cifs: smbd: Indicate to retry on transport sending failureLong Li2019-05-081-2/+3
| | | | | | | | | | | | Failure to send a packet doesn't mean it's a permanent failure, it can't be returned to user process. This I/O should be retried or failed based on server packet response and transport health. This logic is handled by the upper layer. Give this decision to upper layer. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Return EINTR when interruptedLong Li2019-05-081-1/+1
| | | | | | | | When packets are waiting for outbound I/O and interrupted, return the proper error code to user process. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Don't destroy transport on RDMA disconnectLong Li2019-05-081-113/+7
| | | | | | | | Now upper layer is handling the transport shutdown and reconnect, remove the code that handling transport shutdown on RDMA disconnect. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* smbd: Make upper layer decide when to destroy the transportLong Li2019-05-081-20/+94
| | | | | | | | | | | | | | | | | | | | | | | On transport recoonect, upper layer CIFS code destroys the current transport and then recoonect. This code path is not used by SMBD, in that SMBD destroys its transport on RDMA disconnect notification independent of CIFS upper layer behavior. This approach adds some costs to SMBD layer to handle transport shutdown and restart, and to deal with several racing conditions on reconnecting transport. Re-work this code path by introducing a new smbd_destroy. This function is called form upper layer to ask SMBD to destroy the transport. SMBD will no longer need to destroy the transport by itself while worrying about data transfer is in progress. The upper layer guarantees the transport is locked. change log: v2: fix build errors when CONFIG_CIFS_SMB_DIRECT is not configured Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: replace snprintf with scnprintfRonnie Sahlberg2019-03-051-3/+3
| | | | | | | | | | | | | | | | | | | a trivial patch that replaces all use of snprintf with scnprintf. scnprintf() is generally seen as a safer function to use than snprintf for many use cases. In our case, there is no actual difference between the two since we never look at the return value. Thus we did not have any of the bugs that scnprintf protects against and the patch does nothing. However, for people reading our code it will be a receipt that we have done our due dilligence and checked our code for this type of bugs. See the presentation "Making C Less Dangerous In The Linux Kernel" at this years LCA Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* RDMA: Start use ib_device_opsKamal Heib2018-12-121-1/+1
| | | | | | | Make all the required change to start use the ib_device_ops structure. Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* Merge branch 'work.afs' of ↵Linus Torvalds2018-11-021-4/+13
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs Pull AFS updates from Al Viro: "AFS series, with some iov_iter bits included" * 'work.afs' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (26 commits) missing bits of "iov_iter: Separate type from direction and use accessor functions" afs: Probe multiple fileservers simultaneously afs: Fix callback handling afs: Eliminate the address pointer from the address list cursor afs: Allow dumping of server cursor on operation failure afs: Implement YFS support in the fs client afs: Expand data structure fields to support YFS afs: Get the target vnode in afs_rmdir() and get a callback on it afs: Calc callback expiry in op reply delivery afs: Fix FS.FetchStatus delivery from updating wrong vnode afs: Implement the YFS cache manager service afs: Remove callback details from afs_callback_break struct afs: Commit the status on a new file/dir/symlink afs: Increase to 64-bit volume ID and 96-bit vnode ID for YFS afs: Don't invoke the server to read data beyond EOF afs: Add a couple of tracepoints to log I/O errors afs: Handle EIO from delivery function afs: Fix TTL on VL server and address lists afs: Implement VL server rotation afs: Improve FS server rotation error handling ...
| * iov_iter: Use accessor functionDavid Howells2018-10-241-4/+13
| | | | | | | | | | | | | | | | | | Use accessor functions to access an iterator's type and direction. This allows for the possibility of using some other method of determining the type of iterator than if-chains with bitwise-AND conditions. Signed-off-by: David Howells <dhowells@redhat.com>
* | CIFS: SMBD: Do not call ib_dereg_mr on invalidated memory registrationLong Li2018-10-241-19/+19
|/ | | | | | | | It is not necessary to deregister a memory registration after it has been successfully invalidated. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* Merge tag 'v4.18' into rdma.git for-nextJason Gunthorpe2018-08-161-15/+7
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | Resolve merge conflicts from the -rc cycle against the rdma.git tree: Conflicts: drivers/infiniband/core/uverbs_cmd.c - New ifs added to ib_uverbs_ex_create_flow in -rc and for-next - Merge removal of file->ucontext in for-next with new code in -rc drivers/infiniband/core/uverbs_main.c - for-next removed code from ib_uverbs_write() that was modified in for-rc Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * cifs: fix SMB1 breakageRonnie Sahlberg2018-07-051-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | SMB1 mounting broke in commit 35e2cc1ba755 ("cifs: Use correct packet length in SMB2_TRANSFORM header") Fix it and also rename smb2_rqst_len to smb_rqst_len to make it less unobvious that the function is also called from CIFS/SMB1 Good job by Paulo reviewing and cleaning up Ronnie's original patch. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Paulo Alcantara <palcantara@suse.de> Signed-off-by: Steve French <stfrench@microsoft.com>
| * cifs: Use correct packet length in SMB2_TRANSFORM headerPaulo Alcantara2018-06-161-14/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In smb3_init_transform_rq(), 'orig_len' was only counting the request length, but forgot to count any data pages in the request. Writing or creating files with the 'seal' mount option was broken. In addition, do some code refactoring by exporting smb2_rqst_len() to calculate the appropriate packet size and avoid duplicating the same calculation all over the code. The start of the io vector is either the rfc1002 length (4 bytes) or a SMB2 header which is always > 4. Use this fact to check and skip the rfc1002 length if requested. Signed-off-by: Paulo Alcantara <palcantara@suse.de> Reviewed-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* | fs/cifs: Simplify ib_post_(send|recv|srq_recv)() callsBart Van Assche2018-07-251-10/+9
| | | | | | | | | | | | | | | | Instead of declaring and passing a dummy 'bad_wr' pointer, pass NULL as third argument to ib_post_(send|recv|srq_recv)(). Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* | IB/core: add max_send_sge and max_recv_sge attributesSteve Wise2018-06-181-3/+10
|/ | | | | | | | | | | | | | | | | | | This patch replaces the ib_device_attr.max_sge with max_send_sge and max_recv_sge. It allows ulps to take advantage of devices that have very different send and recv sge depths. For example cxgb4 has a max_recv_sge of 4, yet a max_send_sge of 16. Splitting out these attributes allows much more efficient use of the SQ for cxgb4 with ulps that use the RDMA_RW API. Consider a large RDMA WRITE that has 16 scattergather entries. With max_sge of 4, the ulp would send 4 WRITE WRs, but with max_sge of 16, it can be done with 1 WRITE WR. Acked-by: Sagi Grimberg <sagi@grimberg.me> Acked-by: Christoph Hellwig <hch@lst.de> Acked-by: Selvin Xavier <selvin.xavier@broadcom.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Acked-by: Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* CIFS: SMBD: Support page offset in memory registrationLong Li2018-06-061-31/+45
| | | | | | | | Change code to pass the correct page offset during memory registration for RDMA read/write. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* CIFS: SMBD: Support page offset in RDMA recvLong Li2018-06-061-7/+11
| | | | | | | | RDMA recv function needs to place data to the correct place starting at page offset. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* CIFS: SMBD: Support page offset in RDMA sendLong Li2018-06-061-8/+19
| | | | | | | | The RDMA send function needs to look at offset in the request pages, and send data starting from there. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
* cifs: smbd: Avoid allocating iov on the stackLong Li2018-04-251-24/+12
| | | | | | | | | | | | | | | It's not necessary to allocate another iov when going through the buffers in smbd_send() through RDMA send. Remove it to reduce stack size. Thanks to Matt for spotting a printk typo in the earlier version of this. CC: Matt Redfearn <matt.redfearn@mips.com> Signed-off-by: Long Li <longli@microsoft.com> Acked-by: Ronnie Sahlberg <lsahlber@redhat.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <smfrench@gmail.com>
* cifs: smbd: Dump SMB packet when configuredLong Li2018-04-201-1/+5
| | | | | | | | | | | When sending through SMB Direct, also dump the packet in SMB send path. Also fixed a typo in debug message. Signed-off-by: Long Li <longli@microsoft.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <smfrench@gmail.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
* cifs: smbd: Check for iov length on sending the last iovLong Li2018-04-191-0/+2
| | | | | | | | | | | | When sending the last iov that breaks into smaller buffers to fit the transfer size, it's necessary to check if this is the last iov. If this is the latest iov, stop and proceed to send pages. Signed-off-by: Long Li <longli@microsoft.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
* cifs: smbd: disconnect transport on RDMA errorsLong Li2018-04-021-1/+6
| | | | | | | | | | On RDMA errors, transport should disconnect the RDMA CM connection. This will notify the upper layer, and it will attempt transport reconnect. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <smfrench@gmail.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> CC: Stable <stable@vger.kernel.org>
* cifs: smbd: avoid reconnect lockupLong Li2018-04-021-4/+8
| | | | | | | | | | | | | | | | | | During transport reconnect, other processes may have registered memory and blocked on transport. This creates a deadlock situation because the transport resources can't be freed, and reconnect is blocked. Fix this by returning to upper layer on timeout. Before returning, transport status is set to reconnecting so other processes will release memory registration resources. Upper layer will retry the reconnect. This is not in fast I/O path so setting the timeout to 5 seconds. Signed-off-by: Long Li <longli@microsoft.com> Signed-off-by: Steve French <smfrench@gmail.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> CC: Stable <stable@vger.kernel.org>
* CIFS: SMBD: fix spelling mistake: "faield" and "legnth"Colin Ian King2018-04-021-2/+2
| | | | | | | | | Trivial fix to spelling mistake in log_rdma_send and log_rdma_mr message text. Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Steve French <smfrench@gmail.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
* address lock imbalance warnings in smbdirect.cSteve French2018-02-071-7/+9
| | | | | | | | Although at least one of these was an overly strict sparse warning in the new smbdirect code, it is cleaner to fix - so no warnings. Signed-off-by: Steve French <smfrench@gmail.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
* Cleanup some minor endian issues in smb3 rdmaSteve French2018-01-271-2/+3
| | | | | | | | Minor cleanup of some sparse warnings (including a few misc endian fixes for the new smb3 rdma code) Signed-off-by: Steve French <smfrench@gmail.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
* CIFS: SMBD: work around gcc -Wmaybe-uninitialized warningArnd Bergmann2018-01-251-9/+6
| | | | | | | | | | | | | | | | | | | | | GCC versions from 4.9 to 6.3 produce a false-positive warning when dealing with a conditional spin_lock_irqsave(): fs/cifs/smbdirect.c: In function 'smbd_recv_buf': include/linux/spinlock.h:260:3: warning: 'flags' may be used uninitialized in this function [-Wmaybe-uninitialized] This function calls some sleeping interfaces, so it is clear that it does not get called with interrupts disabled and there is no need to save the irq state before taking the spinlock. This lets us remove the variable, which makes the function slightly more efficient and avoids the warning. A further cleanup could do the same change for other functions in this file, but I did not want to take this too far for now. Fixes: ac69f66e54ca ("CIFS: SMBD: Implement function to receive data via RDMA receive") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Steve French <smfrench@gmail.com>
* CIFS: SMBD: _smbd_get_connection() can be statickbuild test robot2018-01-251-1/+1
| | | | | | | Fixes: 07495ff5d9bc ("CIFS: SMBD: Establish SMB Direct connection") Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: Steve French <smfrench@gmail.com> Acked-by: Long Li <longli@microsoft.com>
* CIFS: SMBD: Implement RDMA memory registrationLong Li2018-01-251-0/+421
| | | | | | | | | | | Memory registration is used for transferring payload via RDMA read or write. After I/O is done, memory registrations are recovered and reused. This process can be time consuming and is done in a work queue. Signed-off-by: Long Li <longli@microsoft.com> Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <smfrench@gmail.com>