- 17 Aug, 2015 17 commits
-
-
git://git.linux-nfs.org/projects/anna/nfs-rdmaTrond Myklebust authored
NFS: NFS over RDMA Client Side Changes These patches improve both client performance and scalability, most notably by increasing the maixmum allowed rsize and wsize and by increasing the number of RDMA "credits". There are also several bugfixes, such as correcting how WRITE compounds are encoded and fixing large NFS symlink operations. Signed-off-by:Anna Schumaker <Anna.Schumaker@Netapp.com>
-
Anna Schumaker authored
And call nfs_file_clear_open_context() directly. This makes it obvious that nfs_file_release() will always return 0. Signed-off-by:
Trond Myklebust <trond.myklebust@primarydata.com>
-
Anna Schumaker authored
All nfs_write_inode() does is pass its arguments to nfs_commit_unstable_pages(). Let's cut out the middle man and have nfs_write_pages() do the work directly. Signed-off-by:
-
Anna Schumaker authored
All these functions do is call nfs41_ping_server() without adding anything. Let's remove them and give nfs41_ping_server() a better name instead. Signed-off-by:
-
Anna Schumaker authored
The idmap_init() and idmap_quit() functions only exist to call the _keyring() version. Let's just call the keyring() functions directly. Signed-off-by:
-
Anna Schumaker authored
They already exist and do the exact same thing. Let's save ourselves several lines of code! Signed-off-by:
-
Anna Schumaker authored
This function is to help determine if two sockaddrs are really the same socket. Signed-off-by:
-
Anna Schumaker authored
I'm planning on using these functions inside the client, so remove the underscores to make it feel like I'm using a public interface. Signed-off-by:
-
Anna Schumaker authored
nfs_readdir_xdr_to_array() uses both a cache array and an array of pages, so I rename these functions to make it clearer how the code works. nfs_readdir_large_page() becomes nfs_readdir_alloc_pages() because this function has absolutely nothing to do with setting up a large page. nfs_readdir_free_pagearray() becomes nfs_readdir_free_pages() to stay consistent with the new alloc_pages() function. Signed-off-by:
-
Anna Schumaker authored
This variable is initialized to NULL and is never modified before being passed to nfs_readdir_free_large_page(). But that's okay, because nfs_readdir_free_large_page() only seems to exist as a way of calling nfs_readdir_free_pagearray() without this parameter. Let's simplify by removing pages_ptr and nfs_readdir_free_pagearray(). Signed-off-by:
-
Jeff Layton authored
We already know that pg_lseg is NULL here. Signed-off-by:
Jeff Layton <jeff.layton@primarydata.com> Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Christoph Hellwig authored
We generally want to read and write to a block device that's used by the pNFS block layout client (and even if it's read only the server has no way of telling us). Add FMODE_WRITE to the mode argument so that we don't incorrectly tell the block driver that we want a read-only open. Signed-off-by:
-
Christoph Hellwig authored
Instead of overwriting kernel memory reject too long signatures. Signed-off-by:
-
Christoph Hellwig authored
We need to replace the __be32 with a void pointer to do proper arithmentics on the virtual addresses so that we can get the right page pointers. Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Christoph Hellwig authored
We need to include the first u32 for the number of entries. Add a helper for the calculation instead of opencoding it so that it's in one place. Signed-off-by:
-
Kinglong Mee authored
---Steps to Reproduce-- <nfs-server> # cat /etc/exports /nfs/referal *(rw,insecure,no_subtree_check,no_root_squash,crossmnt) /nfs/old *(ro,insecure,subtree_check,root_squash,crossmnt) <nfs-client> # mount -t nfs nfs-server:/nfs/ /mnt/ # ll /mnt/*/ <nfs-server> # cat /etc/exports /nfs/referal *(rw,insecure,no_subtree_check,no_root_squash,crossmnt,refer=/nfs/old/@nfs-server) /nfs/old *(ro,insecure,subtree_check,root_squash,crossmnt) # service nfs restart <nfs-client> # ll /mnt/*/ --->>>>> oops here [ 5123.102925] BUG: unable to handle kernel NULL pointer dereference at (null) [ 5123.103363] IP: [<ffffffffa03ed38b>] nfs4_proc_get_locations+0x9b/0x120 [nfsv4] [ 5123.103752] PGD 587b9067 PUD 3cbf5067 PMD 0 [ 5123.104131] Oops: 0000 [#1] [ 5123.104529] Modules linked in: nfsv4(OE) nfs(OE) fscache(E) nfsd(OE) xfs libcrc32c iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi coretemp crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel ppdev vmw_balloon parport_pc parport i2c_piix4 shpchp auth_rpcgss nfs_acl vmw_vmci lockd grace sunrpc vmwgfx drm_kms_helper ttm drm mptspi serio_raw scsi_transport_spi e1000 mptscsih mptbase ata_generic pata_acpi [last unloaded: nfsd] [ 5123.105887] CPU: 0 PID: 15853 Comm: ::1-manager Tainted: G OE 4.2.0-rc6+ #214 [ 5123.106358] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 05/20/2014 [ 5123.106860] task: ffff88007620f300 ti: ffff88005877c000 task.ti: ffff88005877c000 [ 5123.107363] RIP: 0010:[<ffffffffa03ed38b>] [<ffffffffa03ed38b>] nfs4_proc_get_locations+0x9b/0x120 [nfsv4] [ 5123.107909] RSP: 0018:ffff88005877fdb8 EFLAGS: 00010246 [ 5123.108435] RAX: ffff880053f3bc00 RBX: ffff88006ce6c908 RCX: ffff880053a0d240 [ 5123.108968] RDX: ffffea0000e6d940 RSI: ffff8800399a0000 RDI: ffff88006ce6c908 [ 5123.109503] RBP: ffff88005877fe28 R08: ffffffff81c708a0 R09: 0000000000000000 [ 5123.110045] R10: 00000000000001a2 R11: ffff88003ba7f5c8 R12: ffff880054c55800 [ 5123.110618] R13: 0000000000000000 R14: ffff880053a0d240 R15: ffff880053a0d240 [ 5123.111169] FS: 0000000000000000(0000) GS:ffffffff81c27000(0000) knlGS:0000000000000000 [ 5123.111726] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 5123.112286] CR2: 0000000000000000 CR3: 0000000054cac000 CR4: 00000000001406f0 [ 5123.112888] Stack: [ 5123.113458] ffffea0000e6d940 ffff8800399a0000 00000000000167d0 0000000000000000 [ 5123.114049] 0000000000000000 0000000000000000 0000000000000000 00000000a7ec82c6 [ 5123.114662] ffff88005877fe18 ffffea0000e6d940 ffff8800399a0000 ffff880054c55800 [ 5123.115264] Call Trace: [ 5123.115868] [<ffffffffa03fb44b>] nfs4_try_migration+0xbb/0x220 [nfsv4] [ 5123.116487] [<ffffffffa03fcb3b>] nfs4_run_state_manager+0x4ab/0x7b0 [nfsv4] [ 5123.117104] [<ffffffffa03fc690>] ? nfs4_do_reclaim+0x510/0x510 [nfsv4] [ 5123.117813] [<ffffffff810a4527>] kthread+0xd7/0xf0 [ 5123.118456] [<ffffffff810a4450>] ? kthread_worker_fn+0x160/0x160 [ 5123.119108] [<ffffffff816d9cdf>] ret_from_fork+0x3f/0x70 [ 5123.119723] [<ffffffff810a4450>] ? kthread_worker_fn+0x160/0x160 [ 5123.120329] Code: 4c 8b 6a 58 74 17 eb 52 48 8d 55 a8 89 c6 4c 89 e7 e8 4a b5 ff ff 8b 45 b0 85 c0 74 1c 4c 89 f9 48 8b 55 90 48 8b 75 98 48 89 df <41> ff 55 00 3d e8 d8 ff ff 41 89 c6 74 cf 48 8b 4d c8 65 48 33 [ 5123.121643] RIP [<ffffffffa03ed38b>] nfs4_proc_get_locations+0x9b/0x120 [nfsv4] [ 5123.122308] RSP <ffff88005877fdb8> [ 5123.122942] CR2: 0000000000000000 Fixes: ec011fe8 ("NFS: Introduce a vector of migration recovery ops") Cc: stable@vger.kernel.org # v3.13+ Signed-off-by:
Kinglong Mee <kinglongmee@gmail.com> Signed-off-by:
-
- 12 Aug, 2015 6 commits
-
-
Kinglong Mee authored
The xprt created by svc_create_xprt have be added to serv->sv_permsocks. So putting the xprt directly is useless. Otherwise, there is a more svc_xprt_put after the xprt be freed. v2, same as v1. Signed-off-by:
-
Kinglong Mee authored
Commit 1d3d4437 "vmscan: per-node deferred work" have made register_shrinker can return an intergater error. If register_shrinker() fail, the later unregister_shrinker() will cause a NULL pointer access. v2, same as v1. Signed-off-by:
-
NeilBrown authored
It is unusual to combine the open flags O_RDONLY and O_EXCL, but it appears that libre-office does just that. [pid 3250] stat("/home/USER/.config", {st_mode=S_IFDIR|0700, st_size=8192, ...}) = 0 [pid 3250] open("/home/USER/.config/libreoffice/4-suse/user/extensions/buildid", O_RDONLY|O_EXCL <unfinished ...> NFSv4 takes O_EXCL as a sign that a setattr command should be sent, probably to reset the timestamps. When it was an O_RDONLY open, the SETATTR command does not identify any actual attributes to change. If no delegation was provided to the open, the SETATTR uses the all-zeros stateid and the request is accepted (at least by the Linux NFS server - no harm, no foul). If a read-delegation was provided, this is used in the SETATTR request, and a Netapp filer will justifiably claim NFS4ERR_BAD_STATEID, which the Linux client takes as a sign to retry - indefinitely. So only treat O_EXCL specially if O_CREAT was also given. Signed-off-by:NeilBrown <neilb@suse.com> Cc: stable@vger.kernel.org Signed-off-by:
-
Jeff Layton authored
The current limit of 32 bytes artificially limits the name string that we end up stuffing into NFSv4.x client ID blobs. If you have multiple hosts with long hostnames that only differ near the end, then this can cause NFSv4 client ID collisions. Linux nodenames are actually limited to __NEW_UTS_LEN bytes (64), so use that as the limit instead. Also, use XDR_QUADLEN to specify the slack length, just for clarity and in case someone in the future changes this to something not evenly divisible by 4. Reported-by:
Michael Skralivetsky <michael.skralivetsky@primarydata.com> Signed-off-by:
-
Trond Myklebust authored
Prevent a potential deadlock. Signed-off-by: -
Peng Tao authored
Turned out I misinterpreted the spec... Cc: Tom Haynes <thomas.haynes@primarydata.com> Reported-by:
Jean Spector <jean@primarydata.com> Signed-off-by:
Peng Tao <tao.peng@primarydata.com> Signed-off-by:
-
- 10 Aug, 2015 1 commit
-
-
Trond Myklebust authored
pnfs_layout_mark_request_commit() needs to ensure that it adds the request to the commit list atomically with all the other updates in order to prevent corruption to buckets[ds_commit_idx].wlseg due to races with pnfs_generic_clear_request_commit(). Fixes: 338d00cf ("pnfs: Refactor the *_layout_mark_request_commit...") Cc: stable@vger.kernel.org # v4.0+ Signed-off-by:
-
- 05 Aug, 2015 16 commits
-
-
Devesh Sharma authored
This is a rework of the following patch sent almost a year back: http://www.mail-archive.com/linux-rdma%40vger.kernel.org/msg20730.html In presence of active mount if someone tries to rmmod vendor-driver, the command remains stuck forever waiting for destruction of all rdma-cm-id. in worst case client can crash during shutdown with active mounts. The existing code assumes that ia->ri_id->device cannot change during the lifetime of a transport. xprtrdma do not have support for DEVICE_REMOVAL event either. Lifting that assumption and adding support for DEVICE_REMOVAL event is a long chain of work, and is in plan. The community decided that preventing the hang right now is more important than waiting for architectural changes. Thus, this patch introduces a temporary workaround to acquire HCA driver module reference count during the mount of a nfs-rdma mount point. Signed-off-by:
Devesh Sharma <devesh.sharma@avagotech.com> Signed-off-by:
Chuck Lever <chuck.lever@oracle.com> Reviewed-by:
Sagi Grimberg <sagig@dev.mellanox.co.il> Signed-off-by:
-
Chuck Lever authored
The verbs are obsolete. The ib_rereg_phys_mr() verb is not used by kernel ULPs, and the last ib_reg_phys_mr() call site in the kernel tree has now been removed. Two staging tree call sites remain in the Lustre client. The Lustre team has been notified of the deprecation of reg_phys_mr. Signed-off-by:
Doug Ledford <dledford@redhat.com> Signed-off-by:
-
Chuck Lever authored
RDMA_NOMSG type calls are less efficient than RDMA_MSG. Count NOMSG calls so administrators can tell if they happen to be used more than expected. Signed-off-by:
-
Chuck Lever authored
checkpatch.pl complained about the seq_printf() format string split across lines and the use of %Lu. Signed-off-by:
-
Chuck Lever authored
Repair how rpcrdma_marshal_req() chooses which RDMA message type to use for large non-WRITE operations so that it picks RDMA_NOMSG in the correct situations, and sets up the marshaling logic to SEND only the RPC/RDMA header. Large NFSv2 SYMLINK requests now use RDMA_NOMSG calls. The Linux NFS server XDR decoder for NFSv2 SYMLINK does not handle having the pathname argument arrive in a separate buffer. The decoder could be fixed, but this is simpler and RDMA_NOMSG can be used in a variety of other situations. Ensure that the Linux client continues to use "RDMA_MSG + read list" when sending large NFSv3 SYMLINK requests, which is more efficient than using RDMA_NOMSG. Large NFSv4 CREATE(NF4LNK) requests are changed to use "RDMA_MSG + read list" just like NFSv3 (see Section 5 of RFC 5667). Before, these did not work at all. Signed-off-by:
-
Chuck Lever authored
Currently xprtrdma appends an extra chunk element to the RPC/RDMA read chunk list of each NFSv4 WRITE compound. The extra element contains the final GETATTR operation in the compound. The result is an extra RDMA READ operation to transfer a very short piece of each NFS WRITE compound (typically 16 bytes). This is inefficient. It is also incorrect. The client is sending the trailing GETATTR at the same Position as the preceding WRITE data payload. Whether or not RFC 5667 allows the GETATTR to appear in a read chunk, RFC 5666 requires that these two separate RPC arguments appear at two distinct Positions. It can also be argued that the GETATTR operation is not bulk data, and therefore RFC 5667 forbids its appearance in a read chunk at all. Although RFC 5667 is not precise about when using a read list with NFSv4 COMPOUND is allowed, the intent is that only data arguments not touched by NFS (ie, read and write payloads) are to be sent using RDMA READ or WRITE. The NFS client constructs GETATTR arguments itself, and therefore is required to send the trailing GETATTR operation as additional inline content, not as a data payload. NB: This change is not backwards compatible. Some older servers do not accept inline content following the read list. The Linux NFS server should handle this content correctly as of commit a97c331f ("svcrdma: Handle additional inline content"). Signed-off-by:
-
Chuck Lever authored
Currently Linux always offers a reply chunk, even when the reply can be sent inline (ie. is smaller than 1KB). On the client, registering a memory region can be expensive. A server may choose not to use the reply chunk, wasting the cost of the registration. This is a change only for RPC replies smaller than 1KB which the server constructs in the RPC reply send buffer. Because the elements of the reply must be XDR encoded, a copy-free data transfer has no benefit in this case. Signed-off-by:
Sagi Grimberg <sagig@mellanox.com> Tested-by:
-
Chuck Lever authored
The client has been setting up a reply chunk for NFS READs that are smaller than the inline threshold. This is not efficient: both the server and client CPUs have to copy the reply's data payload into and out of the memory region that is then transferred via RDMA. Using the write list, the data payload is moved by the device and no extra data copying is necessary. Signed-off-by:
-
Chuck Lever authored
When the size of the RPC message is near the inline threshold (1KB), the client would allow messages to be sent that were a few bytes too large. When marshaling RPC/RDMA requests, ensure the combined size of RPC/RDMA header and RPC header do not exceed the inline threshold. Endpoints typically reject RPC/RDMA messages that exceed the size of their receive buffers. The two server implementations I test with (Linux and Solaris) use receive buffers that are larger than the client’s inline threshold. Thus so far this has been benign, observed only by code inspection. Signed-off-by:
-
Chuck Lever authored
RDMA_MSGP type calls insert a zero pad in the middle of the RPC message to align the RPC request's data payload to the server's alignment preferences. A server can then "page flip" the payload into place to avoid a data copy in certain circumstances. However: 1. The client has to have a priori knowledge of the server's preferred alignment 2. Requests eligible for RDMA_MSGP are requests that are small enough to have been sent inline, and convey a data payload at the _end_ of the RPC message Today 1. is done with a sysctl, and is a global setting that is copied during mount. Linux does not support CCP to query the server's preferences (RFC 5666, Section 6). A small-ish NFSv3 WRITE might use RDMA_MSGP, but no NFSv4 compound fits bullet 2. Thus the Linux client currently leaves RDMA_MSGP disabled. The Linux server handles RDMA_MSGP, but does not use any special page flipping, so it confers no benefit. Clean up the marshaling code by removing the logic that constructs RDMA_MSGP type calls. This also reduces the maximum send iovec size from four to just two elements. /proc/sys/sunrpc/rdma_inline_write_padding is a kernel API, and thus is left in place. Signed-off-by:
-
Chuck Lever authored
Untangle the end of rpcrdma_ia_open() by moving DMA MR set-up, which is different for each registration method, to the .ro_open functions. This is refactoring only. No behavior change is expected. Signed-off-by:
-
Chuck Lever authored
All HCA providers have an ib_get_dma_mr() verb. Thus rpcrdma_ia_open() will either grab the device's local_dma_key if one is available, or it will call ib_get_dma_mr(). If ib_get_dma_mr() fails, rpcrdma_ia_open() fails and no transport is created. Therefore execution never reaches the ib_reg_phys_mr() call site in rpcrdma_register_internal(), so it can be removed. The remaining logic in rpcrdma_{de}register_internal() is folded into rpcrdma_{alloc,free}_regbuf(). This is clean up only. No behavior change is expected. Signed-off-by: -
Chuck Lever authored
PHYSICAL memory registration uses a single rkey for all of the client's memory, thus is insecure. It is still useful in some cases for testing. Retain the ability to select PHYSICAL memory registration capability via /proc/sys/sunrpc/rdma_memreg_strategy, but don't fall back to it if the HCA does not support FRWR or FMR. This means amso1100 no longer works out of the box with NFS/RDMA. When using amso1100 HCAs, set the memreg_strategy sysctl to 6 before performing NFS/RDMA mounts. Signed-off-by:
-
Chuck Lever authored
In preparation for similar increases on NFS/RDMA servers, bump the advertised credit limit for RPC/RDMA to 128. This allocates some extra resources, but the client will continue to allow only the number of RPCs in flight that the server requests via its advertised credit limit. Signed-off-by:
-
Chuck Lever authored
The point of larger rsize and wsize is to reduce the per-byte cost of memory registration and deregistration. Modern HCAs can typically handle a megabyte or more with a single registration operation. Signed-off-by:
-
Chuck Lever authored
In particular, recognize when an IPv6 connection is bound. Signed-off-by:
-
