15 Jan, 2020
40 commits
-
Trace layout errors for pNFS/flexfiles on read/write/commit operations.
Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Clean up the generic file commit tracepoints to use a 64-bit value
for the verifier, and to display the pNFS filehandle, if it exists.Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Clean up the generic writeback tracepoints so they do pass the
full structures as arguments. Also ensure we report the number
of bytes actually written.Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Clean up the generic file read tracepoints so they do pass the
full structures as arguments. Also ensure we report the number
of bytes actually read.Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
If the attempt to do pNFS fails, then record what action we
take to recover (resend, reset to pnfs or reset to mds).Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Casting a negative value to an unsigned long is not the same as
converting it to its absolute value.Fixes: 96650e2effa2 ("NFS: Fix show_nfs_errors macros again")
Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Ensure we always return the number of bytes read/written. Also display
the pnfs filehandle if it is in use.Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Instead of making assumptions about the commit verifier contents, change
the commit code to ensure we always check that the verifier was set
by the XDR code.Fixes: f54bcf2ecee9 ("pnfs: Prepare for flexfiles by pulling out common code")
Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Don't clear the NFS_CONTEXT_RESEND_WRITES flag until after calling
nfs_commit_inode(). Otherwise, if nfs_commit_inode() returns an
error, we end up with dirty pages in the page cache, but no tag
to tell us that those pages need resending.Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Remove gss_mech_list_pseudoflavors() and its callers. This is part of
an unused API, and could leak an RCU reference if it were ever called.Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
If a write or commit failed, and the mapping sees a fatal error, we
need to revalidate the contents of that mapping.Fixes: 06c9fdf3b9f1 ("NFS: On fatal writeback errors, we need to call nfs_inode_remove_request()")
Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
If we suffer a fatal error upon writing a file, which causes us to
need to revalidate the entire mapping, then we should also revalidate
the file size.Fixes: d2ceb7e57086 ("NFS: Don't use page_file_mapping after removing the page")
Signed-off-by: Trond Myklebust
Signed-off-by: Anna Schumaker -
Clean up: This simplifies the logic in rpcrdma_post_recvs.
Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
To safely get rid of all rpcrdma_reps from a particular connection
instance, xprtrdma has to wait until each of those reps is finished
being used. A rep may be backing the rq_rcv_buf of an RPC that has
just completed, for example.Since it is safe to invoke rpcrdma_rep_destroy() only in the Receive
completion handler, simply mark reps remaining in the rb_all_reps
list after the transport is drained. These will then be deleted as
rpcrdma_post_recvs pulls them off the rep free list.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
This reduces the hardware and memory footprint of an unconnected
transport.At some point in the future, transport reconnect will allow
resolving the destination IP address through a different device. The
current change enables reps for the new connection to be allocated
on whichever NUMA node the new device affines to after a reconnect.Note that this does not destroy _all_ the transport's reps... there
will be a few that are still part of a running RPC completion.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
Currently the underlying RDMA device is chosen at transport set-up
time. But it will soon be at connect time instead.The maximum size of a transport header is based on device
capabilities. Thus transport header buffers have to be allocated
_after_ the underlying device has been chosen (via address and route
resolution); ie, in the connect worker.Thus, move the allocation of transport header buffers to the connect
worker, after the point at which the underlying RDMA device has been
chosen.This also means the RDMA device is available to do a DMA mapping of
these buffers at connect time, instead of in the hot I/O path. Make
that optimization as well.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
Refactor: Perform the "is supported" check in rpcrdma_ep_create()
instead of in rpcrdma_ia_open(). frwr_open() is where most of the
logic to query device attributes is already located.The current code displays a redundant error message when the device
does not support FRWR. As an additional clean-up, this patch removes
the extra message.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
To support device hotplug and migrating a connection between devices
of different capabilities, we have to guarantee that all in-kernel
devices can support the same max NFS payload size (1 megabyte).This means that possibly one or two in-tree devices are no longer
supported for NFS/RDMA because they cannot support 1MB rsize/wsize.
The only one I confirmed was cxgb3, but it has already been removed
from the kernel.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
Clean up: there is no need to keep two copies of the same value.
Also, in subsequent patches, rpcrdma_ep_create() will be called in
the connect worker rather than at set-up time.Minor fix: Initialize the transport's sendctx to the value based on
the capabilities of the underlying device, not the maximum setting.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
The size of the sendctx queue depends on the value stored in
ia->ri_max_send_sges. This value is determined by querying the
underlying device.Eventually, rpcrdma_ia_open() and rpcrdma_ep_create() will be called
in the connect worker rather than at transport set-up time. The
underlying device will not have been chosen device set-up time.The sendctx queue will thus have to be created after the underlying
device has been chosen via address and route resolution; in other
words, in the connect worker.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
Clean-up. The max_send_sge value also happens to be stored in
ep->rep_attr. Let's keep just a single copy.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
Currently the allocation of buf is not being null checked and
a null pointer dereference can occur when the memory allocation fails.
Fix this by adding a check and returning -ENOMEM.Addresses-Coverity: ("Dereference null return")
Fixes: 6d972518b821 ("NFS: Add fs_context support.")
Signed-off-by: Colin Ian King
Signed-off-by: Anna Schumaker -
If CONFIG_SWAP=n, it does not make much sense to offer the user the
option to enable support for swapping over NFS, as that will still fail
at run time:# swapon /swap
swapon: /swap: swapon failed: Function not implementedFix this by adding a dependency on CONFIG_SWAP.
Fixes: a564b8f0398636ba ("nfs: enable swap on NFS")
Signed-off-by: Geert Uytterhoeven
Signed-off-by: Anna Schumaker -
The empty_iov structure is only copied into another structure,
so make it const.The opportunity for this change was found using Coccinelle.
Signed-off-by: Julia Lawall
Signed-off-by: Anna Schumaker -
swapon over NFS does not go through generic_swapfile_activate
code path when setting up extents. This makes holes in NFS
swapfiles possible which is not expected for swapon.Signed-off-by: Murphy Zhou
Signed-off-by: Anna Schumaker -
The xprtrdma connect logic can return -EPROTO if the underlying
device or network path does not support RDMA. This can happen
after a device removal/insertion.- When SOFTCONN is set, EPROTO is a permanent error.
- When SOFTCONN is not set, EPROTO is treated as a temporary error.
Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
This seems to be a somewhat common issue with Kerberos NFSv4.0
set-ups.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
Try to capture the reason for the writeback path tagging an error on
a page.Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
Signed-off-by: Chuck Lever
Signed-off-by: Anna Schumaker -
In nfs3_proc_lookup, if nfs_alloc_fattr fails, will only print
"NFS call lookup". This may be confusing, move dprintk after
nfs_alloc_fattr.Reported-by: Hulk Robot
Signed-off-by: zhengbin
Signed-off-by: Anna Schumaker -
Fixes coccicheck warning:
fs/nfs/nfs4state.c:1138:2-3: Unneeded semicolon
fs/nfs/nfs4proc.c:6862:2-3: Unneeded semicolon
fs/nfs/nfs4proc.c:8629:2-3: Unneeded semicolonReported-by: Hulk Robot
Signed-off-by: zhengbin
Signed-off-by: Anna Schumaker -
On 32-bit architectures, xdr_encode_nfstime4() needlessly
truncates timestamps to a 32-bit value in the range between
year 1902 and 2038.Change it to use 'struct timespec64' to allow the entire range
of values supported by the server.Signed-off-by: Arnd Bergmann
Signed-off-by: Anna Schumaker -
For NFSv2 and NFSv3, timestamps are stored using 32-bit entities
and overflow in y2038. For historic reasons we truncate the
64-bit timestamps by converting from a timespec64 to a timespec
first.Remove this unnecessary conversion step and do the truncation
in the final functions that take a timestamp.This is transparent to users, but avoids one of the last uses
of 'timespec' and lets us remove it later.Signed-off-by: Arnd Bergmann
Signed-off-by: Anna Schumaker -
nfs currently behaves differently on 32-bit and 64-bit kernels regarding
the on-disk format of nfs_fscache_inode_auxdata.That format should really be the same on any kernel, and we should avoid
the 'timespec' type in order to remove that from the kernel later on.Using plain 'timespec64' would not be good here, since that includes
implied padding and would possibly leak kernel stack data to the on-disk
format on 32-bit architectures.struct __kernel_timespec would work as a replacement, but open-coding
the two struct members in nfs_fscache_inode_auxdata makes it more
obvious what's going on here, and keeps the current format for 64-bit
architectures.Cc: David Howells
Signed-off-by: Arnd Bergmann
Signed-off-by: Anna Schumaker -
Push down the use of timespec64 into NFS nfs_fattr, to avoid needless
conversions, and get closer to having 64-bit time_t support on 32-bit
NFSv4 and removing some old interfaces from the kernel.Signed-off-by: Arnd Bergmann
Signed-off-by: Anna Schumaker -
Using signed 32-bit types for UTC time leads to the y2038 overflow,
which is what happens in the sunrpc code at the moment.This changes the sunrpc code over to use time64_t where possible.
The one exception is the gss_import_v{1,2}_context() function for
kerberos5, which uses 32-bit timestamps in the protocol. Here,
we can at least treat the numbers as 'unsigned', which extends the
range from 2038 to 2106.Signed-off-by: Arnd Bergmann
Signed-off-by: Anna Schumaker -
Split out from commit "NFS: Add fs_context support."
Add wrappers nfs_errorf(), nfs_invalf(), and nfs_warnf() which log error
information to the fs_context. Convert some printk's to use these new
wrappers instead.Signed-off-by: Scott Mayhew
Signed-off-by: Anna Schumaker -
Split out from commit "NFS: Add fs_context support."
This patch adds additional refactoring for the conversion of NFS to use
fs_context, namely:(*) Merge nfs_mount_info and nfs_clone_mount into nfs_fs_context.
nfs_clone_mount has had several fields removed, and nfs_mount_info
has been removed altogether.
(*) Various functions now take an fs_context as an argument instead
of nfs_mount_info, nfs_fs_context, etc.Signed-off-by: Scott Mayhew
Signed-off-by: Anna Schumaker -
Add filesystem context support to NFS, parsing the options in advance and
attaching the information to struct nfs_fs_context. The highlights are:(*) Merge nfs_mount_info and nfs_clone_mount into nfs_fs_context. This
structure represents NFS's superblock config.(*) Make use of the VFS's parsing support to split comma-separated lists
(*) Pin the NFS protocol module in the nfs_fs_context.
(*) Attach supplementary error information to fs_context. This has the
downside that these strings must be static and can't be formatted.(*) Remove the auxiliary file_system_type structs since the information
necessary can be conveyed in the nfs_fs_context struct instead.(*) Root mounts are made by duplicating the config for the requested mount
so as to have the same parameters. Submounts pick up their parameters
from the parent superblock.[AV -- retrans is u32, not string]
[SM -- Renamed cfg to ctx in a few functions in an earlier patch]
[SM -- Moved fs_context mount option parsing to an earlier patch]
[SM -- Moved fs_context error logging to a later patch]
[SM -- Fixed printks in nfs4_try_get_tree() and nfs4_get_referral_tree()]
[SM -- Added is_remount_fc() helper]
[SM -- Deferred some refactoring to a later patch]
[SM -- Fixed referral mounts, which were broken in the original patch]
[SM -- Fixed leak of nfs_fattr when fs_context is freed]Signed-off-by: David Howells
Signed-off-by: Al Viro
Signed-off-by: Scott Mayhew
Signed-off-by: Anna Schumaker -
Split out from commit "NFS: Add fs_context support."
Convert existing mount option definitions to fs_parameter_enum's and
fs_parameter_spec's. Parse mount options using fs_parse() and
lookup_constant().Notes:
1) Fixed a typo in the udp6 definition in nfs_xprt_protocol_tokens
from the original commit.2) fs_parse() expects an fs_context as the first arg so that any
errors can be logged to the fs_context. We're passing NULL for the
fs_context (this will change in commit "NFS: Add fs_context support.")
which is okay as it will cause logfc() to do a printk() instead.3) fs_parse() expects an fs_paramter as the third arg. We're
building an fs_parameter manually in nfs_fs_context_parse_option(),
which will go away in commit "NFS: Add fs_context support.".Signed-off-by: Scott Mayhew
Signed-off-by: Anna Schumaker
