Mainly focus on files that use BSD 2-Clause license, however the tool I
was using misidentified many licenses so this was mostly a manual - error
prone - task.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Mainly focus on files that use BSD 3-Clause license.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
When the NFSv4.1 pNFS client is using a Flexible File Layout specifying
mirrored Data Servers, it must do the writes and commits to all mirrors.
This patch modifies the client to use a taskqueue to perform these writes
and commits concurrently.
The number of threads can't be changed for taskqueue(9), so it is set
to 4 * mp_ncpus by default, but this can be overridden by setting the
sysctl vfs.nfs.pnfsiothreads.
Differential Revision: https://reviews.freebsd.org/D12632
This patch adds support for the Flexible File Layout to the pNFS client.
Although the patch is rather large, it should only affect NFS mounts
using the "pnfs" option against pNFS servers that do not support File
Layout.
There are still a couple of things missing from the Flexible File Layout
client implementation:
- The code does not yet do a LayoutReturn with I/O error stats when
I/O error(s) occur when attempting to do I/O on a DS.
This will be fixed in a future commit, since it is important for the
MDS to know that I/O on a DS is failing.
- The current code does writes and commits to mirror DSs serially.
Making them happen concurrently will be done in a future commit,
after discussion on freebsd-current@ on the best way to do this.
- The code does not handle NFSv4.0 DSs. Since there is no extant pNFS
server that implements NFSv4.0 DSs and NFSv4.1 DSs makes more sense
now, I don't intend to implement this until there is a need for it.
There is support for NFSv4.1 and NFSv3 DSs.
This patch changes nfsv4_getipaddr() and nfsrpc_fillsa() to use
a sockaddr_in * and sockaddr_in6 * instead of sockaddr_storage, to
avoid allocating the latter on the stack. It also moves the nfsrpc_fillsa()
call to after the completion of parsing of the DeviceInfo reply from
the server. This patch is in preparation for addition of Flex File
Layout support in a future commit.
It only affects the "pnfs" NFSv4.1 client mount option and should not
have changed its semantics.
When a "pnfs" NFSv4.1 mount was unmounted, it didn't free up the layouts
and deviceinfo structures. This leak only affects "pnfs" mounts and only
when the mount is umounted.
Found while testing the pNFS Flexible File layout client code.
MFC after: 2 weeks
This patch adds "vers" and "minorvers" arguments to nfscl_reqstart().
The patch always passes them in as "0" and that implies no change
in semantics. These arguments will be used by a future commit that
adds support for the Flexible File Layout.
nfsm_uiombuflist() zero filled the mbuf list to a multiple of 4bytes
as required for XDR. Unfortunately that modified an mbuf list after
it was m_copym()'d and was broken. This patch removes the zero filling code.
Since nfsm_uiombuflist() is not yet used in head/current, this has no
effect on users.
The function will be used by a future commit of code that adds Flex
File Layout support.
Make the NFSv4 pNFS client function nfsrpc_layoutget() a static, since it
is only used in sys/fs/nfsclient/nfs_clrpcops.c.
This prepares the code for future patches that add Flex File layout
support.
This patch adds a new function called nfsm_uiombuflist(), which is
similar to nfsm_uiombuf(), but doesn't not use the fields in
struct nfsrv_descript. This new function will be used by the pNFS client
for writing to mirrors using Flex Files layout.
The function is not yet called anywhere.
Also, get rid of #ifndef APPLE, which is ancient cruft left over from
the Mac OSX port of the NFSv4 client.
Simplify nfsrpc_layoutreturn() args. in preparation for the addition
of Flex File layout support, since File layout uses a 0 length field.
Flex Files does use a longer field, but that will be added in a
subsequent commit.
The code in nfscl_doflayoutio() bogusly used FREAD instead of
NFSV4OPEN_ACCESSREAD. Since both happen to be defined as "1", this
worked and the patch doesn't result in a functional change.
Found by inspection during development of Flex File Layout support.
MFC after: 2 weeks
Currently several paths in the NFS client upgrade the shared vnode
lock to exclusive, which might cause temporal dropping of the lock.
This action appears to be fatal for nullfs mounts over NFS. If the
operation is performed over nullfs vnode, then bypassed down to NFS
VOP, and the lock is dropped, other thread might reclaim the upper
nullfs vnode. Since on reclaim the nullfs vnode lock and NFS vnode
lock are split, the original lock state of the nullfs vnode is not
restored. As result, VFS operations receive not locked vnode after a
VOP call.
Stop upgrading the vnode lock when we check the consistency or flush
buffers as result of detected inconsistency. Instead, allocate a new
lockmgr lock for each NFS node, which is locked exclusive instead of
the vnode lock upgrade. In other words, the other parallel
modification of the vnode are excluded by either vnode lock conflict
or exclusivity of the new lock when the vnode lock is shared.
Also revert r316529 because now the vnode cannot be reclaimed during
ncl_vinvalbuf().
In collaboration with: pho
Reviewed by: rmacklem
Reported and tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D12083
When an NFS mount is hung against an unresponsive NFS server, the "umount -f"
option can be used to dismount the mount. Unfortunately, "umount -f" gets
hung as well if a "umount" without "-f" has already been done. Usually,
this is because of a vnode lock being held by the "umount" for the mounted-on
vnode.
This patch adds kernel code so that a new "-N" option can be added to "umount",
allowing it to avoid getting hung for this case.
It adds two flags. One indicates that a forced dismount is about to happen
and the other is used, along with setting mnt_data == NULL, to handshake
with the nfs_unmount() VFS call.
It includes a slight change to the interface used between the client and
common NFS modules, so I bumped __FreeBSD_version to ensure both modules are
rebuilt.
Tested by: pho
Reviewed by: kib
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D11735
If the nfsrpc_createlayoutrpc() call in nfsrpc_getcreatelayout() fails,
the code used nfhpp when it might be set NULL. This patch checks for
the error cases (laystat != 0) and avoids using nfhpp for the failure case.
This would only affect NFSv4.1 mounts with the "pnfs" option.
Found while testing the "umount -N" patch not yet in head.
MFC after: 2 weeks
This patch defines a macro that checks for MNTK_UNMOUNTF and replaces
explicit checks with this macro. It has no effect on semantics, but
prepares the code for a future patch where there will also be a
NFS specific flag for "forced dismount about to occur".
Suggested by: kib
MFC after: 2 weeks
Suppose that a file on NFS has partially filled last page, and this
page is dirty. NFS VOP_PAGEOUT() method only marks the the page clean
up to the block of the last written byte, leaving other blocks dirty.
Also any page which erronously exists in the vnode vm_object past EOF
is also left marked as dirty.
With the introduction of the buf-cache coherent pager, each pass of
syncer over the object with such page results in creation of B_DELWRI
buffer due to VOP_WRITE() call. This buffer is noted on next syncer
pass, which results e.g. a visible manifestation of shutdown never
finishing vnode sync. Note that before buf-cache coherency commit, a
dirty page might left never synced to server if a partial writes
occur.
Fix this by clearing dirty bits after EOF. Only blocks of the partial
page which are completely after EOF are marked clean, to avoid
possible user data loss.
Reported by: mav
Reviewed by: alc, markj
Tested by: mav, pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D11697
r320062 used nm_rsize, nm_wsize to set the maximum request/response sizes for
the NFSv4.1 session. If rsize,wsize are not specified as options, the
value of nm_rsize, nm_wsize is 0 at session creation, resulting in
values for request/response that are too small.
This patch fixes the problem. A workaround is to specify rsize=N,wsize=N
mount options explicitly, so they are set before session creation.
This bug only affects NFSv4.1 mounts against some non-FreeBSD servers.
MFC after: 1 week
r320062 used nm_rsize, nm_wsize to set the maximum request/response sizes for
the NFSv4.1 session. If rsize,wsize are not specified as options, the
value of nm_rsize, nm_wsize is 0 at session creation, resulting in
values for request/response that are too small.
This patch fixes the problem. A workaround is to specify rsize=N,wsize=N
mount options explicitly, so they are set before session creation.
This bug only affects NFSv4.1 mounts against some non-FreeBSD servers.
MFC after: 1 week
Update filesystems not currently using vop_stdpathconf() in pathconf
VOPs to use vop_stdpathconf() for any configuration variables that do
not have filesystem-specific values. vop_stdpathconf() is used for
variables that have system-wide settings as well as providing default
values for some values based on system limits. Filesystems can still
explicitly override individual settings.
PR: 219851
Reported by: cem
Reviewed by: cem, kib, ngie
MFC after: 1 month
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D11541
If an NFSv3 server were to reply with weak cache consistency attributes,
but not post operation attributes, the client would use garbage attributes
from memory. This was spotted during work on the code for the NFSv4.1 client.
I have never seen evidence that this happens and it wouldn't make sense
for an NFSv3 server to do this, so this patch is basically "theoretical",
but does fix the problem if a server were to do the above.
PR: 219552
MFC after: 2 weeks
A NFSv4.1/pNFS server using File Layout can specify that Commit operations
are to be done against the DS instead of MDS. Since no extant pNFS
server did this, the code was untested and "#ifdef notyet".
The FreeBSD pNFS server I am developing does specify that Commits be done
through the DS, so the code has been enabled/tested.
This patch should only affect the case of a pNFS server that specfies
Commits through the DS.
PR: 219551
MFC after: 2 weeks
When the NFSv4.1 client is doing pNFS, it needs to get an Open and
a Layout for every file it will be doing I/O on. The current code
does two separate RPCs to get these. This patch adds two new compounds
that do the both the Open and LayoutGet in the same RPC, reducing the
RPC count.
It also factors out the code that sets up and parses the LayoutGet operation
into separate functions, so that the code doesn't get duplicated for
these new RPCs.
This patch is fairly large, but should only affect the NFSv4.1 client
when the "pnfs" option is specified.
PR: 219550
MFC after: 2 weeks
initialized.
bdrewery@ has reported panics "newnfs_copycred: negative nfsc_ngroups".
The only way I can see that this occurs is that the credentials field of
the open structure gets used before being filled in.
I am not sure quite how this happens, but for the file create case, the
code is serialized via the vnode lock on the directory. If, somehow, a
link to the same file gets created just after file creation, this might
occur.
This patch ensures that the credentials field is initialized to a reasonable
set of credentials before the structure is linked into any list, so I
this should ensure it is initialized before use.
I am committing the patch now, since bdrewery@ notes that the panics
are intermittent and it may be months before he knows if the patch fixes
his problem.
Reported by: bdrewery
MFC after: 2 weeks
r320070 removed the definition of maxbcachebuf from sys/param.h to
fix the build for arm.
This patch adds the definition of maxbcachebuf to sys/buf.h, which
should be ok, since sys/buf.h is not being included in arm/arm/elf_note.S.
Suggested by: kib
MFC after: 2 weeks
The code still doesn't use d_off. That will come in a future commit.
The code also removes the checks for servers returning a fileno that
doesn't fit in 32bits, since that should work ok now.
Bump __FreeBSD_version since this patch changes the interface between
the NFS kernel modules.
Reviewed by: kib
arm build.
In the arm build, elf_note.S includes sys/param.h and then does an
elf macro called ELFNOTE(). Although the compile error doesn't make
sense to me, I believe it just means that an "extern ..." can't exist
in param.h for this inclusion case.
I suspect adding #if !defined(LOCORE) might fix the build, but this
commit just takes the definition out.
I will ask freebsd-current@ what is the best was to deal with this
and do a subsequent commit after that.
Reported by: melounmichal@gmail.com
By making MAXBCACHEBUF a tunable, it can be increased to allow for
larger read/write data sizes for the NFS client.
The tunable is limited to MAXPHYS, which is currently 128K.
Making MAXPHYS a tunable or increasing its value is being discussed,
since it would be nice to support a read/write data size of 1Mbyte
for the NFS client when mounting the AmazonEFS file service.
Reviewed by: kib
MFC after: 2 weeks
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D10991
Since ino64 expanded dev_t to 64bit, make VOP_GETATTR(9) provide all
bits of mnt_stat.f_fsid as va_fsid for vnodes on filesystems which use
f_fsid. In particular, NFSv3 and sometimes NFSv4, and ZFS use this
method or reporting st_dev by stat(2).
Provide a new helper vn_fsid() to avoid duplicating code to copy
f_fsid to va_fsid.
Note that the change is mostly cosmetic. Its motivation is to avoid
sign-extension of f_fsid[0] into 64bit dev_t value which happens after
dev_t becomes 64bit..
Reviewed by: avg(zfs), rmacklem (nfs) (both for previous version)
Sponsored by: The FreeBSD Foundation
Extend the ino_t, dev_t, nlink_t types to 64-bit ints. Modify
struct dirent layout to add d_off, increase the size of d_fileno
to 64-bits, increase the size of d_namlen to 16-bits, and change
the required alignment. Increase struct statfs f_mntfromname[] and
f_mntonname[] array length MNAMELEN to 1024.
ABI breakage is mitigated by providing compatibility using versioned
symbols, ingenious use of the existing padding in structures, and
by employing other tricks. Unfortunately, not everything can be
fixed, especially outside the base system. For instance, third-party
APIs which pass struct stat around are broken in backward and
forward incompatible ways.
Kinfo sysctl MIBs ABI is changed in backward-compatible way, but
there is no general mechanism to handle other sysctl MIBS which
return structures where the layout has changed. It was considered
that the breakage is either in the management interfaces, where we
usually allow ABI slip, or is not important.
Struct xvnode changed layout, no compat shims are provided.
For struct xtty, dev_t tty device member was reduced to uint32_t.
It was decided that keeping ABI compat in this case is more useful
than reporting 64-bit dev_t, for the sake of pstat.
Update note: strictly follow the instructions in UPDATING. Build
and install the new kernel with COMPAT_FREEBSD11 option enabled,
then reboot, and only then install new world.
Credits: The 64-bit inode project, also known as ino64, started life
many years ago as a project by Gleb Kurtsou (gleb). Kirk McKusick
(mckusick) then picked up and updated the patch, and acted as a
flag-waver. Feedback, suggestions, and discussions were carried
by Ed Maste (emaste), John Baldwin (jhb), Jilles Tjoelker (jilles),
and Rick Macklem (rmacklem). Kris Moore (kris) performed an initial
ports investigation followed by an exp-run by Antoine Brodin (antoine).
Essential and all-embracing testing was done by Peter Holm (pho).
The heavy lifting of coordinating all these efforts and bringing the
project to completion were done by Konstantin Belousov (kib).
Sponsored by: The FreeBSD Foundation (emaste, kib)
Differential revision: https://reviews.freebsd.org/D10439
The code specified the length of a layout as INT64_MAX instead of
UINT64_MAX. This could result in getting a layout for less than the
full file for extremely large files. Although having little practical
effect, this patch corrects this in the code.
Detected during recent testing of the pNFS server.
MFC after: 2 weeks
The nfsv4_seqsession() call returns NFSERR_REPLYFROMCACHE when it has a
reply in the session, due to a requestor retry. The code erroneously
assumed a return of 0 for this case. This patch fixes this and adds
a KASSERT(). This would be an extremely rare occurrence. It was found
during code inspection during the pNFS server development.
MFC after: 2 weeks
An NFSv4 server has the option of allowing a Read to be done using a Write
Open. If this is not allowed, the server will return NFSERR_OPENMODE.
This patch attempts the read with a write open and then disables this
if the server replies NFSERR_OPENMODE.
This change will avoid some uses of the special stateids. This will be
useful for pNFS/DS Reads, since they cannot use special stateids.
It will also be useful for any NFSv4 server that does not support reading
via the special stateids. It has been tested against both types of NFSv4 server.
MFC after: 2 weeks
The NFSv4.1/pNFS client does not use/need a backchannel for the Data Server (DS)
sessions, so the flag should only be set for MetaData Server (MDS) sessions.
This patch should have been a part of r317275.
MFC after: 2 weeks
The "return layout on close" case in the pNFS client was badly broken.
Fortunately, extant pNFS servers that I have tested against do not
do this. This patch fixes it. It also changes the way the layout stateid.seqid
is set for LayoutReturn. I think this change is correct w.r.t. the RFC,
but I am not 100% sure.
This was found during recent testing of the pNFS server under development.
MFC after: 2 weeks
The NFSv4.1/pNFS client wasn't doing a newnfs_disconnect() call for the
connection to the Data Server (DS) under some circumstances. The main
effect of this was a leak of malloc'd structures in the krpc. This patch
adds the newnfs_disconnect() calls to fix this.
Detected during recent testing against the pNFS server under development.
MFC after: 2 weeks
The nfscl_mtofh() function didn't check for failed operations and, as such,
would have returned EBADRPC for these cases, due to parsing failure.
This patch adds checks, so that it returns with ND_NOMOREDATA set.
This is needed for future use in the pNFS server and acts as a safety
belt in the meantime.
MFC after: 2 weeks
in place. To do per-cpu stats, convert all fields that previously were
maintained in the vmmeters that sit in pcpus to counter(9).
- Since some vmmeter stats may be touched at very early stages of boot,
before we have set up UMA and we can do counter_u64_alloc(), provide an
early counter mechanism:
o Leave one spare uint64_t in struct pcpu, named pc_early_dummy_counter.
o Point counter(9) fields of vmmeter to pcpu[0].pc_early_dummy_counter,
so that at early stages of boot, before counters are allocated we already
point to a counter that can be safely written to.
o For sparc64 that required a whole dummy pcpu[MAXCPU] array.
Further related changes:
- Don't include vmmeter.h into pcpu.h.
- vm.stats.vm.v_swappgsout and vm.stats.vm.v_swappgsin changed to 64-bit,
to match kernel representation.
- struct vmmeter hidden under _KERNEL, and only vmstat(1) is an exclusion.
This is based on benno@'s 4-year old patch:
https://lists.freebsd.org/pipermail/freebsd-arch/2013-July/014471.html
Reviewed by: kib, gallatin, marius, lidl
Differential Revision: https://reviews.freebsd.org/D10156
The "cred" argument of ncl_flush() is unused and it was confusing to have
the code passing in NULL for this argument in some cases. This patch deletes
this argument.
There is no semantic change because of this patch.
MFC after: 2 weeks
Some NFSv4.1 servers such as AmazonEFS can only support a small fixed number
of open_owner4s. This patch adds a mount option called "oneopenown" that
can be used for NFSv4.1 mounts to make the client do all Opens with the
same open_owner4 string. This option can only be used with NFSv4.1 and
may not work correctly when Delegations are is use.
Reported by: cperciva
Tested by: cperciva
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D8988
A function called svcpool_close() was added to the server side krpc by
r313735, so that a pool could be closed without destroying the data structures.
This little patch adds a call to it for the callback pool (svcpool_nfscbd),
so that the nfscbd daemon can be killed/restarted and continue to work
correctly.
MFC after: 2 weeks
When an mmap'd text file is written and then executed immediately
afterwards, it was possible that the modify time would change after the
text file was executing, resulting in the process executing the file
being killed. This was usually only observed when the file system's
times were set to higher resolution, but could have occurred for any
time resolution.
This was reported on a recent email list discussion.
This patch adds a VOP_SET_TEXT() to the NFS client which flushed all
dirty pages to the NFS server and then makes sure that n_mtime is up
to date to avoid this from occurring.
Thanks go to kib@ and pho@ for their help with developing this patch.
Tested by: pho
Reviewed by: kib
MFC after: 2 weeks
If the ExchangeID/CreateSession operations done by an NFSv4.1 client
after the server crashes/reboots fails, it is possible that some process/thread
is waiting for an open_owner lock. If the client state is free'd, this
can cause a crash.
This would not normally happen, but has been observed on a mount of the
AmazonEFS service.
Reported by: cperciva
Tested by: cperciva
PR: 216086
MFC after: 2 weeks
during recovery.
If the NFSv4.1 client gets a NFSv4.1 NFSERR_BADSESSION reply to an Open/Lock
operation while recovering from the server crash/reboot, allow the opens
to be retained for a subsequent recovery attempt. Since NFSv4.1 servers
should only reply NFSERR_BADSESSION after a crash/reboot that has lost
state, this case should almost never happen.
However, for the AmazonEFS file service, this has been observed when
the client does a fresh TCP connection for RPCs.
Reported by: cperciva
Tested by: cperciva
PR: 216088
MFC after: 2 weeks
Instead, issue a diagnostic and return appropriate error if
ncl_flush() was unable to clean buffer queue after the specified
number or retries.
Reviewed by: rmacklem
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
This patch gives a requestor of the exclusive lock on the client state
in the NFSv4 client priority over shared lock requestors. This avoids
the server crash recovery thread being starved out by other threads doing
RPCs.
Tested by: cperciva
PR: 216087
MFC after: 2 weeks
When the NFSv4 client Commit operation encountered a stale write verifier,
it erroneously mapped that to EIO. This could have caused recently written
data to be lost when a server crashes/reboots between an UNSTABLE write
and the subsequent commit. This patch fixes this.
The bug was only for the NFSv4 client and did not affect NFSv3.
Tested by: cperciva
PR: 215887
MFC after: 2 weeks
If an operation that preceeds a Setattr in an NFSv4 compound fails,
there is no bitmap of attributes to parse. Without this patch, the
parsing would fail and return EBADRPC instead of the correct failure
error. This could break recovery from a server crash/reboot.
Tested by: cperciva
PR: 215883
MFC after: 2 weeks
Write out the dirty pages using VOP_WRITE() instead of directly
calling ncl_writerpc(). The state of the buffers now reflects the
write, fixing some hard to diagnose consistency and write order
issues. The change also allowed to remove remapping of paged out
pages into kernel space and related allocation of the phys buffer.
Reviewed by: markj, rmacklem
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D10241
ncl_vinvalbuf() might need to upgrade vnode lock, allowing the vnode
to be reclaimed by other thread. Handle the situation, indicated by
the returned error zero and VI_DOOMED iflag set, converting it into
EBADF. Handle all calls, even where the vnode is exclusively locked
right now.
Reviewed by: markj, rmacklem
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
X-Differential revision: https://reviews.freebsd.org/D10241
Renumber cluase 4 to 3, per what everybody else did when BSD granted
them permission to remove clause 3. My insistance on keeping the same
numbering for legal reasons is too pedantic, so give up on that point.
Submitted by: Jan Schaumann <jschauma@stevens.edu>
Pull Request: https://github.com/freebsd/freebsd/pull/96
Right now this is not critical, but will be after planned increase of
MNAMELEN from 88 to 1k.
Reported and tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
For most NFSv4.1 servers, a NFS4ERR_BAD_SESSION error is a rare failure
that indicates that the server has lost session/open/lock state.
However, recent testing by cperciva@ against the AmazonEFS server found
several problems with client recovery from this due to it generating this
failure frequently.
Briefly, the problems fixed are:
- If all session slots were in use at the time of the failure, some processes
would continue to loop waiting for a slot on the old session forever.
- If an RPC that doesn't use open/lock state failed with NFS4ERR_BAD_SESSION,
it would fail the RPC/syscall instead of initiating recovery and then
looping to retry the RPC.
- If a successful reply to an RPC for an old session wasn't processed
until after a new session was created for a NFS4ERR_BAD_SESSION error,
it would erroneously update the new session and corrupt it.
- The use of the first element of the session list in the nfs mount
structure (which is always the current metadata session) was slightly
racey. With changes for the above problems it became more racey, so all
uses of this head pointer was wrapped with a NFSLOCKMNT()/NFSUNLOCKMNT().
- Although the kernel malloc() usually allocates more bytes than requested
and, as such, this wouldn't have caused problems, the allocation of a
session structure was 1 byte smaller than it should have been.
(Null termination byte for the string not included in byte count.)
There are probably still problems with a pNFS data server that fails
with NFS4ERR_BAD_SESSION, but I have no server that does this to test
against (the AmazonEFS server doesn't do pNFS), so I can't fix these yet.
Although this patch is fairly large, it should only affect the handling
of NFS4ERR_BAD_SESSION error replies from an NFSv4.1 server.
Thanks go to cperciva@ for the extension testing he did to help isolate/fix
these problems.
Reported by: cperciva
Tested by: cperciva
MFC after: 3 months
Differential Revision: https://reviews.freebsd.org/D8745
the vnode is inactivated. This contradicts with the nullfs caching
which keeps upper vnode around, as consequence keeping the use
reference to lower vnode.
Add a filesystem flag to request nullfs to not cache when mounted over
that filesystem, and set the flag for nfs v4 mounts.
Reported by: asomers
Reviewed by: rmacklem
Tested by: asomers, rmacklem
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
The pager, due to its construction, implements clustering for the
page-ins. In particular, buildworld load demonstrates reduction of
the READ RPCs from 39k down to 24k. No change in real or CPU time was
observed.
Discussed with, and measured by: bde
No objections from: rmacklem
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
the patch in D1626 plus changes so that it includes counts for
NFSv4.1 (and the draft of NFSv4.2).
Also, make all the counts uint64_t and add a vers field at the
beginning, so that future revisions can easily be implemented.
There is code in place to handle the old vesion of the nfsstats
structure for backwards binary compatibility.
Subsequent commits will update nfsstat(8) to use the new fields.
Submitted by: will (earlier version)
Reviewed by: ken
MFC after: 1 month
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D1626
executed by inactive methods, must be repeated on reclaim. In
particular, unlink and free sillyrenamed vnode both on inactivation
and reclaim.
Reported and tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Approved by: re (gjb)
If strlen(hostp) was zero, the stack array 'nam' would never be initialized
before being strdup()ed. Fix this by initializing it to the empty string.
It's possible some external condition makes this case impossible, in which
case, an assertion instead of this workaround is appropriate.
Introduced in r299848.
Reported by: Coverity
CID: 1355336
Sponsored by: EMC / Isilon Storage Division
eg an NFSv4 root over WiFi: boot from md_root (small rootfs image
preloaded by loader(8)), setup WiFi, and then reroot into the actual
root, over NFS.
Note that it's currently limited to NFSv4, and due to problems with
nfsuserd(8) it requres a workaround on the server side: one needs
to set the vfs.nfsd.enable_stringtouid=1 sysctl and not run nfsuserd(8)
on either the server or the client side.
Reviewed by: rmacklem@
MFC after: 1 month
Relnotes: yes
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D6347
consequence, the nfs client override of VOP_LOCK1() is no longer
needed.
Reviewed and tested by: rmacklem
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
the old and new NFS clients. He did a good job of isolating the problem
which was caused by the new NFS client not setting the post write mtime
correctly. The new NFS client code was cloned from the old client, but
was incorrect, because the mtime in the nfs vnode's cache wasn't yet
updated. This patch fixes this problem. The patch also adds missing mutex
locking.
Reported and tested by: bde
MFC after: 2 weeks
* Use standard IPv6 SAS instead of rt->rt_ifa address.
* Make address lookup work for IPv6 LLA.
* Save address into buffer provided by caller instead of using static vars.
Discussed with: rmacklem
With the new VOP_GETPAGES() KPI the "count" argument counts pages already,
and doesn't need to be translated from bytes to pages.
While here make it consistent that *rbehind and *rahead are updated only
if we doesn't return error.
Pointy hat to: glebius
o With new KPI consumers can request contiguous ranges of pages, and
unlike before, all pages will be kept busied on return, like it was
done before with the 'reqpage' only. Now the reqpage goes away. With
new interface it is easier to implement code protected from race
conditions.
Such arrayed requests for now should be preceeded by a call to
vm_pager_haspage() to make sure that request is possible. This
could be improved later, making vm_pager_haspage() obsolete.
Strenghtening the promises on the business of the array of pages
allows us to remove such hacks as swp_pager_free_nrpage() and
vm_pager_free_nonreq().
o New KPI accepts two integer pointers that may optionally point at
values for read ahead and read behind, that a pager may do, if it
can. These pages are completely owned by pager, and not controlled
by the caller.
This shifts the UFS-specific readahead logic from vm_fault.c, which
should be file system agnostic, into vnode_pager.c. It also removes
one VOP_BMAP() request per hard fault.
Discussed with: kib, alc, jeff, scottl
Sponsored by: Nginx, Inc.
Sponsored by: Netflix
the name of a filesystem when setting it as the first parameter to the
getnewvnode() function. Most filesystems call getnewvnode from just one
place so can use a literal string as the first parameter. However, NFS
calls getnewvnode from two places, so we create a global constant string
that can be used by the two instances. This change also collapses two
instances of getnewvnode() in the UFS filesystem to a single call.
Reviewed by: kib
Tested by: Peter Holm
At this time I cannot see a way to fix directory caching when it
has partial blocks in the buffer cache, due to the fact that the
syscall's uio_offset won't stay the same as the lblkno * NFS_DIRBLKSIZ
offset.
Reported by: bde
MFC after: 2 weeks
the largest size of buffer cache block or the mapping of the buffer
is bogus. When a mount with rsize=4096,wsize=4096 was done, f_iosize
would be set to 4096. This resulted in corrupted directory data, since
the buffer cache block size for directories is NFS_DIRBLKSIZ (8192).
This patch fixes the code so that it always sets f_iosize to at least
NFS_DIRBLKSIZ.
Tested by: krichy@cflinux.hu
PR: 177971
MFC after: 2 weeks
BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache
NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests
with attributes for totally different files).
Warn very verbosely when this is detected. Linux' NFS client has a similar
warning.
Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity
of warnings; default to 10. (Zero disables; -1 is unlimited.)
Adds a failpoint to aid in validating the warning / behavior with a
non-broken server. Use something like:
sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)'
Reviewed by: rmacklem
Approved by: markj (mentor)
Sponsored by: EMC / Isilon Storage Division
Differential Revision: https://reviews.freebsd.org/D3304
the kernel would generate a bogus one with a ":/<path>" suffix.
This would only occur for the case where there was no explicit
"principal" argument and the getaddrinfo() call in mount_nfs.c failed to a
return a cannonical name for the server.
This patch fixes this unusual case.
PR: 201073
Submitted by: masato@itc.naist.jp
MFC after: 2 weeks
compared to the old NFS client via email to the freebsd-fs@ mailing list.
For the new client, when multiple clients attempted to create a symbolic
link concurrently, more that one client would report success instead of
EEXIST. This was caused by code in the new client that mapped EEXIST to
OK assuming it was caused by a retried RPC request.
Since the old client did not do this, the patch defaults to the old
behaviour and permits the new behaviour to be enabled via a sysctl.
Reported by: alex.burlyga.ietf@gmail.com
Tested by: alex.burlyga.ietf@gmail.com
MFC after: 2 weeks
o Provide an extensive set of assertions for input array of pages.
o Remove now duplicate assertions from different pagers.
Sponsored by: Nginx, Inc.
Sponsored by: Netflix
so that it would not return less data than requested.
Since returning less directory data than requested is not a problem
for FreeBSD and even UFS no longer returns directory structures
with d_fileno == 0, this patch stops the client from doing this.
Although entries with d_fileno == 0 should not be a problem,
the man pages no longer document that these entries should be
ignored, so there was a concern that these entries might be an
issue in the future.
Suggested by: trasz
Tested by: trasz
MFC after: 2 weeks
that are not an exact multiple of DIRBLKSIZ correctly. Fortunately
readdir(3) always uses an exact multiple of DIRBLKSIZ, so few applications
were affected. This patch fixes this problem by reducing the size
of the directory read to an exact multiple of DIRBLKSIZ.
Tested by: trasz
Reported by: trasz
Reviewed by: trasz
MFC after: 2 weeks
Present implementation of large sync writes is too strict and so can be
quite slow. Instead of doing that, execute large async write in chunks,
syncing each chunk separately.
It would be good to fix large sync writes too, but I leave it to somebody
with more skills in this area.
Reviewed by: rmacklem
MFC after: 1 week
largest size for a buffer in the buffer cache. This patch
defines a new constant MAXBCACHEBUF, which is the largest
size for a buffer in the buffer cache. Having a separate
constant allows MAXBCACHEBUF to be set larger than MAXBSIZE
on a per-architecture basis, so that NFS can do larger read/writes
for these architectures. It modifies sys/param.h so that BKVASIZE
can also be set on a per-architecture basis.
A couple of cases where NFS used MAXBSIZE instead of NFS_MAXBSIZE
is fixed as well.
Differential Revision: https://reviews.freebsd.org/D2330
Reviewed by: mav, kib
MFC after: 2 weeks
This is similar to r281756 so set the ptr NULL after free as a safety belt
against future changes.
Obtained from: HardenedBSD (b2e77ced9ae213d358b44d98f552d9ae4636ecac)
Submitted by: Oliver Pinter
Revewed by: rmacklem
use VOP_FSYNC() to perform the NFS server's Commit operation.
This patch adds a mnt_kern_flag called MNTK_USES_BCACHE which
is set by file systems that use the buffer cache. If this flag
is not set, the NFS server always does a VOP_FSYNC().
This should be ok for old file system modules that do not set
MNTK_USES_BCACHE, since calling VOP_FSYNC() is correct, although
it might not be optimal for file systems that use the buffer cache.
Reviewed by: kib
MFC after: 2 weeks
- Allow to call the function with vm object lock held.
- Allow to specify reqpage that doesn't match any page in the region,
meaning freeing all pages.
o Utilize the new function in couple more places in vnode pager.
Reviewed by: alc, kib
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
use VA_UTIMES_NULL to indicate whether it should
set the time to the current tod on the server.
This had the side effect of making the NFS client
use the client's timestamp for exclusive create,
starting with FreeBSD9.2.
Unfortunately a bug in some Solaris NFS servers
causes these servers to return NFS_OK to the
Setattr RPC done during exclusive create, but not
actually set the file's mode, leaving the file's
mode == 0.
This patch restores the NFS client's behaviour to
use the server's tod for the exclusive open's
Setattr RPC, to avoid the Solaris server bug and
to restore the pre-FreeBSD9.2 NFS behaviour.
Discussed on: freebsd-fs
PR: 186293
MFC after: 3 months
exactly the same code is at the end of the nfscl_checksattr()
function that is called just before it. As such, this code
had already been executed and didn't do anything.
MFC after: 1 week
which means that the NFSCLIENT and NFSSERVER
kernel options will no longer work. This commit
only removes the kernel components. Removal of
unused code in the user utilities will be done
later. This commit does not include an addition
to UPDATING, but that will be committed in a
few minutes.
Discussed on: freebsd-fs
into namecache, to avoid cache trashing when doing large operations.
E.g., tar archive extraction is not usually followed by access to many
of the files created.
Right now, each VOP_LOOKUP() implementation explicitely knowns about
this quirk and tests for both MAKEENTRY flag presence and op != CREATE
to make the call to cache_enter(). Centralize the handling of the
quirk into VFS, by deciding to cache only by MAKEENTRY flag in VOP.
VFS now sets NOCACHE flag for CREATE namei() calls.
Note that the change in semantic is backward-compatible and could be
merged to the stable branch, and is compatible with non-changed
third-party filesystems which correctly handle MAKEENTRY.
Suggested by: Chris Torek <torek@pi-coral.com>
Reviewed by: mckusick
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
to mount_nfs(8). They are implemented on Linux, OS X, and Solaris,
and thus can be expected to appear in automounter maps.
Reviewed by: rmacklem@
MFC after: 1 month
Sponsored by: The FreeBSD Foundation
to a power of 2. For non-power of 2 settings, intermittent
page faults have been reported. Although the bug that causes
these page faults/crashes has not been identified, it does
not appear to occur when rsize, wsize is a power of 2.
Reported by: tcberner@gmail.com
MFC after: 2 weeks
to a power of 2. For non-power of 2 settings, intermittent
page faults have been reported. Although the bug that causes
these page faults/crashes has not been identified, it does
not appear to occur when rsize, wsize is a power of 2.
Reported by: tcberner@gmail.com
MFC after: 2 weeks
in userland rename in-kernel getenv()/setenv() to kern_setenv()/kern_getenv().
This fixes a namespace collision with libc symbols.
Submitted by: kmacy
Tested by: make universe
path through the NFS clients' getpages functions.
Introduce vm_pager_free_nonreq(). This function can be used to eliminate
code that is duplicated in many getpages functions. Also, in contrast to
the code that currently appears in those getpages functions,
vm_pager_free_nonreq() avoids acquiring an exclusive object lock in one
case.
Reviewed by: kib
MFC after: 6 weeks
Sponsored by: EMC / Isilon Storage Division
forcing filesystem VOP_LINK() methods to repeat the code. In
tmpfs_link(), remove redundand check for the type of the source,
already done by VFS.
Note that NFS server already performs this check before calling
VOP_LINK().
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
into head. The code is not believed to have any effect
on the semantics of non-NFSv4.1 server behaviour.
It is a rather large merge, but I am hoping that there will
not be any regressions for the NFS server.
MFC after: 1 month
when a newly created file has another open done on it that
update the open mode. This patch moves the code that updates
the open mode up into the block where the mutex is held to
ensure this cannot happen. No bug caused by this potential
race has been observed, but this fix is a safety belt to ensure
it cannot happen.
MFC after: 2 weeks
post-create/mkdir directory attributes. This allows the RPC to
name cache the newly created directory and reduces the lookup RPC
count for applications creating a lot of directories.
MFC after: 2 weeks
post-open/create directory attributes. This allows the RPC to
name cache the newly created file and reduces the lookup RPC
count by about 10% for software builds.
MFC after: 2 weeks
attributes. This allows the client to cache directory names
when they are looked up, reducing the Lookup RPC count by
about 40% for software builds.
MFC after: 2 weeks
further refinement is required as some device drivers intended to be
portable over FreeBSD versions rely on __FreeBSD_version to decide whether
to include capability.h.
MFC after: 3 weeks
nfsv4_fillattr() as NULLs for the Getattr callback. This caused
nfsv4_fillattr() to not fill in the Change attribute for the reply.
I believe this was a violation of the RFC, but had little effect on
server behaviour. This patch passes a non-NULL p argument to fix this.
MFC after: 1 week
for the Getattr and Recall callbacks. This patch fixes it.
Since the NFSv4.1 specific error codes would only happen for
abnormal circumstances, this patch has little effect, in practice.
MFC after: 1 week
synchronous (with FILE_SYNC) writes because non-contiguous
byte ranges in the same buffer cache block are being
written. This patch adds a new mount option "noncontigwr"
which allows the non-contiguous byte ranges to be combined,
with the dirty byte range becoming the superset of the bytes
that are dirty, if the file has not been file locked.
This reduces the number of writes significantly for software
builds. The only case where this change might break existing
applications is where an application is writing
non-overlapping byte ranges within the same buffer cache block
of a file from multiple clients concurrently.
Since such an application would normally do file locking on
the file, avoiding the byte range merge for files that have
been file locked should be sufficient for most (maybe all?) cases.
Submitted by: jhb (earlier version)
Reviewed by: kib
MFC after: 3 weeks
option, unbreak the lock tracing release semantic by embedding
calls to LOCKSTAT_PROFILE_RELEASE_LOCK() direclty in the inlined
version of the releasing functions for mutex, rwlock and sxlock.
Failing to do so skips the lockstat_probe_func invokation for
unlocking.
- As part of the LOCKSTAT support is inlined in mutex operation, for
kernel compiled without lock debugging options, potentially every
consumer must be compiled including opt_kdtrace.h.
Fix this by moving KDTRACE_HOOKS into opt_global.h and remove the
dependency by opt_kdtrace.h for all files, as now only KDTRACE_FRAMES
is linked there and it is only used as a compile-time stub [0].
[0] immediately shows some new bug as DTRACE-derived support for debug
in sfxge is broken and it was never really tested. As it was not
including correctly opt_kdtrace.h before it was never enabled so it
was kept broken for a while. Fix this by using a protection stub,
leaving sfxge driver authors the responsibility for fixing it
appropriately [1].
Sponsored by: EMC / Isilon storage division
Discussed with: rstone
[0] Reported by: rstone
[1] Discussed with: philip
The hang occurred in nfsv4_setsequence() when it couldn't find an
available session slot and is fixed by checking for a forced dismount
in progress and just returning for this case.
MFC after: 1 month
in the future in a backward compatible (API and ABI) way.
The cap_rights_t represents capability rights. We used to use one bit to
represent one right, but we are running out of spare bits. Currently the new
structure provides place for 114 rights (so 50 more than the previous
cap_rights_t), but it is possible to grow the structure to hold at least 285
rights, although we can make it even larger if 285 rights won't be enough.
The structure definition looks like this:
struct cap_rights {
uint64_t cr_rights[CAP_RIGHTS_VERSION + 2];
};
The initial CAP_RIGHTS_VERSION is 0.
The top two bits in the first element of the cr_rights[] array contain total
number of elements in the array - 2. This means if those two bits are equal to
0, we have 2 array elements.
The top two bits in all remaining array elements should be 0.
The next five bits in all array elements contain array index. Only one bit is
used and bit position in this five-bits range defines array index. This means
there can be at most five array elements in the future.
To define new right the CAPRIGHT() macro must be used. The macro takes two
arguments - an array index and a bit to set, eg.
#define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL)
We still support aliases that combine few rights, but the rights have to belong
to the same array element, eg:
#define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL)
#define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL)
#define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP)
There is new API to manage the new cap_rights_t structure:
cap_rights_t *cap_rights_init(cap_rights_t *rights, ...);
void cap_rights_set(cap_rights_t *rights, ...);
void cap_rights_clear(cap_rights_t *rights, ...);
bool cap_rights_is_set(const cap_rights_t *rights, ...);
bool cap_rights_is_valid(const cap_rights_t *rights);
void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src);
void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src);
bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little);
Capability rights to the cap_rights_init(), cap_rights_set(),
cap_rights_clear() and cap_rights_is_set() functions are provided by
separating them with commas, eg:
cap_rights_t rights;
cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT);
There is no need to terminate the list of rights, as those functions are
actually macros that take care of the termination, eg:
#define cap_rights_set(rights, ...) \
__cap_rights_set((rights), __VA_ARGS__, 0ULL)
void __cap_rights_set(cap_rights_t *rights, ...);
Thanks to using one bit as an array index we can assert in those functions that
there are no two rights belonging to different array elements provided
together. For example this is illegal and will be detected, because CAP_LOOKUP
belongs to element 0 and CAP_PDKILL to element 1:
cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL);
Providing several rights that belongs to the same array's element this way is
correct, but is not advised. It should only be used for aliases definition.
This commit also breaks compatibility with some existing Capsicum system calls,
but I see no other way to do that. This should be fine as Capsicum is still
experimental and this change is not going to 9.x.
Sponsored by: The FreeBSD Foundation
is being shut down which were caused by the nfscbd_pool being
destroyed before the backchannel is disabled. This patch is
believed to fix the problem, by simply avoiding ever destroying
the nfscbd_pool. Since the NFS client module cannot be unloaded,
this should not cause a memory leak.
MFC after: 2 weeks
waiting for an RPC reply from the server while holding the mount
point busy (mnt_lockref incremented). This happens because dounmount()
msleep()s waiting for mnt_lockref to become 0, before calling
VFS_UNMOUNT(). This patch adds a new VFS operation called VFS_PURGE(),
which the NFS client implements as purging RPCs in progress. Making
this call before checking mnt_lockref fixes the problem, by ensuring
that the VOP_xxx() calls will fail and unbusy the mount point.
Reported by: sbruno
Reviewed by: kib
MFC after: 2 weeks
credentials to the kernel rpc. Modify the NFSv4 client to add
support for the gssname and allgssname mount options to use this
capability. Requires the gssd daemon to be running with the "-h" option.
Reviewed by: jhb
would sometimes result in a corrupted file was reported via email.
This problem appears to have been caused by r251719 (reverting
r251719 fixed the problem). Although I have not been able to
reproduce this problem, I suspect it is caused by another thread
increasing np->n_size after the mtx_unlock(&np->n_mtx) but before
the vnode_pager_setsize() call. Since the np->n_mtx mutex serializes
updates to np->n_size, doing the vnode_pager_setsize() with the
mutex locked appears to avoid the problem.
Unfortunately, vnode_pager_setsize() where the new size is smaller,
cannot be called with a mutex held.
This patch returns the semantics to be close to pre-r251719 (actually
pre-r248567, r248581, r248567 for the new client) such that the call to
vnode_pager_setsize() is only delayed until after the mutex is
unlocked when np->n_size is shrinking. Since the file is growing
when being written, I believe this will fix the corruption.
A better solution might be to replace the mutex with a sleep lock,
but that is a non-trivial conversion, so this fix is hoped to be
sufficient in the meantime.
Reported by: David G. Lawrence (dg@dglawrence.com)
Tested by: David G. Lawrence (to be done soon)
Reviewed by: kib
MFC after: 1 week
file's size attribute does not get updated. As such, it is necessary
to invalidate the attribute cache before clearing NMODIFIED for pNFS.
MFC after: 2 weeks
enable use of the (no)resvport mount option for NFSv4. I had thought
that the RFC required that non-reserved port #s be allowed, but I couldn't
find it in the RFC.
MFC after: 2 weeks
- Use a shared bufobj lock in getblk() and inmem().
- Convert softdep's lk to rwlock to match the bufobj lock.
- Move INFREECNT to b_flags and protect it with the buf lock.
- Remove unnecessary locking around bremfree() and BKGRDINPROG.
Sponsored by: EMC / Isilon Storage Division
Discussed with: mckusick, kib, mdf
truncated directory for some NFS servers. This turned out to
be because the size of a directory reported by an NFS server
can be smaller that the ufs-like directory created from the
RPC XDR in the client. This patch fixes the problem by changing
r248567 so that vnode_pager_setsize() is only done for regular files.
Reported and tested by: hartmut.brandt@dlr.de
Reviewed by: kib
MFC after: 1 week
buffer for the last vnode on the mount back to the server, it
returns. At that point, the code continues with the unmount,
including freeing up the nfs specific part of the mount structure.
It is possible that an nfsiod thread will try to check for an
empty I/O queue in the nfs specific part of the mount structure
after it has been free'd by the unmount. This patch avoids this problem by
setting the iodmount entries for the mount back to NULL while holding the
mutex in the unmount and checking the appropriate entry is non-NULL after
acquiring the mutex in the nfsiod thread.
Reported and tested by: pho
Reviewed by: kib
MFC after: 2 weeks
option. This can occur when an nfsiod thread that already holds
a buffer lock attempts to acquire a vnode lock on an entry in
the directory (a LOR) when another thread holding the vnode lock
is waiting on an nfsiod thread. This patch avoids the deadlock by disabling
readahead for this case, so the nfsiod threads never do readdirplus.
Since readaheads for directories need the directory offset cookie
from the previous read, they cannot normally happen in parallel.
As such, testing by jhb@ and myself didn't find any performance
degredation when this patch is applied. If there is a case where
this results in a significant performance degradation, mounting
without the "rdirplus" option can be done to re-enable readahead
for directories.
Reported and tested by: jhb
Reviewed by: jhb
MFC after: 2 weeks
it will work with either the old or new server.
The FHA code keeps a cache of currently active file handles for
NFSv2 and v3 requests, so that read and write requests for the same
file are directed to the same group of threads (reads) or thread
(writes). It does not currently work for NFSv4 requests. They are
more complex, and will take more work to support.
This improves read-ahead performance, especially with ZFS, if the
FHA tuning parameters are configured appropriately. Without the
FHA code, concurrent reads that are part of a sequential read from
a file will be directed to separate NFS threads. This has the
effect of confusing the ZFS zfetch (prefetch) code and makes
sequential reads significantly slower with clients like Linux that
do a lot of prefetching.
The FHA code has also been updated to direct write requests to nearby
file offsets to the same thread in the same way it batches reads,
and the FHA code will now also send writes to multiple threads when
needed.
This improves sequential write performance in ZFS, because writes
to a file are now more ordered. Since NFS writes (generally
less than 64K) are smaller than the typical ZFS record size
(usually 128K), out of order NFS writes to the same block can
trigger a read in ZFS. Sending them down the same thread increases
the odds of their being in order.
In order for multiple write threads per file in the FHA code to be
useful, writes in the NFS server have been changed to use a LK_SHARED
vnode lock, and upgrade that to LK_EXCLUSIVE if the filesystem
doesn't allow multiple writers to a file at once. ZFS is currently
the only filesystem that allows multiple writers to a file, because
it has internal file range locking. This change does not affect the
NFSv4 code.
This improves random write performance to a single file in ZFS, since
we can now have multiple writers inside ZFS at one time.
I have changed the default tuning parameters to a 22 bit (4MB)
window size (from 256K) and unlimited commands per thread as a
result of my benchmarking with ZFS.
The FHA code has been updated to allow configuring the tuning
parameters from loader tunable variables in addition to sysctl
variables. The read offset window calculation has been slightly
modified as well. Instead of having separate bins, each file
handle has a rolling window of bin_shift size. This minimizes
glitches in throughput when shifting from one bin to another.
sys/conf/files:
Add nfs_fha_new.c and nfs_fha_old.c. Compile nfs_fha.c
when either the old or the new NFS server is built.
sys/fs/nfs/nfsport.h,
sys/fs/nfs/nfs_commonport.c:
Bring in changes from Rick Macklem to newnfs_realign that
allow it to operate in blocking (M_WAITOK) or non-blocking
(M_NOWAIT) mode.
sys/fs/nfs/nfs_commonsubs.c,
sys/fs/nfs/nfs_var.h:
Bring in a change from Rick Macklem to allow telling
nfsm_dissect() whether or not to wait for mallocs.
sys/fs/nfs/nfsm_subs.h:
Bring in changes from Rick Macklem to create a new
nfsm_dissect_nonblock() inline function and
NFSM_DISSECT_NONBLOCK() macro.
sys/fs/nfs/nfs_commonkrpc.c,
sys/fs/nfsclient/nfs_clkrpc.c:
Add the malloc wait flag to a newnfs_realign() call.
sys/fs/nfsserver/nfs_nfsdkrpc.c:
Setup the new NFS server's RPC thread pool so that it will
call the FHA code.
Add the malloc flag argument to newnfs_realign().
Unstaticize newnfs_nfsv3_procid[] so that we can use it in
the FHA code.
sys/fs/nfsserver/nfs_nfsdsocket.c:
In nfsrvd_dorpc(), add NFSPROC_WRITE to the list of RPC types
that use the LK_SHARED lock type.
sys/fs/nfsserver/nfs_nfsdport.c:
In nfsd_fhtovp(), if we're starting a write, check to see
whether the underlying filesystem supports shared writes.
If not, upgrade the lock type from LK_SHARED to LK_EXCLUSIVE.
sys/nfsserver/nfs_fha.c:
Remove all code that is specific to the NFS server
implementation. Anything that is server-specific is now
accessed through a callback supplied by that server's FHA
shim in the new softc.
There are now separate sysctls and tunables for the FHA
implementations for the old and new NFS servers. The new
NFS server has its tunables under vfs.nfsd.fha, the old
NFS server's tunables are under vfs.nfsrv.fha as before.
In fha_extract_info(), use callouts for all server-specific
code. Getting file handles and offsets is now done in the
individual server's shim module.
In fha_hash_entry_choose_thread(), change the way we decide
whether two reads are in proximity to each other.
Previously, the calculation was a simple shift operation to
see whether the offsets were in the same power of 2 bucket.
The issue was that there would be a bucket (and therefore
thread) transition, even if the reads were in close
proximity. When there is a thread transition, reads wind
up going somewhat out of order, and ZFS gets confused.
The new calculation simply tries to see whether the offsets
are within 1 << bin_shift of each other. If they are, the
reads will be sent to the same thread.
The effect of this change is that for sequential reads, if
the client doesn't exceed the max_reqs_per_nfsd parameter
and the bin_shift is set to a reasonable value (22, or
4MB works well in my tests), the reads in any sequential
stream will largely be confined to a single thread.
Change fha_assign() so that it takes a softc argument. It
is now called from the individual server's shim code, which
will pass in the softc.
Change fhe_stats_sysctl() so that it takes a softc
parameter. It is now called from the individual server's
shim code. Add the current offset to the list of things
printed out about each active thread.
Change the num_reads and num_writes counters in the
fha_hash_entry structure to 32-bit values, and rename them
num_rw and num_exclusive, respectively, to reflect their
changed usage.
Add an enable sysctl and tunable that allows the user to
disable the FHA code (when vfs.XXX.fha.enable = 0). This
is useful for before/after performance comparisons.
nfs_fha.h:
Move most structure definitions out of nfs_fha.c and into
the header file, so that the individual server shims can
see them.
Change the default bin_shift to 22 (4MB) instead of 18
(256K). Allow unlimited commands per thread.
sys/nfsserver/nfs_fha_old.c,
sys/nfsserver/nfs_fha_old.h,
sys/fs/nfsserver/nfs_fha_new.c,
sys/fs/nfsserver/nfs_fha_new.h:
Add shims for the old and new NFS servers to interface with
the FHA code, and callbacks for the
The shims contain all of the code and definitions that are
specific to the NFS servers.
They setup the server-specific callbacks and set the server
name for the sysctl and loader tunable variables.
sys/nfsserver/nfs_srvkrpc.c:
Configure the RPC code to call fhaold_assign() instead of
fha_assign().
sys/modules/nfsd/Makefile:
Add nfs_fha.c and nfs_fha_new.c.
sys/modules/nfsserver/Makefile:
Add nfs_fha_old.c.
Reviewed by: rmacklem
Sponsored by: Spectra Logic
MFC after: 2 weeks
locked. vnode_pager_setsize() might sleep waiting for the page after
EOF be unbusied.
Call vnode_pager_setsize() both for the regular and directory vnodes.
Reported by: mich
Reviewed by: rmacklem
Discussed with: avg, jhb
MFC after: 2 weeks
In common configurations biosize is a power of two, but is not required to
be so. Thanks to markj@ for spotting an additional case beyond my original
patch.
Reviewed by: rmacklem@
future further optimizations where the vm_object lock will be held
in read mode most of the time the page cache resident pool of pages
are accessed for reading purposes.
The change is mostly mechanical but few notes are reported:
* The KPI changes as follow:
- VM_OBJECT_LOCK() -> VM_OBJECT_WLOCK()
- VM_OBJECT_TRYLOCK() -> VM_OBJECT_TRYWLOCK()
- VM_OBJECT_UNLOCK() -> VM_OBJECT_WUNLOCK()
- VM_OBJECT_LOCK_ASSERT(MA_OWNED) -> VM_OBJECT_ASSERT_WLOCKED()
(in order to avoid visibility of implementation details)
- The read-mode operations are added:
VM_OBJECT_RLOCK(), VM_OBJECT_TRYRLOCK(), VM_OBJECT_RUNLOCK(),
VM_OBJECT_ASSERT_RLOCKED(), VM_OBJECT_ASSERT_LOCKED()
* The vm/vm_pager.h namespace pollution avoidance (forcing requiring
sys/mutex.h in consumers directly to cater its inlining functions
using VM_OBJECT_LOCK()) imposes that all the vm/vm_pager.h
consumers now must include also sys/rwlock.h.
* zfs requires a quite convoluted fix to include FreeBSD rwlocks into
the compat layer because the name clash between FreeBSD and solaris
versions must be avoided.
At this purpose zfs redefines the vm_object locking functions
directly, isolating the FreeBSD components in specific compat stubs.
The KPI results heavilly broken by this commit. Thirdy part ports must
be updated accordingly (I can think off-hand of VirtualBox, for example).
Sponsored by: EMC / Isilon storage division
Reviewed by: jeff
Reviewed by: pjd (ZFS specific review)
Discussed with: alc
Tested by: pho
- Capability is no longer separate descriptor type. Now every descriptor
has set of its own capability rights.
- The cap_new(2) system call is left, but it is no longer documented and
should not be used in new code.
- The new syscall cap_rights_limit(2) should be used instead of
cap_new(2), which limits capability rights of the given descriptor
without creating a new one.
- The cap_getrights(2) syscall is renamed to cap_rights_get(2).
- If CAP_IOCTL capability right is present we can further reduce allowed
ioctls list with the new cap_ioctls_limit(2) syscall. List of allowed
ioctls can be retrived with cap_ioctls_get(2) syscall.
- If CAP_FCNTL capability right is present we can further reduce fcntls
that can be used with the new cap_fcntls_limit(2) syscall and retrive
them with cap_fcntls_get(2).
- To support ioctl and fcntl white-listing the filedesc structure was
heavly modified.
- The audit subsystem, kdump and procstat tools were updated to
recognize new syscalls.
- Capability rights were revised and eventhough I tried hard to provide
backward API and ABI compatibility there are some incompatible changes
that are described in detail below:
CAP_CREATE old behaviour:
- Allow for openat(2)+O_CREAT.
- Allow for linkat(2).
- Allow for symlinkat(2).
CAP_CREATE new behaviour:
- Allow for openat(2)+O_CREAT.
Added CAP_LINKAT:
- Allow for linkat(2). ABI: Reuses CAP_RMDIR bit.
- Allow to be target for renameat(2).
Added CAP_SYMLINKAT:
- Allow for symlinkat(2).
Removed CAP_DELETE. Old behaviour:
- Allow for unlinkat(2) when removing non-directory object.
- Allow to be source for renameat(2).
Removed CAP_RMDIR. Old behaviour:
- Allow for unlinkat(2) when removing directory.
Added CAP_RENAMEAT:
- Required for source directory for the renameat(2) syscall.
Added CAP_UNLINKAT (effectively it replaces CAP_DELETE and CAP_RMDIR):
- Allow for unlinkat(2) on any object.
- Required if target of renameat(2) exists and will be removed by this
call.
Removed CAP_MAPEXEC.
CAP_MMAP old behaviour:
- Allow for mmap(2) with any combination of PROT_NONE, PROT_READ and
PROT_WRITE.
CAP_MMAP new behaviour:
- Allow for mmap(2)+PROT_NONE.
Added CAP_MMAP_R:
- Allow for mmap(PROT_READ).
Added CAP_MMAP_W:
- Allow for mmap(PROT_WRITE).
Added CAP_MMAP_X:
- Allow for mmap(PROT_EXEC).
Added CAP_MMAP_RW:
- Allow for mmap(PROT_READ | PROT_WRITE).
Added CAP_MMAP_RX:
- Allow for mmap(PROT_READ | PROT_EXEC).
Added CAP_MMAP_WX:
- Allow for mmap(PROT_WRITE | PROT_EXEC).
Added CAP_MMAP_RWX:
- Allow for mmap(PROT_READ | PROT_WRITE | PROT_EXEC).
Renamed CAP_MKDIR to CAP_MKDIRAT.
Renamed CAP_MKFIFO to CAP_MKFIFOAT.
Renamed CAP_MKNODE to CAP_MKNODEAT.
CAP_READ old behaviour:
- Allow pread(2).
- Disallow read(2), readv(2) (if there is no CAP_SEEK).
CAP_READ new behaviour:
- Allow read(2), readv(2).
- Disallow pread(2) (CAP_SEEK was also required).
CAP_WRITE old behaviour:
- Allow pwrite(2).
- Disallow write(2), writev(2) (if there is no CAP_SEEK).
CAP_WRITE new behaviour:
- Allow write(2), writev(2).
- Disallow pwrite(2) (CAP_SEEK was also required).
Added convinient defines:
#define CAP_PREAD (CAP_SEEK | CAP_READ)
#define CAP_PWRITE (CAP_SEEK | CAP_WRITE)
#define CAP_MMAP_R (CAP_MMAP | CAP_SEEK | CAP_READ)
#define CAP_MMAP_W (CAP_MMAP | CAP_SEEK | CAP_WRITE)
#define CAP_MMAP_X (CAP_MMAP | CAP_SEEK | 0x0000000000000008ULL)
#define CAP_MMAP_RW (CAP_MMAP_R | CAP_MMAP_W)
#define CAP_MMAP_RX (CAP_MMAP_R | CAP_MMAP_X)
#define CAP_MMAP_WX (CAP_MMAP_W | CAP_MMAP_X)
#define CAP_MMAP_RWX (CAP_MMAP_R | CAP_MMAP_W | CAP_MMAP_X)
#define CAP_RECV CAP_READ
#define CAP_SEND CAP_WRITE
#define CAP_SOCK_CLIENT \
(CAP_CONNECT | CAP_GETPEERNAME | CAP_GETSOCKNAME | CAP_GETSOCKOPT | \
CAP_PEELOFF | CAP_RECV | CAP_SEND | CAP_SETSOCKOPT | CAP_SHUTDOWN)
#define CAP_SOCK_SERVER \
(CAP_ACCEPT | CAP_BIND | CAP_GETPEERNAME | CAP_GETSOCKNAME | \
CAP_GETSOCKOPT | CAP_LISTEN | CAP_PEELOFF | CAP_RECV | CAP_SEND | \
CAP_SETSOCKOPT | CAP_SHUTDOWN)
Added defines for backward API compatibility:
#define CAP_MAPEXEC CAP_MMAP_X
#define CAP_DELETE CAP_UNLINKAT
#define CAP_MKDIR CAP_MKDIRAT
#define CAP_RMDIR CAP_UNLINKAT
#define CAP_MKFIFO CAP_MKFIFOAT
#define CAP_MKNOD CAP_MKNODAT
#define CAP_SOCK_ALL (CAP_SOCK_CLIENT | CAP_SOCK_SERVER)
Sponsored by: The FreeBSD Foundation
Reviewed by: Christoph Mallon <christoph.mallon@gmx.de>
Many aspects discussed with: rwatson, benl, jonathan
ABI compatibility discussed with: kib
changes in r246417 were incomplete as they did not add explicit calls to
sigdeferstop() around all the places that previously passed SBDRY to
_sleep(). In addition, nfs_getcacheblk() could trigger a write RPC from
getblk() resulting in sigdeferstop() recursing. Rather than manually
deferring stop signals in specific places, change the VFS_*() and VOP_*()
methods to defer stop signals for filesystems which request this behavior
via a new VFCF_SBDRY flag. Note that this has to be a VFC flag rather than
a MNTK flag so that it works properly with VFS_MOUNT() when the mount is
not yet fully constructed. For now, only the NFS clients are set this new
flag in VFS_SET().
A few other related changes:
- Add an assertion to ensure that TDF_SBDRY doesn't leak to userland.
- When a lookup request uses VOP_READLINK() to follow a symlink, mark
the request as being on behalf of the thread performing the lookup
(cnp_thread) rather than using a NULL thread pointer. This causes
NFS to properly handle signals during this VOP on an interruptible
mount.
PR: kern/176179
Reported by: Russell Cattelan (sigdeferstop() recursion)
Reviewed by: kib
MFC after: 1 month
requested from the server for the read operation. Server shall not
reply with too large size, but client should be resilent too.
Reviewed by: rmacklem
MFC after: 1 week
- Use NFSD_MONOSEC (which maps to time_uptime) instead of the seconds
portion of wall-time stamps to manage timeouts on events.
- Remove unused nd_starttime from the per-request structure in the new
NFS server.
- Use nanotime() for the modification time on a delegation to get as
precise a time as possible.
- Use time_second instead of extracting the second from a call to
getmicrotime().
Submitted by: bde (3)
Reviewed by: bde, rmacklem
MFC after: 2 weeks
to head. I don't think the NFS client behaviour will change unless
the new "minorversion=1" mount option is used. It includes basic
NFSv4.1 support plus support for pNFS using the Files Layout only.
All problems detecting during an NFSv4.1 Bakeathon testing event
in June 2012 have been resolved in this code and it has been tested
against the NFSv4.1 server available to me.
Although not reviewed, I believe that kib@ has looked at it.
which dumps out the actual options being used by an NFS mount.
This will be used to implement a "-m" option for nfsstat(1).
Reviewed by: alfred
MFC after: 2 weeks
received granular locking) but the comment present in UFS has been
copied all over other filesystems code incorrectly for several times.
Removes comments that makes no sense now.
Reviewed by: kib
MFC after: 3 days
zombie list for the pid. This allows several kern.proc sysctls to
report useful information for zombies.
Hold the allproc_lock around all searches instead of relocking it.
Remove private pfind_locked() from the new nfs client code.
Requested and reviewed by: pjd
Tested by: pho
MFC after: 3 weeks
and owner_group strings that consist entirely of
digits, interpreting them as the uid/gid number.
This change was needed since new (>= 3.3) Linux
servers reply with these strings by default.
This change is mandated by the rfc3530bis draft.
Reported on freebsd-stable@ under the Subject
heading "Problem with Linux >= 3.3 as NFSv4 server"
by Norbert Aschendorff on Aug. 20, 2012.
Tested by: norbert.aschendorff at yahoo.de
Reviewed by: jhb
MFC after: 2 weeks
Use it for a printf() that can be harmlessly generated for mmap()'d
files. It will be used extensively for the NFSv4.1 client.
Debugging printf()s are enabled by setting vfs.nfs.debuglevel to
a non-zero value. The higher the value, the more debugging printf()s.
Reviewed by: jhb
MFC after: 2 weeks
network file systems (not only NFS proper). Short reads cause pages
other then the requested one, which were not filled by read response,
to stay invalid.
Change the vm_page_readahead_finish() interface to not take the error
code, but instead to make a decision to free or to (de)activate the
page only by its validity. As result, not requested invalid pages are
freed even if the read RPC indicated success.
Noted and reviewed by: alc
MFC after: 1 week
to pull vm_param.h was removed. Other big dependency of vm_page.h on
vm_param.h are PA_LOCK* definitions, which are only needed for
in-kernel code, because modules use KBI-safe functions to lock the
pages.
Stop including vm_param.h into vm_page.h. Include vm_param.h
explicitely for the kernel code which needs it.
Suggested and reviewed by: alc
MFC after: 2 weeks
vm_page oflags by providing helper function
vm_page_readahead_finish(), which handles completed reads for pages
with indexes other then the requested one, for VOP_GETPAGES().
Reviewed by: alc
MFC after: 1 week
bwrite(). VFS needs to know about EFAULT from uiomove() and does not
care much that partially filled block writeback after EFAULT was
successfull. Early return without error causes short write to be
reported to usermode.
Reported and tested by: andreast
MFC after: 3 weeks
written, but not msync'd by a process. A VOP_PUTPAGES()
called when VOP_RECLAIM() happens will usually fail, since
the NFSv4 Open has already been closed by VOP_INACTIVE().
Add a vm_object_page_clean() call to the NFSv4 client's
VOP_INACTIVE(), so that the write happens before the NFSv4
Open is closed. kib@ suggested using vgone() instead and
I will explore this, but this patch fixes things in the
meantime. For some reason, the VOP_PUTPAGES() is still
attaempted in VOP_RECLAIM(), but having this fail doesn't
cause any problems except a "stateid0 in write" being logged.
Reviewed by: kib
MFC after: 1 week
Do not brelse() the buffer unconditionally with BIO_ERROR set if
uiomove() failed. The brelse() treats most buffers with BIO_ERROR as
B_INVAL, dropping their content. Instead, if the write request
covered the whole buffer, remember the cached state and brelse() with
BIO_ERROR set only if the buffer was not cached previously.
Update the buffer dirtyoff/dirtyend based on the progress recorded by
uiomove() in passed struct uio, even in the presence of
error. Otherwise, usermode could see changed data in the backed pages,
but later the buffer is destroyed without write-back.
If uiomove() failed for IO_UNIT request, try to truncate the vnode
back to the pre-write state, and rewind the progress in passed uio
accordingly, following the FFS behaviour.
Reviewed by: rmacklem (some time ago)
Tested by: pho
MFC after: 1 month
memory mapped pages being written back on an NFS mount.
Since any thread can call VOP_PUTPAGES() to write back a
dirty page, the credentials of that thread may not have
write access to the file on an NFS server. (Often the uid
is 0, which may be mapped to "nobody" in the NFS server.)
Although there is no completely correct fix for this
(NFS servers check access on every write RPC instead of at
open/mmap time), this patch avoids the common cases by
holding onto a credential that recently opened the file
for writing and uses that credential for the write RPCs
being done by VOP_PUTPAGES() for both NFS clients.
Tested by: Joel Ray Holveck (joelh at juniper.net)
PR: kern/165923
Reviewed by: kib
MFC after: 2 weeks
do not include file attributes in the reply to an NFS create RPC
under certain circumstances.
This resulted in a vnode of type VNON that was not usable.
This patch adds an NFS getattr RPC to nfs_create() for this case,
to fix the problem. It was tested by the person that reported
the problem and confirmed to fix this case for their server.
Tested by: Steven Haber (steven.haber at isilon.com)
MFC after: 2 weeks
The primary changes are that the user of the interface no longer
needs to manage the mount-mutex locking and that the vnode that
is returned has its mutex locked (thus avoiding the need to check
to see if its is DOOMED or other possible end of life senarios).
To minimize compatibility issues for third-party developers, the
old MNT_VNODE_FOREACH interface will remain available so that this
change can be MFC'ed to 9. Following the MFC to 9, MNT_VNODE_FOREACH
will be removed in head.
The reason for this update is to prepare for the addition of the
MNT_VNODE_FOREACH_ACTIVE interface that will loop over just the
active vnodes associated with a mount point (typically less than
1% of the vnodes associated with the mount point).
Reviewed by: kib
Tested by: Peter Holm
MFC after: 2 weeks
behaviour on error from write RPC back to behaviour of old nfs client.
When set to not zero, the pages for which write failed are kept dirty.
PR: kern/165927
Reviewed by: alc
MFC after: 2 weeks
significantly. Upon investigation this was caused by name cache
misses for lookups of "..". For name cache entries for non-".."
directories, the cache entry serves double duty. It maps both the
named directory plus ".." for the parent of the directory. As such,
two ctime values (one for each of the directory and its parent) need
to be saved in the name cache entry.
This patch adds an entry for ctime of the parent directory to the
name cache. It also adds an additional uma zone for large entries
with this time value, in order to minimize memory wastage.
As well, it fixes a couple of cases where the mtime of the parent
directory was being saved instead of ctime for positive name cache
entries. With this patch, Lookup RPC counts return to values similar
to pre-r230394 kernels.
Reported by: bde
Discussed with: kib
Reviewed by: jhb
MFC after: 2 weeks
Add the sysctl debug.iosize_max_clamp, enabled by default. Setting the
sysctl to zero allows to perform the SSIZE_MAX-sized i/o requests from
the usermode.
Discussed with: bde, das (previous versions)
MFC after: 1 month
a credential structure would corrupt it. This happened when the
p argument was != NULL. However, I now realize that the copying of
open credentials should only happen for p == NULL, since that indicates
that it is a read-ahead or write-behind. This patch fixes this.
After this commit, r228827 could be reverted, but I think the code is
clearer and safer with the patch, so I am going to leave it in.
Without this patch, it was possible that a NFSv4 VOP_SETATTR() could have
changed the credentials of the caller. This would have happened if
the process doing the VOP_SETATTR() did not have the file open, but
some other process running as a different uid had the file open for writing
at the same time.
MFC after: 5 days
mnt_noasync counter to temporary remove MNTK_ASYNC mount option, which
is needed to guarantee a synchronous completion of the initiated i/o
before syscall or VOP return. Global removal of MNTK_ASYNC option is
harmful because not only i/o started from corresponding thread becomes
synchronous, but all i/o is synchronous on the filesystem which is
initiated during sync(2) or syncer activity.
Instead of removing MNTK_ASYNC from mnt_kern_flag, provide a local
thread flag to disable async i/o for current thread only. Use the
opportunity to move DOINGASYNC() macro into sys/vnode.h and
consistently use it through places which tested for MNTK_ASYNC.
Some testing demonstrated 60-70% improvements in run time for the
metadata-intensive operations on async-mounted UFS volumes, but still
with great deviation due to other reasons.
Reviewed by: mckusick
Tested by: scottl
MFC after: 2 weeks
the original IPv4 implementation from r178888:
- Use RT_DEFAULT_FIB in the IPv4 implementation where noticed.
- Use rt*fib() KPI with explicit RT_DEFAULT_FIB where applicable in
the NFS code.
- Use the new in6_rt* KPI in TCP, gif(4), and the IPv6 network stack
where applicable.
- Split in6_rtqtimo() and in6_mtutimo() as done in IPv4 and equally
prevent multiple initializations of callouts in in6_inithead().
- Use wrapper functions where needed to preserve the current KPI to
ease MFCs. Use BURN_BRIDGES to indicate expected future cleanup.
- Fix (related) comments (both technical or style).
- Convert to rtinit() where applicable and only use custom loops where
currently not possible otherwise.
- Multicast group, most neighbor discovery address actions and faith(4)
are locked to the default FIB. Individual IPv6 addresses will only
appear in the default FIB, however redirect information and prefixes
of connected subnets are automatically propagated to all FIBs by
default (mimicking IPv4 behavior as closely as possible).
Sponsored by: Cisco Systems, Inc.
any thread doing an I/O RPC with a transfer size greater than
NFS_UDPMAXDATA will be hung indefinitely, retrying the RPC.
After a discussion on freebsd-fs@, I decided to add a warning
message for this case, as suggested by Jeremy Chadwick.
Suggested by: freebsd at jdc.parodius.com (Jeremy Chadwick)
MFC after: 2 weeks
NFS clients was reported to freebsd-fs@ under the subject "NFS
corruption in recent HEAD" on Nov. 26, 2011. This problem occurred when
a TCP mounted root fs was changed to using UDP. I believe that this
problem was caused by the change in mnt_stat.f_iosize that occurred
because rsize was decreased to the maximum supported by UDP. This
patch fixes the problem by using v_bufobj.bo_bsize instead of f_iosize,
since the latter is set to f_iosize when the vnode is allocated, but
does not change for a given vnode when f_iosize changes.
Reported by: pjd
Reviewed by: kib
MFC after: 2 weeks
appropriate timestamps. Restore the assertions which verify that
NCF_TS is set when timestamp is asked for.
Reviewed by: jhb (previous version)
MFC after: 2 weeks
we will only trust a positive name cache entry for a specified amount of
time before falling back to a LOOKUP RPC, even if the ctime for the file
handle matches the cached copy in the name cache entry. The timeout is
configured via a new 'nametimeo' mount option and defaults to 60 seconds.
It may be set to zero to disable positive name caching entirely.
Reviewed by: rmacklem
MFC after: 1 week
from TCP to UDP and the rsize/wsize/readdirsize is greater
than NFS_MAXDGRAMDATA, it is possible for a thread doing an
I/O RPC to get stuck repeatedly doing retries. This happens
because the RPC will use a resize/wsize/readdirsize that won't
work for UDP and, as such, it will keep failing indefinitely.
This patch returns an error for this case, to avoid the problem.
A discussion on freebsd-fs@ seemed to indicate that returning
an error was preferable to silently ignoring the "udp"/"mntudp"
option.
This problem was discovered while investigating a problem reported
by pjd@ via email.
MFC after: 2 weeks
entries on one client when a directory was renamed on another client. The
root cause for the stale entry being trusted is that each per-vnode nfsnode
structure has a single 'n_ctime' timestamp used to validate positive name
cache entries. However, if there are multiple entries for a single vnode,
they all share a single timestamp. To fix this, extend the name cache
to allow filesystems to optionally store a timestamp value in each name
cache entry. The NFS clients now fetch the timestamp associated with
each name cache entry and use that to validate cache hits instead of the
timestamps previously stored in the nfsnode. Another part of the fix is
that the NFS clients now use timestamps from the post-op attributes of
RPCs when adding name cache entries rather than pulling the timestamps out
of the file's attribute cache. The latter is subject to races with other
lookups updating the attribute cache concurrently. Some more details:
- Add a variant of nfsm_postop_attr() to the old NFS client that can return
a vattr structure with a copy of the post-op attributes.
- Handle lookups of "." as a special case in the NFS clients since the name
cache does not store name cache entries for ".", so we cannot get a
useful timestamp. It didn't really make much sense to recheck the
attributes on the the directory to validate the namecache hit for "."
anyway.
- ABI compat shims for the name cache routines are present in this commit
so that it is safe to MFC.
MFC after: 2 weeks
The effect of this was, for clients mounted via inet6 addresses,
that the DRC cache would never have a hit in the server. It also
broke NFSv4 callbacks when an inet6 address was the only one available
in the client. This patch fixes the above, plus deletes opt_inet6.h
from a couple of files it is not needed for.
MFC after: 2 weeks
jhb@ spotted that nfscl_getstateid() might modify credentials when
called from nfsrpc_read() for the case where p != NULL, whereas
nfsrpc_read() only did a crdup() to get new credentials for p == NULL.
This bug was introduced by r195510, since pre-r195510 nfscl_getstateid()
only modified credentials for the p == NULL case. This patch modifies
nfsrpc_read()/nfsrpc_write() so that they do crdup() for the p != NULL case.
It is conceivable that this bug caused the crash reported by glebius@, but
that will not be determined for some time, since the crash occurred after
about 1month of operation.
Tested by: glebius
Reviewed by: jhb
MFC after: 2 weeks
of the same lock_owner4 string. As such, the handling of cleanup
of lock_owners could be simplified. This simplification permitted
the client to do a ReleaseLockOwner operation when the process that
the lock_owner4 string represents, has exited. This permits the
server to release any storage related to the lock_owner4 string
before the associated open is closed. Without this change, it
is possible to exhaust a server's storage when a long running
process opens a file and then many child processes do locking
on the file, because the open doesn't get closed. A similar patch
was applied to the Linux NFSv4 client recently so that it wouldn't
exhaust a server's storage.
Reviewed by: zack
MFC after: 2 weeks
client. This does not change the client's behaviour, but prepares
the code so that nfsrpc_rellockown() can be called elsewhere in a
future commit.
MFC after: 2 weeks
head nfsc_defunctlockowner. This patch simply removes the code that
loops through this always empty list, since the code no longer does
anything useful. It should not have any effect on the client's
behaviour.
MFC after: 2 weeks
for regular files. Since other file types don't write into the
buffer cache, calling ncl_flush() is almost a no-op. However, it does
clear the NMODIFIED flag and this shouldn't be done by nfs_fsync() for
directories.
MFC after: 2 weeks
before the nfs_decode_args() call in the new NFS client, so
that a specfied command line value won't be overwritten.
Also, modify the calculation for small values of desiredvnodes
to avoid an unusually large value or a divide by zero crash.
It seems that the default value for nm_wcommitsize is very
conservative and may need to change at some time.
PR: kern/159351
Submitted by: onwahe at gmail.com (earlier version)
Reviewed by: jhb
MFC after: 2 weeks
kernel for FreeBSD 9.0:
Add a new capability mask argument to fget(9) and friends, allowing system
call code to declare what capabilities are required when an integer file
descriptor is converted into an in-kernel struct file *. With options
CAPABILITIES compiled into the kernel, this enforces capability
protection; without, this change is effectively a no-op.
Some cases require special handling, such as mmap(2), which must preserve
information about the maximum rights at the time of mapping in the memory
map so that they can later be enforced in mprotect(2) -- this is done by
narrowing the rights in the existing max_protection field used for similar
purposes with file permissions.
In namei(9), we assert that the code is not reached from within capability
mode, as we're not yet ready to enforce namespace capabilities there.
This will follow in a later commit.
Update two capability names: CAP_EVENT and CAP_KEVENT become
CAP_POST_KEVENT and CAP_POLL_KEVENT to more accurately indicate what they
represent.
Approved by: re (bz)
Submitted by: jonathan
Sponsored by: Google Inc
This was reported to the mailing list freebsd-net@freebsd.org
on July 21, 2011 under the subject "LOR with nfsclient sillyrename".
The LOR occurred when nfs_inactive() called vrele(sp->s_dvp)
while holding the vnode lock on the file in s_dvp. This patch
modifies the client so that it performs the vrele(sp->s_dvp)
as a separate task to avoid the LOR. This fix was discussed
with jhb@ and kib@, who both proposed variations of it.
Tested by: pho, jlott at averesystems.com
Submitted by: jhb (earlier version)
Reviewed by: kib
Approved by: re (kib)
MFC after: 2 weeks
failed after the file was created in nfs_create(). This would
probably only happen during a forced dismount. The old NFS client
does have a vput() for this case. Detected by pho during recent
testing, where an open syscall returned with a vnode still locked.
Tested by: pho
Approved by: re (kib)
MFC after: 2 weeks
loop in nfscl_getcl() when a forced dismount is in progress,
because nfsv4_lock() will return 0 without sleeping when
MNTK_UNMOUNTF is set.
This patch fixes it so it won't loop calling nfsv4_lock()
for this case.
MFC after: 2 weeks
multiple instances of the same lock_owner when a process both
inherited an open file descriptor plus opened the same file itself.
Since some NFSv4 servers cannot handle multiple instances of
the same lock_owner string, this patch changes the algorithm
used by nfscl_getopen() in the new NFSv4 client to keep that
from happening. The new algorithm is simpler, since there is
no longer any need to ascend the process's parentage tree because
all NFSv4 Closes for a file are done at VOP_INACTIVE()/VOP_RECLAIM(),
making the Opens indistinct w.r.t. use with Lock Ops.
This problem was discovered at the recent NFSv4 interoperability
Bakeathon.
MFC after: 2 weeks
to the lock_owner4 string that goes on the wire. Also, add
code to do a ReleaseLockOwner Op on the lock_owner4 string
before a Close. Apparently not all NFSv4 servers handle multiple
instances of the same lock_owner4 string, at least not in a
compatible way. This patch avoids having multiple instances,
except for one unusual case, which will be fixed by a future commit.
Found at the recent NFSv4 interoperability Bakeathon.
Tested by: tdh at excfb.com
MFC after: 2 weeks
mode attribute in as 0 when doing writes. The change adds
the Mode attribute plus the others except Owner and Owner_group
to the list requested by the NFSv4 Write Operation. This fixed
a problem where an executable file built by "cc" would get mode
0111 instead of 0755 for some NFSv4 servers.
Found at the recent NFSv4 interoperability Bakeathon.
Tested by: tdh at excfb.com
MFC after: 2 weeks
the NFS subsystems use five of the rpcsec_gss/kgssapi entry points,
but since it was not obvious which others might be useful, all
nineteen were included. Basically the nineteen entry points are
set in a structure called rpc_gss_entries and inline functions
defined in sys/rpc/rpcsec_gss.h check for the entry points being
non-NULL and then call them. A default value is returned otherwise.
Requested by rwatson.
Reviewed by: jhb
MFC after: 2 weeks
cloned from the old NFS client, plus additions for NFSv4. A
review of this code is in progress, however it was felt by the
reviewer that it could go in now, before code slush. Any changes
required by the review can be committed as bug fixes later.
should be ok, since the client now delays NFSv4 Close operations
until VOP_INACTIVE()/VOP_RECLAIM(). As such, there should be no
risk that the NFSv4 Open is closed while an associated byte range lock
still exists.
Tested by: avg
MFC after: 2 weeks
"p_leader" for the "id" for POSIX byte range locking. I think
this would only have affected processes created by rfork(2)
with the RFTHREAD flag specified. This patch fixes that by
passing the "id" down through the various functions from
nfs_advlock().
MFC after: 2 weeks
VM_PAGER_AGAIN to VM_PAGER_ERROR for the uwritten pages. Return
VM_PAGER_AGAIN for the partially written page. Always forward at least
one page in the loop of vm_object_page_clean().
VM_PAGER_ERROR causes the page reactivation and does not clear the
page dirty state, so the write is not lost.
The change fixes an infinite loop in vm_object_page_clean() when the
filesystem returns permanent errors for some page writes.
Reported and tested by: gavin
Reviewed by: alc, rmacklem
MFC after: 1 week
Pathconf RPC for cases where the reply doesn't include
the answer. This fixes a problem reported by avg@ where
the NFSv3 Pathconf RPC would fail when "ls -l" did an
lpathconf(2) for _PC_ACL_NFS4.
Tested by: avg
MFC after: 2 weeks
correctly during a forced dismount. This required that
the exclusive and shared (refcnt) sleep lock functions check
for MNTK_UMOUNTF before sleeping, so that they won't block
while nfscl_umount() is getting rid of the state. As
such, a "struct mount *" argument was added to the locking
functions. I believe the only remaining case where a forced
dismount can get hung in the kernel is when a thread is
already attempting to do a TCP connect to a dead server
when the krpc client structure called nr_client is NULL.
This will only happen just after a "mount -u" with options
that force a new TCP connection is done, so it shouldn't
be a problem in practice.
MFC after: 2 weeks
in the new NFS client so that a forced dismount doesn't
get stuck in the VFS_SYNC() call that happens before
VFS_UNMOUNT() in dounmount().
Additional changes are needed before forced dismounts will work.
MFC after: 2 weeks
argument for a write RPC when it succeeds for the first one and
fails for a subsequent RPC within the same call to the function.
This makes it compatible with the old NFS client for this case.
MFC after: 2 weeks
to both NFS clients. This avoids the crash reported by
Sergey Kandaurov (pluknet@gmail.com) to the freebsd-fs@
list with subject "[old nfsclient] different nmount()
args passed from mount vs mount_nfs" dated May 17, 2011.
Tested by: pluknet at gmail.com (old nfs client)
MFC after: 2 weeks
new NFS client. It will then be reduced to whatever the
server says it can support. There might be an argument
that this could be one block larger, but since NFS is
a byte granular system, I chose not to do that.
Suggested by: Matt Dillon
Tested by: Daniel Braniss (earlier version)
MFC after: 2 weeks
that are now in "struct statfs" for NFSv3 and NFSv4. Since
the ffiles value is uint64_t on the wire, I clip the value
to INT64_MAX to avoid setting f_ffree negative.
Tested by: kib
MFC after: 2 weeks
checking at open time. It may improve performance for read-only
NFS mounts. Use deliberately.
MFC after: 1 week
Reviewed by: rmacklem, jhb (earlier version)
in mnt_optnew. This is needed so that the old mount(2) syscall
works and that is needed so that amd(8) works. The code was
basically just cribbed from sys/nfsclient/nfs_vfsops.c with minor
changes. This patch is mainly to fix the new NFS client so that
amd(8) works with it. Thanks go to Craig Rodrigues for helping with
this.
Tested by: Craig Rodrigues (for amd)
MFC after: 2 weeks
NFS client (which I guess is no longer experimental). The fstype "newnfs"
is now "nfs" and the regular/old NFS client is now fstype "oldnfs".
Although mounts via fstype "nfs" will usually work without userland
changes, an updated mount_nfs(8) binary is needed for kernels built with
"options NFSCL" but not "options NFSCLIENT". Updated mount_nfs(8) and
mount(8) binaries are needed to do mounts for fstype "oldnfs".
The GENERIC kernel configs have been changed to use options
NFSCL and NFSD (the new client and server) instead of NFSCLIENT and NFSSERVER.
For kernels being used on diskless NFS root systems, "options NFSCL"
must be in the kernel config.
Discussed on freebsd-fs@.
when building kernels that don't have "options NFS_ROOT"
specified. I plan on moving the functions that use these
data structures into the shared code in sys/nfs/nfs_diskless.c
in a future commit. At that time, these definitions will no
longer be needed in nfs_vfsops.c and nfs_clvfsops.c.
MFC after: 2 weeks
same diskless NFS root code as the regular client, which
was moved to sys/nfs by r221032. This fixes the newnfs
client so that it can do an NFSv3 diskless root file system.
MFC after: 2 weeks
