Commit Graph

383 Commits

Author SHA1 Message Date
Robert Watson
c2a9c542a9 Push Giant a bit further off the NFS server in a number of straight
forward cases by converting from unconditional acquisition of Giant
around vnode operations to conditional acquisition:

- Remove nfsrv_access_withgiant(), and cause nfsrv_access() to now
  assert that Giant will be held if it is required for the vnode.

- Add nfsrv_fhtovp_locked(), which will drop the NFS server lock if
  required, and modify nfsrv_fhtovp() to conditionally acquire
  Giant if required.

- In the VOP's not dealing with more than one vnode at a time (i.e.,
  not involving a lookup), conditionally acquire Giant.

This removes Giant use for MPSAFE file systems for a number of quite
important RPCs, including getattr, read, write.  It leaves
unconditional Giant acquisitions in vnode operations that interact
with the name space or more than one vnode at a time as these
require further work.

Tested by:	kris
Reviewed by:	kib
2006-11-24 11:53:16 +00:00
Pawel Jakub Dawidek
270626bc42 Protect nfsm_srvpathsiz() call with the nfsd_mtx lock.
Reviewed by:	mohans
2006-11-20 07:32:52 +00:00
Robert Watson
acd3428b7d Sweep kernel replacing suser(9) calls with priv(9) calls, assigning
specific privilege names to a broad range of privileges.  These may
require some future tweaking.

Sponsored by:           nCircle Network Security, Inc.
Obtained from:          TrustedBSD Project
Discussed on:           arch@
Reviewed (at least in part) by: mlaier, jmg, pjd, bde, ceri,
                        Alex Lyashkov <umka at sevcity dot net>,
                        Skip Ford <skip dot ford at verizon dot net>,
                        Antoine Brodin <antoine dot brodin at laposte dot net>
2006-11-06 13:42:10 +00:00
Konstantin Belousov
1c99939f08 Fix leak in NAMEI zone caused by nfs server when VOP_RENAME fails.
Submitted by:	Padma Bhooma <pbhooma at panasas com>
Reviewed by:	bde
Approved by:	pjd (mentor)
MFC after:	1 week
2006-10-26 12:41:53 +00:00
Robert Watson
aed5570872 Complete break-out of sys/sys/mac.h into sys/security/mac/mac_framework.h
begun with a repo-copy of mac.h to mac_framework.h.  sys/mac.h now
contains the userspace and user<->kernel API and definitions, with all
in-kernel interfaces moved to mac_framework.h, which is now included
across most of the kernel instead.

This change is the first step in a larger cleanup and sweep of MAC
Framework interfaces in the kernel, and will not be MFC'd.

Obtained from:	TrustedBSD Project
Sponsored by:	SPARTA
2006-10-22 11:52:19 +00:00
John Baldwin
6cba7f3609 - Add a new function nfsrv_destroycache() to tear down the server request
cache when unloading the nfsserver module.  This fixes a memory leak and
  a stale pointer.
- Use callout_drain() rather than callout_stop() when unloading the
  nfsserver module.

MFC after:	3 days
2006-08-01 16:27:14 +00:00
John Baldwin
be698f1e7f Use TAILQ_FOREACH_SAFE() in a couple of places. 2006-08-01 15:32:25 +00:00
John Baldwin
af5bf12239 Now that all system calls are MPSAFE, retire the SYF_MPSAFE flag used to
mark system calls as being MPSAFE:
- Stop conditionally acquiring Giant around system call invocations.
- Remove all of the 'M' prefixes from the master system call files.
- Remove support for the 'M' prefix from the script that generates the
  syscall-related files from the master system call files.
- Don't explicitly set SYF_MPSAFE when registering nfssvc.
2006-07-28 19:05:28 +00:00
Robert Watson
b0668f7151 soreceive_generic(), and sopoll_generic(). Add new functions sosend(),
soreceive(), and sopoll(), which are wrappers for pru_sosend,
pru_soreceive, and pru_sopoll, and are now used univerally by socket
consumers rather than either directly invoking the old so*() functions
or directly invoking the protocol switch method (about an even split
prior to this commit).

This completes an architectural change that was begun in 1996 to permit
protocols to provide substitute implementations, as now used by UDP.
Consumers now uniformly invoke sosend(), soreceive(), and sopoll() to
perform these operations on sockets -- in particular, distributed file
systems and socket system calls.

Architectural head nod:	sam, gnn, wollman
2006-07-24 15:20:08 +00:00
Mohan Srinivasan
5943f1a781 Size the NFS server dupreq cache on the basis of nmbclusters. On servers
with low nmbclusters, we tie up too many mbclusters in the NFS duplicate
request cache. This change limits the size of the dupreq cache to 1/2
the nmbclusters (and flaots in a range of [64, 2048]).

MFC after 2 weeks.

Reported by: Steve Kargl, David O'Brien
Tested by:   Steve Kargl
2006-06-23 00:42:26 +00:00
Konstantin Belousov
273147358f Temporary workaround to prevent leak of Giant from nfsd when calling
lookup().

Reviewed by:	tegge
Tested by:	"Arno J. Klaassen" <arno at heho snv jussieu fr>, "Rong-en Fan" <grafan at gmail com>, Dmitriy Kirhlarov <dimma at higis ru>, Dmitry Pryanishnikov <dmitry at atlantis dp ua>
MFC after:	1 week
Approved by:	kan, pjd (mentors)
2006-06-05 14:48:02 +00:00
Mohan Srinivasan
2eafb39b67 Bump up the NFS server dupreq cache limit to 2K (from 64). With a small
duplicate request cache, under heavy load a lot of non-idempotent requests
were getting served again, resulting in errors.

Found by : Kris Kennaway.
2006-04-25 00:21:56 +00:00
Christian S.J. Peron
7935d5382b Introduce a new MAC entry point for label initialization of the NFS daemon's
credential: mac_associate_nfsd_label()

This entry point can be utilized by various Mandatory Access Control policies
so they can properly initialize the label of files which get created
as a result of an NFS operation. This work will be useful for fixing kernel
panics associated with accessing un-initialized or invalid vnode labels.

The implementation of these entry points will come shortly.

Obtained from:	TrustedBSD
Requested by:	mdodd
MFC after:	3 weeks
2006-04-06 23:33:11 +00:00
Chuck Lever
7d8a7e19c7 rick says:
The following bug was just identified in OpenBSD and it looks like the same
bug exists in the other BSDen NFS servers.

A Linux client (don't know which version, but you can look at
	http://bugzilla.kernel.org/show_bug.cgi?id=6256)
does a Setattr of mtime to the server's time, where the file is mode 0664 and
the client user has group access (ie. caller is not the file owner).

The BSD servers fail the Setattr with EPERM, since the VA_UTIMES_NULL flag
isn't set before doing the VOP_SETATTR.

It seems to me that this should be allowed, since it is allowed for a local
utimes(2). If so, the fix is to set VA_UTIMES_NULL for the
"set-time-to-server-time" cases of setting atime and/or mtime.

Submitted by:	rick@snowhite.cis.uoguelph.ca
Reviewed by:	cel
Approved by:	silby
MFC after:	1 week
2006-04-02 04:24:57 +00:00
Jeff Roberson
3bbd6d8ae6 - Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs().
Discussed with:	tegge
Tested by:	kris
Sponsored by:	Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
Jeff Roberson
e64df05c33 - Reorder vrele calls after vput calls to prevent lock order reversals
between leaf and directory locks.

Found by:	kris
Sponsored by:	Isilon Systems, Inc.
2006-03-12 04:59:04 +00:00
Simon L. B. Nielsen
fb46e061e4 When parsing an RPC request in nfsrv_dorec(), KASSERT that there
actually is an mbuf to process.  This catches the missing mbuf before it
would otherwise causes a NULL pointer dereference, which could be
triggered by a 0 length RPC record before the check for such records was
added in rev 1.97.

Approved by:	cperciva (mentor)
2006-03-08 20:21:15 +00:00
Simon L. B. Nielsen
925374d453 Correct a remote kernel panic when processing zero-length RPC records
via TCP. [06:10]

Security:	FreeBSD-SA-06:10.nfs
Approved by:	cperciva
2006-03-01 14:17:32 +00:00
Jeff Roberson
89b0e10910 - Reorder calls to vrele() after calls to vput() when the vrele is a
directory.  vrele() may lock the passed vnode, which in these cases would
   give an invalid lock order of child -> parent.  These situations are
   deadlock prone although do not typically deadlock because the vrele
   is typically not releasing the last reference to the vnode.  Users of
   vrele must consider it as a call to vn_lock() and order it appropriately.

MFC After: 	1 week
Sponsored by:	Isilon Systems, Inc.
Tested by:	kkenn
2006-02-01 00:25:26 +00:00
Christian S.J. Peron
7a3e891951 Manage the ucred for the NFS server using the crget/crfree API defined in
kern_prot.c. This API handles reference counting among many other things.
Notably, if MAC is compiled into the kernel, it will properly initialize the
MAC labels when the ucred is allocated.

This work is in preparation for a new MAC entry point which will be responsible
for properly initializing policy specific labels for the NFS server credential.
Utilization of the crfree/crget APIs reduce the complexity associated with
this label's management.

Submitted by:	green (with changes) [1]
Obtained from:	TrustedBSD Project
Discussed with:	rwatson, alfred

[1] I moved the ucred allocation outside the scope of the NFS server lock to
    prevent M_WAIKOK allocations from occurring with non-sleep-able locks held.
    Additionally, to reduce complexity, the ucred persist as long as the NFS
    server descriptor.
2006-01-28 19:24:40 +00:00
Tom Rhodes
129518ec2c Revert my previous commit.
Proved I'm not that bright at times:	jhb
2006-01-23 21:06:22 +00:00
Tom Rhodes
cb5c1ec725 Fix indentation.
Prodded by:	stefanf, ru, njl (in that order)
2006-01-23 17:41:43 +00:00
Tom Rhodes
9c013503bb Remove some dead code.
Found with:     Coverity Prevent(tm)
2006-01-21 12:10:33 +00:00
Robert Watson
5bb84bc84b Normalize a significant number of kernel malloc type names:
- Prefer '_' to ' ', as it results in more easily parsed results in
  memory monitoring tools such as vmstat.

- Remove punctuation that is incompatible with using memory type names
  as file names, such as '/' characters.

- Disambiguate some collisions by adding subsystem prefixes to some
  memory types.

- Generally prefer lower case to upper case.

- If the same type is defined in multiple architecture directories,
  attempt to use the same name in additional cases.

Not all instances were caught in this change, so more work is required to
finish this conversion.  Similar changes are required for UMA zone names.
2005-10-31 15:41:29 +00:00
Gleb Smirnoff
19bf94288f Keep locks consistent before goto.
Reported by:	pho
Reviewed by:	mohans
2005-10-27 19:02:34 +00:00
John Baldwin
7e9e371f2d Use the refcount API to manage the reference count for user credentials
rather than using pool mutexes.

Tested on:	i386, alpha, sparc64
2005-09-27 18:09:42 +00:00
Robert Watson
ad77d81512 NFS write gathering defers execution of NFS server write requests to wait
to see if additional write requests will arrive that can be coalesced and
clustered with earlier ones.  When doing so, it must determine whether
the two requests are made by credentials with the same access writes, so
as not to coalesce improperly.  NFSW_SAMECRED() implements a test of two
credentials using a binary compare.

Replace NFSW_SAMECRED() macro with nfsrv_samecred() function, which is
aware of the contents and layout of a struct ucred, rather than a simple
binary compare.  While the binary compare works when ucred is simply a
zero'd and embedded 'struct ucred' in the NFS descriptor, it will work
less well when the ucred associated with an NFS descriptor is "real", so
has defined and populated reference count, mutex, etc.

MFC after:	1 week
Obtained from:	TrustedBSD Project
2005-04-17 16:25:36 +00:00
Sam Leffler
bd1da15f2a avoid potential null ptr deref by free'ing excess mbufs instead of
zero'ing their length (copied from m_adj where this code came from
after the equivalent change there has had time to soak)

Noticed by:	Coverity Prevent analysis tool
2005-03-28 18:51:58 +00:00
Xin LI
a431ada687 Do not do write gathering for NFSv3, since it makes no sense unless
the client is broken and does sync writes all the time.

Obtained from:	NetBSD (sys/nfs/nfs_syscalls.c,v 1.44)
Reviewed by:	-arch (bde)
2005-03-26 11:29:02 +00:00
Poul-Henning Kamp
c62801a7f8 Don't try to create vnode_pager objects on other filesystems vnodes,
either they did it themselves or it won't happen.
2005-01-24 22:09:13 +00:00
Paul Saab
f1b3bfb348 Now that we have a non blocking version of nfsm_dissect(), change all the
nfsm_dissect() calls (done under the NFSD lock) to nfsm_dissect_nonblock().

Submitted by:	Mohan Srinivasan
2005-01-19 22:53:40 +00:00
Poul-Henning Kamp
e39db32ab0 Ditch vfs_object_create() and make the callers call VOP_CREATEVOBJECT()
directly.
2005-01-13 12:25:19 +00:00
Poul-Henning Kamp
8df6bac4c7 Remove the unused credential argument from VOP_FSYNC() and VFS_SYNC().
I'm not sure why a credential was added to these in the first place, it is
not used anywhere and it doesn't make much sense:

	The credentials for syncing a file (ability to write to the
	file) should be checked at the system call level.

	Credentials for syncing one or more filesystems ("none")
	should be checked at the system call level as well.

	If the filesystem implementation needs a particular credential
	to carry out the syncing it would logically have to the
	cached mount credential, or a credential cached along with
	any delayed write data.

Discussed with:	rwatson
2005-01-11 07:36:22 +00:00
Warner Losh
c398230b64 /* -> /*- for license, minor formatting changes 2005-01-07 01:45:51 +00:00
Robert Watson
29af382686 Correct a bug in nfsrv_create() where a call to nfsrv_access() might
be made holding the NFS server mutex.  To clean this up, introduce a
version of the function, nfsrv_access_withgiant(), that expects the
NFS server mutex to already have been dropped and Giant acquired.
Wrap nfsrv_access() around this.  This permits callers to more
efficiently check access if they're in a code block performing VFS
operations, and can be substitited for the nfsrv_access() call that
triggered this bug.

PR:		73807, 73208
MFC after:	1 week
2004-11-11 21:30:52 +00:00
Poul-Henning Kamp
494eb176e7 Add b_bufobj to struct buf which eventually will eliminate the need for b_vp.
Initialize b_bufobj for all buffers.

Make incore() and gbincore() take a bufobj instead of a vnode.

Make inmem() local to vfs_bio.c

Change a lot of VI_[UN]LOCK(bp->b_vp) to BO_[UN]LOCK(bp->b_bufobj)
also VI_MTX() to BO_MTX(),

Make buf_vlist_add() take a bufobj instead of a vnode.

Eliminate other uses of bp->b_vp where bp->b_bufobj will do.

Various minor polishing: remove "register", turn panic into KASSERT,
use new function declarations, TAILQ_FOREACH_SAFE() etc.
2004-10-22 08:47:20 +00:00
Robert Watson
ae8c2fa228 Correct several instances where calls to vfs_getvfs() resulting in
failure in the NFS server would result in a leaked instance of the NFS
server subsystem lock.  Liberally sprinkle assertions in all target
labels for error unwinding to assert the desired locking state.

RELENG_5_3 candidate.

MFC after:	3 days
Reported by:	Wilkinson, Alex <alex dot wilkinson at dsto dot defence dot gov dot au>
2004-10-18 11:23:11 +00:00
Robert Watson
e2d2098653 Convert a mtx_lock(&Giant) to a mtx_unlock(&Giant) in nfsrv_link() to
prevent leakage of Giant.  With INVARIANTS, this results in an
assertion failure following execution of the RPC.  Without INVARIANTS,
it could result in problems if the NFS server is killed causing nfsd
to return to user space holding Giant.

Feet provided by:	brueffer
2004-08-25 16:52:59 +00:00
Robert Watson
9e0219d901 If debug.mpsafenet is non-zero, run the NFS server callout without
Giant.
2004-07-24 02:32:27 +00:00
Robert Watson
b0c90b3b8c Remove spl() use from nfsrv_timer. 2004-07-24 02:07:09 +00:00
Poul-Henning Kamp
3e019deaed Do a pass over all modules in the kernel and make them return EOPNOTSUPP
for unknown events.

A number of modules return EINVAL in this instance, and I have left
those alone for now and instead taught MOD_QUIESCE to accept this
as "didn't do anything".
2004-07-15 08:26:07 +00:00
Alfred Perlstein
d83ed0fac0 Do not call sorecieve() in the context of a socket callback as it causes
lock order reversals so->inpcb since we're called with the socket lock
held.
2004-07-13 07:05:38 +00:00
Robert Watson
3a54d6a8a8 Change M_WAITOK argument to sodupsockaddr() to M_NOWAIT. When the call
to dup_sockaddr() was renamed to sodupsockaddr(), the argument was
changed from '1' to 'M_WAITOK', which changed the semantics.  This
resulted in a WITNESS warning about a potential sleep while holding the
NFS server mutex.  Now this will no longer happen, restoring a possible
bug present in the original code (setting RC_NAM even though the malloc
to copy the addres may fail).  bde observes that the flag names here
should probably not be the same as the malloc flags for name space
reasons.

Bumped into by:	kuriyama
2004-07-03 19:17:06 +00:00
Robert Watson
9535efc00d Merge additional socket buffer locking from rwatson_netperf:
- Lock down low hanging fruit use of sb_flags with socket buffer
  lock.

- Lock down low hanging fruit use of so_state with socket lock.

- Lock down low hanging fruit use of so_options.

- Lock down low-hanging fruit use of sb_lowwat and sb_hiwat with
  socket buffer lock.

- Annotate situations in which we unlock the socket lock and then
  grab the receive socket buffer lock, which are currently actually
  the same lock.  Depending on how we want to play our cards, we
  may want to coallesce these lock uses to reduce overhead.

- Convert a if()->panic() into a KASSERT relating to so_state in
  soaccept().

- Remove a number of splnet()/splx() references.

More complex merging of socket and socket buffer locking to
follow.
2004-06-17 22:48:11 +00:00
Poul-Henning Kamp
f3732fd15b Second half of the dev_t cleanup.
The big lines are:
	NODEV -> NULL
	NOUDEV -> NODEV
	udev_t -> dev_t
	udev2dev() -> findcdev()

Various minor adjustments including handling of userland access to kernel
space struct cdev etc.
2004-06-17 17:16:53 +00:00
Bosko Milekic
d1fd2228b8 Giant wasn't dropped here if we have to return EBUSY. This is bad. 2004-05-31 20:21:06 +00:00
Robert Watson
69af1dccdc Release NFS subsystem lock and acquire Giant when calling into
vn_start_write().
2004-05-31 19:08:22 +00:00
Robert Watson
4fa649df8e Add an assertion that nfssvc() isn't called with Giant.
Add two additional pairs of assertions, one at the end of the NFS
server event loop, and one one exit from the NFS daemon, that
assert that if debug.mpsafenet is enabled, Giant is not held, and
that if it is not enabled, Giant will be held.  This is intended
to support debugging scenarios where Giant is "leaked" during NFS
processing.
2004-05-31 16:32:49 +00:00
Robert Watson
30bef9add8 The NFS server modevent code manually patches the system call table to
install nfssvc().  It also updates the argument count, but did so
without setting SYF_MPSAFE, effectively removing the MPSAFE flag even
when syscalls.master indicates it doesn't require Giant.  This change
forces the modevent to set MPSAFE as a flag to its internal notion of
an argument coutn.

Note: this duplication of information is a bad thing, but is a more
general problem I'm not currently willing to address.
2004-05-31 00:59:10 +00:00
Robert Watson
73a4c21f28 One more case where we want to drop the NFS server lock and acquire
Giant when entering VFS.  Discovered by code inspection; still not
hit without debug.mpsafenet=1.

Reported by:	bmilekic
2004-05-30 22:59:54 +00:00