Commit Graph

54 Commits

Author SHA1 Message Date
rmacklem
5120adc280 This patch applies one of the two fixes suggested by
zack.kirsch at isilon.com for a race between nfsrv_freeopen()
and nfsrv_getlockfile() in the experimental NFS server that
he found during testing. Although nfsrv_freeopen() holds a
sleep lock on the lock file structure when called with
cansleep != 0, nfsrv_getlockfile() could still search the
list, once it acquired the NFSLOCKSTATE() mutex. I believe
that acquiring the mutex in nfsrv_freeopen() fixes the race.

MFC after:	2 weeks
2010-09-10 23:49:33 +00:00
rmacklem
ac834a438a Add acquisition of a reference count on nfsv4root_lock to the
nfsd_recalldelegation() function, since this function is called
by nfsd threads when they are handling NFSv2 or NFSv3 RPCs, where
no reference count would have been acquired.

MFC after:	2 weeks
2010-08-28 23:50:09 +00:00
rmacklem
3b7498e206 The timer routine in the experimental NFS server did not acquire
the correct mutex when checking nfsv4root_lock. Although this
could be fixed by adding mutex lock/unlock calls, zack.kirsch at
isilon.com suggested a better fix that uses a non-blocking
acquisition of a reference count on nfsv4root_lock. This fix
allows the weird NFSLOCKSTATE(); NFSUNLOCKSTATE(); synchronization
to be deleted. This patch applies this fix.

Tested by:	zack.kirsch at isilon.com
MFC after:	2 weeks
2010-08-28 21:41:18 +00:00
rmacklem
f06f20d898 Patch the experimental NFSv4 server so that it acquires a reference
count on nfsv4rootfs_lock when dumping state, since these functions
are not called by nfsd threads. Without this reference count, it
is possible for an nfsd thread to acquire an exclusive lock on
nfsv4rootfs_lock while the dump is in progress and then change the
lists, potentially causing a crash.

Reported by:	zack.kirsch at isilon.com
MFC after:	2 weeks
2010-07-16 23:17:05 +00:00
rmacklem
53abeaa9aa Delete comments related to soft clock interrupts that don't apply
to the FreeBSD port of the experimental NFSv4 server.

Submitted by:	zack.kirsch at isilon.com
MFC after:	2 weeks
2010-07-16 01:44:49 +00:00
rmacklem
0224dc3bc1 This patch fixes a bug in the experimental NFSv4 server where it
released a reference count on nfsv4rootfs_lock erroneously when
administrative revocation of state was done.

Submitted by:	zack.kirsch at isilon.com
MFC after:	2 weeks
2010-07-15 03:02:10 +00:00
rmacklem
7dbc0176a4 Fix a bogus comment that mentions lru lists that don't exist.
Reported by:	zack.kirsch at isilon.com
MFC after:	2 weeks
2010-07-13 22:44:14 +00:00
rmacklem
21a9430e2b Add MODULE_DEPEND() macros to the experimental NFS client and
server so that the modules will load when kernels are built with
none of the NFS* configuration options specified. I believe this
resolves the problems reported by PR kern/144458 and the email on
freebsd-stable@ posted by Dmitry Pryanishnikov on June 13.

Tested by:	kib
PR:		kern/144458
Reviewed by:	kib
MFC after:	1 week
2010-06-15 00:25:04 +00:00
kib
9100139edc In NFS clients, instead of inconsistently using #ifdef
DIAGNOSTIC and #ifndef DIAGNOSTIC for debug assertions, prefer
KASSERT(). Also change one #ifdef DIAGNOSTIC in the new nfs server.

Submitted by:	Mikolaj Golub <to.my.trociny gmail com>
MFC after:	2 weeks
2010-06-13 05:24:27 +00:00
rmacklem
2934027caa An NFSv4 server will reply NFSERR_GRACE for non-recovery RPCs
during the grace period after startup. This grace period must
be at least the lease duration, which is typically 1-2 minutes.
It seems prudent for the experimental NFS client to wait a few
seconds before retrying such an RPC, so that the server isn't
flooded with non-recovery RPCs during recovery. This patch adds
an argument to nfs_catnap() to implement a 5 second delay
for this case.

MFC after:	1 week
2010-04-24 22:52:14 +00:00
rmacklem
5ef25d55ac Harden the experimental NFS server a little, by adding range
checks on the length of the client's open/lock owner name. Also,
add free()'s for one case where they were missing and would
have caused a leak if NFSERR_BADXDR had been replied. Probably
never happens, but the leak is now plugged, just in case.

MFC after:	2 weeks
2010-04-06 01:14:49 +00:00
rmacklem
71f4a9b7ef Harden the experimental NFS server a little, by adding extra checks
in the readdir functions for non-positive byte count arguments.
For the negative case, set it to the maximum allowable, since it
was actually a large positive value (unsigned) on the wire.
Also, fix up the readdir function comment a bit.

Suggested by:	dillon AT apollo.backplane.com
MFC after:	2 weeks
2010-04-04 23:19:11 +00:00
rmacklem
7970aa05a8 For the experimental NFS server, add a call to free the lookup
path buffer for one case where it was missing when doing mkdir.
This could have conceivably resulted in a leak of a buffer, but
a leak was never observed during testing, so I suspect it would
have occurred rarely, if ever, in practice.

MFC after:	2 weeks
2010-04-02 02:19:28 +00:00
rmacklem
3c1cb22345 Add SAVENAME to the cn_flags for all cases in the experimental
NFS server for the CREATE cn_nameiop where SAVESTART isn't set.
I was not aware that this needed to be done by the caller until
recently.

Tested by:	lampa AT fit.vutbr.cz (link case)
Submitted by:	lampa AT fit.vutbr.cz (link case)
MFC after:	2 weeks
2010-04-02 01:53:48 +00:00
rmacklem
ad0f267ec8 This patch should fix handling of byte range locks locally
on the server for the experimental nfs server. When enabled
by setting vfs.newnfs.locallocks_enable to non-zero, the
experimental nfs server will now acquire byte range locks
on the file on behalf of NFSv4 clients, such that lock
conflicts between the NFSv4 clients and processes running
locally on the server, will be recognized and handled correctly.

MFC after:	2 weeks
2010-03-30 23:11:50 +00:00
rmacklem
d67e4cffdd Patch the experimental NFS server in a manner analagous to r205661
for the regular NFS server, to ensure that ESTALE is
returned to the client for all errors returned by VFS_FHTOVP().

MFC after:	2 weeks
2010-03-26 01:35:19 +00:00
rwatson
f5e6c09363 Update nfsrv_getsocksndseq() for changes in TCP internals since FreeBSD 6.x:
- so_pcb is now guaranteed to be non-NULL and valid if a valid socket
  reference is held.

- Need to check INP_TIMEWAIT and INP_DROPPED before assuming inp_ppcb is a
  tcpcb, as it might be a tcptw or NULL otherwise.

- tp can never be NULL by the end of the function, so only check
  TCPS_ESTABLISHED before extracting tcpcb fields.

The NFS server arguably incorporates too many assumptions about TCP
internals, but fixing that is left for nother day.

MFC after:		1 week
Reviewed by:		bz
Reviewed and tested by:	rmacklem
Sponsored by:		Juniper Networks
2010-03-11 11:33:04 +00:00
rmacklem
2d2c919325 Change the default value for vfs.newnfs.enable_locallocks to 0 for
the experimental NFS server, since local locking is known to be
broken and the patch to fix it is still a work in progress.

MFC after:	5 days
2010-02-14 00:18:32 +00:00
rmacklem
f9c823994c This fixes the experimental NFS server so that it won't crash in the
caching code for IPv6 by fixing a typo that used the incorrect variable.
It also fixes the indentation of the statement above it.

Reported by:	simon AT comsys.ntu-kpi.kiev.ua
MFC after:	5 days
2010-02-13 23:56:19 +00:00
rmacklem
b51bb2125d The test for "same client" for the experimental nfs server over NFSv4
was broken w.r.t. byte range lock conflicts when it was the same client
and the request used the open_to_lock_owner4 case, since lckstp->ls_clp
was not set. This patch fixes it by using "clp" instead of "lckstp->ls_clp".

MFC after:	2 weeks
2010-01-03 20:08:10 +00:00
rmacklem
0ccdae7b7d Modify the experimental server so that it uses VOP_ACCESSX().
This is necessary in order to enable NFSv4 ACL support. The
argument to nfsvno_accchk() was changed to an accmode_t and
the function nfsrv_aclaccess() was no longer needed and,
therefore, deleted.

Reviewed by:	trasz
MFC after:	2 weeks
2009-12-25 20:44:19 +00:00
delphij
dd5a134c1b Allow using IPv6 in nfsrvd_sentcache() callback.
PR:		kern/141289
Submitted by:	Petr Lampa <lampa fit vutbr cz>
Approved by:	rmacklem
MFC after:	1 week
2009-12-08 23:43:50 +00:00
rmacklem
5ca6063e7a Modify the experimental nfs server so that it falls back to
using VOP_LOOKUP() when VFS_VGET() returns EOPNOTSUPP in the
ReaddirPlus RPC. This patch is based upon one by pjd@ for the
regular nfs server which has not yet been committed. It is needed
when a ZFS volume is exported and ReaddirPlus (which almost
always happens for NFSv4) is performed by a client. The patch
also simplifies vnode lock handling somewhat.

MFC after:	2 weeks
2009-11-23 16:08:15 +00:00
rmacklem
483d9d96d0 Patch the experimental NFS server is a manner analagous to
r197525, so that the creation verifier is handled correctly
in va_atime for 64bit architectures. There were two problems.
One was that the code incorrectly assumed that
sizeof (struct timespec) == 8 and the other was that the tv_sec
field needs to be assigned from a signed 32bit integer, so that
sign extension occurs on 64bit architectures. This is required
for correct operation when exporting ZFS volumes.

Reviewed by:	pjd
MFC after:	2 weeks
2009-11-20 21:21:13 +00:00
rwatson
57ca4583e7 Build on Jeff Roberson's linker-set based dynamic per-CPU allocator
(DPCPU), as suggested by Peter Wemm, and implement a new per-virtual
network stack memory allocator.  Modify vnet to use the allocator
instead of monolithic global container structures (vinet, ...).  This
change solves many binary compatibility problems associated with
VIMAGE, and restores ELF symbols for virtualized global variables.

Each virtualized global variable exists as a "reference copy", and also
once per virtual network stack.  Virtualized global variables are
tagged at compile-time, placing the in a special linker set, which is
loaded into a contiguous region of kernel memory.  Virtualized global
variables in the base kernel are linked as normal, but those in modules
are copied and relocated to a reserved portion of the kernel's vnet
region with the help of a the kernel linker.

Virtualized global variables exist in per-vnet memory set up when the
network stack instance is created, and are initialized statically from
the reference copy.  Run-time access occurs via an accessor macro, which
converts from the current vnet and requested symbol to a per-vnet
address.  When "options VIMAGE" is not compiled into the kernel, normal
global ELF symbols will be used instead and indirection is avoided.

This change restores static initialization for network stack global
variables, restores support for non-global symbols and types, eliminates
the need for many subsystem constructors, eliminates large per-subsystem
structures that caused many binary compatibility issues both for
monitoring applications (netstat) and kernel modules, removes the
per-function INIT_VNET_*() macros throughout the stack, eliminates the
need for vnet_symmap ksym(2) munging, and eliminates duplicate
definitions of virtualized globals under VIMAGE_GLOBALS.

Bump __FreeBSD_version and update UPDATING.

Portions submitted by:  bz
Reviewed by:            bz, zec
Discussed with:         gnn, jamie, jeff, jhb, julian, sam
Suggested by:           peter
Approved by:            re (kensmith)
2009-07-14 22:48:30 +00:00
brooks
f53c1c309d Rework the credential code to support larger values of NGROUPS and
NGROUPS_MAX, eliminate ABI dependencies on them, and raise the to 1024
and 1023 respectively.  (Previously they were equal, but under a close
reading of POSIX, NGROUPS_MAX was defined to be too large by 1 since it
is the number of supplemental groups, not total number of groups.)

The bulk of the change consists of converting the struct ucred member
cr_groups from a static array to a pointer.  Do the equivalent in
kinfo_proc.

Introduce new interfaces crcopysafe() and crsetgroups() for duplicating
a process credential before modifying it and for setting group lists
respectively.  Both interfaces take care for the details of allocating
groups array. crsetgroups() takes care of truncating the group list
to the current maximum (NGROUPS) if necessary.  In the future,
crsetgroups() may be responsible for insuring invariants such as sorting
the supplemental groups to allow groupmember() to be implemented as a
binary search.

Because we can not change struct xucred without breaking application
ABIs, we leave it alone and introduce a new XU_NGROUPS value which is
always 16 and is to be used or NGRPS as appropriate for things such as
NFS which need to use no more than 16 groups.  When feasible, truncate
the group list rather than generating an error.

Minor changes:
  - Reduce the number of hand rolled versions of groupmember().
  - Do not assign to both cr_gid and cr_groups[0].
  - Modify ipfw to cache ucreds instead of part of their contents since
    they are immutable once referenced by more than one entity.

Submitted by:	Isilon Systems (initial implementation)
X-MFC after:	never
PR:		bin/113398 kern/133867
2009-06-19 17:10:35 +00:00
rmacklem
08a5e53744 Add the SVC_RELEASE(xprt), as required by r194407.
Approved by:	kib (mentor)
2009-06-17 22:55:59 +00:00
rmacklem
2afe948156 Remove the "int *" typecast for the aresid argument to vn_rdwr()
and change the type of the argument from size_t to int. This
should avoid issues on 64bit architectures.

Suggested by:	kib
Approved by:	kib (mentor)
2009-06-16 13:52:21 +00:00
rwatson
f4934662e5 Move "options MAC" from opt_mac.h to opt_global.h, as it's now in GENERIC
and used in a large number of files, but also because an increasing number
of incorrect uses of MAC calls were sneaking in due to copy-and-paste of
MAC-aware code without the associated opt_mac.h include.

Discussed with:	pjd
2009-06-05 14:55:22 +00:00
zec
3b5c8e652c Unbreak options VIMAGE kernel builds.
Approved by:	julian (mentor)
2009-05-31 11:57:51 +00:00
rmacklem
c1143c05d4 Fix the experimental nfs subsystem so that it builds with the
current NFSv4 ACLs, as defined in sys/acl.h. It still needs a
way to test a mount point for NFSv4 ACL support before it will
work. Until then, the NFSHASNFS4ACL() macro just always returns 0.

Approved by:	kib (mentor)
2009-05-26 22:21:53 +00:00
rmacklem
82cde148a4 Add two sysctl variables to the experimental nfs server, so
that the range of versions of NFS handled by the server can
be limited. The nfsd daemon must be restarted after these
sysctl variables are changed, in order for the change to take
effect.

Approved by:	kib (mentor)
2009-05-26 01:47:37 +00:00
rmacklem
7373601dad Fix the handling of NFSv4 Illegal Operation number to conform
to RFC3530 (the operation number in the reply must be set to
the value for OP_ILLEGAL). Also cleaned up some indentation.

Approved by:	kib (mentor)
2009-05-26 01:16:09 +00:00
rmacklem
70a64093bf Fix the experimental nfs server's interface to the new krpc so
that it handles the case of a non-exported NFSv4 root correctly.
Also, delete handling for the case where nd_repstat is already
set in nfs_proc(), since that no longer happens.

Approved by:	kib (mentor)
2009-05-26 01:09:33 +00:00
rmacklem
2ebd09c6b9 Add NFSv4 root export checks to the DelegPurge, Renew and
ReleaseLockOwner operations analagous to what is already
in place for SetClientID and SetClientIDConfirm. These are
the five NFSv4 operations that do not use file handle(s),
so the checks are done using the NFSv4 root export entries
in /etc/exports.

Approved by:	kib (mentor)
2009-05-25 01:00:09 +00:00
rmacklem
c220cf0c33 Fix the experimental NFSv4 server so that it handles the case
where a client is not allowed NFSv4 access correctly. This
restriction is specified in the "V4: ..." line(s) in
/etc/exports.

Approved by:	kib (mentor)
2009-05-24 18:49:53 +00:00
rmacklem
4b53f9789d Change the printf of r192595 to identify the function,
as requested by Sam.

Approved by:	kib (mentor)
2009-05-22 19:05:48 +00:00
rmacklem
0d99ad5f66 Modified the printf message of r192590 to remove the
possible DOS attack, as suggested by Sam.

Approved by:	kib (mentor)
2009-05-22 18:10:39 +00:00
rmacklem
b51af5478a Change the comment at the beginning of the function to reflect the
change from panic() to printf() done by r192588.
2009-05-22 16:46:01 +00:00
rmacklem
9bc0f6dfac Change the reboot panic that would have occurred if clientid
numbers wrapped around to a printf() warning of a possible
DOS attack, in the experimental nfsv4 server.

Approved by:	kib (mentor)
2009-05-22 16:41:33 +00:00
rmacklem
fdc5ffc06d Fix the experimental nfs server so that it depends on the nlm,
since it now calls nlm_acquire_next_sysid().

Approved by:	kib (mentor)
2009-05-22 01:15:07 +00:00
rmacklem
44027bf4b7 Fix the comment at line 3711 to be consistent with the change
applied for r192537.

Approved by:	kib (mentor)
2009-05-21 14:52:36 +00:00
rmacklem
573f2880f7 Modify sys/fs/nfsserver/nfs_nfsdport.c to use nlm_acquire_next_sysid()
to set the l_sysid for locks correctly.

Approved by:	kib (mentor)
2009-05-21 01:50:27 +00:00
rmacklem
df90a80062 Although it should never happen, all the nfsv4 server can do
when it runs out of clientids is reboot. I had replaced cpu_reboot()
with printf(), since cpu_reboot() doesn't exist for sparc64.
This change replaces the printf() with panic(), so the reboot
would occur for this highly unlikely occurrence.

Approved by:	kib (mentor)
2009-05-20 18:58:07 +00:00
rmacklem
e0d8f490bb Fix the acquisition of local locks via VOP_ADVLOCK() by the
experimental nfsv4 server. It was setting the a_id argument
to a fixed value, but that wasn't sufficient for FreeBSD8.
Instead, set l_pid and l_sysid to 0 plus set the F_REMOTE
flag to indicate that these fields are used to check for
same lock owner. Since, for NFSv4, a lockowner is a ClientID plus
an up to 1024byte name, it can't be put in l_sysid easily.
I also renamed the p variable to td, since it's a thread ptr.

Approved by:	kib (mentor)
2009-05-17 19:33:48 +00:00
rmacklem
eb2f3f3914 Added a SYSCTL to sys/fs/nfsserver/nfs_nfsdport.c so that the value of
nfsrv_dolocallocks can be changed via sysctl. I also added some non-empty
descriptor strings and reformatted some overly long lines.

Approved by:	kib (mentor)
2009-05-17 17:54:01 +00:00
rmacklem
75b506904d Fixed the Null callback RPCs so that they work with the new krpc. This
required two changes: setting the program and version numbers before
connect and fixing the handling of the Null Rpc case in newnfs_request().

Approved by:	kib (mentor)
2009-05-16 03:12:55 +00:00
rmacklem
ef1be3984d Apply changes to the experimental nfs server so that it uses the security
flavors as exported in FreeBSD-CURRENT. This allows it to use a
slightly modified mountd.c instead of a different utility.

Approved by:	kib (mentor)
2009-05-14 21:39:08 +00:00
rmacklem
e2c7f7ce5b Modify the experimental nfs server to use the new nfsd_nfsd_args
structure for nfsd. Includes a change that clarifies the use of
an empty principal name string to indicate AUTH_SYS only.

Approved by:	kib (mentor)
2009-05-12 16:04:51 +00:00
rmacklem
292f29fdfd Change the name of the nfs server addsock structure from nfsd_args
to nfsd_addsock_args, so that it is consistent with the one in
	sys/nfsserver/nfs.h.

Approved by:	kib (mentor)
2009-05-11 19:37:05 +00:00