Commit Graph

2331 Commits

Author SHA1 Message Date
Rick Macklem
8f63187ec1 Add LK_NOWITNESS to the vn_lock() calls done on newly created nfs
vnodes, since these nodes are not linked into the mount queue and,
as such, the vn_lock() cannot cause a deadlock so LORs are harmless.

Suggested by:	kib
Approved by:	kib (mentor)
MFC after:	3 days
2009-09-09 20:37:49 +00:00
Poul-Henning Kamp
6778431478 Revert previous commit and add myself to the list of people who should
know better than to commit with a cat in the area.
2009-09-08 13:19:05 +00:00
Poul-Henning Kamp
b34421bf9c Add necessary include. 2009-09-08 13:16:55 +00:00
Konstantin Belousov
481208a815 If a race is detected, pfs_vncache_alloc() may reclaim a vnode that had
never been inserted into the pfs_vncache list. Since pfs_vncache_free()
does not anticipate this case, it decrements pfs_vncache_entries
unconditionally; if the vnode was not in the list, pfs_vncache_entries
will no longer reflect the actual number of list entries. This may cause
size of the cache to exceed the configured maximum. It may also trigger
a panic during module unload or system shutdown.

Do not decrement pfs_vncache_entries for the vnode that was not in the
list.

Submitted by:	tegge
Reviewed by:	des
MFC after:	1 week
2009-09-07 12:10:41 +00:00
Konstantin Belousov
6cc745d2d7 insmntque_stddtr() clears vp->v_data and resets vp->v_op to
dead_vnodeops before calling vgone(). Revert r189706 and corresponding
part of the r186560.

Noted and reviewed by:	tegge
Approved by:	des (pseudofs part)
MFC after:	3 days
2009-09-07 11:55:34 +00:00
Konstantin Belousov
34f83c86f6 Remove spurious pfs_unlock().
PR:	kern/137310
Reviewed by:	des
MFC after:	3 days
2009-08-31 09:26:04 +00:00
Jilles Tjoelker
74d1c4927a Fix poll() on half-closed sockets, while retaining POLLHUP for fifos.
This reverts part of r196460, so that sockets only return POLLHUP if both
directions are closed/error. Fifos get POLLHUP by closing the unused
direction immediately after creating the sockets.

The tools/regression/poll/*poll.c tests now pass except for two other things:
- if POLLHUP is returned, POLLIN is always returned as well instead of only
  when there is data left in the buffer to be read
- fifo old/new reader distinction does not work the way POSIX specs it

Reviewed by:	kib, bde
2009-08-25 21:44:14 +00:00
Marko Zec
0348c661d1 Fix NFS panics with options VIMAGE kernels by apropriately setting curvnet
context inside the RPC code.

Temporarily set td's cred to mount's cred before calling socreate() via
__rpc_nconf2socket().

Submitted by:	rmacklem (in part)
Reviewed by:	rmacklem, rwatson
Discussed with:	dfr, bz
Approved by:	re (rwatson), julian (mentor)
MFC after:	3 days
2009-08-24 10:09:30 +00:00
Rick Macklem
60f04e38ca Apply the same patch as r196205 for nfs_upgrade_lock() and
nfs_downgrade_lock() to the experimental nfs client.

Approved by:	re (kensmith), kib (mentor)
2009-08-17 16:12:28 +00:00
Robert Watson
530c006014 Merge the remainder of kern_vimage.c and vimage.h into vnet.c and
vnet.h, we now use jails (rather than vimages) as the abstraction
for virtualization management, and what remained was specific to
virtual network stacks.  Minor cleanups are done in the process,
and comments updated to reflect these changes.

Reviewed by:	bz
Approved by:	re (vimage blanket)
2009-08-01 19:26:27 +00:00
John Baldwin
87eca70e0c Fix some LORs between vnode locks and filedescriptor table locks.
- Don't grab the filedesc lock just to read fd_cmask.
- Drop vnode locks earlier when mounting the root filesystem and before
  sanitizing stdin/out/err file descriptors during execve().

Submitted by:	kib
Approved by:	re (rwatson)
MFC after:	1 week
2009-07-31 13:40:06 +00:00
Rick Macklem
6a795918c1 Fix the experimental nfs client so that it only calls ncl_vinvalbuf()
for NFSv2 and not NFSv4 when nfscl_mustflush() returns 0. Since
nfscl_mustflush() only returns 0 when there is a valid write delegation
issued to the client, it only affects the case of an NFSv4 mount with
callbacks/delegations enabled.

Approved by:	 re (kensmith), kib (mentor)
2009-07-29 14:50:31 +00:00
John Baldwin
013818111a Add a new type of VM object: OBJT_SG. An OBJT_SG object is very similar to
a device pager (OBJT_DEVICE) object in that it uses fictitious pages to
provide aliases to other memory addresses.  The primary difference is that
it uses an sglist(9) to determine the physical addresses for a given offset
into the object instead of invoking the d_mmap() method in a device driver.

Reviewed by:	alc
Approved by:	re (kensmith)
MFC after:	2 weeks
2009-07-24 13:50:29 +00:00
Rick Macklem
93a15e2054 When vfs.newnfs.callback_addr is set to an IPv4 address, the
experimental NFSv4 client might try and use it as an IPv6 address,
breaking callbacks. The fix simply initializes the isinet6 variable
for this case.

Approved by:	re (kensmith), kib (mentor)
2009-07-22 18:10:44 +00:00
Rick Macklem
c79e697621 Add changes to the experimental nfs client to use the PBDRY flag for
msleep(9) when a vnode lock or similar may be held. The changes are
just a clone of the changes applied to the regular nfs client by
r195703.

Approved by:	re (kensmith), kib (mentor)
2009-07-22 14:37:53 +00:00
Rick Macklem
7f1968ba10 When using an NFSv4 mount in the experimental nfs client with delegations
being issued from the server, there was a case where an Open issued locally
based on the delegation would be released before the associated vnode
became inactive. If the delegation was recalled after the open was released,
an Open against the server would not have been acquired and subsequent I/O
operations would need to use the special stateid of all zeros. This patch
fixes that case.

Approved by:	re (kensmith), kib (mentor)
2009-07-22 14:32:28 +00:00
Rick Macklem
80e556956e Fix two bugs in the experimental nfs client:
- When the root vnode was acquired during mounting, mnt_stat.f_iosize was
  still set to 0, so getnewvnode() would set bo_bsize == 0. This would
  confuse getblk(), so that it always returned the first block causing
  the problem when the root directory of the mount point was greater
  than one block in size. It was fixed by setting mnt_stat.f_iosize to
  NFS_DIRBLKSIZ before calling ncl_nget() to acquire the root vnode.
- NFSMNT_INT was being set temporarily while the initial connect to a
  server was being done. This erroneously configured the krpc for
  interruptible RPCs, which caused problems because signals weren't
  being masked off as they would have been for interruptible mounts.
  This code was deleted to fix the problem. Since mount_nfs does an
  NFS null RPC before the mount system call, connections to the server
  should work ok.

Tested by:	swell dot k at gmail dot com
Approved by:	re (kensmith), kib (mentor)
2009-07-19 16:44:26 +00:00
Rick Macklem
405229f913 Fix the experimental nfs client so that it does not cause a
"share->excl" panic when doing a lookup of dotdot at the root
of a server's file system. The patch avoids calling vn_lock()
for that case, since nfscl_nget() has already acquired a lock
for the vnode.

Approved by:	re (kensmith), kib (mentor)
2009-07-14 23:10:23 +00:00
Robert Watson
eddfbb763d 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
Rick Macklem
089f366ab0 Add calls to the experimental nfs client for the case of an "intr" mount,
so that signals that aren't supposed to terminate RPCs in progress are
masked off during the RPC.

Approved by:	re (kensmith), kib (mentor)
2009-07-12 17:07:35 +00:00
Rick Macklem
ad86aef9af Fix the handling of dotdot in lookup for the experimental nfs client
in a manner analagous to the change in r195294 for the regular nfs client.

Approved by:	re (kensmith), kib (mentor)
2009-07-12 17:02:17 +00:00
Rick Macklem
9ca27b565b Since the nfscl_getclose() function both decremented open counts and,
optionally, created a separate list of NFSv4 opens to be closed, it
was possible for the associated OpenOwner to be free'd before the Open
was closed. The problem was that the Open was taken off the OpenOwner
list before the Close RPC was done and OpenOwners can be free'd once the
list is empty. This patch separates out the case of doing the Close RPC
into a separate function called nfscl_doclose() and simplifies nfsrpc_doclose()
so that it closes a single open instead of a list of them. This avoids
removing the Open from the OpenOwner list before doing the Close RPC.

Approved by:	re (kensmith), kib (mentor)
2009-07-09 19:00:29 +00:00
Konstantin Belousov
7f5dff5064 Fix poll(2) and select(2) for named pipes to return "ready for read"
when all writers, observed by reader, exited. Use writer generation
counter for fifo, and store the snapshot of the fifo generation in the
f_seqcount field of struct file, that is otherwise unused for fifos.
Set FreeBSD-undocumented POLLINIGNEOF flag only when file f_seqcount is
equal to fifo' fi_wgen, and revert r89376.

Fix POLLINIGNEOF for sockets and pipes, and return POLLHUP for them.
Note that the patch does not fix not returning POLLHUP for fifos.

PR:	kern/94772
Submitted by:	bde (original version)
Reviewed by:	rwatson, jilles
Approved by:	re (kensmith)
MFC after:	6 weeks (might be)
2009-07-07 09:43:44 +00:00
Konstantin Belousov
f1eccd05ec In vn_vget_ino() and their inline equivalents, mnt_ref() the mount point
around the sequence that drop vnode lock and then busies the mount point.
Not having vlocked node or direct reference to the mp allows for the
forced unmount to proceed, making mp unmounted or reused.

Tested by:	pho
Reviewed by:	jeff
Approved by:	re (kensmith)
MFC after:	2 weeks
2009-07-02 18:02:55 +00:00
Konstantin Belousov
9f80ce043d Change the type of uio_resid member of struct uio from int to ssize_t.
Note that this does not actually enable full-range i/o requests for
64 architectures, and is done now to update KBI only.

Tested by:	pho
Reviewed by:	jhb, bde (as part of the review of the bigger patch)
2009-06-25 18:46:30 +00:00
Robert Watson
2d9cfabad4 Add a new global rwlock, in_ifaddr_lock, which will synchronize use of the
in_ifaddrhead and INADDR_HASH address lists.

Previously, these lists were used unsynchronized as they were effectively
never changed in steady state, but we've seen increasing reports of
writer-writer races on very busy VPN servers as core count has gone up
(and similar configurations where address lists change frequently and
concurrently).

For the time being, use rwlocks rather than rmlocks in order to take
advantage of their better lock debugging support.  As a result, we don't
enable ip_input()'s read-locking of INADDR_HASH until an rmlock conversion
is complete and a performance analysis has been done.  This means that one
class of reader-writer races still exists.

MFC after:      6 weeks
Reviewed by:    bz
2009-06-25 11:52:33 +00:00
Konstantin Belousov
3364c323e6 Implement global and per-uid accounting of the anonymous memory. Add
rlimit RLIMIT_SWAP that limits the amount of swap that may be reserved
for the uid.

The accounting information (charge) is associated with either map entry,
or vm object backing the entry, assuming the object is the first one
in the shadow chain and entry does not require COW. Charge is moved
from entry to object on allocation of the object, e.g. during the mmap,
assuming the object is allocated, or on the first page fault on the
entry. It moves back to the entry on forks due to COW setup.

The per-entry granularity of accounting makes the charge process fair
for processes that change uid during lifetime, and decrements charge
for proper uid when region is unmapped.

The interface of vm_pager_allocate(9) is extended by adding struct ucred *,
that is used to charge appropriate uid when allocation if performed by
kernel, e.g. md(4).

Several syscalls, among them is fork(2), may now return ENOMEM when
global or per-uid limits are enforced.

In collaboration with:	pho
Reviewed by:	alc
Approved by:	re (kensmith)
2009-06-23 20:45:22 +00:00
Konstantin Belousov
c808c9632d Add explicit struct ucred * argument for VOP_VPTOCNP, to be used by
vn_open_cred in default implementation. Valid struct ucred is needed for
audit and MAC, and curthread credentials may be wrong.

This further requires modifying the interface of vn_fullpath(9), but it
is out of scope of this change.

Reviewed by:	rwatson
2009-06-21 19:21:01 +00:00
Roman Divacky
23057f089b In non-debugging mode make this define (void)0 instead of nothing. This
helps to catch bugs like the below with clang.

	if (cond);		<--- note the trailing ;
	   something();

Approved by:	ed (mentor)
Discussed on:	current@
2009-06-21 08:36:30 +00:00
Rick Macklem
65cc6600c5 Replace RPCAUTH_UNIXGIDS with NFS_MAXGRPS so that nfscbd.c will build.
Approved by:	kib (mentor)
2009-06-20 17:11:07 +00:00
Ed Schouten
f8f6146082 Improve nested jail awareness of devfs by handling credentials.
Now that we start to use credentials on character devices more often
(because of MPSAFE TTY), move the prison-checks that are in place in the
TTY code into devfs.

Instead of strictly comparing the prisons, use the more common
prison_check() function to compare credentials. This means that
pseudo-terminals are only visible in devfs by processes within the same
jail and parent jails.

Even though regular users in parent jails can now interact with
pseudo-terminals from child jails, this seems to be the right approach.
These processes are also capable of interacting with the jailed
processes anyway, through signals for example.

Reviewed by:	kib, rwatson (older version)
2009-06-20 14:50:32 +00:00
Rick Macklem
2c1e6cce5b Change the size of the nfsc_groups[] array in the experimental nfs
client to RPCAUTH_UNIXGIDS + 1 (17), since that is what can go on
the wire for AUTH_SYS authentication.

Reviewed by:	brooks
Approved by:	kib (mentor)
2009-06-20 00:54:57 +00:00
Brooks Davis
838d985825 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
Alan Cox
57a7e73261 Fix some of the style errors in *getpages(). 2009-06-18 05:56:24 +00:00
Rick Macklem
76b30a0cd4 Add the SVC_RELEASE(xprt), as required by r194407.
Approved by:	kib (mentor)
2009-06-17 22:55:59 +00:00
Bjoern A. Zeeb
ebd8672cc3 Add explicit includes for jail.h to the files that need them and
remove the "hidden" one from vimage.h.
2009-06-17 15:01:01 +00:00
Rick Macklem
81e3c4fc8e Fix handling of ".." in nfs_lookup() for the forced dismount case
by cribbing the change made to the regular nfs client in r194358.

Approved by:	kib (mentor)
2009-06-17 14:10:18 +00:00
Bjoern A. Zeeb
7654a365db Add the explicit include of vimage.h to another five .c files still
missing it.

Remove the "hidden" kernel only include of vimage.h from ip_var.h added
with the very first Vimage commit r181803 to avoid further kernel poisoning.
2009-06-17 12:44:11 +00:00
Rick Macklem
47b7dc9933 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
Alan Cox
47f11d9a46 Eliminate unnecessary variables. 2009-06-13 20:21:08 +00:00
Jamie Gritton
c1f192193d Rename the host-related prison fields to be the same as the host.*
parameters they represent, and the variables they replaced, instead of
abbreviated versions of them.

Approved by:	bz (mentor)
2009-06-13 15:39:12 +00:00
Jamie Gritton
01de879ac7 Use getcredhostuuid instead of accessing the prison directly.
Approved by:	bz (mentor)
2009-06-13 15:35:22 +00:00
John Baldwin
04f7f4636f Update the inline version of vn_get_ino() for ".." lookups to match the
recentish changes to vn_get_ino().

MFC after:	1 week
2009-06-12 21:19:57 +00:00
Rick Macklem
934a309971 This commit is analagous to r193952, but for the experimental nfs
subsystem. Add a test for VI_DOOMED just after ncl_upgrade_vnlock() in
ncl_bioread_check_cons(). This is required since it is possible
for the vnode to be vgonel()'d while in ncl_upgrade_vnlock() when
a forced dismount is in progress. Also, move the check for VI_DOOMED
in ncl_vinvalbuf() down to after ncl_upgrade_vnlock() and replace the
out of date comment for it.

Approved by:	kib (mentor)
2009-06-10 21:16:39 +00:00
Konstantin Belousov
93bc76dc3e For cd9660_ioctl, check for recycled vnode after locking it.
Noted by:	Jaakko Heinonen <jh saunalahti fi>
MFC after:	2 weeks
2009-06-10 15:48:34 +00:00
Konstantin Belousov
d6da640860 Fix r193923 by noting that type of a_fp is struct file *, not int.
It was assumed that r193923 was trivial change that cannot be done
wrong.

MFC after:	2 weeks
2009-06-10 14:24:31 +00:00
Konstantin Belousov
e4d9bdc105 s/a_fdidx/a_fp/ for VOP_OPEN comments that inline struct vop_open_args
definition.

Discussed with:	bde
MFC after:	2 weeks
2009-06-10 14:09:05 +00:00
Konstantin Belousov
c4702e66f4 Remove unused VOP_IOCTL and VOP_KQFILTER implementations for fifofs.
MFC after:	2 weeks
2009-06-10 14:02:22 +00:00
Konstantin Belousov
c4df27d5c8 VOP_IOCTL takes unlocked vnode as an argument. Due to this, v_data may
be NULL or derefenced memory may become free at arbitrary moment.

Lock the vnode in cd9660, devfs and pseudofs implementation of VOP_IOCTL
to prevent reclaim; check whether the vnode was already reclaimed after
the lock is granted.

Reported by:	georg at dts su
Reviewed by:	des (pseudofs)
MFC after:	2 weeks
2009-06-10 13:57:36 +00:00
Rick Macklem
410654ec1b Since vn_lock() with the LK_RETRY flag never returns an error
for FreeBSD-CURRENT, the code that checked for and returned the
error was broken. Change it to check for VI_DOOMED set after
vn_lock() and return an error for that case. I believe this
should only happen for forced dismounts.

Approved by:	kib (mentor)
2009-06-09 15:18:01 +00:00