d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
12c8434bbd
freebsd-skq/sys/fs/nfsserver/nfs_nfsdsocket.c
ken fc3fc9c036 Revamp the old NFS server's File Handle Affinity (FHA) code so that 
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
2013-04-17 21:00:22 +00:00

956 lines
37 KiB
C
Raw Blame History

/*-
* Copyright (c) 1989, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* Rick Macklem at The University of Guelph.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
/*
* Socket operations for use by the nfs server.
*/
#ifndef APPLEKEXT
#include <fs/nfs/nfsport.h>
extern struct nfsstats newnfsstats;
extern struct nfsrvfh nfs_pubfh, nfs_rootfh;
extern int nfs_pubfhset, nfs_rootfhset;
extern struct nfsv4lock nfsv4rootfs_lock;
extern struct nfsrv_stablefirst nfsrv_stablefirst;
extern struct nfsclienthashhead nfsclienthash[NFSCLIENTHASHSIZE];
extern int nfsrc_floodlevel, nfsrc_tcpsavedreplies;
NFSV4ROOTLOCKMUTEX;
NFSSTATESPINLOCK;
int (*nfsrv3_procs0[NFS_V3NPROCS])(struct nfsrv_descript *,
int, vnode_t , NFSPROC_T *, struct nfsexstuff *) = {
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_getattr,
nfsrvd_setattr,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_access,
nfsrvd_readlink,
nfsrvd_read,
nfsrvd_write,
nfsrvd_create,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_remove,
nfsrvd_remove,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_readdir,
nfsrvd_readdirplus,
nfsrvd_statfs,
nfsrvd_fsinfo,
nfsrvd_pathconf,
nfsrvd_commit,
};
int (*nfsrv3_procs1[NFS_V3NPROCS])(struct nfsrv_descript *,
int, vnode_t , vnode_t *, fhandle_t *,
NFSPROC_T *, struct nfsexstuff *) = {
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_lookup,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_mkdir,
nfsrvd_symlink,
nfsrvd_mknod,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
};
int (*nfsrv3_procs2[NFS_V3NPROCS])(struct nfsrv_descript *,
int, vnode_t , vnode_t , NFSPROC_T *,
struct nfsexstuff *, struct nfsexstuff *) = {
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
nfsrvd_rename,
nfsrvd_link,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
};
int (*nfsrv4_ops0[NFSV4OP_NOPS])(struct nfsrv_descript *,
int, vnode_t , NFSPROC_T *, struct nfsexstuff *) = {
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_access,
nfsrvd_close,
nfsrvd_commit,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_delegpurge,
nfsrvd_delegreturn,
nfsrvd_getattr,
nfsrvd_getfh,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_lock,
nfsrvd_lockt,
nfsrvd_locku,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_verify,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_openconfirm,
nfsrvd_opendowngrade,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_read,
nfsrvd_readdirplus,
nfsrvd_readlink,
nfsrvd_remove,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_renew,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_secinfo,
nfsrvd_setattr,
nfsrvd_setclientid,
nfsrvd_setclientidcfrm,
nfsrvd_verify,
nfsrvd_write,
nfsrvd_releaselckown,
};
int (*nfsrv4_ops1[NFSV4OP_NOPS])(struct nfsrv_descript *,
int, vnode_t , vnode_t *, fhandle_t *,
NFSPROC_T *, struct nfsexstuff *) = {
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_mknod,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_lookup,
nfsrvd_lookup,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
nfsrvd_open,
nfsrvd_openattr,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t *, fhandle_t *, NFSPROC_T *, struct nfsexstuff *))0,
};
int (*nfsrv4_ops2[NFSV4OP_NOPS])(struct nfsrv_descript *,
int, vnode_t , vnode_t , NFSPROC_T *,
struct nfsexstuff *, struct nfsexstuff *) = {
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
nfsrvd_link,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
nfsrvd_rename,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
(int (*)(struct nfsrv_descript *, int, vnode_t , vnode_t , NFSPROC_T *, struct nfsexstuff *, struct nfsexstuff *))0,
};
#endif /* !APPLEKEXT */
/*
* Static array that defines which nfs rpc's are nonidempotent
*/
static int nfsrv_nonidempotent[NFS_V3NPROCS] = {
FALSE,
FALSE,
TRUE,
FALSE,
FALSE,
FALSE,
FALSE,
TRUE,
TRUE,
TRUE,
TRUE,
TRUE,
TRUE,
TRUE,
TRUE,
TRUE,
FALSE,
FALSE,
FALSE,
FALSE,
FALSE,
FALSE,
};
/*
* This static array indicates whether or not the RPC modifies the
* file system.
*/
static int nfs_writerpc[NFS_NPROCS] = { 0, 0, 1, 0, 0, 0, 0,
1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1 };
/* local functions */
static void nfsrvd_compound(struct nfsrv_descript *nd, int isdgram,
NFSPROC_T *p);
/*
* This static array indicates which server procedures require the extra
* arguments to return the current file handle for V2, 3.
*/
static int nfs_retfh[NFS_V3NPROCS] = { 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 1,
1, 0, 0, 2, 2, 0, 0, 0, 0, 0, 0 };
extern struct nfsv4_opflag nfsv4_opflag[NFSV4OP_NOPS];
static int nfsv3to4op[NFS_V3NPROCS] = {
NFSPROC_NULL,
NFSV4OP_GETATTR,
NFSV4OP_SETATTR,
NFSV4OP_LOOKUP,
NFSV4OP_ACCESS,
NFSV4OP_READLINK,
NFSV4OP_READ,
NFSV4OP_WRITE,
NFSV4OP_V3CREATE,
NFSV4OP_MKDIR,
NFSV4OP_SYMLINK,
NFSV4OP_MKNOD,
NFSV4OP_REMOVE,
NFSV4OP_RMDIR,
NFSV4OP_RENAME,
NFSV4OP_LINK,
NFSV4OP_READDIR,
NFSV4OP_READDIRPLUS,
NFSV4OP_FSSTAT,
NFSV4OP_FSINFO,
NFSV4OP_PATHCONF,
NFSV4OP_COMMIT,
};
/*
* Do an RPC. Basically, get the file handles translated to vnode pointers
* and then call the appropriate server routine. The server routines are
* split into groups, based on whether they use a file handle or file
* handle plus name or ...
* The NFS V4 Compound RPC is performed separately by nfsrvd_compound().
*/
APPLESTATIC void
nfsrvd_dorpc(struct nfsrv_descript *nd, int isdgram,
NFSPROC_T *p)
{
int error = 0, lktype;
vnode_t vp;
mount_t mp = NULL;
struct nfsrvfh fh;
struct nfsexstuff nes;
/*
* Get a locked vnode for the first file handle
*/
if (!(nd->nd_flag & ND_NFSV4)) {
KASSERT(nd->nd_repstat == 0, ("nfsrvd_dorpc"));
/*
* For NFSv3, if the malloc/mget allocation is near limits,
* return NFSERR_DELAY.
*/
if ((nd->nd_flag & ND_NFSV3) && nfsrv_mallocmget_limit()) {
nd->nd_repstat = NFSERR_DELAY;
vp = NULL;
} else {
error = nfsrv_mtofh(nd, &fh);
if (error) {
if (error != EBADRPC)
printf("nfs dorpc err1=%d\n", error);
nd->nd_repstat = NFSERR_GARBAGE;
goto out;
}
if (nd->nd_procnum == NFSPROC_READ ||
nd->nd_procnum == NFSPROC_WRITE ||
nd->nd_procnum == NFSPROC_READDIR ||
nd->nd_procnum == NFSPROC_READLINK ||
nd->nd_procnum == NFSPROC_GETATTR ||
nd->nd_procnum == NFSPROC_ACCESS)
lktype = LK_SHARED;
else
lktype = LK_EXCLUSIVE;
if (nd->nd_flag & ND_PUBLOOKUP)
nfsd_fhtovp(nd, &nfs_pubfh, lktype, &vp, &nes,
&mp, nfs_writerpc[nd->nd_procnum], p);
else
nfsd_fhtovp(nd, &fh, lktype, &vp, &nes,
&mp, nfs_writerpc[nd->nd_procnum], p);
if (nd->nd_repstat == NFSERR_PROGNOTV4)
goto out;
}
}
/*
* For V2 and 3, set the ND_SAVEREPLY flag for the recent request
* cache, as required.
* For V4, nfsrvd_compound() does this.
*/
if (!(nd->nd_flag & ND_NFSV4) && nfsrv_nonidempotent[nd->nd_procnum])
nd->nd_flag |= ND_SAVEREPLY;
nfsrvd_rephead(nd);
/*
* If nd_repstat is non-zero, just fill in the reply status
* to complete the RPC reply for V2. Otherwise, you must do
* the RPC.
*/
if (nd->nd_repstat && (nd->nd_flag & ND_NFSV2)) {
*nd->nd_errp = nfsd_errmap(nd);
NFSINCRGLOBAL(newnfsstats.srvrpccnt[nfsv3to4op[nd->nd_procnum]]);
if (mp != NULL && nfs_writerpc[nd->nd_procnum] != 0)
vn_finished_write(mp);
goto out;
}
/*
* Now the procedure can be performed. For V4, nfsrvd_compound()
* works through the sub-rpcs, otherwise just call the procedure.
* The procedures are in three groups with different arguments.
* The group is indicated by the value in nfs_retfh[].
*/
if (nd->nd_flag & ND_NFSV4) {
nfsrvd_compound(nd, isdgram, p);
} else {
if (nfs_retfh[nd->nd_procnum] == 1) {
if (vp)
NFSVOPUNLOCK(vp, 0);
error = (*(nfsrv3_procs1[nd->nd_procnum]))(nd, isdgram,
vp, NULL, (fhandle_t *)fh.nfsrvfh_data, p, &nes);
} else if (nfs_retfh[nd->nd_procnum] == 2) {
error = (*(nfsrv3_procs2[nd->nd_procnum]))(nd, isdgram,
vp, NULL, p, &nes, NULL);
} else {
error = (*(nfsrv3_procs0[nd->nd_procnum]))(nd, isdgram,
vp, p, &nes);
}
if (mp != NULL && nfs_writerpc[nd->nd_procnum] != 0)
vn_finished_write(mp);
NFSINCRGLOBAL(newnfsstats.srvrpccnt[nfsv3to4op[nd->nd_procnum]]);
}
if (error) {
if (error != EBADRPC)
printf("nfs dorpc err2=%d\n", error);
nd->nd_repstat = NFSERR_GARBAGE;
}
*nd->nd_errp = nfsd_errmap(nd);
/*
* Don't cache certain reply status values.
*/
if (nd->nd_repstat && (nd->nd_flag & ND_SAVEREPLY) &&
(nd->nd_repstat == NFSERR_GARBAGE ||
nd->nd_repstat == NFSERR_BADXDR ||
nd->nd_repstat == NFSERR_MOVED ||
nd->nd_repstat == NFSERR_DELAY ||
nd->nd_repstat == NFSERR_BADSEQID ||
nd->nd_repstat == NFSERR_RESOURCE ||
nd->nd_repstat == NFSERR_SERVERFAULT ||
nd->nd_repstat == NFSERR_STALECLIENTID ||
nd->nd_repstat == NFSERR_STALESTATEID ||
nd->nd_repstat == NFSERR_OLDSTATEID ||
nd->nd_repstat == NFSERR_BADSTATEID ||
nd->nd_repstat == NFSERR_GRACE ||
nd->nd_repstat == NFSERR_NOGRACE))
nd->nd_flag &= ~ND_SAVEREPLY;
out:
NFSEXITCODE2(0, nd);
}
/*
* Breaks down a compound RPC request and calls the server routines for
* the subprocedures.
* Some suboperations are performed directly here to simplify file handle<-->
* vnode pointer handling.
*/
static void
nfsrvd_compound(struct nfsrv_descript *nd, int isdgram,
NFSPROC_T *p)
{
int i, op;
u_int32_t *tl;
struct nfsclient *clp, *nclp;
int numops, taglen = -1, error = 0, igotlock;
u_int32_t minorvers, retops = 0, *retopsp = NULL, *repp;
u_char tag[NFSV4_SMALLSTR + 1], *tagstr;
vnode_t vp, nvp, savevp;
struct nfsrvfh fh;
mount_t new_mp, temp_mp = NULL;
struct ucred *credanon;
struct nfsexstuff nes, vpnes, savevpnes;
fsid_t cur_fsid, save_fsid;
static u_int64_t compref = 0;
NFSVNO_EXINIT(&vpnes);
NFSVNO_EXINIT(&savevpnes);
/*
* Put the seq# of the current compound RPC in nfsrv_descript.
* (This is used by nfsrv_checkgetattr(), to see if the write
* delegation was created by the same compound RPC as the one
* with that Getattr in it.)
* Don't worry about the 64bit number wrapping around. It ain't
* gonna happen before this server gets shut down/rebooted.
*/
nd->nd_compref = compref++;
/*
* Check for and optionally get a lock on the root. This lock means that
* no nfsd will be fiddling with the V4 file system and state stuff. It
* is required when the V4 root is being changed, the stable storage
* restart file is being updated, or callbacks are being done.
* When any of the nfsd are processing an NFSv4 compound RPC, they must
* either hold a reference count (nfs_usecnt) or the lock. When
* nfsrv_unlock() is called to release the lock, it can optionally
* also get a reference count, which saves the need for a call to
* nfsrv_getref() after nfsrv_unlock().
*/
/*
* First, check to see if we need to wait for an update lock.
*/
igotlock = 0;
NFSLOCKV4ROOTMUTEX();
if (nfsrv_stablefirst.nsf_flags & NFSNSF_NEEDLOCK)
igotlock = nfsv4_lock(&nfsv4rootfs_lock, 1, NULL,
NFSV4ROOTLOCKMUTEXPTR, NULL);
else
igotlock = nfsv4_lock(&nfsv4rootfs_lock, 0, NULL,
NFSV4ROOTLOCKMUTEXPTR, NULL);
NFSUNLOCKV4ROOTMUTEX();
if (igotlock) {
/*
* If I got the lock, I can update the stable storage file.
* Done when the grace period is over or a client has long
* since expired.
*/
nfsrv_stablefirst.nsf_flags &= ~NFSNSF_NEEDLOCK;
if ((nfsrv_stablefirst.nsf_flags &
(NFSNSF_GRACEOVER | NFSNSF_UPDATEDONE)) == NFSNSF_GRACEOVER)
nfsrv_updatestable(p);
/*
* If at least one client has long since expired, search
* the client list for them, write a REVOKE record on the
* stable storage file and then remove them from the client
* list.
*/
if (nfsrv_stablefirst.nsf_flags & NFSNSF_EXPIREDCLIENT) {
nfsrv_stablefirst.nsf_flags &= ~NFSNSF_EXPIREDCLIENT;
for (i = 0; i < NFSCLIENTHASHSIZE; i++) {
LIST_FOREACH_SAFE(clp, &nfsclienthash[i], lc_hash,
nclp) {
if (clp->lc_flags & LCL_EXPIREIT) {
if (!LIST_EMPTY(&clp->lc_open) ||
!LIST_EMPTY(&clp->lc_deleg))
nfsrv_writestable(clp->lc_id,
clp->lc_idlen, NFSNST_REVOKE, p);
nfsrv_cleanclient(clp, p);
nfsrv_freedeleglist(&clp->lc_deleg);
nfsrv_freedeleglist(&clp->lc_olddeleg);
LIST_REMOVE(clp, lc_hash);
nfsrv_zapclient(clp, p);
}
}
}
}
NFSLOCKV4ROOTMUTEX();
nfsv4_unlock(&nfsv4rootfs_lock, 1);
NFSUNLOCKV4ROOTMUTEX();
} else {
/*
* If we didn't get the lock, we need to get a refcnt,
* which also checks for and waits for the lock.
*/
NFSLOCKV4ROOTMUTEX();
nfsv4_getref(&nfsv4rootfs_lock, NULL,
NFSV4ROOTLOCKMUTEXPTR, NULL);
NFSUNLOCKV4ROOTMUTEX();
}
/*
* If flagged, search for open owners that haven't had any opens
* for a long time.
*/
if (nfsrv_stablefirst.nsf_flags & NFSNSF_NOOPENS) {
nfsrv_throwawayopens(p);
}
savevp = vp = NULL;
save_fsid.val[0] = save_fsid.val[1] = 0;
cur_fsid.val[0] = cur_fsid.val[1] = 0;
NFSM_DISSECT(tl, u_int32_t *, NFSX_UNSIGNED);
taglen = fxdr_unsigned(int, *tl);
if (taglen < 0) {
error = EBADRPC;
goto nfsmout;
}
if (taglen <= NFSV4_SMALLSTR)
tagstr = tag;
else
tagstr = malloc(taglen + 1, M_TEMP, M_WAITOK);
error = nfsrv_mtostr(nd, tagstr, taglen);
if (error) {
if (taglen > NFSV4_SMALLSTR)
free(tagstr, M_TEMP);
taglen = -1;
goto nfsmout;
}
(void) nfsm_strtom(nd, tag, taglen);
if (taglen > NFSV4_SMALLSTR) {
free(tagstr, M_TEMP);
}
NFSM_BUILD(retopsp, u_int32_t *, NFSX_UNSIGNED);
NFSM_DISSECT(tl, u_int32_t *, 2 * NFSX_UNSIGNED);
minorvers = fxdr_unsigned(u_int32_t, *tl++);
if (minorvers != NFSV4_MINORVERSION)
nd->nd_repstat = NFSERR_MINORVERMISMATCH;
if (nd->nd_repstat)
numops = 0;
else
numops = fxdr_unsigned(int, *tl);
/*
* Loop around doing the sub ops.
* vp - is an unlocked vnode pointer for the CFH
* savevp - is an unlocked vnode pointer for the SAVEDFH
* (at some future date, it might turn out to be more appropriate
* to keep the file handles instead of vnode pointers?)
* savevpnes and vpnes - are the export flags for the above.
*/
for (i = 0; i < numops; i++) {
NFSM_DISSECT(tl, u_int32_t *, NFSX_UNSIGNED);
NFSM_BUILD(repp, u_int32_t *, 2 * NFSX_UNSIGNED);
*repp = *tl;
op = fxdr_unsigned(int, *tl);
if (op < NFSV4OP_ACCESS || op >= NFSV4OP_NOPS) {
nd->nd_repstat = NFSERR_OPILLEGAL;
*repp++ = txdr_unsigned(NFSV4OP_OPILLEGAL);
*repp = nfsd_errmap(nd);
retops++;
break;
} else {
repp++;
}
/*
* Check for a referral on the current FH and, if so, return
* NFSERR_MOVED for all ops that allow it, except Getattr.
*/
if (vp != NULL && op != NFSV4OP_GETATTR &&
nfsv4root_getreferral(vp, NULL, 0) != NULL &&
nfsrv_errmoved(op)) {
nd->nd_repstat = NFSERR_MOVED;
*repp = nfsd_errmap(nd);
retops++;
break;
}
nd->nd_procnum = op;
/*
* If over flood level, reply NFSERR_RESOURCE, if at the first
* Op. (Since a client recovery from NFSERR_RESOURCE can get
* really nasty for certain Op sequences, I'll play it safe
* and only return the error at the beginning.) The cache
* will still function over flood level, but uses lots of
* mbufs.)
* If nfsrv_mallocmget_limit() returns True, the system is near
* to its limit for memory that malloc()/mget() can allocate.
*/
if (i == 0 && nd->nd_rp->rc_refcnt == 0 &&
(nfsrv_mallocmget_limit() ||
nfsrc_tcpsavedreplies > nfsrc_floodlevel)) {
if (nfsrc_tcpsavedreplies > nfsrc_floodlevel) {
printf("nfsd server cache flooded, try to");
printf(" increase nfsrc_floodlevel\n");
}
nd->nd_repstat = NFSERR_RESOURCE;
*repp = nfsd_errmap(nd);
if (op == NFSV4OP_SETATTR) {
/*
* Setattr replies require a bitmap.
* even for errors like these.
*/
NFSM_BUILD(tl, u_int32_t *, NFSX_UNSIGNED);
*tl = 0;
}
retops++;
break;
}
if (nfsv4_opflag[op].savereply)
nd->nd_flag |= ND_SAVEREPLY;
NFSINCRGLOBAL(newnfsstats.srvrpccnt[nd->nd_procnum]);
switch (op) {
case NFSV4OP_PUTFH:
error = nfsrv_mtofh(nd, &fh);
if (error)
goto nfsmout;
if (!nd->nd_repstat)
nfsd_fhtovp(nd, &fh, LK_SHARED, &nvp, &nes,
NULL, 0, p);
/* For now, allow this for non-export FHs */
if (!nd->nd_repstat) {
if (vp)
vrele(vp);
vp = nvp;
cur_fsid = vp->v_mount->mnt_stat.f_fsid;
NFSVOPUNLOCK(vp, 0);
vpnes = nes;
}
break;
case NFSV4OP_PUTPUBFH:
if (nfs_pubfhset)
nfsd_fhtovp(nd, &nfs_pubfh, LK_SHARED, &nvp,
&nes, NULL, 0, p);
else
nd->nd_repstat = NFSERR_NOFILEHANDLE;
if (!nd->nd_repstat) {
if (vp)
vrele(vp);
vp = nvp;
cur_fsid = vp->v_mount->mnt_stat.f_fsid;
NFSVOPUNLOCK(vp, 0);
vpnes = nes;
}
break;
case NFSV4OP_PUTROOTFH:
if (nfs_rootfhset) {
nfsd_fhtovp(nd, &nfs_rootfh, LK_SHARED, &nvp,
&nes, NULL, 0, p);
if (!nd->nd_repstat) {
if (vp)
vrele(vp);
vp = nvp;
cur_fsid = vp->v_mount->mnt_stat.f_fsid;
NFSVOPUNLOCK(vp, 0);
vpnes = nes;
}
} else
nd->nd_repstat = NFSERR_NOFILEHANDLE;
break;
case NFSV4OP_SAVEFH:
if (vp && NFSVNO_EXPORTED(&vpnes)) {
nd->nd_repstat = 0;
/* If vp == savevp, a no-op */
if (vp != savevp) {
if (savevp)
vrele(savevp);
VREF(vp);
savevp = vp;
savevpnes = vpnes;
save_fsid = cur_fsid;
}
} else {
nd->nd_repstat = NFSERR_NOFILEHANDLE;
}
break;
case NFSV4OP_RESTOREFH:
if (savevp) {
nd->nd_repstat = 0;
/* If vp == savevp, a no-op */
if (vp != savevp) {
VREF(savevp);
vrele(vp);
vp = savevp;
vpnes = savevpnes;
cur_fsid = save_fsid;
}
} else {
nd->nd_repstat = NFSERR_RESTOREFH;
}
break;
default:
/*
* Allow a Lookup, Getattr, GetFH, Secinfo on an
* non-exported directory if
* nfs_rootfhset. Do I need to allow any other Ops?
* (You can only have a non-exported vpnes if
* nfs_rootfhset is true. See nfsd_fhtovp())
* Allow AUTH_SYS to be used for file systems
* exported GSS only for certain Ops, to allow
* clients to do mounts more easily.
*/
if (nfsv4_opflag[op].needscfh && vp) {
if (!NFSVNO_EXPORTED(&vpnes) &&
op != NFSV4OP_LOOKUP &&
op != NFSV4OP_GETATTR &&
op != NFSV4OP_GETFH &&
op != NFSV4OP_ACCESS &&
op != NFSV4OP_READLINK &&
op != NFSV4OP_SECINFO)
nd->nd_repstat = NFSERR_NOFILEHANDLE;
else if (nfsvno_testexp(nd, &vpnes) &&
op != NFSV4OP_LOOKUP &&
op != NFSV4OP_GETFH &&
op != NFSV4OP_GETATTR &&
op != NFSV4OP_SECINFO)
nd->nd_repstat = NFSERR_WRONGSEC;
if (nd->nd_repstat) {
if (op == NFSV4OP_SETATTR) {
/*
* Setattr reply requires a bitmap
* even for errors like these.
*/
NFSM_BUILD(tl, u_int32_t *,
NFSX_UNSIGNED);
*tl = 0;
}
break;
}
}
if (nfsv4_opflag[op].retfh == 1) {
if (!vp) {
nd->nd_repstat = NFSERR_NOFILEHANDLE;
break;
}
VREF(vp);
if (nfsv4_opflag[op].modifyfs)
vn_start_write(vp, &temp_mp, V_WAIT);
error = (*(nfsrv4_ops1[op]))(nd, isdgram, vp,
&nvp, (fhandle_t *)fh.nfsrvfh_data, p, &vpnes);
if (!error && !nd->nd_repstat) {
if (op == NFSV4OP_LOOKUP || op == NFSV4OP_LOOKUPP) {
new_mp = nvp->v_mount;
if (cur_fsid.val[0] !=
new_mp->mnt_stat.f_fsid.val[0] ||
cur_fsid.val[1] !=
new_mp->mnt_stat.f_fsid.val[1]) {
/* crossed a server mount point */
nd->nd_repstat = nfsvno_checkexp(new_mp,
nd->nd_nam, &nes, &credanon);
if (!nd->nd_repstat)
nd->nd_repstat = nfsd_excred(nd,
&nes, credanon);
if (credanon != NULL)
crfree(credanon);
if (!nd->nd_repstat) {
vpnes = nes;
cur_fsid = new_mp->mnt_stat.f_fsid;
}
}
/* Lookup ops return a locked vnode */
NFSVOPUNLOCK(nvp, 0);
}
if (!nd->nd_repstat) {
vrele(vp);
vp = nvp;
} else
vrele(nvp);
}
if (nfsv4_opflag[op].modifyfs)
vn_finished_write(temp_mp);
} else if (nfsv4_opflag[op].retfh == 2) {
if (vp == NULL || savevp == NULL) {
nd->nd_repstat = NFSERR_NOFILEHANDLE;
break;
} else if (cur_fsid.val[0] != save_fsid.val[0] ||
cur_fsid.val[1] != save_fsid.val[1]) {
nd->nd_repstat = NFSERR_XDEV;
break;
}
if (nfsv4_opflag[op].modifyfs)
vn_start_write(savevp, &temp_mp, V_WAIT);
if (NFSVOPLOCK(savevp, LK_EXCLUSIVE) == 0) {
VREF(vp);
VREF(savevp);
error = (*(nfsrv4_ops2[op]))(nd, isdgram,
savevp, vp, p, &savevpnes, &vpnes);
} else
nd->nd_repstat = NFSERR_PERM;
if (nfsv4_opflag[op].modifyfs)
vn_finished_write(temp_mp);
} else {
if (nfsv4_opflag[op].retfh != 0)
panic("nfsrvd_compound");
if (nfsv4_opflag[op].needscfh) {
if (vp != NULL) {
if (nfsv4_opflag[op].modifyfs)
vn_start_write(vp, &temp_mp,
V_WAIT);
if (NFSVOPLOCK(vp, nfsv4_opflag[op].lktype)
== 0)
VREF(vp);
else
nd->nd_repstat = NFSERR_PERM;
} else {
nd->nd_repstat = NFSERR_NOFILEHANDLE;
if (op == NFSV4OP_SETATTR) {
/*
* Setattr reply requires a
* bitmap even for errors like
* these.
*/
NFSM_BUILD(tl, u_int32_t *,
NFSX_UNSIGNED);
*tl = 0;
}
break;
}
if (nd->nd_repstat == 0)
error = (*(nfsrv4_ops0[op]))(nd,
isdgram, vp, p, &vpnes);
if (nfsv4_opflag[op].modifyfs)
vn_finished_write(temp_mp);
} else {
error = (*(nfsrv4_ops0[op]))(nd, isdgram,
NULL, p, &vpnes);
}
}
};
if (error) {
if (error == EBADRPC || error == NFSERR_BADXDR) {
nd->nd_repstat = NFSERR_BADXDR;
} else {
nd->nd_repstat = error;
printf("nfsv4 comperr0=%d\n", error);
}
error = 0;
}
retops++;
if (nd->nd_repstat) {
*repp = nfsd_errmap(nd);
break;
} else {
*repp = 0; /* NFS4_OK */
}
}
nfsmout:
if (error) {
if (error == EBADRPC || error == NFSERR_BADXDR)
nd->nd_repstat = NFSERR_BADXDR;
else
printf("nfsv4 comperr1=%d\n", error);
}
if (taglen == -1) {
NFSM_BUILD(tl, u_int32_t *, 2 * NFSX_UNSIGNED);
*tl++ = 0;
*tl = 0;
} else {
*retopsp = txdr_unsigned(retops);
}
if (vp)
vrele(vp);
if (savevp)
vrele(savevp);
NFSLOCKV4ROOTMUTEX();
nfsv4_relref(&nfsv4rootfs_lock);
NFSUNLOCKV4ROOTMUTEX();
NFSEXITCODE2(0, nd);
}
Reference in New Issue View Git Blame Copy Permalink