d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
08e1076cf6
freebsd-nq/sys/fs/nfs/nfs_commonport.c

745 lines
21 KiB
C
Raw Normal View History

Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
/*-
* 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.
Renumber copyright clause 4 Renumber cluase 4 to 3, per what everybody else did when BSD granted them permission to remove clause 3. My insistance on keeping the same numbering for legal reasons is too pedantic, so give up on that point. Submitted by: Jan Schaumann <jschauma@stevens.edu> Pull Request: https://github.com/freebsd/freebsd/pull/96
2017-02-28 23:42:47 +00:00
* 3. Neither the name of the University nor the names of its contributors
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
* 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("$FreeBSD$");
/*
* Functions that need to be different for different versions of BSD
* kernel should be kept here, along with any global storage specific
* to this BSD variant.
*/
#include <fs/nfs/nfsport.h>
#include <sys/sysctl.h>
#include <vm/vm.h>
#include <vm/vm_object.h>
#include <vm/vm_page.h>
#include <vm/vm_param.h>
#include <vm/vm_map.h>
#include <vm/vm_kern.h>
#include <vm/vm_extern.h>
#include <vm/uma.h>
extern int nfscl_ticks;
extern int nfsrv_nfsuserd;
extern struct nfssockreq nfsrv_nfsuserdsock;
extern void (*nfsd_call_recall)(struct vnode *, int, struct ucred *,
struct thread *);
extern int nfsrv_useacl;
struct mount nfsv4root_mnt;
int newnfs_numnfsd = 0;
Update the nfsstats structure to include the changes needed by the patch in D1626 plus changes so that it includes counts for NFSv4.1 (and the draft of NFSv4.2). Also, make all the counts uint64_t and add a vers field at the beginning, so that future revisions can easily be implemented. There is code in place to handle the old vesion of the nfsstats structure for backwards binary compatibility. Subsequent commits will update nfsstat(8) to use the new fields. Submitted by: will (earlier version) Reviewed by: ken MFC after: 1 month Relnotes: yes Differential Revision: https://reviews.freebsd.org/D1626
2016-08-12 22:44:59 +00:00
struct nfsstatsv1 nfsstatsv1;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
int nfs_numnfscbd = 0;
Add a simple printf() based debug facility to the new nfs client. Use it for a printf() that can be harmlessly generated for mmap()'d files. It will be used extensively for the NFSv4.1 client. Debugging printf()s are enabled by setting vfs.nfs.debuglevel to a non-zero value. The higher the value, the more debugging printf()s. Reviewed by: jhb MFC after: 2 weeks
2012-09-09 21:00:45 +00:00
int nfscl_debuglevel = 0;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
char nfsv4_callbackaddr[INET6_ADDRSTRLEN];
struct callout newnfsd_callout;
Add kernel support to the NFS server for the "-manage-gids" option that will be added to the nfsuserd daemon in a future commit. It modifies the cache used by NFSv4 for name<-->id translation (both username/uid and group/gid) to support this. When "-manage-gids" is set, the server looks up each uid for the RPC and uses the list of groups cached in the server instead of the list of groups provided in the RPC request. The cached group list is acquired for the cache by the nfsuserd daemon via getgrouplist(3). This avoids the 16 groups limit for the list in the RPC request. Since the cache is now used for every RPC when "-manage-gids" is enabled, the code also modifies the cache to use a separate mutex for each hash list instead of a single global mutex. Suggested by: jpaetzel Tested by: jpaetzel MFC after: 2 weeks
2015-11-30 21:54:27 +00:00
int nfsrv_lughashsize = 100;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
void (*nfsd_call_servertimer)(void) = NULL;
void (*ncl_call_invalcaches)(struct vnode *) = NULL;
static int nfs_realign_test;
static int nfs_realign_count;
Update the nfsstats structure to include the changes needed by the patch in D1626 plus changes so that it includes counts for NFSv4.1 (and the draft of NFSv4.2). Also, make all the counts uint64_t and add a vers field at the beginning, so that future revisions can easily be implemented. There is code in place to handle the old vesion of the nfsstats structure for backwards binary compatibility. Subsequent commits will update nfsstat(8) to use the new fields. Submitted by: will (earlier version) Reviewed by: ken MFC after: 1 month Relnotes: yes Differential Revision: https://reviews.freebsd.org/D1626
2016-08-12 22:44:59 +00:00
static struct ext_nfsstats oldnfsstats;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
Replace "new NFS" with just "NFS" in some sysctl description strings. Sponsored by: The FreeBSD Foundation
2015-04-19 06:18:41 +00:00
SYSCTL_NODE(_vfs, OID_AUTO, nfs, CTLFLAG_RW, 0, "NFS filesystem");
Change the sysctl naming for the old and new NFS clients to vfs.oldnfs.xxx and vfs.nfs.xxx respectively. This makes the default nfs client use vfs.nfs.xxx after r221124.
2011-05-15 20:52:43 +00:00
SYSCTL_INT(_vfs_nfs, OID_AUTO, realign_test, CTLFLAG_RW, &nfs_realign_test,
Fix up some of the sysctls for the experimental NFS client so that they use the same names as the regular client. Also add string descriptions for them. MFC after: 2 weeks
2011-04-17 18:56:17 +00:00
0, "Number of realign tests done");
Change the sysctl naming for the old and new NFS clients to vfs.oldnfs.xxx and vfs.nfs.xxx respectively. This makes the default nfs client use vfs.nfs.xxx after r221124.
2011-05-15 20:52:43 +00:00
SYSCTL_INT(_vfs_nfs, OID_AUTO, realign_count, CTLFLAG_RW, &nfs_realign_count,
Fix up some of the sysctls for the experimental NFS client so that they use the same names as the regular client. Also add string descriptions for them. MFC after: 2 weeks
2011-04-17 18:56:17 +00:00
0, "Number of mbuf realignments done");
Change the sysctl naming for the old and new NFS clients to vfs.oldnfs.xxx and vfs.nfs.xxx respectively. This makes the default nfs client use vfs.nfs.xxx after r221124.
2011-05-15 20:52:43 +00:00
SYSCTL_STRING(_vfs_nfs, OID_AUTO, callback_addr, CTLFLAG_RW,
Fix up some of the sysctls for the experimental NFS client so that they use the same names as the regular client. Also add string descriptions for them. MFC after: 2 weeks
2011-04-17 18:56:17 +00:00
nfsv4_callbackaddr, sizeof(nfsv4_callbackaddr),
"NFSv4 callback addr for server to use");
Add a simple printf() based debug facility to the new nfs client. Use it for a printf() that can be harmlessly generated for mmap()'d files. It will be used extensively for the NFSv4.1 client. Debugging printf()s are enabled by setting vfs.nfs.debuglevel to a non-zero value. The higher the value, the more debugging printf()s. Reviewed by: jhb MFC after: 2 weeks
2012-09-09 21:00:45 +00:00
SYSCTL_INT(_vfs_nfs, OID_AUTO, debuglevel, CTLFLAG_RW, &nfscl_debuglevel,
Replace "new NFS" with just "NFS" in some sysctl description strings. Sponsored by: The FreeBSD Foundation
2015-04-19 06:18:41 +00:00
0, "Debug level for NFS client");
Add kernel support to the NFS server for the "-manage-gids" option that will be added to the nfsuserd daemon in a future commit. It modifies the cache used by NFSv4 for name<-->id translation (both username/uid and group/gid) to support this. When "-manage-gids" is set, the server looks up each uid for the RPC and uses the list of groups cached in the server instead of the list of groups provided in the RPC request. The cached group list is acquired for the cache by the nfsuserd daemon via getgrouplist(3). This avoids the 16 groups limit for the list in the RPC request. Since the cache is now used for every RPC when "-manage-gids" is enabled, the code also modifies the cache to use a separate mutex for each hash list instead of a single global mutex. Suggested by: jpaetzel Tested by: jpaetzel MFC after: 2 weeks
2015-11-30 21:54:27 +00:00
SYSCTL_INT(_vfs_nfs, OID_AUTO, userhashsize, CTLFLAG_RDTUN, &nfsrv_lughashsize,
0, "Size of hash tables for uid/name mapping");
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
/*
* Defines for malloc
* (Here for FreeBSD, since they allocate storage.)
*/
MALLOC_DEFINE(M_NEWNFSRVCACHE, "NFSD srvcache", "NFSD Server Request Cache");
MALLOC_DEFINE(M_NEWNFSDCLIENT, "NFSD V4client", "NFSD V4 Client Id");
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
MALLOC_DEFINE(M_NEWNFSDSTATE, "NFSD V4state",
"NFSD V4 State (Openowner, Open, Lockowner, Delegation");
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
MALLOC_DEFINE(M_NEWNFSDLOCK, "NFSD V4lock", "NFSD V4 byte range lock");
MALLOC_DEFINE(M_NEWNFSDLOCKFILE, "NFSD lckfile", "NFSD Open/Lock file");
MALLOC_DEFINE(M_NEWNFSSTRING, "NFSD string", "NFSD V4 long string");
MALLOC_DEFINE(M_NEWNFSUSERGROUP, "NFSD usrgroup", "NFSD V4 User/group map");
MALLOC_DEFINE(M_NEWNFSDREQ, "NFS req", "NFS request header");
MALLOC_DEFINE(M_NEWNFSFH, "NFS fh", "NFS file handle");
MALLOC_DEFINE(M_NEWNFSCLOWNER, "NFSCL owner", "NFSCL Open Owner");
MALLOC_DEFINE(M_NEWNFSCLOPEN, "NFSCL open", "NFSCL Open");
MALLOC_DEFINE(M_NEWNFSCLDELEG, "NFSCL deleg", "NFSCL Delegation");
MALLOC_DEFINE(M_NEWNFSCLCLIENT, "NFSCL client", "NFSCL Client");
MALLOC_DEFINE(M_NEWNFSCLLOCKOWNER, "NFSCL lckown", "NFSCL Lock Owner");
MALLOC_DEFINE(M_NEWNFSCLLOCK, "NFSCL lck", "NFSCL Lock");
Replace "new NFS" with just "NFS" in some sysctl description strings. Sponsored by: The FreeBSD Foundation
2015-04-19 06:18:41 +00:00
MALLOC_DEFINE(M_NEWNFSV4NODE, "NEWNFSnode", "NFS vnode");
MALLOC_DEFINE(M_NEWNFSDIRECTIO, "NEWdirectio", "NFS Direct IO buffer");
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
MALLOC_DEFINE(M_NEWNFSDIROFF, "NFSCL diroffdiroff",
Replace "new NFS" with just "NFS" in some sysctl description strings. Sponsored by: The FreeBSD Foundation
2015-04-19 06:18:41 +00:00
"NFS directory offset data");
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
MALLOC_DEFINE(M_NEWNFSDROLLBACK, "NFSD rollback",
Replace "new NFS" with just "NFS" in some sysctl description strings. Sponsored by: The FreeBSD Foundation
2015-04-19 06:18:41 +00:00
"NFS local lock rollback");
Move the NFSv4.1 client patches over from projects/nfsv4.1-client to head. I don't think the NFS client behaviour will change unless the new "minorversion=1" mount option is used. It includes basic NFSv4.1 support plus support for pNFS using the Files Layout only. All problems detecting during an NFSv4.1 Bakeathon testing event in June 2012 have been resolved in this code and it has been tested against the NFSv4.1 server available to me. Although not reviewed, I believe that kib@ has looked at it.
2012-12-08 22:52:39 +00:00
MALLOC_DEFINE(M_NEWNFSLAYOUT, "NFSCL layout", "NFSv4.1 Layout");
MALLOC_DEFINE(M_NEWNFSFLAYOUT, "NFSCL flayout", "NFSv4.1 File Layout");
MALLOC_DEFINE(M_NEWNFSDEVINFO, "NFSCL devinfo", "NFSv4.1 Device Info");
MALLOC_DEFINE(M_NEWNFSSOCKREQ, "NFSCL sockreq", "NFS Sock Req");
MALLOC_DEFINE(M_NEWNFSCLDS, "NFSCL session", "NFSv4.1 Session");
MALLOC_DEFINE(M_NEWNFSLAYRECALL, "NFSCL layrecall", "NFSv4.1 Layout Recall");
Merge the NFSv4.1 server code in projects/nfsv4.1-server over into head. The code is not believed to have any effect on the semantics of non-NFSv4.1 server behaviour. It is a rather large merge, but I am hoping that there will not be any regressions for the NFS server. MFC after: 1 month
2014-07-01 20:47:16 +00:00
MALLOC_DEFINE(M_NEWNFSDSESSION, "NFSD session", "NFSD Session for a client");
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
/*
* Definition of mutex locks.
* newnfsd_mtx is used in nfsrvd_nfsd() to protect the nfs socket list
* and assorted other nfsd structures.
*/
struct mtx newnfsd_mtx;
struct mtx nfs_sockl_mutex;
struct mtx nfs_state_mutex;
struct mtx nfs_nameid_mutex;
struct mtx nfs_req_mutex;
struct mtx nfs_slock_mutex;
Fix NFSv4.1 client recovery from NFS4ERR_BAD_SESSION errors. For most NFSv4.1 servers, a NFS4ERR_BAD_SESSION error is a rare failure that indicates that the server has lost session/open/lock state. However, recent testing by cperciva@ against the AmazonEFS server found several problems with client recovery from this due to it generating this failure frequently. Briefly, the problems fixed are: - If all session slots were in use at the time of the failure, some processes would continue to loop waiting for a slot on the old session forever. - If an RPC that doesn't use open/lock state failed with NFS4ERR_BAD_SESSION, it would fail the RPC/syscall instead of initiating recovery and then looping to retry the RPC. - If a successful reply to an RPC for an old session wasn't processed until after a new session was created for a NFS4ERR_BAD_SESSION error, it would erroneously update the new session and corrupt it. - The use of the first element of the session list in the nfs mount structure (which is always the current metadata session) was slightly racey. With changes for the above problems it became more racey, so all uses of this head pointer was wrapped with a NFSLOCKMNT()/NFSUNLOCKMNT(). - Although the kernel malloc() usually allocates more bytes than requested and, as such, this wouldn't have caused problems, the allocation of a session structure was 1 byte smaller than it should have been. (Null termination byte for the string not included in byte count.) There are probably still problems with a pNFS data server that fails with NFS4ERR_BAD_SESSION, but I have no server that does this to test against (the AmazonEFS server doesn't do pNFS), so I can't fix these yet. Although this patch is fairly large, it should only affect the handling of NFS4ERR_BAD_SESSION error replies from an NFSv4.1 server. Thanks go to cperciva@ for the extension testing he did to help isolate/fix these problems. Reported by: cperciva Tested by: cperciva MFC after: 3 months Differential Revision: https://reviews.freebsd.org/D8745
2016-12-23 23:14:53 +00:00
struct mtx nfs_clstate_mutex;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
/* local functions */
static int nfssvc_call(struct thread *, struct nfssvc_args *, struct ucred *);
Fix the experimental NFS subsystem so that it uses the correct preprocessor macro name for not requiring strict data alignment. Suggested by: marius MFC after: 2 weeks
2010-03-24 02:02:02 +00:00
#ifdef __NO_STRICT_ALIGNMENT
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
/*
* These architectures don't need re-alignment, so just return.
*/
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
int
newnfs_realign(struct mbuf **pm, int how)
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
{
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
return (0);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
}
Fix the experimental NFS subsystem so that it uses the correct preprocessor macro name for not requiring strict data alignment. Suggested by: marius MFC after: 2 weeks
2010-03-24 02:02:02 +00:00
#else /* !__NO_STRICT_ALIGNMENT */
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
/*
Crib the realign function out of nfs_krpc.c and add a call to it for the client side reply. Hopefully this fixes the problem with using the new krpc for arm for the experimental nfs client. Approved by: kib (mentor)
2009-05-24 19:46:12 +00:00
* newnfs_realign:
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
*
* Check for badly aligned mbuf data and realign by copying the unaligned
* portion of the data into a new mbuf chain and freeing the portions
* of the old chain that were replaced.
*
* We cannot simply realign the data within the existing mbuf chain
* because the underlying buffers may contain other rpc commands and
* we cannot afford to overwrite them.
*
* We would prefer to avoid this situation entirely. The situation does
NFS: spelling fixes on comments. No funcional change.
2016-04-29 16:07:25 +00:00
* not occur with NFS/UDP and is supposed to only occasionally occur
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
* with TCP. Use vfs.nfs.realign_count and realign_test to check this.
Crib the realign function out of nfs_krpc.c and add a call to it for the client side reply. Hopefully this fixes the problem with using the new krpc for arm for the experimental nfs client. Approved by: kib (mentor)
2009-05-24 19:46:12 +00:00
*
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
*/
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
int
newnfs_realign(struct mbuf **pm, int how)
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
{
Crib the realign function out of nfs_krpc.c and add a call to it for the client side reply. Hopefully this fixes the problem with using the new krpc for arm for the experimental nfs client. Approved by: kib (mentor)
2009-05-24 19:46:12 +00:00
struct mbuf *m, *n;
int off, space;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
++nfs_realign_test;
while ((m = *pm) != NULL) {
if ((m->m_len & 0x3) || (mtod(m, intptr_t) & 0x3)) {
Crib the realign function out of nfs_krpc.c and add a call to it for the client side reply. Hopefully this fixes the problem with using the new krpc for arm for the experimental nfs client. Approved by: kib (mentor)
2009-05-24 19:46:12 +00:00
/*
* NB: we can't depend on m_pkthdr.len to help us
* decide what to do here. May not be worth doing
* the m_length calculation as m_copyback will
* expand the mbuf chain below as needed.
*/
space = m_length(m, NULL);
if (space >= MINCLSIZE) {
/* NB: m_copyback handles space > MCLBYTES */
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
n = m_getcl(how, MT_DATA, 0);
Crib the realign function out of nfs_krpc.c and add a call to it for the client side reply. Hopefully this fixes the problem with using the new krpc for arm for the experimental nfs client. Approved by: kib (mentor)
2009-05-24 19:46:12 +00:00
} else
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
n = m_get(how, MT_DATA);
Crib the realign function out of nfs_krpc.c and add a call to it for the client side reply. Hopefully this fixes the problem with using the new krpc for arm for the experimental nfs client. Approved by: kib (mentor)
2009-05-24 19:46:12 +00:00
if (n == NULL)
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
return (ENOMEM);
Crib the realign function out of nfs_krpc.c and add a call to it for the client side reply. Hopefully this fixes the problem with using the new krpc for arm for the experimental nfs client. Approved by: kib (mentor)
2009-05-24 19:46:12 +00:00
/*
* Align the remainder of the mbuf chain.
*/
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
n->m_len = 0;
Crib the realign function out of nfs_krpc.c and add a call to it for the client side reply. Hopefully this fixes the problem with using the new krpc for arm for the experimental nfs client. Approved by: kib (mentor)
2009-05-24 19:46:12 +00:00
off = 0;
while (m != NULL) {
m_copyback(n, off, m->m_len, mtod(m, caddr_t));
off += m->m_len;
m = m->m_next;
}
m_freem(*pm);
*pm = n;
++nfs_realign_count;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
break;
}
pm = &m->m_next;
}
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
return (0);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
}
Fix the experimental NFS subsystem so that it uses the correct preprocessor macro name for not requiring strict data alignment. Suggested by: marius MFC after: 2 weeks
2010-03-24 02:02:02 +00:00
#endif /* __NO_STRICT_ALIGNMENT */
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
#ifdef notdef
static void
nfsrv_object_create(struct vnode *vp, struct thread *td)
{
if (vp == NULL || vp->v_type != VREG)
return;
(void) vfs_object_create(vp, td, td->td_ucred);
}
#endif
/*
* Look up a file name. Basically just initialize stuff and call namei().
*/
int
nfsrv_lookupfilename(struct nameidata *ndp, char *fname, NFSPROC_T *p)
{
int error;
Remove the support for using non-mpsafe filesystem modules. In particular, do not lock Giant conditionally when calling into the filesystem module, remove the VFS_LOCK_GIANT() and related macros. Stop handling buffers belonging to non-mpsafe filesystems. The VFS_VERSION is bumped to indicate the interface change which does not result in the interface signatures changes. Conducted and reviewed by: attilio Tested by: pho
2012-10-22 17:50:54 +00:00
NDINIT(ndp, LOOKUP, FOLLOW | LOCKLEAF, UIO_USERSPACE, fname,
For the experimental NFSv4 server's dumplocks operation, add the MPSAFE flag to cn_flags so that it doesn't panic. The panics weren't seen since nfsdumpstate(8) is broken for the "-l" case, so this was never done. I'll do a separate commit to fix nfsdumpstate(8). Submitted by: zack.kirsch at isilon.com MFC after: 2 weeks
2010-07-19 23:33:42 +00:00
p);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
error = namei(ndp);
if (!error) {
NDFREE(ndp, NDF_ONLY_PNBUF);
}
return (error);
}
/*
* Copy NFS uid, gids to the cred structure.
*/
void
newnfs_copycred(struct nfscred *nfscr, struct ucred *cr)
{
The experimental NFS client was not filling in recovery credentials for opens done locally in the client when a delegation for the file was held. This could cause the client to crash in crsetgroups() when recovering from a server crash/reboot. This patch fills in the recovery credentials for this case, in order to avoid the client crash. Also, add KASSERT()s to the credential copy functions, to catch any other cases where the credentials aren't filled in correctly. MFC after: 1 week
2010-04-15 22:57:30 +00:00
KASSERT(nfscr->nfsc_ngroups >= 0,
("newnfs_copycred: negative nfsc_ngroups"));
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
cr->cr_uid = nfscr->nfsc_uid;
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
crsetgroups(cr, nfscr->nfsc_ngroups, nfscr->nfsc_groups);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
}
/*
* Map args from nfsmsleep() to msleep().
*/
int
nfsmsleep(void *chan, void *mutex, int prio, const char *wmesg,
struct timespec *ts)
{
u_int64_t nsecval;
int error, timeo;
if (ts) {
timeo = hz * ts->tv_sec;
nsecval = (u_int64_t)ts->tv_nsec;
nsecval = ((nsecval * ((u_int64_t)hz)) + 500000000) /
1000000000;
timeo += (int)nsecval;
} else {
timeo = 0;
}
error = msleep(chan, (struct mtx *)mutex, prio, wmesg, timeo);
return (error);
}
/*
* Get the file system info for the server. For now, just assume FFS.
*/
void
nfsvno_getfs(struct nfsfsinfo *sip, int isdgram)
{
int pref;
/*
* XXX
* There should be file system VFS OP(s) to get this information.
* For now, assume ufs.
*/
if (isdgram)
pref = NFS_MAXDGRAMDATA;
else
mav@ has found that NFS servers exporting ZFS file systems can perform better when using a 128K read/write data size. This patch changes NFS_MAXDATA from 64K to 128K so that clients can use 128K for NFS mounts to allow this. The patch also renames NFS_MAXDATA to NFS_SRVMAXIO so that it is clear that it applies to the NFS server side only. It also avoids a name conflict with the NFS_MAXDATA defined in rpcsvc/nfs_prot.h, that is used for userland RPC. Tested by: mav Reviewed by: mav MFC after: 2 weeks
2015-04-16 22:35:15 +00:00
pref = NFS_SRVMAXIO;
sip->fs_rtmax = NFS_SRVMAXIO;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
sip->fs_rtpref = pref;
sip->fs_rtmult = NFS_FABLKSIZE;
mav@ has found that NFS servers exporting ZFS file systems can perform better when using a 128K read/write data size. This patch changes NFS_MAXDATA from 64K to 128K so that clients can use 128K for NFS mounts to allow this. The patch also renames NFS_MAXDATA to NFS_SRVMAXIO so that it is clear that it applies to the NFS server side only. It also avoids a name conflict with the NFS_MAXDATA defined in rpcsvc/nfs_prot.h, that is used for userland RPC. Tested by: mav Reviewed by: mav MFC after: 2 weeks
2015-04-16 22:35:15 +00:00
sip->fs_wtmax = NFS_SRVMAXIO;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
sip->fs_wtpref = pref;
sip->fs_wtmult = NFS_FABLKSIZE;
sip->fs_dtpref = pref;
sip->fs_maxfilesize = 0xffffffffffffffffull;
sip->fs_timedelta.tv_sec = 0;
sip->fs_timedelta.tv_nsec = 1;
sip->fs_properties = (NFSV3FSINFO_LINK |
NFSV3FSINFO_SYMLINK | NFSV3FSINFO_HOMOGENEOUS |
NFSV3FSINFO_CANSETTIME);
}
Move nfsvno_pathconf to be accessible to sys/fs/nfs; no functionality change. Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:05:17 +00:00
/*
* Do the pathconf vnode op.
*/
int
nfsvno_pathconf(struct vnode *vp, int flag, register_t *retf,
struct ucred *cred, struct thread *p)
{
int error;
error = VOP_PATHCONF(vp, flag, retf);
if (error == EOPNOTSUPP || error == EINVAL) {
/*
* Some file systems return EINVAL for name arguments not
* supported and some return EOPNOTSUPP for this case.
* So the NFSv3 Pathconf RPC doesn't fail for these cases,
* just fake them.
*/
switch (flag) {
case _PC_LINK_MAX:
*retf = LINK_MAX;
break;
case _PC_NAME_MAX:
*retf = NAME_MAX;
break;
case _PC_CHOWN_RESTRICTED:
*retf = 1;
break;
case _PC_NO_TRUNC:
*retf = 1;
break;
default:
/*
* Only happens if a _PC_xxx is added to the server,
* but this isn't updated.
*/
*retf = 0;
printf("nfsrvd pathconf flag=%d not supp\n", flag);
Cleanup unnecessary semicolons from the kernel. Found with devel/coccinelle.
2016-04-10 23:07:00 +00:00
}
Move nfsvno_pathconf to be accessible to sys/fs/nfs; no functionality change. Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:05:17 +00:00
error = 0;
}
Add DEXITCODE plumbing to NFS. Isilon has the concept of an in-memory exit-code ring that saves the last exit code of a function and allows for stack tracing. This is very helpful when debugging tough issues. This patch is essentially a no-op for BSD at this point, until we upstream the dexitcode logic itself. The patch adds DEXITCODE calls to every NFS function that returns an errno error code. A number of code paths were also reorganized to have single exit paths, to reduce code duplication. Submitted by: David Kwan <dkwan@isilon.com> Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:51:09 +00:00
NFSEXITCODE(error);
Move nfsvno_pathconf to be accessible to sys/fs/nfs; no functionality change. Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:05:17 +00:00
return (error);
}
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
/* Fake nfsrv_atroot. Just return 0 */
int
nfsrv_atroot(struct vnode *vp, long *retp)
{
return (0);
}
/*
* Set the credentials to refer to root.
* If only the various BSDen could agree on whether cr_gid is a separate
* field or cr_groups[0]...
*/
void
newnfs_setroot(struct ucred *cred)
{
cred->cr_uid = 0;
cred->cr_groups[0] = 0;
cred->cr_ngroups = 1;
}
/*
* Get the client credential. Used for Renew and recovery.
*/
struct ucred *
newnfs_getcred(void)
{
struct ucred *cred;
struct thread *td = curthread;
cred = crdup(td->td_ucred);
newnfs_setroot(cred);
return (cred);
}
/*
* Nfs timer routine
* Call the nfsd's timer function once/sec.
*/
void
newnfs_timer(void *arg)
{
static time_t lasttime = 0;
/*
* Call the server timer, if set up.
* The argument indicates if it is the next second and therefore
* leases should be checked.
*/
if (lasttime != NFSD_MONOSEC) {
lasttime = NFSD_MONOSEC;
if (nfsd_call_servertimer != NULL)
(*nfsd_call_servertimer)();
}
callout_reset(&newnfsd_callout, nfscl_ticks, newnfs_timer, NULL);
}
/*
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
* Sleep for a short period of time unless errval == NFSERR_GRACE, where
* the sleep should be for 5 seconds.
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
* Since lbolt doesn't exist in FreeBSD-CURRENT, just use a timeout on
* an event that never gets a wakeup. Only return EINTR or 0.
*/
int
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
nfs_catnap(int prio, int errval, const char *wmesg)
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
{
static int non_event;
int ret;
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
if (errval == NFSERR_GRACE)
ret = tsleep(&non_event, prio, wmesg, 5 * hz);
else
ret = tsleep(&non_event, prio, wmesg, 1);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
if (ret != EINTR)
ret = 0;
return (ret);
}
/*
* Get referral. For now, just fail.
*/
struct nfsreferral *
nfsv4root_getreferral(struct vnode *vp, struct vnode *dvp, u_int32_t fileno)
{
return (NULL);
}
static int
nfssvc_nfscommon(struct thread *td, struct nfssvc_args *uap)
{
int error;
error = nfssvc_call(td, uap, td->td_ucred);
Add DEXITCODE plumbing to NFS. Isilon has the concept of an in-memory exit-code ring that saves the last exit code of a function and allows for stack tracing. This is very helpful when debugging tough issues. This patch is essentially a no-op for BSD at this point, until we upstream the dexitcode logic itself. The patch adds DEXITCODE calls to every NFS function that returns an errno error code. A number of code paths were also reorganized to have single exit paths, to reduce code duplication. Submitted by: David Kwan <dkwan@isilon.com> Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:51:09 +00:00
NFSEXITCODE(error);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
return (error);
}
static int
nfssvc_call(struct thread *p, struct nfssvc_args *uap, struct ucred *cred)
{
Update the nfsstats structure to include the changes needed by the patch in D1626 plus changes so that it includes counts for NFSv4.1 (and the draft of NFSv4.2). Also, make all the counts uint64_t and add a vers field at the beginning, so that future revisions can easily be implemented. There is code in place to handle the old vesion of the nfsstats structure for backwards binary compatibility. Subsequent commits will update nfsstat(8) to use the new fields. Submitted by: will (earlier version) Reviewed by: ken MFC after: 1 month Relnotes: yes Differential Revision: https://reviews.freebsd.org/D1626
2016-08-12 22:44:59 +00:00
int error = EINVAL, i, j;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
struct nfsd_idargs nid;
Add kernel support to the NFS server for the "-manage-gids" option that will be added to the nfsuserd daemon in a future commit. It modifies the cache used by NFSv4 for name<-->id translation (both username/uid and group/gid) to support this. When "-manage-gids" is set, the server looks up each uid for the RPC and uses the list of groups cached in the server instead of the list of groups provided in the RPC request. The cached group list is acquired for the cache by the nfsuserd daemon via getgrouplist(3). This avoids the 16 groups limit for the list in the RPC request. Since the cache is now used for every RPC when "-manage-gids" is enabled, the code also modifies the cache to use a separate mutex for each hash list instead of a single global mutex. Suggested by: jpaetzel Tested by: jpaetzel MFC after: 2 weeks
2015-11-30 21:54:27 +00:00
struct nfsd_oidargs onid;
Update the nfsstats structure to include the changes needed by the patch in D1626 plus changes so that it includes counts for NFSv4.1 (and the draft of NFSv4.2). Also, make all the counts uint64_t and add a vers field at the beginning, so that future revisions can easily be implemented. There is code in place to handle the old vesion of the nfsstats structure for backwards binary compatibility. Subsequent commits will update nfsstat(8) to use the new fields. Submitted by: will (earlier version) Reviewed by: ken MFC after: 1 month Relnotes: yes Differential Revision: https://reviews.freebsd.org/D1626
2016-08-12 22:44:59 +00:00
struct {
int vers; /* Just the first field of nfsstats. */
} nfsstatver;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
if (uap->flag & NFSSVC_IDNAME) {
Add kernel support to the NFS server for the "-manage-gids" option that will be added to the nfsuserd daemon in a future commit. It modifies the cache used by NFSv4 for name<-->id translation (both username/uid and group/gid) to support this. When "-manage-gids" is set, the server looks up each uid for the RPC and uses the list of groups cached in the server instead of the list of groups provided in the RPC request. The cached group list is acquired for the cache by the nfsuserd daemon via getgrouplist(3). This avoids the 16 groups limit for the list in the RPC request. Since the cache is now used for every RPC when "-manage-gids" is enabled, the code also modifies the cache to use a separate mutex for each hash list instead of a single global mutex. Suggested by: jpaetzel Tested by: jpaetzel MFC after: 2 weeks
2015-11-30 21:54:27 +00:00
if ((uap->flag & NFSSVC_NEWSTRUCT) != 0)
error = copyin(uap->argp, &nid, sizeof(nid));
else {
error = copyin(uap->argp, &onid, sizeof(onid));
if (error == 0) {
nid.nid_flag = onid.nid_flag;
nid.nid_uid = onid.nid_uid;
nid.nid_gid = onid.nid_gid;
nid.nid_usermax = onid.nid_usermax;
nid.nid_usertimeout = onid.nid_usertimeout;
nid.nid_name = onid.nid_name;
nid.nid_namelen = onid.nid_namelen;
nid.nid_ngroup = 0;
nid.nid_grps = NULL;
}
}
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
if (error)
Add DEXITCODE plumbing to NFS. Isilon has the concept of an in-memory exit-code ring that saves the last exit code of a function and allows for stack tracing. This is very helpful when debugging tough issues. This patch is essentially a no-op for BSD at this point, until we upstream the dexitcode logic itself. The patch adds DEXITCODE calls to every NFS function that returns an errno error code. A number of code paths were also reorganized to have single exit paths, to reduce code duplication. Submitted by: David Kwan <dkwan@isilon.com> Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:51:09 +00:00
goto out;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
error = nfssvc_idname(&nid);
Add DEXITCODE plumbing to NFS. Isilon has the concept of an in-memory exit-code ring that saves the last exit code of a function and allows for stack tracing. This is very helpful when debugging tough issues. This patch is essentially a no-op for BSD at this point, until we upstream the dexitcode logic itself. The patch adds DEXITCODE calls to every NFS function that returns an errno error code. A number of code paths were also reorganized to have single exit paths, to reduce code duplication. Submitted by: David Kwan <dkwan@isilon.com> Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:51:09 +00:00
goto out;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
} else if (uap->flag & NFSSVC_GETSTATS) {
Update the nfsstats structure to include the changes needed by the patch in D1626 plus changes so that it includes counts for NFSv4.1 (and the draft of NFSv4.2). Also, make all the counts uint64_t and add a vers field at the beginning, so that future revisions can easily be implemented. There is code in place to handle the old vesion of the nfsstats structure for backwards binary compatibility. Subsequent commits will update nfsstat(8) to use the new fields. Submitted by: will (earlier version) Reviewed by: ken MFC after: 1 month Relnotes: yes Differential Revision: https://reviews.freebsd.org/D1626
2016-08-12 22:44:59 +00:00
if ((uap->flag & NFSSVC_NEWSTRUCT) == 0) {
/* Copy fields to the old ext_nfsstat structure. */
oldnfsstats.attrcache_hits =
nfsstatsv1.attrcache_hits;
oldnfsstats.attrcache_misses =
nfsstatsv1.attrcache_misses;
oldnfsstats.lookupcache_hits =
nfsstatsv1.lookupcache_hits;
oldnfsstats.lookupcache_misses =
nfsstatsv1.lookupcache_misses;
oldnfsstats.direofcache_hits =
nfsstatsv1.direofcache_hits;
oldnfsstats.direofcache_misses =
nfsstatsv1.direofcache_misses;
oldnfsstats.accesscache_hits =
nfsstatsv1.accesscache_hits;
oldnfsstats.accesscache_misses =
nfsstatsv1.accesscache_misses;
oldnfsstats.biocache_reads =
nfsstatsv1.biocache_reads;
oldnfsstats.read_bios =
nfsstatsv1.read_bios;
oldnfsstats.read_physios =
nfsstatsv1.read_physios;
oldnfsstats.biocache_writes =
nfsstatsv1.biocache_writes;
oldnfsstats.write_bios =
nfsstatsv1.write_bios;
oldnfsstats.write_physios =
nfsstatsv1.write_physios;
oldnfsstats.biocache_readlinks =
nfsstatsv1.biocache_readlinks;
oldnfsstats.readlink_bios =
nfsstatsv1.readlink_bios;
oldnfsstats.biocache_readdirs =
nfsstatsv1.biocache_readdirs;
oldnfsstats.readdir_bios =
nfsstatsv1.readdir_bios;
for (i = 0; i < NFSV4_NPROCS; i++)
oldnfsstats.rpccnt[i] = nfsstatsv1.rpccnt[i];
oldnfsstats.rpcretries = nfsstatsv1.rpcretries;
for (i = 0; i < NFSV4OP_NOPS; i++)
oldnfsstats.srvrpccnt[i] =
nfsstatsv1.srvrpccnt[i];
for (i = NFSV42_NOPS, j = NFSV4OP_NOPS;
i < NFSV42_NOPS + NFSV4OP_FAKENOPS; i++, j++)
oldnfsstats.srvrpccnt[j] =
nfsstatsv1.srvrpccnt[i];
oldnfsstats.srvrpc_errs = nfsstatsv1.srvrpc_errs;
oldnfsstats.srv_errs = nfsstatsv1.srv_errs;
oldnfsstats.rpcrequests = nfsstatsv1.rpcrequests;
oldnfsstats.rpctimeouts = nfsstatsv1.rpctimeouts;
oldnfsstats.rpcunexpected = nfsstatsv1.rpcunexpected;
oldnfsstats.rpcinvalid = nfsstatsv1.rpcinvalid;
oldnfsstats.srvcache_inproghits =
nfsstatsv1.srvcache_inproghits;
oldnfsstats.srvcache_idemdonehits =
nfsstatsv1.srvcache_idemdonehits;
oldnfsstats.srvcache_nonidemdonehits =
nfsstatsv1.srvcache_nonidemdonehits;
oldnfsstats.srvcache_misses =
nfsstatsv1.srvcache_misses;
oldnfsstats.srvcache_tcppeak =
nfsstatsv1.srvcache_tcppeak;
oldnfsstats.srvcache_size = nfsstatsv1.srvcache_size;
oldnfsstats.srvclients = nfsstatsv1.srvclients;
oldnfsstats.srvopenowners = nfsstatsv1.srvopenowners;
oldnfsstats.srvopens = nfsstatsv1.srvopens;
oldnfsstats.srvlockowners = nfsstatsv1.srvlockowners;
oldnfsstats.srvlocks = nfsstatsv1.srvlocks;
oldnfsstats.srvdelegates = nfsstatsv1.srvdelegates;
for (i = 0; i < NFSV4OP_CBNOPS; i++)
oldnfsstats.cbrpccnt[i] =
nfsstatsv1.cbrpccnt[i];
oldnfsstats.clopenowners = nfsstatsv1.clopenowners;
oldnfsstats.clopens = nfsstatsv1.clopens;
oldnfsstats.cllockowners = nfsstatsv1.cllockowners;
oldnfsstats.cllocks = nfsstatsv1.cllocks;
oldnfsstats.cldelegates = nfsstatsv1.cldelegates;
oldnfsstats.cllocalopenowners =
nfsstatsv1.cllocalopenowners;
oldnfsstats.cllocalopens = nfsstatsv1.cllocalopens;
oldnfsstats.cllocallockowners =
nfsstatsv1.cllocallockowners;
oldnfsstats.cllocallocks = nfsstatsv1.cllocallocks;
error = copyout(&oldnfsstats, uap->argp,
sizeof (oldnfsstats));
} else {
error = copyin(uap->argp, &nfsstatver,
sizeof(nfsstatver));
if (error == 0 && nfsstatver.vers != NFSSTATS_V1)
error = EPERM;
if (error == 0)
error = copyout(&nfsstatsv1, uap->argp,
sizeof (nfsstatsv1));
}
Add kernel support for NFSSVC_ZEROCLTSTATS and NFSSVC_ZEROSRVSTATS so that they can be used by nfsstat(1) to implement the "-z" option for the new NFS subsystem. MFC after: 2 weeks
2011-05-04 13:36:18 +00:00
if (error == 0) {
if ((uap->flag & NFSSVC_ZEROCLTSTATS) != 0) {
Update the nfsstats structure to include the changes needed by the patch in D1626 plus changes so that it includes counts for NFSv4.1 (and the draft of NFSv4.2). Also, make all the counts uint64_t and add a vers field at the beginning, so that future revisions can easily be implemented. There is code in place to handle the old vesion of the nfsstats structure for backwards binary compatibility. Subsequent commits will update nfsstat(8) to use the new fields. Submitted by: will (earlier version) Reviewed by: ken MFC after: 1 month Relnotes: yes Differential Revision: https://reviews.freebsd.org/D1626
2016-08-12 22:44:59 +00:00
nfsstatsv1.attrcache_hits = 0;
nfsstatsv1.attrcache_misses = 0;
nfsstatsv1.lookupcache_hits = 0;
nfsstatsv1.lookupcache_misses = 0;
nfsstatsv1.direofcache_hits = 0;
nfsstatsv1.direofcache_misses = 0;
nfsstatsv1.accesscache_hits = 0;
nfsstatsv1.accesscache_misses = 0;
nfsstatsv1.biocache_reads = 0;
nfsstatsv1.read_bios = 0;
nfsstatsv1.read_physios = 0;
nfsstatsv1.biocache_writes = 0;
nfsstatsv1.write_bios = 0;
nfsstatsv1.write_physios = 0;
nfsstatsv1.biocache_readlinks = 0;
nfsstatsv1.readlink_bios = 0;
nfsstatsv1.biocache_readdirs = 0;
nfsstatsv1.readdir_bios = 0;
nfsstatsv1.rpcretries = 0;
nfsstatsv1.rpcrequests = 0;
nfsstatsv1.rpctimeouts = 0;
nfsstatsv1.rpcunexpected = 0;
nfsstatsv1.rpcinvalid = 0;
bzero(nfsstatsv1.rpccnt,
sizeof(nfsstatsv1.rpccnt));
Add kernel support for NFSSVC_ZEROCLTSTATS and NFSSVC_ZEROSRVSTATS so that they can be used by nfsstat(1) to implement the "-z" option for the new NFS subsystem. MFC after: 2 weeks
2011-05-04 13:36:18 +00:00
}
if ((uap->flag & NFSSVC_ZEROSRVSTATS) != 0) {
Update the nfsstats structure to include the changes needed by the patch in D1626 plus changes so that it includes counts for NFSv4.1 (and the draft of NFSv4.2). Also, make all the counts uint64_t and add a vers field at the beginning, so that future revisions can easily be implemented. There is code in place to handle the old vesion of the nfsstats structure for backwards binary compatibility. Subsequent commits will update nfsstat(8) to use the new fields. Submitted by: will (earlier version) Reviewed by: ken MFC after: 1 month Relnotes: yes Differential Revision: https://reviews.freebsd.org/D1626
2016-08-12 22:44:59 +00:00
nfsstatsv1.srvrpc_errs = 0;
nfsstatsv1.srv_errs = 0;
nfsstatsv1.srvcache_inproghits = 0;
nfsstatsv1.srvcache_idemdonehits = 0;
nfsstatsv1.srvcache_nonidemdonehits = 0;
nfsstatsv1.srvcache_misses = 0;
nfsstatsv1.srvcache_tcppeak = 0;
bzero(nfsstatsv1.srvrpccnt,
sizeof(nfsstatsv1.srvrpccnt));
bzero(nfsstatsv1.cbrpccnt,
sizeof(nfsstatsv1.cbrpccnt));
Add kernel support for NFSSVC_ZEROCLTSTATS and NFSSVC_ZEROSRVSTATS so that they can be used by nfsstat(1) to implement the "-z" option for the new NFS subsystem. MFC after: 2 weeks
2011-05-04 13:36:18 +00:00
}
}
Add DEXITCODE plumbing to NFS. Isilon has the concept of an in-memory exit-code ring that saves the last exit code of a function and allows for stack tracing. This is very helpful when debugging tough issues. This patch is essentially a no-op for BSD at this point, until we upstream the dexitcode logic itself. The patch adds DEXITCODE calls to every NFS function that returns an errno error code. A number of code paths were also reorganized to have single exit paths, to reduce code duplication. Submitted by: David Kwan <dkwan@isilon.com> Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:51:09 +00:00
goto out;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
} else if (uap->flag & NFSSVC_NFSUSERDPORT) {
u_short sockport;
error = copyin(uap->argp, (caddr_t)&sockport,
sizeof (u_short));
if (!error)
error = nfsrv_nfsuserdport(sockport, p);
} else if (uap->flag & NFSSVC_NFSUSERDDELPORT) {
nfsrv_nfsuserddelport();
error = 0;
}
Add DEXITCODE plumbing to NFS. Isilon has the concept of an in-memory exit-code ring that saves the last exit code of a function and allows for stack tracing. This is very helpful when debugging tough issues. This patch is essentially a no-op for BSD at this point, until we upstream the dexitcode logic itself. The patch adds DEXITCODE calls to every NFS function that returns an errno error code. A number of code paths were also reorganized to have single exit paths, to reduce code duplication. Submitted by: David Kwan <dkwan@isilon.com> Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:51:09 +00:00
out:
NFSEXITCODE(error);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
return (error);
}
/*
* called by all three modevent routines, so that it gets things
* initialized soon enough.
*/
void
newnfs_portinit(void)
{
static int inited = 0;
if (inited)
return;
inited = 1;
/* Initialize SMP locks used by both client and server. */
mtx_init(&newnfsd_mtx, "newnfsd_mtx", NULL, MTX_DEF);
mtx_init(&nfs_state_mutex, "nfs_state_mutex", NULL, MTX_DEF);
Fix NFSv4.1 client recovery from NFS4ERR_BAD_SESSION errors. For most NFSv4.1 servers, a NFS4ERR_BAD_SESSION error is a rare failure that indicates that the server has lost session/open/lock state. However, recent testing by cperciva@ against the AmazonEFS server found several problems with client recovery from this due to it generating this failure frequently. Briefly, the problems fixed are: - If all session slots were in use at the time of the failure, some processes would continue to loop waiting for a slot on the old session forever. - If an RPC that doesn't use open/lock state failed with NFS4ERR_BAD_SESSION, it would fail the RPC/syscall instead of initiating recovery and then looping to retry the RPC. - If a successful reply to an RPC for an old session wasn't processed until after a new session was created for a NFS4ERR_BAD_SESSION error, it would erroneously update the new session and corrupt it. - The use of the first element of the session list in the nfs mount structure (which is always the current metadata session) was slightly racey. With changes for the above problems it became more racey, so all uses of this head pointer was wrapped with a NFSLOCKMNT()/NFSUNLOCKMNT(). - Although the kernel malloc() usually allocates more bytes than requested and, as such, this wouldn't have caused problems, the allocation of a session structure was 1 byte smaller than it should have been. (Null termination byte for the string not included in byte count.) There are probably still problems with a pNFS data server that fails with NFS4ERR_BAD_SESSION, but I have no server that does this to test against (the AmazonEFS server doesn't do pNFS), so I can't fix these yet. Although this patch is fairly large, it should only affect the handling of NFS4ERR_BAD_SESSION error replies from an NFSv4.1 server. Thanks go to cperciva@ for the extension testing he did to help isolate/fix these problems. Reported by: cperciva Tested by: cperciva MFC after: 3 months Differential Revision: https://reviews.freebsd.org/D8745
2016-12-23 23:14:53 +00:00
mtx_init(&nfs_clstate_mutex, "nfs_clstate_mutex", NULL, MTX_DEF);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
}
Add a function to the experimental nfs subsystem that tests to see if a local file system supports NFSv4 ACLs. This allows the NFSHASNFS4ACL() macro to be correctly implemented. The NFSv4 ACL support should now work when the server exports a ZFS volume. Approved by: kib (mentor)
2009-05-27 15:16:56 +00:00
/*
* Determine if the file system supports NFSv4 ACLs.
* Return 1 if it does, 0 otherwise.
*/
int
Fix the experimental NFSv4 server so that it uses VOP_PATHCONF() to determine if a file system supports NFSv4 ACLs. Since VOP_PATHCONF() must be called with a locked vnode, the function is called before nfsvno_fillattr() and the result is passed in as an extra argument. MFC after: 2 weeks
2011-04-14 23:46:15 +00:00
nfs_supportsnfsv4acls(struct vnode *vp)
Add a function to the experimental nfs subsystem that tests to see if a local file system supports NFSv4 ACLs. This allows the NFSHASNFS4ACL() macro to be correctly implemented. The NFSv4 ACL support should now work when the server exports a ZFS volume. Approved by: kib (mentor)
2009-05-27 15:16:56 +00:00
{
Fix the experimental NFSv4 server so that it uses VOP_PATHCONF() to determine if a file system supports NFSv4 ACLs. Since VOP_PATHCONF() must be called with a locked vnode, the function is called before nfsvno_fillattr() and the result is passed in as an extra argument. MFC after: 2 weeks
2011-04-14 23:46:15 +00:00
int error;
register_t retval;
Add a function to the experimental nfs subsystem that tests to see if a local file system supports NFSv4 ACLs. This allows the NFSHASNFS4ACL() macro to be correctly implemented. The NFSv4 ACL support should now work when the server exports a ZFS volume. Approved by: kib (mentor)
2009-05-27 15:16:56 +00:00
Fix the experimental NFSv4 server so that it uses VOP_PATHCONF() to determine if a file system supports NFSv4 ACLs. Since VOP_PATHCONF() must be called with a locked vnode, the function is called before nfsvno_fillattr() and the result is passed in as an extra argument. MFC after: 2 weeks
2011-04-14 23:46:15 +00:00
ASSERT_VOP_LOCKED(vp, "nfs supports nfsv4acls");
if (nfsrv_useacl == 0)
Add a function to the experimental nfs subsystem that tests to see if a local file system supports NFSv4 ACLs. This allows the NFSHASNFS4ACL() macro to be correctly implemented. The NFSv4 ACL support should now work when the server exports a ZFS volume. Approved by: kib (mentor)
2009-05-27 15:16:56 +00:00
return (0);
Revert revision 224079 as Rick pointed out that I would be calling VOP_PATHCONF without the vnode lock held. Implicitly approved by: zml (mentor)
2011-07-17 03:44:05 +00:00
error = VOP_PATHCONF(vp, _PC_ACL_NFS4, &retval);
Fix the experimental NFSv4 server so that it uses VOP_PATHCONF() to determine if a file system supports NFSv4 ACLs. Since VOP_PATHCONF() must be called with a locked vnode, the function is called before nfsvno_fillattr() and the result is passed in as an extra argument. MFC after: 2 weeks
2011-04-14 23:46:15 +00:00
if (error == 0 && retval != 0)
Add a function to the experimental nfs subsystem that tests to see if a local file system supports NFSv4 ACLs. This allows the NFSHASNFS4ACL() macro to be correctly implemented. The NFSv4 ACL support should now work when the server exports a ZFS volume. Approved by: kib (mentor)
2009-05-27 15:16:56 +00:00
return (1);
return (0);
}
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
extern int (*nfsd_call_nfscommon)(struct thread *, struct nfssvc_args *);
/*
* Called once to initialize data structures...
*/
static int
nfscommon_modevent(module_t mod, int type, void *data)
{
int error = 0;
static int loaded = 0;
switch (type) {
case MOD_LOAD:
if (loaded)
Add DEXITCODE plumbing to NFS. Isilon has the concept of an in-memory exit-code ring that saves the last exit code of a function and allows for stack tracing. This is very helpful when debugging tough issues. This patch is essentially a no-op for BSD at this point, until we upstream the dexitcode logic itself. The patch adds DEXITCODE calls to every NFS function that returns an errno error code. A number of code paths were also reorganized to have single exit paths, to reduce code duplication. Submitted by: David Kwan <dkwan@isilon.com> Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:51:09 +00:00
goto out;
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
newnfs_portinit();
mtx_init(&nfs_nameid_mutex, "nfs_nameid_mutex", NULL, MTX_DEF);
mtx_init(&nfs_sockl_mutex, "nfs_sockl_mutex", NULL, MTX_DEF);
mtx_init(&nfs_slock_mutex, "nfs_slock_mutex", NULL, MTX_DEF);
mtx_init(&nfs_req_mutex, "nfs_req_mutex", NULL, MTX_DEF);
mtx_init(&nfsrv_nfsuserdsock.nr_mtx, "nfsuserd", NULL,
MTX_DEF);
CALLOUT_MPSAFE has lost its meaning since r141428, i.e., for more than ten years for head. However, it is continuously misused as the mpsafe argument for callout_init(9). Deprecate the flag and clean up callout_init() calls to make them more consistent. Differential Revision: https://reviews.freebsd.org/D2613 Reviewed by: jhb MFC after: 2 weeks
2015-05-22 17:05:21 +00:00
callout_init(&newnfsd_callout, 1);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
newnfs_init();
nfsd_call_nfscommon = nfssvc_nfscommon;
loaded = 1;
break;
case MOD_UNLOAD:
if (newnfs_numnfsd != 0 || nfsrv_nfsuserd != 0 ||
nfs_numnfscbd != 0) {
error = EBUSY;
break;
}
nfsd_call_nfscommon = NULL;
callout_drain(&newnfsd_callout);
Fix the memory leak that occurs when the nfscommon.ko module is unloaded. This leak was introduced by r291527. Since the nfscommon.ko module is rarely unloaded, this leak would not have been much of an issue. MFC after: 2 weeks
2015-12-02 02:47:13 +00:00
/* Clean out the name<-->id cache. */
nfsrv_cleanusergroup();
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
/* and get rid of the mutexes */
mtx_destroy(&nfs_nameid_mutex);
mtx_destroy(&newnfsd_mtx);
mtx_destroy(&nfs_state_mutex);
Fix NFSv4.1 client recovery from NFS4ERR_BAD_SESSION errors. For most NFSv4.1 servers, a NFS4ERR_BAD_SESSION error is a rare failure that indicates that the server has lost session/open/lock state. However, recent testing by cperciva@ against the AmazonEFS server found several problems with client recovery from this due to it generating this failure frequently. Briefly, the problems fixed are: - If all session slots were in use at the time of the failure, some processes would continue to loop waiting for a slot on the old session forever. - If an RPC that doesn't use open/lock state failed with NFS4ERR_BAD_SESSION, it would fail the RPC/syscall instead of initiating recovery and then looping to retry the RPC. - If a successful reply to an RPC for an old session wasn't processed until after a new session was created for a NFS4ERR_BAD_SESSION error, it would erroneously update the new session and corrupt it. - The use of the first element of the session list in the nfs mount structure (which is always the current metadata session) was slightly racey. With changes for the above problems it became more racey, so all uses of this head pointer was wrapped with a NFSLOCKMNT()/NFSUNLOCKMNT(). - Although the kernel malloc() usually allocates more bytes than requested and, as such, this wouldn't have caused problems, the allocation of a session structure was 1 byte smaller than it should have been. (Null termination byte for the string not included in byte count.) There are probably still problems with a pNFS data server that fails with NFS4ERR_BAD_SESSION, but I have no server that does this to test against (the AmazonEFS server doesn't do pNFS), so I can't fix these yet. Although this patch is fairly large, it should only affect the handling of NFS4ERR_BAD_SESSION error replies from an NFSv4.1 server. Thanks go to cperciva@ for the extension testing he did to help isolate/fix these problems. Reported by: cperciva Tested by: cperciva MFC after: 3 months Differential Revision: https://reviews.freebsd.org/D8745
2016-12-23 23:14:53 +00:00
mtx_destroy(&nfs_clstate_mutex);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
mtx_destroy(&nfs_sockl_mutex);
mtx_destroy(&nfs_slock_mutex);
mtx_destroy(&nfs_req_mutex);
mtx_destroy(&nfsrv_nfsuserdsock.nr_mtx);
loaded = 0;
break;
default:
error = EOPNOTSUPP;
break;
}
Add DEXITCODE plumbing to NFS. Isilon has the concept of an in-memory exit-code ring that saves the last exit code of a function and allows for stack tracing. This is very helpful when debugging tough issues. This patch is essentially a no-op for BSD at this point, until we upstream the dexitcode logic itself. The patch adds DEXITCODE calls to every NFS function that returns an errno error code. A number of code paths were also reorganized to have single exit paths, to reduce code duplication. Submitted by: David Kwan <dkwan@isilon.com> Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:51:09 +00:00
out:
NFSEXITCODE(error);
Add the experimental nfs subtree to the kernel, that includes support for NFSv4 as well as NFSv2 and 3. It lives in 3 subdirs under sys/fs: nfs - functions that are common to the client and server nfsclient - a mutation of sys/nfsclient that call generic functions to do RPCs and handle state. As such, it retains the buffer cache handling characteristics and vnode semantics that are found in sys/nfsclient, for the most part. nfsserver - the server. It includes a DRC designed specifically for NFSv4, that is used instead of the generic DRC in sys/rpc. The build glue will be checked in later, so at this point, it consists of 3 new subdirs that should not affect kernel building. Approved by: kib (mentor)
2009-05-04 15:23:58 +00:00
return error;
}
static moduledata_t nfscommon_mod = {
"nfscommon",
nfscommon_modevent,
NULL,
};
DECLARE_MODULE(nfscommon, nfscommon_mod, SI_SUB_VFS, SI_ORDER_ANY);
/* So that loader and kldload(2) can find us, wherever we are.. */
MODULE_VERSION(nfscommon, 1);
MODULE_DEPEND(nfscommon, nfssvc, 1, 1, 1);
MODULE_DEPEND(nfscommon, krpc, 1, 1, 1);
Reference in New Issue Copy Permalink