d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
cbc4d2db75
freebsd-nq/sys/fs/nfsclient/nfs_clsubs.c

403 lines
10 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.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* from nfs_subs.c 8.8 (Berkeley) 5/22/95
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
/*
* These functions support the macros and help fiddle mbuf chains for
* the nfs op functions. They do things like create the rpc header and
* copy data between mbuf chains and uio lists.
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/bio.h>
#include <sys/buf.h>
#include <sys/proc.h>
#include <sys/mount.h>
#include <sys/vnode.h>
#include <sys/namei.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/malloc.h>
#include <sys/sysent.h>
#include <sys/syscall.h>
#include <sys/sysproto.h>
Change the experimental NFS client so that it creates nfsiod threads in the same manner as the regular NFS client after r214026 was committed. This resolves the lors fixed by r214026 and its predecessors for the regular client. Reviewed by: jhb MFC after: 2 weeks
2011-04-15 23:07:48 +00:00
#include <sys/taskqueue.h>
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
#include <vm/vm.h>
#include <vm/vm_object.h>
#include <vm/vm_extern.h>
#include <vm/uma.h>
#include <fs/nfs/nfsport.h>
#include <fs/nfsclient/nfsnode.h>
#include <fs/nfsclient/nfsmount.h>
#include <fs/nfsclient/nfs.h>
Add DTrace support to the new NFS client. This is essentially cloned from the old NFS client, plus additions for NFSv4. A review of this code is in progress, however it was felt by the reviewer that it could go in now, before code slush. Any changes required by the review can be committed as bug fixes later.
2011-06-18 23:02:53 +00:00
#include <fs/nfsclient/nfs_kdtrace.h>
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
#include <netinet/in.h>
/*
* Note that stdarg.h and the ANSI style va_start macro is used for both
* ANSI and traditional C compilers.
*/
#include <machine/stdarg.h>
extern struct mtx ncl_iod_mutex;
Change the experimental NFS client so that it creates nfsiod threads in the same manner as the regular NFS client after r214026 was committed. This resolves the lors fixed by r214026 and its predecessors for the regular client. Reviewed by: jhb MFC after: 2 weeks
2011-04-15 23:07:48 +00:00
extern enum nfsiod_state ncl_iodwant[NFS_MAXASYNCDAEMON];
extern struct nfsmount *ncl_iodmount[NFS_MAXASYNCDAEMON];
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 ncl_numasync;
extern unsigned int ncl_iodmax;
extern struct nfsstats newnfsstats;
Change the experimental NFS client so that it creates nfsiod threads in the same manner as the regular NFS client after r214026 was committed. This resolves the lors fixed by r214026 and its predecessors for the regular client. Reviewed by: jhb MFC after: 2 weeks
2011-04-15 23:07:48 +00:00
struct task ncl_nfsiodnew_task;
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
ncl_uninit(struct vfsconf *vfsp)
{
Unloading of the nfscl module is unsupported because newnfslock doesn't support unloading. It's not trivial to implement newnfslock unloading so for now just admit that unloading is unsupported and refuse to attempt unload in all nfscl module event handlers. Reviewed by: rmacklem Approved by: trasz (mentor)
2009-10-20 15:06:18 +00:00
/*
* XXX: Unloading of nfscl module is unsupported.
*/
#if 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
int i;
/*
* Tell all nfsiod processes to exit. Clear ncl_iodmax, and wakeup
* any sleeping nfsiods so they check ncl_iodmax and exit.
*/
mtx_lock(&ncl_iod_mutex);
ncl_iodmax = 0;
for (i = 0; i < ncl_numasync; i++)
Patch the experimental NFS client in a manner analogous to r203072 for the regular NFS client. Also, delete two fields of struct nfsmount that are not used by the FreeBSD port of the client. MFC after: 2 weeks
2010-01-28 16:17:24 +00:00
if (ncl_iodwant[i] == NFSIOD_AVAILABLE)
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
wakeup(&ncl_iodwant[i]);
/* The last nfsiod to exit will wake us up when ncl_numasync hits 0 */
while (ncl_numasync)
msleep(&ncl_numasync, &ncl_iod_mutex, PWAIT, "ioddie", 0);
mtx_unlock(&ncl_iod_mutex);
ncl_nhuninit();
return (0);
Unloading of the nfscl module is unsupported because newnfslock doesn't support unloading. It's not trivial to implement newnfslock unloading so for now just admit that unloading is unsupported and refuse to attempt unload in all nfscl module event handlers. Reviewed by: rmacklem Approved by: trasz (mentor)
2009-10-20 15:06:18 +00:00
#else
return (EOPNOTSUPP);
#endif
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
ncl_dircookie_lock(struct nfsnode *np)
{
mtx_lock(&np->n_mtx);
while (np->n_flag & NDIRCOOKIELK)
(void) msleep(&np->n_flag, &np->n_mtx, PZERO, "nfsdirlk", 0);
np->n_flag |= NDIRCOOKIELK;
mtx_unlock(&np->n_mtx);
}
void
ncl_dircookie_unlock(struct nfsnode *np)
{
mtx_lock(&np->n_mtx);
np->n_flag &= ~NDIRCOOKIELK;
wakeup(&np->n_flag);
mtx_unlock(&np->n_mtx);
}
int
ncl_upgrade_vnlock(struct vnode *vp)
{
int old_lock;
Apply the same patch as r196205 for nfs_upgrade_lock() and nfs_downgrade_lock() to the experimental nfs client. Approved by: re (kensmith), kib (mentor)
2009-08-17 16:12:28 +00:00
ASSERT_VOP_LOCKED(vp, "ncl_upgrade_vnlock");
Simple find/replace of VOP_ISLOCKED -> NFSVOPISLOCKED. This is done so that NFSVOPISLOCKED can be modified later to add enhanced logging and assertions. Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:05:41 +00:00
old_lock = NFSVOPISLOCKED(vp);
Apply the same patch as r196205 for nfs_upgrade_lock() and nfs_downgrade_lock() to the experimental nfs client. Approved by: re (kensmith), kib (mentor)
2009-08-17 16:12:28 +00:00
if (old_lock != LK_EXCLUSIVE) {
KASSERT(old_lock == LK_SHARED,
("ncl_upgrade_vnlock: wrong old_lock %d", old_lock));
/* Upgrade to exclusive lock, this might block */
Simple find/replace of vn_lock -> NFSVOPLOCK. This is done so that NFSVOPLOCK can be modified later to add enhanced logging and assertions. Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:05:31 +00:00
NFSVOPLOCK(vp, LK_UPGRADE | LK_RETRY);
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
}
Apply the same patch as r196205 for nfs_upgrade_lock() and nfs_downgrade_lock() to the experimental nfs client. Approved by: re (kensmith), kib (mentor)
2009-08-17 16:12:28 +00:00
return (old_lock);
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
ncl_downgrade_vnlock(struct vnode *vp, int old_lock)
{
if (old_lock != LK_EXCLUSIVE) {
Apply the same patch as r196205 for nfs_upgrade_lock() and nfs_downgrade_lock() to the experimental nfs client. Approved by: re (kensmith), kib (mentor)
2009-08-17 16:12:28 +00:00
KASSERT(old_lock == LK_SHARED, ("wrong old_lock %d", old_lock));
/* Downgrade from exclusive lock. */
Simple find/replace of vn_lock -> NFSVOPLOCK. This is done so that NFSVOPLOCK can be modified later to add enhanced logging and assertions. Reviewed by: rmacklem Approved by: zml (mentor) MFC after: 2 weeks
2011-07-16 08:05:31 +00:00
NFSVOPLOCK(vp, LK_DOWNGRADE | LK_RETRY);
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
ncl_printf(const char *fmt, ...)
{
va_list ap;
mtx_lock(&Giant);
va_start(ap, fmt);
In ncl_printf(), call vprintf() instead of printf(). MFC after: 3 days
2011-05-04 11:22:52 +00:00
vprintf(fmt, ap);
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
va_end(ap);
mtx_unlock(&Giant);
}
#ifdef NFS_ACDEBUG
#include <sys/sysctl.h>
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_DECL(_vfs_nfs);
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 nfs_acdebug;
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, acdebug, CTLFLAG_RW, &nfs_acdebug, 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
#endif
/*
* Check the time stamp
* If the cache is valid, copy contents to *vap and return 0
* otherwise return an error
*/
int
ncl_getattrcache(struct vnode *vp, struct vattr *vaper)
{
struct nfsnode *np;
struct vattr *vap;
struct nfsmount *nmp;
Modify the return value for nfscl_mustflush() from boolean_t, which I mistakenly thought was correct w.r.t. style(9), back to int and add the checks for != 0. This is just a stylistic modification. MFC after: 1 week
2010-08-03 01:49:28 +00:00
int timeo, mustflush;
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
np = VTONFS(vp);
vap = &np->n_vattr.na_vattr;
nmp = VFSTONFS(vp->v_mount);
Allow the experimental NFSv4 client to use cached attributes when a write delegation is held. Also, add a missing mtx_unlock() call for the ACL debugging code. MFC after: 5 days
2010-05-18 05:18:21 +00:00
mustflush = nfscl_mustflush(vp); /* must be before mtx_lock() */
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 NFS_ACDEBUG
mtx_lock(&Giant); /* ncl_printf() */
#endif
mtx_lock(&np->n_mtx);
/* XXX n_mtime doesn't seem to be updated on a miss-and-reload */
timeo = (time_second - np->n_mtime.tv_sec) / 10;
#ifdef NFS_ACDEBUG
if (nfs_acdebug>1)
ncl_printf("nfs_getattrcache: initial timeo = %d\n", timeo);
#endif
if (vap->va_type == VDIR) {
if ((np->n_flag & NMODIFIED) || timeo < nmp->nm_acdirmin)
timeo = nmp->nm_acdirmin;
else if (timeo > nmp->nm_acdirmax)
timeo = nmp->nm_acdirmax;
} else {
if ((np->n_flag & NMODIFIED) || timeo < nmp->nm_acregmin)
timeo = nmp->nm_acregmin;
else if (timeo > nmp->nm_acregmax)
timeo = nmp->nm_acregmax;
}
#ifdef NFS_ACDEBUG
if (nfs_acdebug > 2)
ncl_printf("acregmin %d; acregmax %d; acdirmin %d; acdirmax %d\n",
nmp->nm_acregmin, nmp->nm_acregmax,
nmp->nm_acdirmin, nmp->nm_acdirmax);
if (nfs_acdebug)
ncl_printf("nfs_getattrcache: age = %d; final timeo = %d\n",
(time_second - np->n_attrstamp), timeo);
#endif
Allow the experimental NFSv4 client to use cached attributes when a write delegation is held. Also, add a missing mtx_unlock() call for the ACL debugging code. MFC after: 5 days
2010-05-18 05:18:21 +00:00
if ((time_second - np->n_attrstamp) >= timeo &&
Modify the return value for nfscl_mustflush() from boolean_t, which I mistakenly thought was correct w.r.t. style(9), back to int and add the checks for != 0. This is just a stylistic modification. MFC after: 1 week
2010-08-03 01:49:28 +00:00
(mustflush != 0 || np->n_attrstamp == 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
newnfsstats.attrcache_misses++;
mtx_unlock(&np->n_mtx);
Allow the experimental NFSv4 client to use cached attributes when a write delegation is held. Also, add a missing mtx_unlock() call for the ACL debugging code. MFC after: 5 days
2010-05-18 05:18:21 +00:00
#ifdef NFS_ACDEBUG
mtx_unlock(&Giant); /* ncl_printf() */
#endif
Add DTrace support to the new NFS client. This is essentially cloned from the old NFS client, plus additions for NFSv4. A review of this code is in progress, however it was felt by the reviewer that it could go in now, before code slush. Any changes required by the review can be committed as bug fixes later.
2011-06-18 23:02:53 +00:00
KDTRACE_NFS_ATTRCACHE_GET_MISS(vp);
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( ENOENT);
}
newnfsstats.attrcache_hits++;
if (vap->va_size != np->n_size) {
if (vap->va_type == VREG) {
if (np->n_flag & NMODIFIED) {
if (vap->va_size < np->n_size)
vap->va_size = np->n_size;
else
np->n_size = vap->va_size;
} else {
np->n_size = vap->va_size;
}
vnode_pager_setsize(vp, np->n_size);
} else {
np->n_size = vap->va_size;
}
}
bcopy((caddr_t)vap, (caddr_t)vaper, sizeof(struct vattr));
if (np->n_flag & NCHG) {
if (np->n_flag & NACC)
vaper->va_atime = np->n_atim;
if (np->n_flag & NUPD)
vaper->va_mtime = np->n_mtim;
}
mtx_unlock(&np->n_mtx);
#ifdef NFS_ACDEBUG
mtx_unlock(&Giant); /* ncl_printf() */
#endif
Add DTrace support to the new NFS client. This is essentially cloned from the old NFS client, plus additions for NFSv4. A review of this code is in progress, however it was felt by the reviewer that it could go in now, before code slush. Any changes required by the review can be committed as bug fixes later.
2011-06-18 23:02:53 +00:00
KDTRACE_NFS_ATTRCACHE_GET_HIT(vp, vap);
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 (0);
}
static nfsuint64 nfs_nullcookie = { { 0, 0 } };
/*
* This function finds the directory cookie that corresponds to the
* logical byte offset given.
*/
nfsuint64 *
ncl_getcookie(struct nfsnode *np, off_t off, int add)
{
struct nfsdmap *dp, *dp2;
int pos;
nfsuint64 *retval = NULL;
pos = (uoff_t)off / NFS_DIRBLKSIZ;
if (pos == 0 || off < 0) {
In NFS clients, instead of inconsistently using #ifdef DIAGNOSTIC and #ifndef DIAGNOSTIC for debug assertions, prefer KASSERT(). Also change one #ifdef DIAGNOSTIC in the new nfs server. Submitted by: Mikolaj Golub <to.my.trociny gmail com> MFC after: 2 weeks
2010-06-13 05:24:27 +00:00
KASSERT(!add, ("nfs getcookie add at <= 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
return (&nfs_nullcookie);
}
pos--;
dp = LIST_FIRST(&np->n_cookies);
if (!dp) {
if (add) {
MALLOC(dp, struct nfsdmap *, sizeof (struct nfsdmap),
M_NFSDIROFF, M_WAITOK);
dp->ndm_eocookie = 0;
LIST_INSERT_HEAD(&np->n_cookies, dp, ndm_list);
} else
goto out;
}
while (pos >= NFSNUMCOOKIES) {
pos -= NFSNUMCOOKIES;
if (LIST_NEXT(dp, ndm_list)) {
if (!add && dp->ndm_eocookie < NFSNUMCOOKIES &&
pos >= dp->ndm_eocookie)
goto out;
dp = LIST_NEXT(dp, ndm_list);
} else if (add) {
MALLOC(dp2, struct nfsdmap *, sizeof (struct nfsdmap),
M_NFSDIROFF, M_WAITOK);
dp2->ndm_eocookie = 0;
LIST_INSERT_AFTER(dp, dp2, ndm_list);
dp = dp2;
} else
goto out;
}
if (pos >= dp->ndm_eocookie) {
if (add)
dp->ndm_eocookie = pos + 1;
else
goto out;
}
retval = &dp->ndm_cookies[pos];
out:
return (retval);
}
/*
* Invalidate cached directory information, except for the actual directory
* blocks (which are invalidated separately).
* Done mainly to avoid the use of stale offset cookies.
*/
void
ncl_invaldir(struct vnode *vp)
{
struct nfsnode *np = VTONFS(vp);
In NFS clients, instead of inconsistently using #ifdef DIAGNOSTIC and #ifndef DIAGNOSTIC for debug assertions, prefer KASSERT(). Also change one #ifdef DIAGNOSTIC in the new nfs server. Submitted by: Mikolaj Golub <to.my.trociny gmail com> MFC after: 2 weeks
2010-06-13 05:24:27 +00:00
KASSERT(vp->v_type == VDIR, ("nfs: invaldir not dir"));
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
ncl_dircookie_lock(np);
np->n_direofoffset = 0;
np->n_cookieverf.nfsuquad[0] = 0;
np->n_cookieverf.nfsuquad[1] = 0;
if (LIST_FIRST(&np->n_cookies))
LIST_FIRST(&np->n_cookies)->ndm_eocookie = 0;
ncl_dircookie_unlock(np);
}
/*
* The write verifier has changed (probably due to a server reboot), so all
* B_NEEDCOMMIT blocks will have to be written again. Since they are on the
* dirty block list as B_DELWRI, all this takes is clearing the B_NEEDCOMMIT
* and B_CLUSTEROK flags. Once done the new write verifier can be set for the
* mount point.
*
* B_CLUSTEROK must be cleared along with B_NEEDCOMMIT because stage 1 data
* writes are not clusterable.
*/
void
ncl_clearcommit(struct mount *mp)
{
struct vnode *vp, *nvp;
struct buf *bp, *nbp;
struct bufobj *bo;
Replace the MNT_VNODE_FOREACH interface with MNT_VNODE_FOREACH_ALL. The primary changes are that the user of the interface no longer needs to manage the mount-mutex locking and that the vnode that is returned has its mutex locked (thus avoiding the need to check to see if its is DOOMED or other possible end of life senarios). To minimize compatibility issues for third-party developers, the old MNT_VNODE_FOREACH interface will remain available so that this change can be MFC'ed to 9. Following the MFC to 9, MNT_VNODE_FOREACH will be removed in head. The reason for this update is to prepare for the addition of the MNT_VNODE_FOREACH_ACTIVE interface that will loop over just the active vnodes associated with a mount point (typically less than 1% of the vnodes associated with the mount point). Reviewed by: kib Tested by: Peter Holm MFC after: 2 weeks
2012-04-17 16:28:22 +00:00
MNT_VNODE_FOREACH_ALL(vp, mp, nvp) {
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
bo = &vp->v_bufobj;
vholdl(vp);
VI_UNLOCK(vp);
BO_LOCK(bo);
TAILQ_FOREACH_SAFE(bp, &bo->bo_dirty.bv_hd, b_bobufs, nbp) {
if (!BUF_ISLOCKED(bp) &&
(bp->b_flags & (B_DELWRI | B_NEEDCOMMIT))
== (B_DELWRI | B_NEEDCOMMIT))
bp->b_flags &= ~(B_NEEDCOMMIT | B_CLUSTEROK);
}
BO_UNLOCK(bo);
vdrop(vp);
}
}
/*
* Called once to initialize data structures...
*/
int
ncl_init(struct vfsconf *vfsp)
{
int i;
/* Ensure async daemons disabled */
Change the experimental NFS client so that it creates nfsiod threads in the same manner as the regular NFS client after r214026 was committed. This resolves the lors fixed by r214026 and its predecessors for the regular client. Reviewed by: jhb MFC after: 2 weeks
2011-04-15 23:07:48 +00:00
for (i = 0; i < NFS_MAXASYNCDAEMON; i++) {
Patch the experimental NFS client in a manner analogous to r203072 for the regular NFS client. Also, delete two fields of struct nfsmount that are not used by the FreeBSD port of the client. MFC after: 2 weeks
2010-01-28 16:17:24 +00:00
ncl_iodwant[i] = NFSIOD_NOT_AVAILABLE;
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
ncl_iodmount[i] = NULL;
}
Change the experimental NFS client so that it creates nfsiod threads in the same manner as the regular NFS client after r214026 was committed. This resolves the lors fixed by r214026 and its predecessors for the regular client. Reviewed by: jhb MFC after: 2 weeks
2011-04-15 23:07:48 +00:00
TASK_INIT(&ncl_nfsiodnew_task, 0, ncl_nfsiodnew_tq, 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
ncl_nhinit(); /* Init the nfsnode table */
return (0);
}
Reference in New Issue Copy Permalink