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

1421 lines
38 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
/*-
sys: further adoption of SPDX licensing ID tags. Mainly focus on files that use BSD 3-Clause license. The Software Package Data Exchange (SPDX) group provides a specification to make it easier for automated tools to detect and summarize well known opensource licenses. We are gradually adopting the specification, noting that the tags are considered only advisory and do not, in any way, superceed or replace the license texts. Special thanks to Wind River for providing access to "The Duke of Highlander" tool: an older (2014) run over FreeBSD tree was useful as a starting point.
2017-11-20 19:43:44 +00:00
* SPDX-License-Identifier: BSD-3-Clause
*
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("$FreeBSD$");
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
#include "opt_inet.h"
opt_inet6.h was missing from some files in the new NFS subsystem. The effect of this was, for clients mounted via inet6 addresses, that the DRC cache would never have a hit in the server. It also broke NFSv4 callbacks when an inet6 address was the only one available in the client. This patch fixes the above, plus deletes opt_inet6.h from a couple of files it is not needed for. MFC after: 2 weeks
2012-01-08 01:54:46 +00:00
#include "opt_inet6.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
Update kernel inclusions of capability.h to use capsicum.h instead; some further refinement is required as some device drivers intended to be portable over FreeBSD versions rely on __FreeBSD_version to decide whether to include capability.h. MFC after: 3 weeks
2014-03-16 10:55:57 +00:00
#include <sys/capsicum.h>
Second-to-last commit implementing Capsicum capabilities in the FreeBSD kernel for FreeBSD 9.0: Add a new capability mask argument to fget(9) and friends, allowing system call code to declare what capabilities are required when an integer file descriptor is converted into an in-kernel struct file *. With options CAPABILITIES compiled into the kernel, this enforces capability protection; without, this change is effectively a no-op. Some cases require special handling, such as mmap(2), which must preserve information about the maximum rights at the time of mapping in the memory map so that they can later be enforced in mprotect(2) -- this is done by narrowing the rights in the existing max_protection field used for similar purposes with file permissions. In namei(9), we assert that the code is not reached from within capability mode, as we're not yet ready to enforce namespace capabilities there. This will follow in a later commit. Update two capability names: CAP_EVENT and CAP_KEVENT become CAP_POST_KEVENT and CAP_POLL_KEVENT to more accurately indicate what they represent. Approved by: re (bz) Submitted by: jonathan Sponsored by: Google Inc
2011-08-11 12:30:23 +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
/*
* generally, I don't like #includes inside .h files, but it seems to
* be the easiest way to handle the port.
*/
nfsclient: Protest loudly when GETATTR responses are invalid BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests with attributes for totally different files). Warn very verbosely when this is detected. Linux' NFS client has a similar warning. Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity of warnings; default to 10. (Zero disables; -1 is unlimited.) Adds a failpoint to aid in validating the warning / behavior with a non-broken server. Use something like: sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)' Reviewed by: rmacklem Approved by: markj (mentor) Sponsored by: EMC / Isilon Storage Division Differential Revision: https://reviews.freebsd.org/D3304
2015-08-05 22:27:30 +00:00
#include <sys/fail.h>
Fix the experimental NFSv4 client so that it recognizes server mount point crossings correctly. It was testing the wrong flag. Also, try harder to make sure that the fsid is different than the one assigned to the client mount point, by hashing the server's fsid (just to create a different value deterministically) when it is the same. MFC after: 2 weeks
2011-04-13 22:16:52 +00:00
#include <sys/hash.h>
nfsclient: Protest loudly when GETATTR responses are invalid BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests with attributes for totally different files). Warn very verbosely when this is detected. Linux' NFS client has a similar warning. Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity of warnings; default to 10. (Zero disables; -1 is unlimited.) Adds a failpoint to aid in validating the warning / behavior with a non-broken server. Use something like: sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)' Reviewed by: rmacklem Approved by: markj (mentor) Sponsored by: EMC / Isilon Storage Division Differential Revision: https://reviews.freebsd.org/D3304
2015-08-05 22:27:30 +00:00
#include <sys/sysctl.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 <fs/nfs/nfsport.h>
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
#include <netinet/in_fib.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/if_ether.h>
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
#include <netinet6/ip6_var.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 <net/if_types.h>
Make NFS address selection use fib4_lookup(). fib4_lookup_nh_ represents pre-epoch generation of fib api, providing less guarantees over pointer validness and requiring on-stack data copying. Switch call to use new fib4_lookup(), allowing to eventually deprecate old api. Differential Revision: https://reviews.freebsd.org/D24977
2020-05-28 07:35:07 +00:00
#include <net/route/nhop.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
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>
#ifdef KDTRACE_HOOKS
dtrace_nfsclient_attrcache_flush_probe_func_t
dtrace_nfscl_attrcache_flush_done_probe;
uint32_t nfscl_attrcache_flush_done_id;
dtrace_nfsclient_attrcache_get_hit_probe_func_t
dtrace_nfscl_attrcache_get_hit_probe;
uint32_t nfscl_attrcache_get_hit_id;
dtrace_nfsclient_attrcache_get_miss_probe_func_t
dtrace_nfscl_attrcache_get_miss_probe;
uint32_t nfscl_attrcache_get_miss_id;
dtrace_nfsclient_attrcache_load_probe_func_t
dtrace_nfscl_attrcache_load_done_probe;
uint32_t nfscl_attrcache_load_done_id;
#endif /* !KDTRACE_HOOKS */
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 u_int32_t newnfs_true, newnfs_false, newnfs_xdrneg1;
extern struct vop_vector newnfs_vnodeops;
extern struct vop_vector newnfs_fifoops;
extern uma_zone_t newnfsnode_zone;
extern struct buf_ops buf_ops_newnfs;
Allocate pager bufs from UMA instead of 80-ish mutex protected linked list. o In vm_pager_bufferinit() create pbuf_zone and start accounting on how many pbufs are we going to have set. In various subsystems that are going to utilize pbufs create private zones via call to pbuf_zsecond_create(). The latter calls uma_zsecond_create(), and sets a limit on created zone. After startup preallocate pbufs according to requirements of all pbuf zones. Subsystems that used to have a private limit with old allocator now have private pbuf zones: md(4), fusefs, NFS client, smbfs, VFS cluster, FFS, swap, vnode pager. The following subsystems use shared pbuf zone: cam(4), nvme(4), physio(9), aio(4). They should have their private limits, but changing that is out of scope of this commit. o Fetch tunable value of kern.nswbuf from init_param2() and while here move NSWBUF_MIN to opt_param.h and eliminate opt_swap.h, that was holding only this option. Default values aren't touched by this commit, but they probably should be reviewed wrt to modern hardware. This change removes a tight bottleneck from sendfile(2) operation, that uses pbufs in vnode pager. Other pagers also would benefit from faster allocation. Together with: gallatin Tested by: pho
2019-01-15 01:02:16 +00:00
extern uma_zone_t ncl_pbuf_zone;
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 short nfsv4_cbport;
extern int nfscl_enablecallb;
extern int nfs_numnfscbd;
extern int nfscl_inited;
struct mtx ncl_iod_mutex;
NFSDLOCKMUTEX;
Merge the pNFS server code from projects/pnfs-planb-server into head. This code merge adds a pNFS service to the NFSv4.1 server. Although it is a large commit it should not affect behaviour for a non-pNFS NFS server. Some documentation on how this works can be found at: http://people.freebsd.org/~rmacklem/pnfs-planb-setup.txt and will hopefully be turned into a proper document soon. This is a merge of the kernel code. Userland and man page changes will come soon, once the dust settles on this merge. It has passed a "make universe", so I hope it will not cause build problems. It also adds NFSv4.1 server support for the "current stateid". Here is a brief overview of the pNFS service: A pNFS service separates the Read/Write oeprations from all the other NFSv4.1 Metadata operations. It is hoped that this separation allows a pNFS service to be configured that exceeds the limits of a single NFS server for either storage capacity and/or I/O bandwidth. It is possible to configure mirroring within the data servers (DSs) so that the data storage file for an MDS file will be mirrored on two or more of the DSs. When this is used, failure of a DS will not stop the pNFS service and a failed DS can be recovered once repaired while the pNFS service continues to operate. Although two way mirroring would be the norm, it is possible to set a mirroring level of up to four or the number of DSs, whichever is less. The Metadata server will always be a single point of failure, just as a single NFS server is. A Plan B pNFS service consists of a single MetaData Server (MDS) and K Data Servers (DS), all of which are recent FreeBSD systems. Clients will mount the MDS as they would a single NFS server. When files are created, the MDS creates a file tree identical to what a single NFS server creates, except that all the regular (VREG) files will be empty. As such, if you look at the exported tree on the MDS directly on the MDS server (not via an NFS mount), the files will all be of size 0. Each of these files will also have two extended attributes in the system attribute name space: pnfsd.dsfile - This extended attrbute stores the information that the MDS needs to find the data storage file(s) on DS(s) for this file. pnfsd.dsattr - This extended attribute stores the Size, AccessTime, ModifyTime and Change attributes for the file, so that the MDS doesn't need to acquire the attributes from the DS for every Getattr operation. For each regular (VREG) file, the MDS creates a data storage file on one (or more if mirroring is enabled) of the DSs in one of the "dsNN" subdirectories. The name of this file is the file handle of the file on the MDS in hexadecimal so that the name is unique. The DSs use subdirectories named "ds0" to "dsN" so that no one directory gets too large. The value of "N" is set via the sysctl vfs.nfsd.dsdirsize on the MDS, with the default being 20. For production servers that will store a lot of files, this value should probably be much larger. It can be increased when the "nfsd" daemon is not running on the MDS, once the "dsK" directories are created. For pNFS aware NFSv4.1 clients, the FreeBSD server will return two pieces of information to the client that allows it to do I/O directly to the DS. DeviceInfo - This is relatively static information that defines what a DS is. The critical bits of information returned by the FreeBSD server is the IP address of the DS and, for the Flexible File layout, that NFSv4.1 is to be used and that it is "tightly coupled". There is a "deviceid" which identifies the DeviceInfo. Layout - This is per file and can be recalled by the server when it is no longer valid. For the FreeBSD server, there is support for two types of layout, call File and Flexible File layout. Both allow the client to do I/O on the DS via NFSv4.1 I/O operations. The Flexible File layout is a more recent variant that allows specification of mirrors, where the client is expected to do writes to all mirrors to maintain them in a consistent state. The Flexible File layout also allows the client to report I/O errors for a DS back to the MDS. The Flexible File layout supports two variants referred to as "tightly coupled" vs "loosely coupled". The FreeBSD server always uses the "tightly coupled" variant where the client uses the same credentials to do I/O on the DS as it would on the MDS. For the "loosely coupled" variant, the layout specifies a synthetic user/group that the client uses to do I/O on the DS. The FreeBSD server does not do striping and always returns layouts for the entire file. The critical information in a layout is Read vs Read/Writea and DeviceID(s) that identify which DS(s) the data is stored on. At this time, the MDS generates File Layout layouts to NFSv4.1 clients that know how to do pNFS for the non-mirrored DS case unless the sysctl vfs.nfsd.default_flexfile is set non-zero, in which case Flexible File layouts are generated. The mirrored DS configuration always generates Flexible File layouts. For NFS clients that do not support NFSv4.1 pNFS, all I/O operations are done against the MDS which acts as a proxy for the appropriate DS(s). When the MDS receives an I/O RPC, it will do the RPC on the DS as a proxy. If the DS is on the same machine, the MDS/DS will do the RPC on the DS as a proxy and so on, until the machine runs out of some resource, such as session slots or mbufs. As such, DSs must be separate systems from the MDS. Tested by: james.rose@framestore.com Relnotes: yes
2018-06-12 19:36:32 +00:00
extern struct mtx nfsrv_dslock_mtx;
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 void (*ncl_call_invalcaches)(struct vnode *);
nfsclient: Protest loudly when GETATTR responses are invalid BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests with attributes for totally different files). Warn very verbosely when this is detected. Linux' NFS client has a similar warning. Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity of warnings; default to 10. (Zero disables; -1 is unlimited.) Adds a failpoint to aid in validating the warning / behavior with a non-broken server. Use something like: sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)' Reviewed by: rmacklem Approved by: markj (mentor) Sponsored by: EMC / Isilon Storage Division Differential Revision: https://reviews.freebsd.org/D3304
2015-08-05 22:27:30 +00:00
SYSCTL_DECL(_vfs_nfs);
static int ncl_fileid_maxwarnings = 10;
SYSCTL_INT(_vfs_nfs, OID_AUTO, fileid_maxwarnings, CTLFLAG_RWTUN,
&ncl_fileid_maxwarnings, 0,
"Limit fileid corruption warnings; 0 is off; -1 is unlimited");
static volatile int ncl_fileid_nwarnings;
static void nfscl_warn_fileid(struct nfsmount *, struct nfsvattr *,
struct nfsvattr *);
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
/*
* Comparison function for vfs_hash functions.
*/
int
newnfs_vncmpf(struct vnode *vp, void *arg)
{
struct nfsfh *nfhp = (struct nfsfh *)arg;
struct nfsnode *np = VTONFS(vp);
if (np->n_fhp->nfh_len != nfhp->nfh_len ||
NFSBCMP(np->n_fhp->nfh_fh, nfhp->nfh_fh, nfhp->nfh_len))
return (1);
return (0);
}
/*
* Look up a vnode/nfsnode by file handle.
* Callers must check for mount points!!
* In all cases, a pointer to a
* nfsnode structure is returned.
* This variant takes a "struct nfsfh *" as second argument and uses
* that structure up, either by hanging off the nfsnode or FREEing it.
*/
int
nfscl_nget(struct mount *mntp, struct vnode *dvp, struct nfsfh *nfhp,
struct componentname *cnp, struct thread *td, struct nfsnode **npp,
Add a lktype flags argument to nfscl_nget() and ncl_nget() in the experimental NFS client so that its nfs_lookup() function can use cn_lkflags in a manner analagous to the regular NFS client. MFC after: 2 weeks
2011-04-16 23:20:21 +00:00
void *stuff, int lkflags)
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 nfsnode *np, *dnp;
struct vnode *vp, *nvp;
struct nfsv4node *newd, *oldd;
int error;
u_int hash;
struct nfsmount *nmp;
nmp = VFSTONFS(mntp);
dnp = VTONFS(dvp);
*npp = NULL;
hash = fnv_32_buf(nfhp->nfh_fh, nfhp->nfh_len, FNV1_32_INIT);
Add a lktype flags argument to nfscl_nget() and ncl_nget() in the experimental NFS client so that its nfs_lookup() function can use cn_lkflags in a manner analagous to the regular NFS client. MFC after: 2 weeks
2011-04-16 23:20:21 +00:00
error = vfs_hash_get(mntp, hash, lkflags,
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
td, &nvp, newnfs_vncmpf, nfhp);
if (error == 0 && nvp != NULL) {
/*
* I believe there is a slight chance that vgonel() could
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
* get called on this vnode between when NFSVOPLOCK() drops
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
* the VI_LOCK() and vget() acquires it again, so that it
* hasn't yet had v_usecount incremented. If this were to
vfs: introduce v_irflag and make v_type smaller The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time. v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED. Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
2019-12-08 21:30:04 +00:00
* happen, the VIRF_DOOMED flag would be set, so check for
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
* that here. Since we now have the v_usecount incremented,
vfs: introduce v_irflag and make v_type smaller The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time. v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED. Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
2019-12-08 21:30:04 +00:00
* we should be ok until we vrele() it, if the VIRF_DOOMED
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
* flag isn't set now.
*/
VI_LOCK(nvp);
vfs: introduce v_irflag and make v_type smaller The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time. v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED. Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
2019-12-08 21:30:04 +00:00
if (VN_IS_DOOMED(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
VI_UNLOCK(nvp);
vrele(nvp);
error = ENOENT;
} else {
VI_UNLOCK(nvp);
}
}
if (error) {
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
free(nfhp, M_NFSFH);
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);
}
if (nvp != NULL) {
np = VTONFS(nvp);
/*
* For NFSv4, check to see if it is the same name and
* replace the name, if it is different.
*/
oldd = newd = NULL;
if ((nmp->nm_flag & NFSMNT_NFSV4) && np->n_v4 != NULL &&
nvp->v_type == VREG &&
(np->n_v4->n4_namelen != cnp->cn_namelen ||
NFSBCMP(cnp->cn_nameptr, NFS4NODENAME(np->n_v4),
cnp->cn_namelen) ||
dnp->n_fhp->nfh_len != np->n_v4->n4_fhlen ||
NFSBCMP(dnp->n_fhp->nfh_fh, np->n_v4->n4_data,
dnp->n_fhp->nfh_len))) {
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
newd = malloc(
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
sizeof (struct nfsv4node) + dnp->n_fhp->nfh_len +
+ cnp->cn_namelen - 1, M_NFSV4NODE, M_WAITOK);
NFSLOCKNODE(np);
if (newd != NULL && np->n_v4 != NULL && nvp->v_type == VREG
&& (np->n_v4->n4_namelen != cnp->cn_namelen ||
NFSBCMP(cnp->cn_nameptr, NFS4NODENAME(np->n_v4),
cnp->cn_namelen) ||
dnp->n_fhp->nfh_len != np->n_v4->n4_fhlen ||
NFSBCMP(dnp->n_fhp->nfh_fh, np->n_v4->n4_data,
dnp->n_fhp->nfh_len))) {
oldd = np->n_v4;
np->n_v4 = newd;
newd = NULL;
np->n_v4->n4_fhlen = dnp->n_fhp->nfh_len;
np->n_v4->n4_namelen = cnp->cn_namelen;
NFSBCOPY(dnp->n_fhp->nfh_fh, np->n_v4->n4_data,
dnp->n_fhp->nfh_len);
NFSBCOPY(cnp->cn_nameptr, NFS4NODENAME(np->n_v4),
cnp->cn_namelen);
}
NFSUNLOCKNODE(np);
}
if (newd != NULL)
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
free(newd, M_NFSV4NODE);
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 (oldd != NULL)
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
free(oldd, M_NFSV4NODE);
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
*npp = np;
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
free(nfhp, M_NFSFH);
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);
}
np = uma_zalloc(newnfsnode_zone, M_WAITOK | M_ZERO);
For performance reasons, it is useful to have a single string used as the name of a filesystem when setting it as the first parameter to the getnewvnode() function. Most filesystems call getnewvnode from just one place so can use a literal string as the first parameter. However, NFS calls getnewvnode from two places, so we create a global constant string that can be used by the two instances. This change also collapses two instances of getnewvnode() in the UFS filesystem to a single call. Reviewed by: kib Tested by: Peter Holm
2015-11-29 21:01:02 +00:00
error = getnewvnode(nfs_vnode_tag, mntp, &newnfs_vnodeops, &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
if (error) {
uma_zfree(newnfsnode_zone, np);
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
free(nfhp, M_NFSFH);
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);
}
vp = nvp;
A problem with respect to data read through the buffer cache for both NFS clients was reported to freebsd-fs@ under the subject "NFS corruption in recent HEAD" on Nov. 26, 2011. This problem occurred when a TCP mounted root fs was changed to using UDP. I believe that this problem was caused by the change in mnt_stat.f_iosize that occurred because rsize was decreased to the maximum supported by UDP. This patch fixes the problem by using v_bufobj.bo_bsize instead of f_iosize, since the latter is set to f_iosize when the vnode is allocated, but does not change for a given vnode when f_iosize changes. Reported by: pjd Reviewed by: kib MFC after: 2 weeks
2012-01-27 02:46:12 +00:00
KASSERT(vp->v_bufobj.bo_bsize != 0, ("nfscl_nget: bo_bsize == 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
vp->v_bufobj.bo_ops = &buf_ops_newnfs;
vp->v_data = np;
np->n_vnode = vp;
/*
* Initialize the mutex even if the vnode is going to be a loser.
* This simplifies the logic in reclaim, which can then unconditionally
* destroy the mutex (in the case of the loser, or if hash_insert
* happened to return an error no special casing is needed).
*/
mtx_init(&np->n_mtx, "NEWNFSnode lock", NULL, MTX_DEF | MTX_DUPOK);
Do not drop NFS vnode lock when performing consistency checks. Currently several paths in the NFS client upgrade the shared vnode lock to exclusive, which might cause temporal dropping of the lock. This action appears to be fatal for nullfs mounts over NFS. If the operation is performed over nullfs vnode, then bypassed down to NFS VOP, and the lock is dropped, other thread might reclaim the upper nullfs vnode. Since on reclaim the nullfs vnode lock and NFS vnode lock are split, the original lock state of the nullfs vnode is not restored. As result, VFS operations receive not locked vnode after a VOP call. Stop upgrading the vnode lock when we check the consistency or flush buffers as result of detected inconsistency. Instead, allocate a new lockmgr lock for each NFS node, which is locked exclusive instead of the vnode lock upgrade. In other words, the other parallel modification of the vnode are excluded by either vnode lock conflict or exclusivity of the new lock when the vnode lock is shared. Also revert r316529 because now the vnode cannot be reclaimed during ncl_vinvalbuf(). In collaboration with: pho Reviewed by: rmacklem Reported and tested by: pho Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D12083
2017-08-20 10:08:45 +00:00
lockinit(&np->n_excl, PVFS, "nfsupg", VLKTIMEOUT, LK_NOSHARE |
LK_CANRECURSE);
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
/*
* Are we getting the root? If so, make sure the vnode flags
* are correct
*/
if ((nfhp->nfh_len == nmp->nm_fhsize) &&
!bcmp(nfhp->nfh_fh, nmp->nm_fh, nfhp->nfh_len)) {
if (vp->v_type == VNON)
vp->v_type = VDIR;
vp->v_vflag |= VV_ROOT;
}
Fix interface between nfsclient and vnode pager. Make the nfsclient always call vnode_pager_setsize() with the vnode exclusively locked. This ensures that page fault always can find the backing page if the object size check succeeded. Set VV_VMSIZEVNLOCK flag on NFS nodes. The main offender breaking the interface in nfsclient is nfs_loadattrcache(), which is used whenever server responded with updated attributes, which can happen on non-changing operations as well. Also, iod threads only have buffers locked (and even that is LK_KERNPROC), but they still may call nfs_loadattrcache() on RPC response. Instead of immediately calling vnode_pager_setsize() if server response indicated changed file size, but the vnode is not exclusively locked, set a new node flag NVNSETSZSKIP. When the vnode exclusively locked, or when we can temporary upgrade the lock to exclusive, call vnode_pager_setsize(), by providing the nfsclient VOP_LOCK() implementation. Tested by: pho Discussed with: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D21883
2019-10-22 16:17:38 +00:00
vp->v_vflag |= VV_VMSIZEVNLOCK;
fs: clean up empty lines in .c and .h files
2020-09-01 21:18:40 +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
np->n_fhp = nfhp;
/*
* For NFSv4, we have to attach the directory file handle and
* file name, so that Open Ops can be done later.
*/
if (nmp->nm_flag & NFSMNT_NFSV4) {
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
np->n_v4 = malloc(sizeof (struct nfsv4node)
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
+ dnp->n_fhp->nfh_len + cnp->cn_namelen - 1, M_NFSV4NODE,
M_WAITOK);
np->n_v4->n4_fhlen = dnp->n_fhp->nfh_len;
np->n_v4->n4_namelen = cnp->cn_namelen;
NFSBCOPY(dnp->n_fhp->nfh_fh, np->n_v4->n4_data,
dnp->n_fhp->nfh_len);
NFSBCOPY(cnp->cn_nameptr, NFS4NODENAME(np->n_v4),
cnp->cn_namelen);
} else {
np->n_v4 = NULL;
}
/*
* NFS supports recursive and shared locking.
*/
Add dedicated routines to toggle lockmgr flags such as LK_NOSHARE and LK_CANRECURSE after a lock is created. Use them to implement macros that otherwise manipulated the flags directly. Assert that the associated lockmgr lock is exclusively locked by the current thread when manipulating these flags to ensure the flag updates are safe. This last change required some minor shuffling in a few filesystems to exclusively lock a brand new vnode slightly earlier. Reviewed by: kib MFC after: 3 days
2010-08-20 19:46:50 +00:00
lockmgr(vp->v_vnlock, LK_EXCLUSIVE | LK_NOWITNESS, 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
VN_LOCK_AREC(vp);
VN_LOCK_ASHARE(vp);
error = insmntque(vp, mntp);
if (error != 0) {
*npp = NULL;
mtx_destroy(&np->n_mtx);
Do not drop NFS vnode lock when performing consistency checks. Currently several paths in the NFS client upgrade the shared vnode lock to exclusive, which might cause temporal dropping of the lock. This action appears to be fatal for nullfs mounts over NFS. If the operation is performed over nullfs vnode, then bypassed down to NFS VOP, and the lock is dropped, other thread might reclaim the upper nullfs vnode. Since on reclaim the nullfs vnode lock and NFS vnode lock are split, the original lock state of the nullfs vnode is not restored. As result, VFS operations receive not locked vnode after a VOP call. Stop upgrading the vnode lock when we check the consistency or flush buffers as result of detected inconsistency. Instead, allocate a new lockmgr lock for each NFS node, which is locked exclusive instead of the vnode lock upgrade. In other words, the other parallel modification of the vnode are excluded by either vnode lock conflict or exclusivity of the new lock when the vnode lock is shared. Also revert r316529 because now the vnode cannot be reclaimed during ncl_vinvalbuf(). In collaboration with: pho Reviewed by: rmacklem Reported and tested by: pho Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D12083
2017-08-20 10:08:45 +00:00
lockdestroy(&np->n_excl);
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
free(nfhp, M_NFSFH);
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 (np->n_v4 != NULL)
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
free(np->n_v4, M_NFSV4NODE);
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
uma_zfree(newnfsnode_zone, np);
return (error);
}
Add a lktype flags argument to nfscl_nget() and ncl_nget() in the experimental NFS client so that its nfs_lookup() function can use cn_lkflags in a manner analagous to the regular NFS client. MFC after: 2 weeks
2011-04-16 23:20:21 +00:00
error = vfs_hash_insert(vp, hash, lkflags,
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
td, &nvp, newnfs_vncmpf, nfhp);
if (error)
return (error);
if (nvp != NULL) {
*npp = VTONFS(nvp);
/* vfs_hash_insert() vput()'s the losing vnode */
return (0);
}
*npp = np;
return (0);
}
/*
Use vfs_hash_ref(9) to eliminate LK_EXCLOTHER kludge. As a consequence, the nfs client override of VOP_LOCK1() is no longer needed. Reviewed and tested by: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 1 week
2016-05-11 06:35:46 +00:00
* Another variant of nfs_nget(). This one is only used by reopen. It
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
* takes almost the same args as nfs_nget(), but only succeeds if an entry
* exists in the cache. (Since files should already be "open" with a
* vnode ref cnt on the node when reopen calls this, it should always
* succeed.)
* Also, don't get a vnode lock, since it may already be locked by some
* other process that is handling it. This is ok, since all other threads
* on the client are blocked by the nfsc_lock being exclusively held by the
* caller of this function.
*/
int
nfscl_ngetreopen(struct mount *mntp, u_int8_t *fhp, int fhsize,
struct thread *td, struct nfsnode **npp)
{
struct vnode *nvp;
u_int hash;
struct nfsfh *nfhp;
int error;
*npp = NULL;
/* For forced dismounts, just return error. */
Replace the checks for MNTK_UNMOUNTF with a macro that does the same thing. This patch defines a macro that checks for MNTK_UNMOUNTF and replaces explicit checks with this macro. It has no effect on semantics, but prepares the code for a future patch where there will also be a NFS specific flag for "forced dismount about to occur". Suggested by: kib MFC after: 2 weeks
2017-07-27 20:55:31 +00:00
if (NFSCL_FORCEDISM(mntp))
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 (EINTR);
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
nfhp = malloc(sizeof (struct nfsfh) + fhsize,
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
M_NFSFH, M_WAITOK);
bcopy(fhp, &nfhp->nfh_fh[0], fhsize);
nfhp->nfh_len = fhsize;
hash = fnv_32_buf(fhp, fhsize, FNV1_32_INIT);
/*
* First, try to get the vnode locked, but don't block for the lock.
*/
error = vfs_hash_get(mntp, hash, (LK_EXCLUSIVE | LK_NOWAIT), td, &nvp,
newnfs_vncmpf, nfhp);
if (error == 0 && nvp != NULL) {
vfs: drop the mostly unused flags argument from VOP_UNLOCK Filesystems which want to use it in limited capacity can employ the VOP_UNLOCK_FLAGS macro. Reviewed by: kib (previous version) Differential Revision: https://reviews.freebsd.org/D21427
2020-01-03 22:29:58 +00:00
NFSVOPUNLOCK(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
} else if (error == EBUSY) {
/*
Use vfs_hash_ref(9) to eliminate LK_EXCLOTHER kludge. As a consequence, the nfs client override of VOP_LOCK1() is no longer needed. Reviewed and tested by: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 1 week
2016-05-11 06:35:46 +00:00
* It is safe so long as a vflush() with
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
* FORCECLOSE has not been done. Since the Renew thread is
* stopped and the MNTK_UNMOUNTF flag is set before doing
* a vflush() with FORCECLOSE, we should be ok here.
*/
Replace the checks for MNTK_UNMOUNTF with a macro that does the same thing. This patch defines a macro that checks for MNTK_UNMOUNTF and replaces explicit checks with this macro. It has no effect on semantics, but prepares the code for a future patch where there will also be a NFS specific flag for "forced dismount about to occur". Suggested by: kib MFC after: 2 weeks
2017-07-27 20:55:31 +00:00
if (NFSCL_FORCEDISM(mntp))
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 = EINTR;
Use vfs_hash_ref(9) to eliminate LK_EXCLOTHER kludge. As a consequence, the nfs client override of VOP_LOCK1() is no longer needed. Reviewed and tested by: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 1 week
2016-05-11 06:35:46 +00:00
else {
vfs_hash_ref(mntp, hash, td, &nvp, newnfs_vncmpf, nfhp);
if (nvp == NULL) {
error = ENOENT;
vfs: introduce v_irflag and make v_type smaller The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time. v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED. Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
2019-12-08 21:30:04 +00:00
} else if (VN_IS_DOOMED(nvp)) {
Use vfs_hash_ref(9) to eliminate LK_EXCLOTHER kludge. As a consequence, the nfs client override of VOP_LOCK1() is no longer needed. Reviewed and tested by: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 1 week
2016-05-11 06:35:46 +00:00
error = ENOENT;
vrele(nvp);
} else {
error = 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
}
style: Remove remaining deprecated MALLOC/FREE macros Mechanically replace uses of MALLOC/FREE with appropriate invocations of malloc(9) / free(9) (a series of sed expressions). Something like: * MALLOC(a, b, ... -> a = malloc(... * FREE( -> free( * free((caddr_t) -> free( No functional change. For now, punt on modifying contrib ipfilter code, leaving a definition of the macro in its KMALLOC(). Reported by: jhb Reviewed by: cy, imp, markj, rmacklem Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D14035
2018-01-25 22:25:13 +00:00
free(nfhp, M_NFSFH);
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)
return (error);
if (nvp != NULL) {
*npp = VTONFS(nvp);
return (0);
}
return (EINVAL);
}
nfsclient: Protest loudly when GETATTR responses are invalid BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests with attributes for totally different files). Warn very verbosely when this is detected. Linux' NFS client has a similar warning. Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity of warnings; default to 10. (Zero disables; -1 is unlimited.) Adds a failpoint to aid in validating the warning / behavior with a non-broken server. Use something like: sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)' Reviewed by: rmacklem Approved by: markj (mentor) Sponsored by: EMC / Isilon Storage Division Differential Revision: https://reviews.freebsd.org/D3304
2015-08-05 22:27:30 +00:00
static void
nfscl_warn_fileid(struct nfsmount *nmp, struct nfsvattr *oldnap,
struct nfsvattr *newnap)
{
int off;
if (ncl_fileid_maxwarnings >= 0 &&
ncl_fileid_nwarnings >= ncl_fileid_maxwarnings)
return;
off = 0;
if (ncl_fileid_maxwarnings >= 0) {
if (++ncl_fileid_nwarnings >= ncl_fileid_maxwarnings)
off = 1;
}
printf("newnfs: server '%s' error: fileid changed. "
"fsid %jx:%jx: expected fileid %#jx, got %#jx. "
"(BROKEN NFS SERVER OR MIDDLEWARE)\n",
nmp->nm_com.nmcom_hostname,
(uintmax_t)nmp->nm_fsid[0],
(uintmax_t)nmp->nm_fsid[1],
(uintmax_t)oldnap->na_fileid,
(uintmax_t)newnap->na_fileid);
if (off)
printf("newnfs: Logged %d times about fileid corruption; "
"going quiet to avoid spamming logs excessively. (Limit "
"is: %d).\n", ncl_fileid_nwarnings,
ncl_fileid_maxwarnings);
}
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
/*
* Load the attribute cache (that lives in the nfsnode entry) with
* the attributes of the second argument and
* Iff vaper not NULL
* copy the attributes to *vaper
* Similar to nfs_loadattrcache(), except the attributes are passed in
* instead of being parsed out of the mbuf list.
*/
int
nfscl_loadattrcache(struct vnode **vpp, struct nfsvattr *nap, void *nvaper,
void *stuff, int writeattr, int dontshrink)
{
struct vnode *vp = *vpp;
struct vattr *vap, *nvap = &nap->na_vattr, *vaper = nvaper;
struct nfsnode *np;
struct nfsmount *nmp;
struct timespec mtime_save;
Further refine r352393, only call vnode_pager_setsize() outside the node lock when shrinking. This is similar to r252528, applied to the above commit. Apparently there is a race which makes necessary at least to keep the n_size and pager size consistent when extending. Current suspect is that iod threads perform vnode_pager_setsize() without taking the vnode lock, which corrupts the file content. Reported and tested by: Masachika ISHIZUKA <ish@amail.plala.or.jp> Discussed with: rmacklem (related issues) Sponsored by: The FreeBSD Foundation MFC after: 1 week
2019-09-17 18:41:39 +00:00
int error, force_fid_err;
nfsclient: Protest loudly when GETATTR responses are invalid BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests with attributes for totally different files). Warn very verbosely when this is detected. Linux' NFS client has a similar warning. Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity of warnings; default to 10. (Zero disables; -1 is unlimited.) Adds a failpoint to aid in validating the warning / behavior with a non-broken server. Use something like: sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)' Reviewed by: rmacklem Approved by: markj (mentor) Sponsored by: EMC / Isilon Storage Division Differential Revision: https://reviews.freebsd.org/D3304
2015-08-05 22:27:30 +00:00
error = 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
/*
* If v_type == VNON it is a new node, so fill in the v_type,
* n_mtime fields. Check to see if it represents a special
* device, and if so, check for a possible alias. Once the
* correct vnode has been obtained, fill in the rest of the
* information.
*/
np = VTONFS(vp);
NFSLOCKNODE(np);
if (vp->v_type != nvap->va_type) {
vp->v_type = nvap->va_type;
if (vp->v_type == VFIFO)
vp->v_op = &newnfs_fifoops;
np->n_mtime = nvap->va_mtime;
}
nmp = VFSTONFS(vp->v_mount);
vap = &np->n_vattr.na_vattr;
mtime_save = vap->va_mtime;
if (writeattr) {
np->n_vattr.na_filerev = nap->na_filerev;
np->n_vattr.na_size = nap->na_size;
np->n_vattr.na_mtime = nap->na_mtime;
np->n_vattr.na_ctime = nap->na_ctime;
np->n_vattr.na_fsid = nap->na_fsid;
Fix the new NFSv4 client so that it doesn't fill the cached mode attribute in as 0 when doing writes. The change adds the Mode attribute plus the others except Owner and Owner_group to the list requested by the NFSv4 Write Operation. This fixed a problem where an executable file built by "cc" would get mode 0111 instead of 0755 for some NFSv4 servers. Found at the recent NFSv4 interoperability Bakeathon. Tested by: tdh at excfb.com MFC after: 2 weeks
2011-06-28 22:52:38 +00:00
np->n_vattr.na_mode = nap->na_mode;
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 {
nfsclient: Protest loudly when GETATTR responses are invalid BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests with attributes for totally different files). Warn very verbosely when this is detected. Linux' NFS client has a similar warning. Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity of warnings; default to 10. (Zero disables; -1 is unlimited.) Adds a failpoint to aid in validating the warning / behavior with a non-broken server. Use something like: sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)' Reviewed by: rmacklem Approved by: markj (mentor) Sponsored by: EMC / Isilon Storage Division Differential Revision: https://reviews.freebsd.org/D3304
2015-08-05 22:27:30 +00:00
force_fid_err = 0;
KFAIL_POINT_ERROR(DEBUG_FP, nfscl_force_fileid_warning,
force_fid_err);
/*
* BROKEN NFS SERVER OR MIDDLEWARE
*
* Certain NFS servers (certain old proprietary filers ca.
* 2006) or broken middleboxes (e.g. WAN accelerator products)
* will respond to GETATTR requests with results for a
* different fileid.
*
* The WAN accelerator we've observed not only serves stale
* cache results for a given file, it also occasionally serves
* results for wholly different files. This causes surprising
* problems; for example the cached size attribute of a file
* may truncate down and then back up, resulting in zero
* regions in file contents read by applications. We observed
* this reliably with Clang and .c files during parallel build.
* A pcap revealed packet fragmentation and GETATTR RPC
* responses with wholly wrong fileids.
*/
if ((np->n_vattr.na_fileid != 0 &&
np->n_vattr.na_fileid != nap->na_fileid) ||
force_fid_err) {
nfscl_warn_fileid(nmp, &np->n_vattr, nap);
error = EIDRM;
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
NFSBCOPY((caddr_t)nap, (caddr_t)&np->n_vattr,
sizeof (struct nfsvattr));
}
/*
* For NFSv4, if the node's fsid is not equal to the mount point's
* fsid, return the low order 32bits of the node's fsid. This
* allows getcwd(3) to work. There is a chance that the fsid might
* be the same as a local fs, but since this is in an NFS mount
* point, I don't think that will cause any problems?
*/
Fix the experimental NFSv4 client so that it recognizes server mount point crossings correctly. It was testing the wrong flag. Also, try harder to make sure that the fsid is different than the one assigned to the client mount point, by hashing the server's fsid (just to create a different value deterministically) when it is the same. MFC after: 2 weeks
2011-04-13 22:16:52 +00:00
if (NFSHASNFSV4(nmp) && NFSHASHASSETFSID(nmp) &&
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
(nmp->nm_fsid[0] != np->n_vattr.na_filesid[0] ||
Fix the experimental NFSv4 client so that it recognizes server mount point crossings correctly. It was testing the wrong flag. Also, try harder to make sure that the fsid is different than the one assigned to the client mount point, by hashing the server's fsid (just to create a different value deterministically) when it is the same. MFC after: 2 weeks
2011-04-13 22:16:52 +00:00
nmp->nm_fsid[1] != np->n_vattr.na_filesid[1])) {
/*
* va_fsid needs to be set to some value derived from
* np->n_vattr.na_filesid that is not equal
* vp->v_mount->mnt_stat.f_fsid[0], so that it changes
* from the value used for the top level server volume
* in the mounted subtree.
*/
Use whole mnt_stat.f_fsid bits for st_dev. Since ino64 expanded dev_t to 64bit, make VOP_GETATTR(9) provide all bits of mnt_stat.f_fsid as va_fsid for vnodes on filesystems which use f_fsid. In particular, NFSv3 and sometimes NFSv4, and ZFS use this method or reporting st_dev by stat(2). Provide a new helper vn_fsid() to avoid duplicating code to copy f_fsid to va_fsid. Note that the change is mostly cosmetic. Its motivation is to avoid sign-extension of f_fsid[0] into 64bit dev_t value which happens after dev_t becomes 64bit.. Reviewed by: avg(zfs), rmacklem (nfs) (both for previous version) Sponsored by: The FreeBSD Foundation
2017-05-27 17:00:30 +00:00
vn_fsid(vp, vap);
Fix bug in r318997: remove the line which overrides vn_fsid() calculation. Noted by: jhb Reviewed by: rmacklem Sponsored by: The FreeBSD Foundation
2017-05-30 21:20:54 +00:00
if ((uint32_t)vap->va_fsid == np->n_vattr.na_filesid[0])
Use whole mnt_stat.f_fsid bits for st_dev. Since ino64 expanded dev_t to 64bit, make VOP_GETATTR(9) provide all bits of mnt_stat.f_fsid as va_fsid for vnodes on filesystems which use f_fsid. In particular, NFSv3 and sometimes NFSv4, and ZFS use this method or reporting st_dev by stat(2). Provide a new helper vn_fsid() to avoid duplicating code to copy f_fsid to va_fsid. Note that the change is mostly cosmetic. Its motivation is to avoid sign-extension of f_fsid[0] into 64bit dev_t value which happens after dev_t becomes 64bit.. Reviewed by: avg(zfs), rmacklem (nfs) (both for previous version) Sponsored by: The FreeBSD Foundation
2017-05-27 17:00:30 +00:00
vap->va_fsid = hash32_buf(
Fix the experimental NFSv4 client so that it recognizes server mount point crossings correctly. It was testing the wrong flag. Also, try harder to make sure that the fsid is different than the one assigned to the client mount point, by hashing the server's fsid (just to create a different value deterministically) when it is the same. MFC after: 2 weeks
2011-04-13 22:16:52 +00:00
np->n_vattr.na_filesid, 2 * sizeof(uint64_t), 0);
} else
Use whole mnt_stat.f_fsid bits for st_dev. Since ino64 expanded dev_t to 64bit, make VOP_GETATTR(9) provide all bits of mnt_stat.f_fsid as va_fsid for vnodes on filesystems which use f_fsid. In particular, NFSv3 and sometimes NFSv4, and ZFS use this method or reporting st_dev by stat(2). Provide a new helper vn_fsid() to avoid duplicating code to copy f_fsid to va_fsid. Note that the change is mostly cosmetic. Its motivation is to avoid sign-extension of f_fsid[0] into 64bit dev_t value which happens after dev_t becomes 64bit.. Reviewed by: avg(zfs), rmacklem (nfs) (both for previous version) Sponsored by: The FreeBSD Foundation
2017-05-27 17:00:30 +00:00
vn_fsid(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
np->n_attrstamp = time_second;
if (vap->va_size != np->n_size) {
if (vap->va_type == VREG) {
if (dontshrink && vap->va_size < np->n_size) {
/*
* We've been told not to shrink the file;
* zero np->n_attrstamp to indicate that
* the attributes are stale.
*/
Further refine r352393, only call vnode_pager_setsize() outside the node lock when shrinking. This is similar to r252528, applied to the above commit. Apparently there is a race which makes necessary at least to keep the n_size and pager size consistent when extending. Current suspect is that iod threads perform vnode_pager_setsize() without taking the vnode lock, which corrupts the file content. Reported and tested by: Masachika ISHIZUKA <ish@amail.plala.or.jp> Discussed with: rmacklem (related issues) Sponsored by: The FreeBSD Foundation MFC after: 1 week
2019-09-17 18:41:39 +00:00
vap->va_size = np->n_size;
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->n_attrstamp = 0;
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_FLUSH_DONE(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
} else if (np->n_flag & NMODIFIED) {
/*
* We've modified the file: Use the larger
* of our size, and the server's size.
*/
if (vap->va_size < np->n_size) {
vap->va_size = np->n_size;
} else {
np->n_size = vap->va_size;
np->n_flag |= NSIZECHANGED;
}
} else {
Further refine r352393, only call vnode_pager_setsize() outside the node lock when shrinking. This is similar to r252528, applied to the above commit. Apparently there is a race which makes necessary at least to keep the n_size and pager size consistent when extending. Current suspect is that iod threads perform vnode_pager_setsize() without taking the vnode lock, which corrupts the file content. Reported and tested by: Masachika ISHIZUKA <ish@amail.plala.or.jp> Discussed with: rmacklem (related issues) Sponsored by: The FreeBSD Foundation MFC after: 1 week
2019-09-17 18:41:39 +00:00
np->n_size = vap->va_size;
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->n_flag |= NSIZECHANGED;
}
Post-r248567, there were times when the client would return a truncated directory for some NFS servers. This turned out to be because the size of a directory reported by an NFS server can be smaller that the ufs-like directory created from the RPC XDR in the client. This patch fixes the problem by changing r248567 so that vnode_pager_setsize() is only done for regular files. Reported and tested by: hartmut.brandt@dlr.de Reviewed by: kib MFC after: 1 week
2013-05-28 22:36:01 +00:00
} else {
np->n_size = vap->va_size;
Do not call vnode_pager_setsize() while a NFS node mutex is locked. vnode_pager_setsize() might sleep waiting for the page after EOF be unbusied. Call vnode_pager_setsize() both for the regular and directory vnodes. Reported by: mich Reviewed by: rmacklem Discussed with: avg, jhb MFC after: 2 weeks
2013-03-21 07:25:08 +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
}
/*
* The following checks are added to prevent a race between (say)
* a READDIR+ and a WRITE.
* READDIR+, WRITE requests sent out.
* READDIR+ resp, WRITE resp received on client.
* However, the WRITE resp was handled before the READDIR+ resp
* causing the post op attrs from the write to be loaded first
* and the attrs from the READDIR+ to be loaded later. If this
* happens, we have stale attrs loaded into the attrcache.
* We detect this by for the mtime moving back. We invalidate the
* attrcache when this happens.
*/
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
if (timespeccmp(&mtime_save, &vap->va_mtime, >)) {
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
/* Size changed or mtime went backwards */
np->n_attrstamp = 0;
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_FLUSH_DONE(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
if (vaper != NULL) {
NFSBCOPY((caddr_t)vap, (caddr_t)vaper, sizeof(*vap));
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;
}
}
nfsclient: Protest loudly when GETATTR responses are invalid BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests with attributes for totally different files). Warn very verbosely when this is detected. Linux' NFS client has a similar warning. Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity of warnings; default to 10. (Zero disables; -1 is unlimited.) Adds a failpoint to aid in validating the warning / behavior with a non-broken server. Use something like: sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)' Reviewed by: rmacklem Approved by: markj (mentor) Sponsored by: EMC / Isilon Storage Division Differential Revision: https://reviews.freebsd.org/D3304
2015-08-05 22:27:30 +00:00
out:
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
#ifdef KDTRACE_HOOKS
if (np->n_attrstamp != 0)
nfsclient: Protest loudly when GETATTR responses are invalid BROKEN NFS SERVER OR MIDDLEWARE: Certain WAN "accelerators" attempt to cache NFS GETATTR traffic, but actually corrupt it (e.g., responding to requests with attributes for totally different files). Warn very verbosely when this is detected. Linux' NFS client has a similar warning. Adds a sysctl/tunable (vfs.nfs.fileid_maxwarnings) to configure the quantity of warnings; default to 10. (Zero disables; -1 is unlimited.) Adds a failpoint to aid in validating the warning / behavior with a non-broken server. Use something like: sysctl 'debug.fail_point.nfscl_force_fileid_warning=10%return(1)' Reviewed by: rmacklem Approved by: markj (mentor) Sponsored by: EMC / Isilon Storage Division Differential Revision: https://reviews.freebsd.org/D3304
2015-08-05 22:27:30 +00:00
KDTRACE_NFS_ATTRCACHE_LOAD_DONE(vp, vap, error);
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
#endif
Fix interface between nfsclient and vnode pager. Make the nfsclient always call vnode_pager_setsize() with the vnode exclusively locked. This ensures that page fault always can find the backing page if the object size check succeeded. Set VV_VMSIZEVNLOCK flag on NFS nodes. The main offender breaking the interface in nfsclient is nfs_loadattrcache(), which is used whenever server responded with updated attributes, which can happen on non-changing operations as well. Also, iod threads only have buffers locked (and even that is LK_KERNPROC), but they still may call nfs_loadattrcache() on RPC response. Instead of immediately calling vnode_pager_setsize() if server response indicated changed file size, but the vnode is not exclusively locked, set a new node flag NVNSETSZSKIP. When the vnode exclusively locked, or when we can temporary upgrade the lock to exclusive, call vnode_pager_setsize(), by providing the nfsclient VOP_LOCK() implementation. Tested by: pho Discussed with: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D21883
2019-10-22 16:17:38 +00:00
(void)ncl_pager_setsize(vp, NULL);
return (error);
}
/*
* Call vnode_pager_setsize() if the size of the node changed, as
* recorded in nfsnode vs. v_object, or delay the call if notifying
* the pager is not possible at the moment.
*
* If nsizep is non-NULL, the call is delayed and the new node size is
* provided. Caller should itself call vnode_pager_setsize() if
* function returned true. If nsizep is NULL, function tries to call
* vnode_pager_setsize() itself if needed and possible, and the nfs
* node is unlocked unconditionally, the return value is not useful.
*/
bool
ncl_pager_setsize(struct vnode *vp, u_quad_t *nsizep)
{
struct nfsnode *np;
vm_object_t object;
struct vattr *vap;
u_quad_t nsize;
bool setnsize;
np = VTONFS(vp);
NFSASSERTNODE(np);
vap = &np->n_vattr.na_vattr;
Further refine r352393, only call vnode_pager_setsize() outside the node lock when shrinking. This is similar to r252528, applied to the above commit. Apparently there is a race which makes necessary at least to keep the n_size and pager size consistent when extending. Current suspect is that iod threads perform vnode_pager_setsize() without taking the vnode lock, which corrupts the file content. Reported and tested by: Masachika ISHIZUKA <ish@amail.plala.or.jp> Discussed with: rmacklem (related issues) Sponsored by: The FreeBSD Foundation MFC after: 1 week
2019-09-17 18:41:39 +00:00
nsize = vap->va_size;
object = vp->v_object;
setnsize = false;
Fix interface between nfsclient and vnode pager. Make the nfsclient always call vnode_pager_setsize() with the vnode exclusively locked. This ensures that page fault always can find the backing page if the object size check succeeded. Set VV_VMSIZEVNLOCK flag on NFS nodes. The main offender breaking the interface in nfsclient is nfs_loadattrcache(), which is used whenever server responded with updated attributes, which can happen on non-changing operations as well. Also, iod threads only have buffers locked (and even that is LK_KERNPROC), but they still may call nfs_loadattrcache() on RPC response. Instead of immediately calling vnode_pager_setsize() if server response indicated changed file size, but the vnode is not exclusively locked, set a new node flag NVNSETSZSKIP. When the vnode exclusively locked, or when we can temporary upgrade the lock to exclusive, call vnode_pager_setsize(), by providing the nfsclient VOP_LOCK() implementation. Tested by: pho Discussed with: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D21883
2019-10-22 16:17:38 +00:00
if (object != NULL && nsize != object->un_pager.vnp.vnp_size) {
Fix NFS client deadlock when read reports truncated node. If node attribute returned in the reply for read rpc indicate truncation, and it happens that the vnode is exclusively locked, update of the node attributes would try to shrink vnode size. Since during the read some vnode pages were busied by the reading thread, vnode_pager_setsize() deadlocks waiting for the busy state owned by the caller. Use a thread-local flag to indicate that NFS read owns some (s)busy pages states and postpone the call to vnode_pager_setsize() until the thread relinguishes the ownership. Diagnosed by: rlibby Tested by: pho, rlibby Sponsored by: The FreeBSD Foundation MFC after: 1 week
2020-02-22 20:50:30 +00:00
if (VOP_ISLOCKED(vp) == LK_EXCLUSIVE &&
(curthread->td_pflags2 & TDP2_SBPAGES) == 0)
Further refine r352393, only call vnode_pager_setsize() outside the node lock when shrinking. This is similar to r252528, applied to the above commit. Apparently there is a race which makes necessary at least to keep the n_size and pager size consistent when extending. Current suspect is that iod threads perform vnode_pager_setsize() without taking the vnode lock, which corrupts the file content. Reported and tested by: Masachika ISHIZUKA <ish@amail.plala.or.jp> Discussed with: rmacklem (related issues) Sponsored by: The FreeBSD Foundation MFC after: 1 week
2019-09-17 18:41:39 +00:00
setnsize = true;
Fix interface between nfsclient and vnode pager. Make the nfsclient always call vnode_pager_setsize() with the vnode exclusively locked. This ensures that page fault always can find the backing page if the object size check succeeded. Set VV_VMSIZEVNLOCK flag on NFS nodes. The main offender breaking the interface in nfsclient is nfs_loadattrcache(), which is used whenever server responded with updated attributes, which can happen on non-changing operations as well. Also, iod threads only have buffers locked (and even that is LK_KERNPROC), but they still may call nfs_loadattrcache() on RPC response. Instead of immediately calling vnode_pager_setsize() if server response indicated changed file size, but the vnode is not exclusively locked, set a new node flag NVNSETSZSKIP. When the vnode exclusively locked, or when we can temporary upgrade the lock to exclusive, call vnode_pager_setsize(), by providing the nfsclient VOP_LOCK() implementation. Tested by: pho Discussed with: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D21883
2019-10-22 16:17:38 +00:00
else
np->n_flag |= NVNSETSZSKIP;
}
if (nsizep == NULL) {
NFSUNLOCKNODE(np);
if (setnsize)
Further refine r352393, only call vnode_pager_setsize() outside the node lock when shrinking. This is similar to r252528, applied to the above commit. Apparently there is a race which makes necessary at least to keep the n_size and pager size consistent when extending. Current suspect is that iod threads perform vnode_pager_setsize() without taking the vnode lock, which corrupts the file content. Reported and tested by: Masachika ISHIZUKA <ish@amail.plala.or.jp> Discussed with: rmacklem (related issues) Sponsored by: The FreeBSD Foundation MFC after: 1 week
2019-09-17 18:41:39 +00:00
vnode_pager_setsize(vp, nsize);
Fix interface between nfsclient and vnode pager. Make the nfsclient always call vnode_pager_setsize() with the vnode exclusively locked. This ensures that page fault always can find the backing page if the object size check succeeded. Set VV_VMSIZEVNLOCK flag on NFS nodes. The main offender breaking the interface in nfsclient is nfs_loadattrcache(), which is used whenever server responded with updated attributes, which can happen on non-changing operations as well. Also, iod threads only have buffers locked (and even that is LK_KERNPROC), but they still may call nfs_loadattrcache() on RPC response. Instead of immediately calling vnode_pager_setsize() if server response indicated changed file size, but the vnode is not exclusively locked, set a new node flag NVNSETSZSKIP. When the vnode exclusively locked, or when we can temporary upgrade the lock to exclusive, call vnode_pager_setsize(), by providing the nfsclient VOP_LOCK() implementation. Tested by: pho Discussed with: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D21883
2019-10-22 16:17:38 +00:00
setnsize = false;
} else {
*nsizep = nsize;
Further refine r352393, only call vnode_pager_setsize() outside the node lock when shrinking. This is similar to r252528, applied to the above commit. Apparently there is a race which makes necessary at least to keep the n_size and pager size consistent when extending. Current suspect is that iod threads perform vnode_pager_setsize() without taking the vnode lock, which corrupts the file content. Reported and tested by: Masachika ISHIZUKA <ish@amail.plala.or.jp> Discussed with: rmacklem (related issues) Sponsored by: The FreeBSD Foundation MFC after: 1 week
2019-09-17 18:41:39 +00:00
}
Fix interface between nfsclient and vnode pager. Make the nfsclient always call vnode_pager_setsize() with the vnode exclusively locked. This ensures that page fault always can find the backing page if the object size check succeeded. Set VV_VMSIZEVNLOCK flag on NFS nodes. The main offender breaking the interface in nfsclient is nfs_loadattrcache(), which is used whenever server responded with updated attributes, which can happen on non-changing operations as well. Also, iod threads only have buffers locked (and even that is LK_KERNPROC), but they still may call nfs_loadattrcache() on RPC response. Instead of immediately calling vnode_pager_setsize() if server response indicated changed file size, but the vnode is not exclusively locked, set a new node flag NVNSETSZSKIP. When the vnode exclusively locked, or when we can temporary upgrade the lock to exclusive, call vnode_pager_setsize(), by providing the nfsclient VOP_LOCK() implementation. Tested by: pho Discussed with: rmacklem Sponsored by: The FreeBSD Foundation MFC after: 2 weeks Differential revision: https://reviews.freebsd.org/D21883
2019-10-22 16:17:38 +00:00
return (setnsize);
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
}
/*
* Fill in the client id name. For these bytes:
* 1 - they must be unique
* 2 - they should be persistent across client reboots
* 1 is more critical than 2
* Use the mount point's unique id plus either the uuid or, if that
* isn't set, random junk.
*/
void
nfscl_fillclid(u_int64_t clval, char *uuid, u_int8_t *cp, u_int16_t idlen)
{
int uuidlen;
/*
* First, put in the 64bit mount point identifier.
*/
if (idlen >= sizeof (u_int64_t)) {
NFSBCOPY((caddr_t)&clval, cp, sizeof (u_int64_t));
cp += sizeof (u_int64_t);
idlen -= sizeof (u_int64_t);
}
/*
* If uuid is non-zero length, use it.
*/
uuidlen = strlen(uuid);
if (uuidlen > 0 && idlen >= uuidlen) {
NFSBCOPY(uuid, cp, uuidlen);
cp += uuidlen;
idlen -= uuidlen;
}
/*
* This only normally happens if the uuid isn't set.
*/
while (idlen > 0) {
*cp++ = (u_int8_t)(arc4random() % 256);
idlen--;
}
}
/*
* Fill in a lock owner name. For now, pid + the process's creation time.
*/
void
The new NFSv4 client was erroneously using "p" instead of "p_leader" for the "id" for POSIX byte range locking. I think this would only have affected processes created by rfork(2) with the RFTHREAD flag specified. This patch fixes that by passing the "id" down through the various functions from nfs_advlock(). MFC after: 2 weeks
2011-06-05 18:17:37 +00:00
nfscl_filllockowner(void *id, u_int8_t *cp, int flags)
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
{
union {
u_int32_t lval;
u_int8_t cval[4];
} tl;
struct proc *p;
The new NFSv4 client was erroneously using "p" instead of "p_leader" for the "id" for POSIX byte range locking. I think this would only have affected processes created by rfork(2) with the RFTHREAD flag specified. This patch fixes that by passing the "id" down through the various functions from nfs_advlock(). MFC after: 2 weeks
2011-06-05 18:17:37 +00:00
if (id == NULL) {
Add an NFSv4.1 mount option for "use one openowner". Some NFSv4.1 servers such as AmazonEFS can only support a small fixed number of open_owner4s. This patch adds a mount option called "oneopenown" that can be used for NFSv4.1 mounts to make the client do all Opens with the same open_owner4 string. This option can only be used with NFSv4.1 and may not work correctly when Delegations are is use. Reported by: cperciva Tested by: cperciva MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D8988
2017-04-13 21:54:19 +00:00
/* Return the single open_owner of all 0 bytes. */
The new NFSv4 client was erroneously using "p" instead of "p_leader" for the "id" for POSIX byte range locking. I think this would only have affected processes created by rfork(2) with the RFTHREAD flag specified. This patch fixes that by passing the "id" down through the various functions from nfs_advlock(). MFC after: 2 weeks
2011-06-05 18:17:37 +00:00
bzero(cp, NFSV4CL_LOCKNAMELEN);
return;
}
if ((flags & F_POSIX) != 0) {
p = (struct proc *)id;
tl.lval = p->p_pid;
*cp++ = tl.cval[0];
*cp++ = tl.cval[1];
*cp++ = tl.cval[2];
*cp++ = tl.cval[3];
tl.lval = p->p_stats->p_start.tv_sec;
*cp++ = tl.cval[0];
*cp++ = tl.cval[1];
*cp++ = tl.cval[2];
*cp++ = tl.cval[3];
tl.lval = p->p_stats->p_start.tv_usec;
*cp++ = tl.cval[0];
*cp++ = tl.cval[1];
*cp++ = tl.cval[2];
*cp = tl.cval[3];
Add support for flock(2) locks to the new NFSv4 client. I think this should be ok, since the client now delays NFSv4 Close operations until VOP_INACTIVE()/VOP_RECLAIM(). As such, there should be no risk that the NFSv4 Open is closed while an associated byte range lock still exists. Tested by: avg MFC after: 2 weeks
2011-06-05 20:22:56 +00:00
} else if ((flags & F_FLOCK) != 0) {
bcopy(&id, cp, sizeof(id));
bzero(&cp[sizeof(id)], NFSV4CL_LOCKNAMELEN - sizeof(id));
The new NFSv4 client was erroneously using "p" instead of "p_leader" for the "id" for POSIX byte range locking. I think this would only have affected processes created by rfork(2) with the RFTHREAD flag specified. This patch fixes that by passing the "id" down through the various functions from nfs_advlock(). MFC after: 2 weeks
2011-06-05 18:17:37 +00:00
} else {
Add support for flock(2) locks to the new NFSv4 client. I think this should be ok, since the client now delays NFSv4 Close operations until VOP_INACTIVE()/VOP_RECLAIM(). As such, there should be no risk that the NFSv4 Open is closed while an associated byte range lock still exists. Tested by: avg MFC after: 2 weeks
2011-06-05 20:22:56 +00:00
printf("nfscl_filllockowner: not F_POSIX or F_FLOCK\n");
The new NFSv4 client was erroneously using "p" instead of "p_leader" for the "id" for POSIX byte range locking. I think this would only have affected processes created by rfork(2) with the RFTHREAD flag specified. This patch fixes that by passing the "id" down through the various functions from nfs_advlock(). MFC after: 2 weeks
2011-06-05 18:17:37 +00:00
bzero(cp, NFSV4CL_LOCKNAMELEN);
}
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
}
/*
* Find the parent process for the thread passed in as an argument.
* If none exists, return NULL, otherwise return a thread for the parent.
* (Can be any of the threads, since it is only used for td->td_proc.)
*/
NFSPROC_T *
nfscl_getparent(struct thread *td)
{
struct proc *p;
struct thread *ptd;
if (td == NULL)
return (NULL);
p = td->td_proc;
if (p->p_pid == 0)
return (NULL);
p = p->p_pptr;
if (p == NULL)
return (NULL);
ptd = TAILQ_FIRST(&p->p_threads);
return (ptd);
}
/*
* Start up the renew kernel thread.
*/
static void
start_nfscl(void *arg)
{
struct nfsclclient *clp;
struct thread *td;
clp = (struct nfsclclient *)arg;
td = TAILQ_FIRST(&clp->nfsc_renewthread->p_threads);
nfscl_renewthread(clp, td);
kproc_exit(0);
}
void
nfscl_start_renewthread(struct nfsclclient *clp)
{
kproc_create(start_nfscl, (void *)clp, &clp->nfsc_renewthread, 0, 0,
"nfscl");
}
/*
* Handle wcc_data.
* For NFSv4, it assumes that nfsv4_wccattr() was used to set up the getattr
* as the first Op after PutFH.
* (For NFSv4, the postop attributes are after the Op, so they can't be
* parsed here. A separate call to nfscl_postop_attr() is required.)
*/
int
nfscl_wcc_data(struct nfsrv_descript *nd, struct vnode *vp,
struct nfsvattr *nap, int *flagp, int *wccflagp, void *stuff)
{
u_int32_t *tl;
struct nfsnode *np = VTONFS(vp);
struct nfsvattr nfsva;
int error = 0;
if (wccflagp != NULL)
*wccflagp = 0;
if (nd->nd_flag & ND_NFSV3) {
*flagp = 0;
NFSM_DISSECT(tl, u_int32_t *, NFSX_UNSIGNED);
if (*tl == newnfs_true) {
NFSM_DISSECT(tl, u_int32_t *, 6 * NFSX_UNSIGNED);
if (wccflagp != NULL) {
Replace all mtx_lock()/mtx_unlock() on n_mtx with the macros. For a long time, some places in the NFS code have locked/unlocked the NFS node lock with the macros NFSLOCKNODE()/NFSUNLOCKNODE() whereas others have simply used mtx_lock()/mtx_unlock(). Since the NFS node mutex needs to change to an sx lock so it can be held when vnode_pager_setsize() is called, replace all occurrences of mtx_lock/mtx_unlock with the macros to simply making the change to an sx lock in future commit. There is no semantic change as a result of this commit. I am not sure if the change to an sx lock will be MFC'd soon, so I put an MFC of 1 week on this commit so that it could be MFC'd with that commit. Suggested by: kib MFC after: 1 week
2019-09-24 01:58:54 +00:00
NFSLOCKNODE(np);
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
*wccflagp = (np->n_mtime.tv_sec ==
fxdr_unsigned(u_int32_t, *(tl + 2)) &&
np->n_mtime.tv_nsec ==
fxdr_unsigned(u_int32_t, *(tl + 3)));
Replace all mtx_lock()/mtx_unlock() on n_mtx with the macros. For a long time, some places in the NFS code have locked/unlocked the NFS node lock with the macros NFSLOCKNODE()/NFSUNLOCKNODE() whereas others have simply used mtx_lock()/mtx_unlock(). Since the NFS node mutex needs to change to an sx lock so it can be held when vnode_pager_setsize() is called, replace all occurrences of mtx_lock/mtx_unlock with the macros to simply making the change to an sx lock in future commit. There is no semantic change as a result of this commit. I am not sure if the change to an sx lock will be MFC'd soon, so I put an MFC of 1 week on this commit so that it could be MFC'd with that commit. Suggested by: kib MFC after: 1 week
2019-09-24 01:58:54 +00:00
NFSUNLOCKNODE(np);
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 = nfscl_postop_attr(nd, nap, flagp, stuff);
Fix an NFSv3 client case that probably never happens. If an NFSv3 server were to reply with weak cache consistency attributes, but not post operation attributes, the client would use garbage attributes from memory. This was spotted during work on the code for the NFSv4.1 client. I have never seen evidence that this happens and it wouldn't make sense for an NFSv3 server to do this, so this patch is basically "theoretical", but does fix the problem if a server were to do the above. PR: 219552 MFC after: 2 weeks
2017-06-28 21:37:08 +00:00
if (wccflagp != NULL && *flagp == 0)
*wccflagp = 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
} else if ((nd->nd_flag & (ND_NOMOREDATA | ND_NFSV4 | ND_V4WCCATTR))
== (ND_NFSV4 | ND_V4WCCATTR)) {
error = nfsv4_loadattr(nd, NULL, &nfsva, NULL,
NULL, 0, NULL, NULL, NULL, NULL, NULL, 0,
NULL, NULL, NULL, NULL, NULL);
if (error)
return (error);
/*
* Get rid of Op# and status for next op.
*/
NFSM_DISSECT(tl, u_int32_t *, 2 * NFSX_UNSIGNED);
if (*++tl)
nd->nd_flag |= ND_NOMOREDATA;
if (wccflagp != NULL &&
nfsva.na_vattr.va_mtime.tv_sec != 0) {
Replace all mtx_lock()/mtx_unlock() on n_mtx with the macros. For a long time, some places in the NFS code have locked/unlocked the NFS node lock with the macros NFSLOCKNODE()/NFSUNLOCKNODE() whereas others have simply used mtx_lock()/mtx_unlock(). Since the NFS node mutex needs to change to an sx lock so it can be held when vnode_pager_setsize() is called, replace all occurrences of mtx_lock/mtx_unlock with the macros to simply making the change to an sx lock in future commit. There is no semantic change as a result of this commit. I am not sure if the change to an sx lock will be MFC'd soon, so I put an MFC of 1 week on this commit so that it could be MFC'd with that commit. Suggested by: kib MFC after: 1 week
2019-09-24 01:58:54 +00:00
NFSLOCKNODE(np);
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
*wccflagp = (np->n_mtime.tv_sec ==
nfsva.na_vattr.va_mtime.tv_sec &&
np->n_mtime.tv_nsec ==
nfsva.na_vattr.va_mtime.tv_sec);
Replace all mtx_lock()/mtx_unlock() on n_mtx with the macros. For a long time, some places in the NFS code have locked/unlocked the NFS node lock with the macros NFSLOCKNODE()/NFSUNLOCKNODE() whereas others have simply used mtx_lock()/mtx_unlock(). Since the NFS node mutex needs to change to an sx lock so it can be held when vnode_pager_setsize() is called, replace all occurrences of mtx_lock/mtx_unlock with the macros to simply making the change to an sx lock in future commit. There is no semantic change as a result of this commit. I am not sure if the change to an sx lock will be MFC'd soon, so I put an MFC of 1 week on this commit so that it could be MFC'd with that commit. Suggested by: kib MFC after: 1 week
2019-09-24 01:58:54 +00:00
NFSUNLOCKNODE(np);
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
}
}
nfsmout:
return (error);
}
/*
* Get postop attributes.
*/
int
nfscl_postop_attr(struct nfsrv_descript *nd, struct nfsvattr *nap, int *retp,
void *stuff)
{
u_int32_t *tl;
int error = 0;
*retp = 0;
if (nd->nd_flag & ND_NOMOREDATA)
return (error);
if (nd->nd_flag & ND_NFSV3) {
NFSM_DISSECT(tl, u_int32_t *, NFSX_UNSIGNED);
*retp = fxdr_unsigned(int, *tl);
} else if (nd->nd_flag & ND_NFSV4) {
/*
* For NFSv4, the postop attr are at the end, so no point
* in looking if nd_repstat != 0.
*/
if (!nd->nd_repstat) {
NFSM_DISSECT(tl, u_int32_t *, 2 * NFSX_UNSIGNED);
if (*(tl + 1))
/* should never happen since nd_repstat != 0 */
nd->nd_flag |= ND_NOMOREDATA;
else
*retp = 1;
}
} else if (!nd->nd_repstat) {
/* For NFSv2, the attributes are here iff nd_repstat == 0 */
*retp = 1;
}
if (*retp) {
error = nfsm_loadattr(nd, nap);
if (error)
*retp = 0;
}
nfsmout:
return (error);
}
/*
* nfscl_request() - mostly a wrapper for newnfs_request().
*/
int
nfscl_request(struct nfsrv_descript *nd, struct vnode *vp, NFSPROC_T *p,
struct ucred *cred, void *stuff)
{
int ret, vers;
struct nfsmount *nmp;
nmp = VFSTONFS(vp->v_mount);
if (nd->nd_flag & ND_NFSV4)
vers = NFS_VER4;
else if (nd->nd_flag & ND_NFSV3)
vers = NFS_VER3;
else
vers = NFS_VER2;
ret = newnfs_request(nd, nmp, NULL, &nmp->nm_sockreq, vp, p, cred,
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
NFS_PROG, vers, NULL, 1, NULL, 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
return (ret);
}
/*
* fill in this bsden's variant of statfs using nfsstatfs.
*/
void
nfscl_loadsbinfo(struct nfsmount *nmp, struct nfsstatfs *sfp, void *statfs)
{
struct statfs *sbp = (struct statfs *)statfs;
if (nmp->nm_flag & (NFSMNT_NFSV3 | NFSMNT_NFSV4)) {
sbp->f_bsize = NFS_FABLKSIZE;
Fix the new NFS client so that it handles the 64bit fields that are now in "struct statfs" for NFSv3 and NFSv4. Since the ffiles value is uint64_t on the wire, I clip the value to INT64_MAX to avoid setting f_ffree negative. Tested by: kib MFC after: 2 weeks
2011-05-05 00:11:09 +00:00
sbp->f_blocks = sfp->sf_tbytes / NFS_FABLKSIZE;
sbp->f_bfree = sfp->sf_fbytes / NFS_FABLKSIZE;
/*
* Although sf_abytes is uint64_t and f_bavail is int64_t,
* the value after dividing by NFS_FABLKSIZE is small
* enough that it will fit in 63bits, so it is ok to
* assign it to f_bavail without fear that it will become
* negative.
*/
sbp->f_bavail = sfp->sf_abytes / NFS_FABLKSIZE;
sbp->f_files = sfp->sf_tfiles;
/* Since f_ffree is int64_t, clip it to 63bits. */
if (sfp->sf_ffiles > INT64_MAX)
sbp->f_ffree = INT64_MAX;
else
sbp->f_ffree = sfp->sf_ffiles;
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 ((nmp->nm_flag & NFSMNT_NFSV4) == 0) {
Fix the new NFS client so that it handles the 64bit fields that are now in "struct statfs" for NFSv3 and NFSv4. Since the ffiles value is uint64_t on the wire, I clip the value to INT64_MAX to avoid setting f_ffree negative. Tested by: kib MFC after: 2 weeks
2011-05-05 00:11:09 +00:00
/*
* The type casts to (int32_t) ensure that this code is
* compatible with the old NFS client, in that it will
* propagate bit31 to the high order bits. This may or may
* not be correct for NFSv2, but since it is a legacy
* environment, I'd rather retain backwards compatibility.
*/
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
sbp->f_bsize = (int32_t)sfp->sf_bsize;
sbp->f_blocks = (int32_t)sfp->sf_blocks;
sbp->f_bfree = (int32_t)sfp->sf_bfree;
sbp->f_bavail = (int32_t)sfp->sf_bavail;
sbp->f_files = 0;
sbp->f_ffree = 0;
}
}
/*
* Use the fsinfo stuff to update the mount point.
*/
void
nfscl_loadfsinfo(struct nfsmount *nmp, struct nfsfsinfo *fsp)
{
if ((nmp->nm_wsize == 0 || fsp->fs_wtpref < nmp->nm_wsize) &&
fsp->fs_wtpref >= NFS_FABLKSIZE)
nmp->nm_wsize = (fsp->fs_wtpref + NFS_FABLKSIZE - 1) &
~(NFS_FABLKSIZE - 1);
if (fsp->fs_wtmax < nmp->nm_wsize && fsp->fs_wtmax > 0) {
nmp->nm_wsize = fsp->fs_wtmax & ~(NFS_FABLKSIZE - 1);
if (nmp->nm_wsize == 0)
nmp->nm_wsize = fsp->fs_wtmax;
}
if (nmp->nm_wsize < NFS_FABLKSIZE)
nmp->nm_wsize = NFS_FABLKSIZE;
if ((nmp->nm_rsize == 0 || fsp->fs_rtpref < nmp->nm_rsize) &&
fsp->fs_rtpref >= NFS_FABLKSIZE)
nmp->nm_rsize = (fsp->fs_rtpref + NFS_FABLKSIZE - 1) &
~(NFS_FABLKSIZE - 1);
if (fsp->fs_rtmax < nmp->nm_rsize && fsp->fs_rtmax > 0) {
nmp->nm_rsize = fsp->fs_rtmax & ~(NFS_FABLKSIZE - 1);
if (nmp->nm_rsize == 0)
nmp->nm_rsize = fsp->fs_rtmax;
}
if (nmp->nm_rsize < NFS_FABLKSIZE)
nmp->nm_rsize = NFS_FABLKSIZE;
if ((nmp->nm_readdirsize == 0 || fsp->fs_dtpref < nmp->nm_readdirsize)
&& fsp->fs_dtpref >= NFS_DIRBLKSIZ)
nmp->nm_readdirsize = (fsp->fs_dtpref + NFS_DIRBLKSIZ - 1) &
~(NFS_DIRBLKSIZ - 1);
if (fsp->fs_rtmax < nmp->nm_readdirsize && fsp->fs_rtmax > 0) {
nmp->nm_readdirsize = fsp->fs_rtmax & ~(NFS_DIRBLKSIZ - 1);
if (nmp->nm_readdirsize == 0)
nmp->nm_readdirsize = fsp->fs_rtmax;
}
if (nmp->nm_readdirsize < NFS_DIRBLKSIZ)
nmp->nm_readdirsize = NFS_DIRBLKSIZ;
if (fsp->fs_maxfilesize > 0 &&
fsp->fs_maxfilesize < nmp->nm_maxfilesize)
nmp->nm_maxfilesize = fsp->fs_maxfilesize;
nmp->nm_mountp->mnt_stat.f_iosize = newnfs_iosize(nmp);
nmp->nm_state |= NFSSTA_GOTFSINFO;
}
/*
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
* Lookups source address which should be used to communicate with
* @nmp and stores it inside @pdst.
*
* Returns 0 on success.
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
*/
u_int8_t *
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
nfscl_getmyip(struct nfsmount *nmp, struct in6_addr *paddr, int *isinet6p)
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
{
Unbreak NOIP builds after r294084.
2016-01-15 16:45:36 +00:00
#if defined(INET6) || defined(INET)
Fix NOINET6 build broken by r361575. Reported by: ci, hps
2020-05-28 09:52:28 +00:00
int fibnum;
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
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
fibnum = curthread->td_proc->p_fibnum;
Unbreak NOIP builds after r294084.
2016-01-15 16:45:36 +00:00
#endif
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
#ifdef INET
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 (nmp->nm_nam->sa_family == AF_INET) {
Make NFS address selection use fib4_lookup(). fib4_lookup_nh_ represents pre-epoch generation of fib api, providing less guarantees over pointer validness and requiring on-stack data copying. Switch call to use new fib4_lookup(), allowing to eventually deprecate old api. Differential Revision: https://reviews.freebsd.org/D24977
2020-05-28 07:35:07 +00:00
struct epoch_tracker et;
struct nhop_object *nh;
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
struct sockaddr_in *sin;
Make NFS address selection use fib4_lookup(). fib4_lookup_nh_ represents pre-epoch generation of fib api, providing less guarantees over pointer validness and requiring on-stack data copying. Switch call to use new fib4_lookup(), allowing to eventually deprecate old api. Differential Revision: https://reviews.freebsd.org/D24977
2020-05-28 07:35:07 +00:00
struct in_addr addr = {};
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +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
sin = (struct sockaddr_in *)nmp->nm_nam;
Make NFS address selection use fib4_lookup(). fib4_lookup_nh_ represents pre-epoch generation of fib api, providing less guarantees over pointer validness and requiring on-stack data copying. Switch call to use new fib4_lookup(), allowing to eventually deprecate old api. Differential Revision: https://reviews.freebsd.org/D24977
2020-05-28 07:35:07 +00:00
NET_EPOCH_ENTER(et);
Fix the new NFSv4 client so that it doesn't crash when a mount is done for a VIMAGE kernel. Tested by: glz at hidden-powers dot com Reviewed by: bz MFC after: 2 weeks
2011-06-05 17:31:44 +00:00
CURVNET_SET(CRED_TO_VNET(nmp->nm_sockreq.nr_cred));
Make NFS address selection use fib4_lookup(). fib4_lookup_nh_ represents pre-epoch generation of fib api, providing less guarantees over pointer validness and requiring on-stack data copying. Switch call to use new fib4_lookup(), allowing to eventually deprecate old api. Differential Revision: https://reviews.freebsd.org/D24977
2020-05-28 07:35:07 +00:00
nh = fib4_lookup(fibnum, sin->sin_addr, 0, NHR_NONE, 0);
Fix the new NFSv4 client so that it doesn't crash when a mount is done for a VIMAGE kernel. Tested by: glz at hidden-powers dot com Reviewed by: bz MFC after: 2 weeks
2011-06-05 17:31:44 +00:00
CURVNET_RESTORE();
Make NFS address selection use fib4_lookup(). fib4_lookup_nh_ represents pre-epoch generation of fib api, providing less guarantees over pointer validness and requiring on-stack data copying. Switch call to use new fib4_lookup(), allowing to eventually deprecate old api. Differential Revision: https://reviews.freebsd.org/D24977
2020-05-28 07:35:07 +00:00
if (nh != NULL)
addr = IA_SIN(ifatoia(nh->nh_ifa))->sin_addr;
NET_EPOCH_EXIT(et);
if (nh == NULL)
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
return (NULL);
Make NFS address selection use fib4_lookup(). fib4_lookup_nh_ represents pre-epoch generation of fib api, providing less guarantees over pointer validness and requiring on-stack data copying. Switch call to use new fib4_lookup(), allowing to eventually deprecate old api. Differential Revision: https://reviews.freebsd.org/D24977
2020-05-28 07:35:07 +00:00
if (IN_LOOPBACK(ntohl(addr.s_addr))) {
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
/* Ignore loopback addresses */
return (NULL);
}
*isinet6p = 0;
Make NFS address selection use fib4_lookup(). fib4_lookup_nh_ represents pre-epoch generation of fib api, providing less guarantees over pointer validness and requiring on-stack data copying. Switch call to use new fib4_lookup(), allowing to eventually deprecate old api. Differential Revision: https://reviews.freebsd.org/D24977
2020-05-28 07:35:07 +00:00
*((struct in_addr *)paddr) = addr;
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
return (u_int8_t *)paddr;
}
#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
#ifdef INET6
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
if (nmp->nm_nam->sa_family == AF_INET6) {
struct sockaddr_in6 *sin6;
Fix NOINET6 build broken by r361575. Reported by: ci, hps
2020-05-28 09:52:28 +00:00
int 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
sin6 = (struct sockaddr_in6 *)nmp->nm_nam;
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
Fix the new NFSv4 client so that it doesn't crash when a mount is done for a VIMAGE kernel. Tested by: glz at hidden-powers dot com Reviewed by: bz MFC after: 2 weeks
2011-06-05 17:31:44 +00:00
CURVNET_SET(CRED_TO_VNET(nmp->nm_sockreq.nr_cred));
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
error = in6_selectsrc_addr(fibnum, &sin6->sin6_addr,
sin6->sin6_scope_id, NULL, paddr, NULL);
Fix the new NFSv4 client so that it doesn't crash when a mount is done for a VIMAGE kernel. Tested by: glz at hidden-powers dot com Reviewed by: bz MFC after: 2 weeks
2011-06-05 17:31:44 +00:00
CURVNET_RESTORE();
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
if (error != 0)
return (NULL);
if (IN6_IS_ADDR_LOOPBACK(paddr))
return (NULL);
/* Scope is embedded in */
*isinet6p = 1;
return (u_int8_t *)paddr;
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
}
Make nfscl_getmyip() use new routing KPI. * Use standard IPv6 SAS instead of rt->rt_ifa address. * Make address lookup work for IPv6 LLA. * Save address into buffer provided by caller instead of using static vars. Discussed with: rmacklem
2016-01-15 09:05:14 +00:00
#endif
return (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
}
/*
* Copy NFS uid, gids from the cred structure.
*/
void
newnfs_copyincred(struct ucred *cr, struct nfscred *nfscr)
{
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
int i;
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
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(cr->cr_ngroups >= 0,
("newnfs_copyincred: negative cr_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
nfscr->nfsc_uid = cr->cr_uid;
Replace RPCAUTH_UNIXGIDS with NFS_MAXGRPS so that nfscbd.c will build. Approved by: kib (mentor)
2009-06-20 17:11:07 +00:00
nfscr->nfsc_ngroups = MIN(cr->cr_ngroups, NFS_MAXGRPS + 1);
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
for (i = 0; i < nfscr->nfsc_ngroups; i++)
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
nfscr->nfsc_groups[i] = cr->cr_groups[i];
}
/*
* Do any client specific initialization.
*/
void
nfscl_init(void)
{
static int inited = 0;
if (inited)
return;
inited = 1;
nfscl_inited = 1;
Allocate pager bufs from UMA instead of 80-ish mutex protected linked list. o In vm_pager_bufferinit() create pbuf_zone and start accounting on how many pbufs are we going to have set. In various subsystems that are going to utilize pbufs create private zones via call to pbuf_zsecond_create(). The latter calls uma_zsecond_create(), and sets a limit on created zone. After startup preallocate pbufs according to requirements of all pbuf zones. Subsystems that used to have a private limit with old allocator now have private pbuf zones: md(4), fusefs, NFS client, smbfs, VFS cluster, FFS, swap, vnode pager. The following subsystems use shared pbuf zone: cam(4), nvme(4), physio(9), aio(4). They should have their private limits, but changing that is out of scope of this commit. o Fetch tunable value of kern.nswbuf from init_param2() and while here move NSWBUF_MIN to opt_param.h and eliminate opt_swap.h, that was holding only this option. Default values aren't touched by this commit, but they probably should be reviewed wrt to modern hardware. This change removes a tight bottleneck from sendfile(2) operation, that uses pbufs in vnode pager. Other pagers also would benefit from faster allocation. Together with: gallatin Tested by: pho
2019-01-15 01:02:16 +00:00
ncl_pbuf_zone = pbuf_zsecond_create("nfspbuf", nswbuf / 2);
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 each of the attributes to be set, to ensure they aren't already
* the correct value. Disable setting ones already correct.
*/
int
nfscl_checksattr(struct vattr *vap, struct nfsvattr *nvap)
{
if (vap->va_mode != (mode_t)VNOVAL) {
if (vap->va_mode == nvap->na_mode)
vap->va_mode = (mode_t)VNOVAL;
}
if (vap->va_uid != (uid_t)VNOVAL) {
if (vap->va_uid == nvap->na_uid)
vap->va_uid = (uid_t)VNOVAL;
}
if (vap->va_gid != (gid_t)VNOVAL) {
if (vap->va_gid == nvap->na_gid)
vap->va_gid = (gid_t)VNOVAL;
}
if (vap->va_size != VNOVAL) {
if (vap->va_size == nvap->na_size)
vap->va_size = VNOVAL;
}
/*
* We are normally called with only a partially initialized
* VAP. Since the NFSv3 spec says that server may use the
* file attributes to store the verifier, the spec requires
* us to do a SETATTR RPC. FreeBSD servers store the verifier
* in atime, but we can't really assume that all servers will
* so we ensure that our SETATTR sets both atime and mtime.
r245508 modified the NFS client's Setattr RPC to use VA_UTIMES_NULL to indicate whether it should set the time to the current tod on the server. This had the side effect of making the NFS client use the client's timestamp for exclusive create, starting with FreeBSD9.2. Unfortunately a bug in some Solaris NFS servers causes these servers to return NFS_OK to the Setattr RPC done during exclusive create, but not actually set the file's mode, leaving the file's mode == 0. This patch restores the NFS client's behaviour to use the server's tod for the exclusive open's Setattr RPC, to avoid the Solaris server bug and to restore the pre-FreeBSD9.2 NFS behaviour. Discussed on: freebsd-fs PR: 186293 MFC after: 3 months
2014-12-28 21:13:52 +00:00
* Set the VA_UTIMES_NULL flag for this case, so that
* the server's time will be used. This is needed to
* work around a bug in some Solaris servers, where
* setting the time TOCLIENT causes the Setattr RPC
* to return NFS_OK, but not set va_mode.
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
*/
r245508 modified the NFS client's Setattr RPC to use VA_UTIMES_NULL to indicate whether it should set the time to the current tod on the server. This had the side effect of making the NFS client use the client's timestamp for exclusive create, starting with FreeBSD9.2. Unfortunately a bug in some Solaris NFS servers causes these servers to return NFS_OK to the Setattr RPC done during exclusive create, but not actually set the file's mode, leaving the file's mode == 0. This patch restores the NFS client's behaviour to use the server's tod for the exclusive open's Setattr RPC, to avoid the Solaris server bug and to restore the pre-FreeBSD9.2 NFS behaviour. Discussed on: freebsd-fs PR: 186293 MFC after: 3 months
2014-12-28 21:13:52 +00:00
if (vap->va_mtime.tv_sec == VNOVAL) {
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
vfs_timestamp(&vap->va_mtime);
r245508 modified the NFS client's Setattr RPC to use VA_UTIMES_NULL to indicate whether it should set the time to the current tod on the server. This had the side effect of making the NFS client use the client's timestamp for exclusive create, starting with FreeBSD9.2. Unfortunately a bug in some Solaris NFS servers causes these servers to return NFS_OK to the Setattr RPC done during exclusive create, but not actually set the file's mode, leaving the file's mode == 0. This patch restores the NFS client's behaviour to use the server's tod for the exclusive open's Setattr RPC, to avoid the Solaris server bug and to restore the pre-FreeBSD9.2 NFS behaviour. Discussed on: freebsd-fs PR: 186293 MFC after: 3 months
2014-12-28 21:13:52 +00:00
vap->va_vaflags |= VA_UTIMES_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 (vap->va_atime.tv_sec == VNOVAL)
vap->va_atime = vap->va_mtime;
return (1);
}
/*
* Map nfsv4 errors to errno.h errors.
* The uid and gid arguments are only used for NFSERR_BADOWNER and that
* error should only be returned for the Open, Create and Setattr Ops.
* As such, most calls can just pass in 0 for those arguments.
*/
nfs: Remove APPLESTATIC macro It is no longer useful. Reviewed by: rmacklem Approved by: mav (mentor) MFC after: 1 week Sponsored by: iXsystems, Inc. Differential Revision: https://reviews.freebsd.org/D24811
2020-05-12 13:23:25 +00:00
int
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
nfscl_maperr(struct thread *td, int error, uid_t uid, gid_t gid)
{
struct proc *p;
Fix the NFSv4 client hndling of a stale write verifier in the Commit operation. When the NFSv4 client Commit operation encountered a stale write verifier, it erroneously mapped that to EIO. This could have caused recently written data to be lost when a server crashes/reboots between an UNSTABLE write and the subsequent commit. This patch fixes this. The bug was only for the NFSv4 client and did not affect NFSv3. Tested by: cperciva PR: 215887 MFC after: 2 weeks
2017-04-09 21:50:21 +00:00
if (error < 10000 || error >= NFSERR_STALEWRITEVERF)
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);
if (td != NULL)
p = td->td_proc;
else
p = NULL;
switch (error) {
case NFSERR_BADOWNER:
tprintf(p, LOG_INFO,
"No name and/or group mapping for uid,gid:(%d,%d)\n",
uid, gid);
return (EPERM);
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
case NFSERR_BADNAME:
case NFSERR_BADCHAR:
printf("nfsv4 char/name not handled by server\n");
return (ENOENT);
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
case NFSERR_STALECLIENTID:
case NFSERR_STALESTATEID:
case NFSERR_EXPIRED:
case NFSERR_BADSTATEID:
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
case NFSERR_BADSESSION:
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
printf("nfsv4 recover err returned %d\n", error);
return (EIO);
case NFSERR_BADHANDLE:
case NFSERR_SERVERFAULT:
case NFSERR_BADTYPE:
case NFSERR_FHEXPIRED:
case NFSERR_RESOURCE:
case NFSERR_MOVED:
case NFSERR_NOFILEHANDLE:
case NFSERR_MINORVERMISMATCH:
case NFSERR_OLDSTATEID:
case NFSERR_BADSEQID:
case NFSERR_LEASEMOVED:
case NFSERR_RECLAIMBAD:
case NFSERR_BADXDR:
case NFSERR_OPILLEGAL:
printf("nfsv4 client/server protocol prob err=%d\n",
error);
return (EIO);
default:
tprintf(p, LOG_INFO, "nfsv4 err=%d\n", error);
return (EIO);
};
}
/*
* Check to see if the process for this owner exists. Return 1 if it doesn't
* and 0 otherwise.
*/
int
nfscl_procdoesntexist(u_int8_t *own)
{
union {
u_int32_t lval;
u_int8_t cval[4];
} tl;
struct proc *p;
pid_t pid;
Add an NFSv4.1 mount option for "use one openowner". Some NFSv4.1 servers such as AmazonEFS can only support a small fixed number of open_owner4s. This patch adds a mount option called "oneopenown" that can be used for NFSv4.1 mounts to make the client do all Opens with the same open_owner4 string. This option can only be used with NFSv4.1 and may not work correctly when Delegations are is use. Reported by: cperciva Tested by: cperciva MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D8988
2017-04-13 21:54:19 +00:00
int i, ret = 0;
/* For the single open_owner of all 0 bytes, just return 0. */
for (i = 0; i < NFSV4CL_LOCKNAMELEN; i++)
if (own[i] != 0)
break;
if (i == NFSV4CL_LOCKNAMELEN)
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
tl.cval[0] = *own++;
tl.cval[1] = *own++;
tl.cval[2] = *own++;
tl.cval[3] = *own++;
pid = tl.lval;
Fix the NFSv4 client to safely find processes. r340744 broke the NFSv4 client, because it replaced pfind_locked() with a call to pfind(), since pfind() acquires the sx lock for the pid hash and the NFSv4 already holds a mutex when it does the call. The patch fixes the problem by recreating a pfind_any_locked() and adding the functions pidhash_slockall() and pidhash_sunlockall to acquire/release all of the pid hash locks. These functions are then used by the NFSv4 client instead of acquiring the allproc_lock and calling pfind(). Reviewed by: kib, mjg MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D19887
2019-04-15 01:27:15 +00:00
p = pfind_any_locked(pid);
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 (p == NULL)
return (1);
if (p->p_stats == NULL) {
PROC_UNLOCK(p);
return (0);
}
tl.cval[0] = *own++;
tl.cval[1] = *own++;
tl.cval[2] = *own++;
tl.cval[3] = *own++;
if (tl.lval != p->p_stats->p_start.tv_sec) {
ret = 1;
} else {
tl.cval[0] = *own++;
tl.cval[1] = *own++;
tl.cval[2] = *own++;
tl.cval[3] = *own;
if (tl.lval != p->p_stats->p_start.tv_usec)
ret = 1;
}
PROC_UNLOCK(p);
return (ret);
}
/*
* - nfs pseudo system call for the client
*/
/*
* MPSAFE
*/
static int
nfssvc_nfscl(struct thread *td, struct nfssvc_args *uap)
{
struct file *fp;
struct nfscbd_args nfscbdarg;
struct nfsd_nfscbd_args nfscbdarg2;
Add an nfssvc() option to the kernel for the new NFS client which dumps out the actual options being used by an NFS mount. This will be used to implement a "-m" option for nfsstat(1). Reviewed by: alfred MFC after: 2 weeks
2012-12-02 01:16:04 +00:00
struct nameidata nd;
struct nfscl_dumpmntopts dumpmntopts;
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
cap_rights_t rights;
Add an nfssvc() option to the kernel for the new NFS client which dumps out the actual options being used by an NFS mount. This will be used to implement a "-m" option for nfsstat(1). Reviewed by: alfred MFC after: 2 weeks
2012-12-02 01:16:04 +00:00
char *buf;
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
int error;
Add kernel support for the NFS client forced dismount "umount -N" option. When an NFS mount is hung against an unresponsive NFS server, the "umount -f" option can be used to dismount the mount. Unfortunately, "umount -f" gets hung as well if a "umount" without "-f" has already been done. Usually, this is because of a vnode lock being held by the "umount" for the mounted-on vnode. This patch adds kernel code so that a new "-N" option can be added to "umount", allowing it to avoid getting hung for this case. It adds two flags. One indicates that a forced dismount is about to happen and the other is used, along with setting mnt_data == NULL, to handshake with the nfs_unmount() VFS call. It includes a slight change to the interface used between the client and common NFS modules, so I bumped __FreeBSD_version to ensure both modules are rebuilt. Tested by: pho Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D11735
2017-07-29 19:52:47 +00:00
struct mount *mp;
struct nfsmount *nmp;
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_CBADDSOCK) {
error = copyin(uap->argp, (caddr_t)&nfscbdarg, sizeof(nfscbdarg));
if (error)
return (error);
Second-to-last commit implementing Capsicum capabilities in the FreeBSD kernel for FreeBSD 9.0: Add a new capability mask argument to fget(9) and friends, allowing system call code to declare what capabilities are required when an integer file descriptor is converted into an in-kernel struct file *. With options CAPABILITIES compiled into the kernel, this enforces capability protection; without, this change is effectively a no-op. Some cases require special handling, such as mmap(2), which must preserve information about the maximum rights at the time of mapping in the memory map so that they can later be enforced in mprotect(2) -- this is done by narrowing the rights in the existing max_protection field used for similar purposes with file permissions. In namei(9), we assert that the code is not reached from within capability mode, as we're not yet ready to enforce namespace capabilities there. This will follow in a later commit. Update two capability names: CAP_EVENT and CAP_KEVENT become CAP_POST_KEVENT and CAP_POLL_KEVENT to more accurately indicate what they represent. Approved by: re (bz) Submitted by: jonathan Sponsored by: Google Inc
2011-08-11 12:30:23 +00:00
/*
* Since we don't know what rights might be required,
* pretend that we need them all. It is better to be too
* careful than too reckless.
*/
Change the cap_rights_t type from uint64_t to a structure that we can extend in the future in a backward compatible (API and ABI) way. The cap_rights_t represents capability rights. We used to use one bit to represent one right, but we are running out of spare bits. Currently the new structure provides place for 114 rights (so 50 more than the previous cap_rights_t), but it is possible to grow the structure to hold at least 285 rights, although we can make it even larger if 285 rights won't be enough. The structure definition looks like this: struct cap_rights { uint64_t cr_rights[CAP_RIGHTS_VERSION + 2]; }; The initial CAP_RIGHTS_VERSION is 0. The top two bits in the first element of the cr_rights[] array contain total number of elements in the array - 2. This means if those two bits are equal to 0, we have 2 array elements. The top two bits in all remaining array elements should be 0. The next five bits in all array elements contain array index. Only one bit is used and bit position in this five-bits range defines array index. This means there can be at most five array elements in the future. To define new right the CAPRIGHT() macro must be used. The macro takes two arguments - an array index and a bit to set, eg. #define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL) We still support aliases that combine few rights, but the rights have to belong to the same array element, eg: #define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL) #define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL) #define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP) There is new API to manage the new cap_rights_t structure: cap_rights_t *cap_rights_init(cap_rights_t *rights, ...); void cap_rights_set(cap_rights_t *rights, ...); void cap_rights_clear(cap_rights_t *rights, ...); bool cap_rights_is_set(const cap_rights_t *rights, ...); bool cap_rights_is_valid(const cap_rights_t *rights); void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src); void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src); bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little); Capability rights to the cap_rights_init(), cap_rights_set(), cap_rights_clear() and cap_rights_is_set() functions are provided by separating them with commas, eg: cap_rights_t rights; cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT); There is no need to terminate the list of rights, as those functions are actually macros that take care of the termination, eg: #define cap_rights_set(rights, ...) \ __cap_rights_set((rights), __VA_ARGS__, 0ULL) void __cap_rights_set(cap_rights_t *rights, ...); Thanks to using one bit as an array index we can assert in those functions that there are no two rights belonging to different array elements provided together. For example this is illegal and will be detected, because CAP_LOOKUP belongs to element 0 and CAP_PDKILL to element 1: cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL); Providing several rights that belongs to the same array's element this way is correct, but is not advised. It should only be used for aliases definition. This commit also breaks compatibility with some existing Capsicum system calls, but I see no other way to do that. This should be fine as Capsicum is still experimental and this change is not going to 9.x. Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
error = fget(td, nfscbdarg.sock,
cap_rights_init(&rights, CAP_SOCK_CLIENT), &fp);
if (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);
if (fp->f_type != DTYPE_SOCKET) {
fdrop(fp, td);
return (EPERM);
}
error = nfscbd_addsock(fp);
fdrop(fp, td);
if (!error && nfscl_enablecallb == 0) {
nfsv4_cbport = nfscbdarg.port;
nfscl_enablecallb = 1;
}
} else if (uap->flag & NFSSVC_NFSCBD) {
if (uap->argp == NULL)
return (EINVAL);
error = copyin(uap->argp, (caddr_t)&nfscbdarg2,
sizeof(nfscbdarg2));
if (error)
return (error);
error = nfscbd_nfsd(td, &nfscbdarg2);
Add an nfssvc() option to the kernel for the new NFS client which dumps out the actual options being used by an NFS mount. This will be used to implement a "-m" option for nfsstat(1). Reviewed by: alfred MFC after: 2 weeks
2012-12-02 01:16:04 +00:00
} else if (uap->flag & NFSSVC_DUMPMNTOPTS) {
error = copyin(uap->argp, &dumpmntopts, sizeof(dumpmntopts));
if (error == 0 && (dumpmntopts.ndmnt_blen < 256 ||
dumpmntopts.ndmnt_blen > 1024))
error = EINVAL;
if (error == 0)
error = nfsrv_lookupfilename(&nd,
dumpmntopts.ndmnt_fname, td);
if (error == 0 && strcmp(nd.ni_vp->v_mount->mnt_vfc->vfc_name,
"nfs") != 0) {
vput(nd.ni_vp);
error = EINVAL;
}
if (error == 0) {
buf = malloc(dumpmntopts.ndmnt_blen, M_TEMP, M_WAITOK);
nfscl_retopts(VFSTONFS(nd.ni_vp->v_mount), buf,
dumpmntopts.ndmnt_blen);
vput(nd.ni_vp);
error = copyout(buf, dumpmntopts.ndmnt_buf,
dumpmntopts.ndmnt_blen);
free(buf, M_TEMP);
}
Add kernel support for the NFS client forced dismount "umount -N" option. When an NFS mount is hung against an unresponsive NFS server, the "umount -f" option can be used to dismount the mount. Unfortunately, "umount -f" gets hung as well if a "umount" without "-f" has already been done. Usually, this is because of a vnode lock being held by the "umount" for the mounted-on vnode. This patch adds kernel code so that a new "-N" option can be added to "umount", allowing it to avoid getting hung for this case. It adds two flags. One indicates that a forced dismount is about to happen and the other is used, along with setting mnt_data == NULL, to handshake with the nfs_unmount() VFS call. It includes a slight change to the interface used between the client and common NFS modules, so I bumped __FreeBSD_version to ensure both modules are rebuilt. Tested by: pho Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D11735
2017-07-29 19:52:47 +00:00
} else if (uap->flag & NFSSVC_FORCEDISM) {
buf = malloc(MNAMELEN + 1, M_TEMP, M_WAITOK);
error = copyinstr(uap->argp, buf, MNAMELEN + 1, NULL);
if (error == 0) {
nmp = NULL;
mtx_lock(&mountlist_mtx);
TAILQ_FOREACH(mp, &mountlist, mnt_list) {
if (strcmp(mp->mnt_stat.f_mntonname, buf) ==
0 && strcmp(mp->mnt_stat.f_fstypename,
"nfs") == 0 && mp->mnt_data != NULL) {
nmp = VFSTONFS(mp);
Merge the pNFS server code from projects/pnfs-planb-server into head. This code merge adds a pNFS service to the NFSv4.1 server. Although it is a large commit it should not affect behaviour for a non-pNFS NFS server. Some documentation on how this works can be found at: http://people.freebsd.org/~rmacklem/pnfs-planb-setup.txt and will hopefully be turned into a proper document soon. This is a merge of the kernel code. Userland and man page changes will come soon, once the dust settles on this merge. It has passed a "make universe", so I hope it will not cause build problems. It also adds NFSv4.1 server support for the "current stateid". Here is a brief overview of the pNFS service: A pNFS service separates the Read/Write oeprations from all the other NFSv4.1 Metadata operations. It is hoped that this separation allows a pNFS service to be configured that exceeds the limits of a single NFS server for either storage capacity and/or I/O bandwidth. It is possible to configure mirroring within the data servers (DSs) so that the data storage file for an MDS file will be mirrored on two or more of the DSs. When this is used, failure of a DS will not stop the pNFS service and a failed DS can be recovered once repaired while the pNFS service continues to operate. Although two way mirroring would be the norm, it is possible to set a mirroring level of up to four or the number of DSs, whichever is less. The Metadata server will always be a single point of failure, just as a single NFS server is. A Plan B pNFS service consists of a single MetaData Server (MDS) and K Data Servers (DS), all of which are recent FreeBSD systems. Clients will mount the MDS as they would a single NFS server. When files are created, the MDS creates a file tree identical to what a single NFS server creates, except that all the regular (VREG) files will be empty. As such, if you look at the exported tree on the MDS directly on the MDS server (not via an NFS mount), the files will all be of size 0. Each of these files will also have two extended attributes in the system attribute name space: pnfsd.dsfile - This extended attrbute stores the information that the MDS needs to find the data storage file(s) on DS(s) for this file. pnfsd.dsattr - This extended attribute stores the Size, AccessTime, ModifyTime and Change attributes for the file, so that the MDS doesn't need to acquire the attributes from the DS for every Getattr operation. For each regular (VREG) file, the MDS creates a data storage file on one (or more if mirroring is enabled) of the DSs in one of the "dsNN" subdirectories. The name of this file is the file handle of the file on the MDS in hexadecimal so that the name is unique. The DSs use subdirectories named "ds0" to "dsN" so that no one directory gets too large. The value of "N" is set via the sysctl vfs.nfsd.dsdirsize on the MDS, with the default being 20. For production servers that will store a lot of files, this value should probably be much larger. It can be increased when the "nfsd" daemon is not running on the MDS, once the "dsK" directories are created. For pNFS aware NFSv4.1 clients, the FreeBSD server will return two pieces of information to the client that allows it to do I/O directly to the DS. DeviceInfo - This is relatively static information that defines what a DS is. The critical bits of information returned by the FreeBSD server is the IP address of the DS and, for the Flexible File layout, that NFSv4.1 is to be used and that it is "tightly coupled". There is a "deviceid" which identifies the DeviceInfo. Layout - This is per file and can be recalled by the server when it is no longer valid. For the FreeBSD server, there is support for two types of layout, call File and Flexible File layout. Both allow the client to do I/O on the DS via NFSv4.1 I/O operations. The Flexible File layout is a more recent variant that allows specification of mirrors, where the client is expected to do writes to all mirrors to maintain them in a consistent state. The Flexible File layout also allows the client to report I/O errors for a DS back to the MDS. The Flexible File layout supports two variants referred to as "tightly coupled" vs "loosely coupled". The FreeBSD server always uses the "tightly coupled" variant where the client uses the same credentials to do I/O on the DS as it would on the MDS. For the "loosely coupled" variant, the layout specifies a synthetic user/group that the client uses to do I/O on the DS. The FreeBSD server does not do striping and always returns layouts for the entire file. The critical information in a layout is Read vs Read/Writea and DeviceID(s) that identify which DS(s) the data is stored on. At this time, the MDS generates File Layout layouts to NFSv4.1 clients that know how to do pNFS for the non-mirrored DS case unless the sysctl vfs.nfsd.default_flexfile is set non-zero, in which case Flexible File layouts are generated. The mirrored DS configuration always generates Flexible File layouts. For NFS clients that do not support NFSv4.1 pNFS, all I/O operations are done against the MDS which acts as a proxy for the appropriate DS(s). When the MDS receives an I/O RPC, it will do the RPC on the DS as a proxy. If the DS is on the same machine, the MDS/DS will do the RPC on the DS as a proxy and so on, until the machine runs out of some resource, such as session slots or mbufs. As such, DSs must be separate systems from the MDS. Tested by: james.rose@framestore.com Relnotes: yes
2018-06-12 19:36:32 +00:00
NFSDDSLOCK();
if (nfsv4_findmirror(nmp) != NULL) {
NFSDDSUNLOCK();
error = ENXIO;
nmp = NULL;
break;
}
Add kernel support for the NFS client forced dismount "umount -N" option. When an NFS mount is hung against an unresponsive NFS server, the "umount -f" option can be used to dismount the mount. Unfortunately, "umount -f" gets hung as well if a "umount" without "-f" has already been done. Usually, this is because of a vnode lock being held by the "umount" for the mounted-on vnode. This patch adds kernel code so that a new "-N" option can be added to "umount", allowing it to avoid getting hung for this case. It adds two flags. One indicates that a forced dismount is about to happen and the other is used, along with setting mnt_data == NULL, to handshake with the nfs_unmount() VFS call. It includes a slight change to the interface used between the client and common NFS modules, so I bumped __FreeBSD_version to ensure both modules are rebuilt. Tested by: pho Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D11735
2017-07-29 19:52:47 +00:00
mtx_lock(&nmp->nm_mtx);
if ((nmp->nm_privflag &
NFSMNTP_FORCEDISM) == 0) {
nmp->nm_privflag |=
(NFSMNTP_FORCEDISM |
NFSMNTP_CANCELRPCS);
mtx_unlock(&nmp->nm_mtx);
} else {
mtx_unlock(&nmp->nm_mtx);
Fix use of pointer after being set NULL. Using a pointer after setting it NULL is probably not a good plan. Spotted by inspection during changes for Flexible File Layout Ioerr handling. This code path obviously isn't normally executed. MFC after: 1 week
2018-04-20 11:38:29 +00:00
nmp = NULL;
Add kernel support for the NFS client forced dismount "umount -N" option. When an NFS mount is hung against an unresponsive NFS server, the "umount -f" option can be used to dismount the mount. Unfortunately, "umount -f" gets hung as well if a "umount" without "-f" has already been done. Usually, this is because of a vnode lock being held by the "umount" for the mounted-on vnode. This patch adds kernel code so that a new "-N" option can be added to "umount", allowing it to avoid getting hung for this case. It adds two flags. One indicates that a forced dismount is about to happen and the other is used, along with setting mnt_data == NULL, to handshake with the nfs_unmount() VFS call. It includes a slight change to the interface used between the client and common NFS modules, so I bumped __FreeBSD_version to ensure both modules are rebuilt. Tested by: pho Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D11735
2017-07-29 19:52:47 +00:00
}
Merge the pNFS server code from projects/pnfs-planb-server into head. This code merge adds a pNFS service to the NFSv4.1 server. Although it is a large commit it should not affect behaviour for a non-pNFS NFS server. Some documentation on how this works can be found at: http://people.freebsd.org/~rmacklem/pnfs-planb-setup.txt and will hopefully be turned into a proper document soon. This is a merge of the kernel code. Userland and man page changes will come soon, once the dust settles on this merge. It has passed a "make universe", so I hope it will not cause build problems. It also adds NFSv4.1 server support for the "current stateid". Here is a brief overview of the pNFS service: A pNFS service separates the Read/Write oeprations from all the other NFSv4.1 Metadata operations. It is hoped that this separation allows a pNFS service to be configured that exceeds the limits of a single NFS server for either storage capacity and/or I/O bandwidth. It is possible to configure mirroring within the data servers (DSs) so that the data storage file for an MDS file will be mirrored on two or more of the DSs. When this is used, failure of a DS will not stop the pNFS service and a failed DS can be recovered once repaired while the pNFS service continues to operate. Although two way mirroring would be the norm, it is possible to set a mirroring level of up to four or the number of DSs, whichever is less. The Metadata server will always be a single point of failure, just as a single NFS server is. A Plan B pNFS service consists of a single MetaData Server (MDS) and K Data Servers (DS), all of which are recent FreeBSD systems. Clients will mount the MDS as they would a single NFS server. When files are created, the MDS creates a file tree identical to what a single NFS server creates, except that all the regular (VREG) files will be empty. As such, if you look at the exported tree on the MDS directly on the MDS server (not via an NFS mount), the files will all be of size 0. Each of these files will also have two extended attributes in the system attribute name space: pnfsd.dsfile - This extended attrbute stores the information that the MDS needs to find the data storage file(s) on DS(s) for this file. pnfsd.dsattr - This extended attribute stores the Size, AccessTime, ModifyTime and Change attributes for the file, so that the MDS doesn't need to acquire the attributes from the DS for every Getattr operation. For each regular (VREG) file, the MDS creates a data storage file on one (or more if mirroring is enabled) of the DSs in one of the "dsNN" subdirectories. The name of this file is the file handle of the file on the MDS in hexadecimal so that the name is unique. The DSs use subdirectories named "ds0" to "dsN" so that no one directory gets too large. The value of "N" is set via the sysctl vfs.nfsd.dsdirsize on the MDS, with the default being 20. For production servers that will store a lot of files, this value should probably be much larger. It can be increased when the "nfsd" daemon is not running on the MDS, once the "dsK" directories are created. For pNFS aware NFSv4.1 clients, the FreeBSD server will return two pieces of information to the client that allows it to do I/O directly to the DS. DeviceInfo - This is relatively static information that defines what a DS is. The critical bits of information returned by the FreeBSD server is the IP address of the DS and, for the Flexible File layout, that NFSv4.1 is to be used and that it is "tightly coupled". There is a "deviceid" which identifies the DeviceInfo. Layout - This is per file and can be recalled by the server when it is no longer valid. For the FreeBSD server, there is support for two types of layout, call File and Flexible File layout. Both allow the client to do I/O on the DS via NFSv4.1 I/O operations. The Flexible File layout is a more recent variant that allows specification of mirrors, where the client is expected to do writes to all mirrors to maintain them in a consistent state. The Flexible File layout also allows the client to report I/O errors for a DS back to the MDS. The Flexible File layout supports two variants referred to as "tightly coupled" vs "loosely coupled". The FreeBSD server always uses the "tightly coupled" variant where the client uses the same credentials to do I/O on the DS as it would on the MDS. For the "loosely coupled" variant, the layout specifies a synthetic user/group that the client uses to do I/O on the DS. The FreeBSD server does not do striping and always returns layouts for the entire file. The critical information in a layout is Read vs Read/Writea and DeviceID(s) that identify which DS(s) the data is stored on. At this time, the MDS generates File Layout layouts to NFSv4.1 clients that know how to do pNFS for the non-mirrored DS case unless the sysctl vfs.nfsd.default_flexfile is set non-zero, in which case Flexible File layouts are generated. The mirrored DS configuration always generates Flexible File layouts. For NFS clients that do not support NFSv4.1 pNFS, all I/O operations are done against the MDS which acts as a proxy for the appropriate DS(s). When the MDS receives an I/O RPC, it will do the RPC on the DS as a proxy. If the DS is on the same machine, the MDS/DS will do the RPC on the DS as a proxy and so on, until the machine runs out of some resource, such as session slots or mbufs. As such, DSs must be separate systems from the MDS. Tested by: james.rose@framestore.com Relnotes: yes
2018-06-12 19:36:32 +00:00
NFSDDSUNLOCK();
Add kernel support for the NFS client forced dismount "umount -N" option. When an NFS mount is hung against an unresponsive NFS server, the "umount -f" option can be used to dismount the mount. Unfortunately, "umount -f" gets hung as well if a "umount" without "-f" has already been done. Usually, this is because of a vnode lock being held by the "umount" for the mounted-on vnode. This patch adds kernel code so that a new "-N" option can be added to "umount", allowing it to avoid getting hung for this case. It adds two flags. One indicates that a forced dismount is about to happen and the other is used, along with setting mnt_data == NULL, to handshake with the nfs_unmount() VFS call. It includes a slight change to the interface used between the client and common NFS modules, so I bumped __FreeBSD_version to ensure both modules are rebuilt. Tested by: pho Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D11735
2017-07-29 19:52:47 +00:00
break;
}
}
mtx_unlock(&mountlist_mtx);
if (nmp != NULL) {
/*
* Call newnfs_nmcancelreqs() to cause
* any RPCs in progress on the mount point to
* fail.
* This will cause any process waiting for an
* RPC to complete while holding a vnode lock
* on the mounted-on vnode (such as "df" or
* a non-forced "umount") to fail.
* This will unlock the mounted-on vnode so
* a forced dismount can succeed.
* Then clear NFSMNTP_CANCELRPCS and wakeup(),
* so that nfs_unmount() can complete.
*/
newnfs_nmcancelreqs(nmp);
mtx_lock(&nmp->nm_mtx);
nmp->nm_privflag &= ~NFSMNTP_CANCELRPCS;
wakeup(nmp);
mtx_unlock(&nmp->nm_mtx);
Merge the pNFS server code from projects/pnfs-planb-server into head. This code merge adds a pNFS service to the NFSv4.1 server. Although it is a large commit it should not affect behaviour for a non-pNFS NFS server. Some documentation on how this works can be found at: http://people.freebsd.org/~rmacklem/pnfs-planb-setup.txt and will hopefully be turned into a proper document soon. This is a merge of the kernel code. Userland and man page changes will come soon, once the dust settles on this merge. It has passed a "make universe", so I hope it will not cause build problems. It also adds NFSv4.1 server support for the "current stateid". Here is a brief overview of the pNFS service: A pNFS service separates the Read/Write oeprations from all the other NFSv4.1 Metadata operations. It is hoped that this separation allows a pNFS service to be configured that exceeds the limits of a single NFS server for either storage capacity and/or I/O bandwidth. It is possible to configure mirroring within the data servers (DSs) so that the data storage file for an MDS file will be mirrored on two or more of the DSs. When this is used, failure of a DS will not stop the pNFS service and a failed DS can be recovered once repaired while the pNFS service continues to operate. Although two way mirroring would be the norm, it is possible to set a mirroring level of up to four or the number of DSs, whichever is less. The Metadata server will always be a single point of failure, just as a single NFS server is. A Plan B pNFS service consists of a single MetaData Server (MDS) and K Data Servers (DS), all of which are recent FreeBSD systems. Clients will mount the MDS as they would a single NFS server. When files are created, the MDS creates a file tree identical to what a single NFS server creates, except that all the regular (VREG) files will be empty. As such, if you look at the exported tree on the MDS directly on the MDS server (not via an NFS mount), the files will all be of size 0. Each of these files will also have two extended attributes in the system attribute name space: pnfsd.dsfile - This extended attrbute stores the information that the MDS needs to find the data storage file(s) on DS(s) for this file. pnfsd.dsattr - This extended attribute stores the Size, AccessTime, ModifyTime and Change attributes for the file, so that the MDS doesn't need to acquire the attributes from the DS for every Getattr operation. For each regular (VREG) file, the MDS creates a data storage file on one (or more if mirroring is enabled) of the DSs in one of the "dsNN" subdirectories. The name of this file is the file handle of the file on the MDS in hexadecimal so that the name is unique. The DSs use subdirectories named "ds0" to "dsN" so that no one directory gets too large. The value of "N" is set via the sysctl vfs.nfsd.dsdirsize on the MDS, with the default being 20. For production servers that will store a lot of files, this value should probably be much larger. It can be increased when the "nfsd" daemon is not running on the MDS, once the "dsK" directories are created. For pNFS aware NFSv4.1 clients, the FreeBSD server will return two pieces of information to the client that allows it to do I/O directly to the DS. DeviceInfo - This is relatively static information that defines what a DS is. The critical bits of information returned by the FreeBSD server is the IP address of the DS and, for the Flexible File layout, that NFSv4.1 is to be used and that it is "tightly coupled". There is a "deviceid" which identifies the DeviceInfo. Layout - This is per file and can be recalled by the server when it is no longer valid. For the FreeBSD server, there is support for two types of layout, call File and Flexible File layout. Both allow the client to do I/O on the DS via NFSv4.1 I/O operations. The Flexible File layout is a more recent variant that allows specification of mirrors, where the client is expected to do writes to all mirrors to maintain them in a consistent state. The Flexible File layout also allows the client to report I/O errors for a DS back to the MDS. The Flexible File layout supports two variants referred to as "tightly coupled" vs "loosely coupled". The FreeBSD server always uses the "tightly coupled" variant where the client uses the same credentials to do I/O on the DS as it would on the MDS. For the "loosely coupled" variant, the layout specifies a synthetic user/group that the client uses to do I/O on the DS. The FreeBSD server does not do striping and always returns layouts for the entire file. The critical information in a layout is Read vs Read/Writea and DeviceID(s) that identify which DS(s) the data is stored on. At this time, the MDS generates File Layout layouts to NFSv4.1 clients that know how to do pNFS for the non-mirrored DS case unless the sysctl vfs.nfsd.default_flexfile is set non-zero, in which case Flexible File layouts are generated. The mirrored DS configuration always generates Flexible File layouts. For NFS clients that do not support NFSv4.1 pNFS, all I/O operations are done against the MDS which acts as a proxy for the appropriate DS(s). When the MDS receives an I/O RPC, it will do the RPC on the DS as a proxy. If the DS is on the same machine, the MDS/DS will do the RPC on the DS as a proxy and so on, until the machine runs out of some resource, such as session slots or mbufs. As such, DSs must be separate systems from the MDS. Tested by: james.rose@framestore.com Relnotes: yes
2018-06-12 19:36:32 +00:00
} else if (error == 0)
Add kernel support for the NFS client forced dismount "umount -N" option. When an NFS mount is hung against an unresponsive NFS server, the "umount -f" option can be used to dismount the mount. Unfortunately, "umount -f" gets hung as well if a "umount" without "-f" has already been done. Usually, this is because of a vnode lock being held by the "umount" for the mounted-on vnode. This patch adds kernel code so that a new "-N" option can be added to "umount", allowing it to avoid getting hung for this case. It adds two flags. One indicates that a forced dismount is about to happen and the other is used, along with setting mnt_data == NULL, to handshake with the nfs_unmount() VFS call. It includes a slight change to the interface used between the client and common NFS modules, so I bumped __FreeBSD_version to ensure both modules are rebuilt. Tested by: pho Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D11735
2017-07-29 19:52:47 +00:00
error = EINVAL;
}
free(buf, M_TEMP);
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 {
error = EINVAL;
}
return (error);
}
extern int (*nfsd_call_nfscl)(struct thread *, struct nfssvc_args *);
/*
* Called once to initialize data structures...
*/
static int
nfscl_modevent(module_t mod, int type, void *data)
{
int error = 0;
static int loaded = 0;
switch (type) {
case MOD_LOAD:
if (loaded)
return (0);
newnfs_portinit();
mtx_init(&ncl_iod_mutex, "ncl_iod_mutex", NULL, MTX_DEF);
nfscl_init();
NFSD_LOCK();
nfsrvd_cbinit(0);
NFSD_UNLOCK();
ncl_call_invalcaches = ncl_invalcaches;
nfsd_call_nfscl = nfssvc_nfscl;
loaded = 1;
break;
case MOD_UNLOAD:
if (nfs_numnfscbd != 0) {
error = EBUSY;
break;
}
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
ncl_call_invalcaches = NULL;
nfsd_call_nfscl = NULL;
Allocate pager bufs from UMA instead of 80-ish mutex protected linked list. o In vm_pager_bufferinit() create pbuf_zone and start accounting on how many pbufs are we going to have set. In various subsystems that are going to utilize pbufs create private zones via call to pbuf_zsecond_create(). The latter calls uma_zsecond_create(), and sets a limit on created zone. After startup preallocate pbufs according to requirements of all pbuf zones. Subsystems that used to have a private limit with old allocator now have private pbuf zones: md(4), fusefs, NFS client, smbfs, VFS cluster, FFS, swap, vnode pager. The following subsystems use shared pbuf zone: cam(4), nvme(4), physio(9), aio(4). They should have their private limits, but changing that is out of scope of this commit. o Fetch tunable value of kern.nswbuf from init_param2() and while here move NSWBUF_MIN to opt_param.h and eliminate opt_swap.h, that was holding only this option. Default values aren't touched by this commit, but they probably should be reviewed wrt to modern hardware. This change removes a tight bottleneck from sendfile(2) operation, that uses pbufs in vnode pager. Other pagers also would benefit from faster allocation. Together with: gallatin Tested by: pho
2019-01-15 01:02:16 +00:00
uma_zdestroy(ncl_pbuf_zone);
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(&ncl_iod_mutex);
loaded = 0;
break;
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
/* FALLTHROUGH */
#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
default:
error = EOPNOTSUPP;
break;
}
return error;
}
static moduledata_t nfscl_mod = {
"nfscl",
nfscl_modevent,
NULL,
};
Fix ordering of nfscl_modevent() and ncl_uninit(). nfscl_modevent() must be called after ncl_uninit() when unloading the nfscl module because ncl_uninit() uses ncl_iod_mutex which is destroyed in nfscl_modevent(). Reviewed by: rmacklem Approved by: trasz (mentor)
2009-10-20 15:01:46 +00:00
DECLARE_MODULE(nfscl, nfscl_mod, SI_SUB_VFS, SI_ORDER_FIRST);
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
/* So that loader and kldload(2) can find us, wherever we are.. */
MODULE_VERSION(nfscl, 1);
Fix the name of the module common to the client and server in the experimental nfs subsystem to the correct one for the MODULE_DEPEND() macro. Approved by: kib (mentor)
2009-05-22 20:55:29 +00:00
MODULE_DEPEND(nfscl, nfscommon, 1, 1, 1);
Add MODULE_DEPEND() macros to the experimental NFS client and server so that the modules will load when kernels are built with none of the NFS* configuration options specified. I believe this resolves the problems reported by PR kern/144458 and the email on freebsd-stable@ posted by Dmitry Pryanishnikov on June 13. Tested by: kib PR: kern/144458 Reviewed by: kib MFC after: 1 week
2010-06-15 00:25:04 +00:00
MODULE_DEPEND(nfscl, krpc, 1, 1, 1);
MODULE_DEPEND(nfscl, nfssvc, 1, 1, 1);
Reference in New Issue Copy Permalink